1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
8 (During the 1.9 series, we will keep an incremental NEWS for the latest
9 prerelease, and a full NEWS corresponding to 1.8 -> 2.0.)
11 Changes in 1.9.11 (since the 1.9.10 prerelease):
13 ** Renamed module: (rnrs bytevectors)
15 This module was called (rnrs bytevector), its name from earlier drafts
16 of the R6RS. Its name has been changed. Users of previous 1.9 preleases
17 may want to search for any stale rnrs/bytevector .go or .scm file, and
20 ** New module: (sxml match)
22 Guile has incorporated Jim Bender's `sxml-match' library. See
23 "sxml-match' in the manual for more information. Thanks, Jim!
25 ** New module: (srfi srfi-9 gnu)
27 This module adds an extension to srfi-9, `set-record-type-printer!'. See
28 "SRFI-9" in the manual for more information.
30 ** Support for R6RS libraries
32 The `library' and `import' forms from the latest Scheme report have been
33 added to Guile, in such a way that R6RS libraries share a namespace with
34 Guile modules. R6RS modules may import Guile modules, and are available
35 for Guile modules to import via use-modules and all the rest. See "R6RS
36 Libraries" in the manual for more information.
38 ** Implementations of R6RS libraries
40 Guile now has implementations for all of the libraries defined in the
41 R6RS. Thanks to Julian Graham for this excellent hack. See "R6RS
42 Standard Libraries" in the manual for a full list of libraries.
44 ** Partial R6RS compatibility
46 Guile now has enough support for R6RS to run a reasonably large subset
49 Guile is not fully R6RS compatible. Many incompatibilities are simply
50 bugs, though some parts of Guile will remain R6RS-incompatible for the
51 foreseeable future. See "R6RS Incompatibilities" in the manual, for more
54 Please contact bug-guile@gnu.org if you have found an issue not
55 mentioned in that compatibility list.
57 ** Macro expansion produces structures instead of s-expressions
59 In the olden days, macroexpanding an s-expression would yield another
60 s-expression. Though the lexical variables were renamed, expansions of
61 core forms like `if' and `begin' were still non-hygienic, as they relied
62 on the toplevel definitions of `if' et al being the conventional ones.
64 The solution is to expand to structures instead of s-expressions. There
65 is an `if' structure, a `begin' structure, a `toplevel-ref' structure,
66 etc. The expander already did this for compilation, producing Tree-IL
67 directly; it has been changed now to do so when expanding for the
70 The real truth is somewhat more involved: Tree-IL doesn't exist until
71 modules have been booted, but we need the expander to boot modules, and
72 additionally we need a boot expander before psyntax is loaded. So a
73 subset of Tree-IL is defined in C, and the boot expander produces these
74 "macroexpanded" structures. Psyntax has been modified to produce those
75 structures as well. When Tree-IL loads, it incorporates those structures
76 directly as part of its language.
78 Finally, the evaluator has been adapted to accept these "expanded"
79 structures, and enhanced to better support the gamut of this subset of
80 Tree-IL, including `lambda*' and `case-lambda'. This was a much-needed
81 harmonization between the compiler, expander, and evaluator.
83 ** Deprecated `scm_badargsp'
85 This function is unused in Guile, but was part of its API.
87 ** `sxml->xml' enhancement
89 `sxml->xml' from `(sxml simple)' can now handle the result of
90 `xml->sxml'. See bug #29260 for more information.
92 ** New module: (system vm coverage)
94 This new module can produce code coverage reports for compiled Scheme
95 code on a line-by-line level. See "Code Coverage" in the manual for more
100 The frame objects passed to VM hook procedures are now allocated on the
101 stack instead of the heap, making the next-instruction hook practical to
104 ** New `eval-when' situation: `expand'
106 Sometimes it's important to cause side-effects while expanding an
107 expression, even in eval mode. This situation is used in
108 `define-module', `use-modules', et al, in order to affect the current
109 module and its set of syntax expanders.
111 ** Better module-level hygiene
113 Instead of attempting to track changes to the current module when
114 expanding toplevel sequences, we instead preserve referential
115 transparency relative to where the macro itself was defined. If the
116 macro should expand to expressions in the context of the new module, it
117 should wrap those expressions in `@@', which has been enhanced to accept
118 generic expressions, not just identifier references. For example, part
119 of the definition of the R6RS `library' form:
122 (define-module (name name* ...) #:pure #:version (version ...))
124 (re-export r ...) (export e ...)
125 (@@ (name name* ...) body)
128 In this example the `import' refers to the `import' definition in the
129 module where the `library' macro is defined, not in the new module.
131 ** Module system macros rewritten as hygienic macros
133 `define-module', `use-modules', `export', and other such macros have
134 been rewritten as hygienic macros. This allows the necessary referential
135 transparency for the R6RS `library' form to do the right thing.
137 ** Compiler and VM documentation updated
139 The documentation for the compiler and VM had slipped out of date; it
140 has been brought back... to the future!
142 ** Tree-IL field renaming: `vars' -> `gensyms'
144 The `vars' fields of <let>, <letrec>, <fix>, and <lambda-case> has been
145 renamed to `gensyms', for clarity, and to match <lexical-ref>.
147 ** Removed `version' field from <language>
149 Language versions weren't being updated or used in any worthwhile way;
150 they have been removed, for now at least.
152 ** New procedure: `module-export-all!'
154 This procedure exports all current and future bindings from a module.
155 Use as `(module-export-all! (current-module))'.
159 The introductory sections of the manual have been reorganized
160 significantly, making it more accessible to new users of Guile. Check it
163 ** The module namespace is now separate from the value namespace
165 It was a little-known implementation detail of Guile's module system
166 that it was built on a single hierarchical namespace of values -- that
167 if there was a module named `(foo bar)', then in the module named
168 `(foo)' there was a binding from `bar' to the `(foo bar)' module.
170 This was a neat trick, but presented a number of problems. One problem
171 was that the bindings in a module were not apparent from the module
172 itself; perhaps the `(foo)' module had a private binding for `bar', and
173 then an external contributor defined `(foo bar)'. In the end there can
174 be only one binding, so one of the two will see the wrong thing, and
175 produce an obtuse error of unclear provenance.
177 Also, the public interface of a module was also bound in the value
178 namespace, as `%module-public-interface'. This was a hack from the early
179 days of Guile's modules.
181 Both of these warts have been fixed by the addition of fields in the
182 `module' data type. Access to modules and their interfaces from the
183 value namespace has been deprecated, and all accessors use the new
184 record accessors appropriately.
186 When Guile is built with support for deprecated code, as is the default,
187 the value namespace is still searched for modules and public interfaces,
188 and a deprecation warning is raised as appropriate.
190 Finally, to support lazy loading of modules as one used to be able to do
191 with module binder procedures, Guile now has submodule binders, called
192 if a given submodule is not found. See boot-9.scm for more information.
194 ** New procedures: module-ref-submodule, module-define-submodule,
195 nested-ref-module, nested-define-module!, local-ref-module,
198 These new accessors are like their bare variants, but operate on
199 namespaces instead of values.
201 ** The (app modules) module tree is officially deprecated
203 It used to be that one could access a module named `(foo bar)' via
204 `(nested-ref the-root-module '(app modules foo bar))'. The `(app
205 modules)' bit was a never-used and never-documented abstraction, and has
206 been deprecated. See the following mail for a full discussion:
208 http://lists.gnu.org/archive/html/guile-devel/2010-04/msg00168.html
210 The `%app' binding is also deprecated.
212 ** Deprecated `@bind' syntax
214 `@bind' was part of an older implementation of the Emacs Lisp language,
215 and is no longer used.
217 ** New fluid: `%file-port-name-canonicalization'
219 This fluid parameterizes the file names that are associated with file
220 ports. If %file-port-name-canonicalization is 'absolute, then file names
221 are canonicalized to be absolute paths. If it is 'relative, then the
222 name is canonicalized, but any prefix corresponding to a member of
223 `%load-path' is stripped off. Otherwise the names are passed through
226 ** Source file name canonicalization in `compile-file', `compile-and-load'
228 These file-compiling procedures now bind
229 %file-port-name-canonicalization to their `#:canonicalization' keyword
230 argument, which defaults to 'relative. In this way, one might compile
231 "../module/ice-9/boot-9.scm", but the path that gets residualized into
232 the .go is "ice-9/boot-9.scm".
234 ** Deprecate arity access via (procedure-properties proc 'arity)
236 Instead of accessing a procedure's arity as a property, use the new
237 `procedure-minimum-arity' function, which gives the most permissive
238 arity that the the function has, in the same format as the old arity
241 ** Remove redundant accessors: program-name, program-documentation,
242 program-properties, program-property
244 Instead, just use procedure-name, procedure-documentation,
245 procedure-properties, and procedure-property.
247 ** Enhance documentation for support of Emacs Lisp's `nil'
249 See "Nil" in the manual, for more details.
251 ** Enhance documentation for support of other languages
253 See "Other Languages" in the manual, for more details.
255 ** And of course, the usual collection of bugfixes
257 Interested users should see the ChangeLog for more information.
261 Changes in 1.9.x (since the 1.8.x series):
263 * New modules (see the manual for details)
265 ** `(srfi srfi-18)', more sophisticated multithreading support
266 ** `(ice-9 i18n)', internationalization support
267 ** `(rnrs bytevectors)', the R6RS bytevector API
268 ** `(rnrs io ports)', a subset of the R6RS I/O port API
269 ** `(system xref)', a cross-referencing facility (FIXME undocumented)
270 ** `(ice-9 vlist)', lists with constant-time random access; hash lists
271 ** `(system foreign)', foreign function interface
273 ** Imported statprof, SSAX, and texinfo modules from Guile-Lib
275 The statprof statistical profiler, the SSAX XML toolkit, and the texinfo
276 toolkit from Guile-Lib have been imported into Guile proper. See
277 "Standard Library" in the manual for more details.
279 ** Integration of lalr-scm, a parser generator
281 Guile has included Dominique Boucher's fine `lalr-scm' parser generator
282 as `(system base lalr)'. See "LALR(1) Parsing" in the manual, for more
285 * Changes to the stand-alone interpreter
287 ** Guile now can compile Scheme to bytecode for a custom virtual machine.
289 Compiled code loads much faster than Scheme source code, and runs around
290 3 or 4 times as fast, generating much less garbage in the process.
292 ** Evaluating Scheme code does not use the C stack.
294 Besides when compiling Guile itself, Guile no longer uses a recursive C
295 function as an evaluator. This obviates the need to check the C stack
296 pointer for overflow. Continuations still capture the C stack, however.
298 ** New environment variables: GUILE_LOAD_COMPILED_PATH,
299 GUILE_SYSTEM_LOAD_COMPILED_PATH
301 GUILE_LOAD_COMPILED_PATH is for compiled files what GUILE_LOAD_PATH is
302 for source files. It is a different path, however, because compiled
303 files are architecture-specific. GUILE_SYSTEM_LOAD_COMPILED_PATH is like
306 ** New read-eval-print loop (REPL) implementation
308 Running Guile with no arguments drops the user into the new REPL. While
309 it is self-documenting to an extent, the new REPL has not yet been
310 documented in the manual. This will be fixed before 2.0.
312 ** New reader options: `square-brackets' and `r6rs-hex-escapes'
314 The reader supports a new option (changeable via `read-options'),
315 `square-brackets', which instructs it to interpret square brackets as
316 parentheses. This option is on by default.
318 When the new `r6rs-hex-escapes' reader option is enabled, the reader
319 will recognize string escape sequences as defined in R6RS.
321 ** Function profiling and tracing at the REPL
323 The `,profile FORM' REPL meta-command can now be used to statistically
324 profile execution of a form, to see which functions are taking the most
325 time. See `,help profile' for more information.
327 Similarly, `,trace FORM' traces all function applications that occur
328 during the execution of `FORM'. See `,help trace' for more information.
332 By default, if an exception is raised at the REPL and not caught by user
333 code, Guile will drop the user into a debugger. The user may request a
334 backtrace, inspect frames, or continue raising the exception. Full
335 documentation is available from within the debugger.
337 ** New `guile-tools' commands: `compile', `disassemble'
339 Pass the `--help' command-line option to these commands for more
342 ** Guile now adds its install prefix to the LTDL_LIBRARY_PATH
344 Users may now install Guile to nonstandard prefixes and just run
345 `/path/to/bin/guile', instead of also having to set LTDL_LIBRARY_PATH to
346 include `/path/to/lib'.
348 ** Guile's Emacs integration is now more keyboard-friendly
350 Backtraces may now be disclosed with the keyboard in addition to the
353 ** Load path change: search in version-specific paths before site paths
355 When looking for a module, Guile now searches first in Guile's
356 version-specific path (the library path), *then* in the site dir. This
357 allows Guile's copy of SSAX to override any Guile-Lib copy the user has
358 installed. Also it should cut the number of `stat' system calls by half,
361 ** Interactive Guile follows GNU conventions
363 As recommended by the GPL, Guile now shows a brief copyright and
364 warranty disclaimer on startup, along with pointers to more information.
366 * Changes to Scheme functions and syntax
368 ** New implementation of `primitive-eval'
370 Guile's `primitive-eval' is now implemented in Scheme. Actually there is
371 still a C evaluator, used when building a fresh Guile to interpret the
372 compiler, so we can compile eval.scm. Thereafter all calls to
373 primitive-eval are implemented by VM-compiled code.
375 This allows all of Guile's procedures, be they interpreted or compiled,
376 to execute on the same stack, unifying multiple-value return semantics,
377 providing for proper tail recursion between interpreted and compiled
378 code, and simplifying debugging.
380 As part of this change, the evaluator no longer mutates the internal
381 representation of the code being evaluated in a thread-unsafe manner.
383 There are two negative aspects of this change, however. First, Guile
384 takes a lot longer to compile now. Also, there is less debugging
385 information available for debugging interpreted code. We hope to improve
386 both of these situations.
388 There are many changes to the internal C evalator interface, but all
389 public interfaces should be the same. See the ChangeLog for details. If
390 we have inadvertantly changed an interface that you were using, please
391 contact bug-guile@gnu.org.
393 ** Procedure removed: `the-environment'
395 This procedure was part of the interpreter's execution model, and does
396 not apply to the compiler.
398 ** No more `local-eval'
400 `local-eval' used to exist so that one could evaluate code in the
401 lexical context of a function. Since there is no way to get the lexical
402 environment any more, as that concept has no meaning for the compiler,
403 and a different meaning for the interpreter, we have removed the
406 If you think you need `local-eval', you should probably implement your
407 own metacircular evaluator. It will probably be as fast as Guile's
410 ** Scheme source files will now be compiled automatically.
412 If a compiled .go file corresponding to a .scm file is not found or is
413 not fresh, the .scm file will be compiled on the fly, and the resulting
414 .go file stored away. An advisory note will be printed on the console.
416 Note that this mechanism depends on preservation of the .scm and .go
417 modification times; if the .scm or .go files are moved after
418 installation, care should be taken to preserve their original
421 Autocompiled files will be stored in the $XDG_CACHE_HOME/guile/ccache
422 directory, where $XDG_CACHE_HOME defaults to ~/.cache. This directory
423 will be created if needed.
425 To inhibit autocompilation, set the GUILE_AUTO_COMPILE environment
426 variable to 0, or pass --no-autocompile on the Guile command line.
428 ** New POSIX procedures: `getrlimit' and `setrlimit'
430 Note however that the interface of these functions is likely to change
431 in the next prerelease.
433 ** New POSIX procedure: `getsid'
435 Scheme binding for the `getsid' C library call.
437 ** New POSIX procedure: `getaddrinfo'
439 Scheme binding for the `getaddrinfo' C library function.
441 ** New procedure in `(oops goops)': `method-formals'
443 ** New procedures in (ice-9 session): `add-value-help-handler!',
444 `remove-value-help-handler!', `add-name-help-handler!'
445 `remove-name-help-handler!', `procedure-arguments'
447 The value and name help handlers provide some minimal extensibility to
448 the help interface. Guile-lib's `(texinfo reflection)' uses them, for
449 example, to make stexinfo help documentation available. See those
450 procedures' docstrings for more information.
452 `procedure-arguments' describes the arguments that a procedure can take,
453 combining arity and formals. For example:
455 (procedure-arguments resolve-interface)
456 => ((required . (name)) (rest . args))
458 Additionally, `module-commentary' is now publically exported from
461 ** Removed: `procedure->memoizing-macro', `procedure->syntax'
463 These procedures created primitive fexprs for the old evaluator, and are
464 no longer supported. If you feel that you need these functions, you
465 probably need to write your own metacircular evaluator (which will
466 probably be as fast as Guile's, anyway).
468 ** New language: ECMAScript
470 Guile now ships with one other high-level language supported,
471 ECMAScript. The goal is to support all of version 3.1 of the standard,
472 but not all of the libraries are there yet. This support is not yet
473 documented; ask on the mailing list if you are interested.
475 ** New language: Brainfuck
477 Brainfuck is a toy language that closely models Turing machines. Guile's
478 brainfuck compiler is meant to be an example of implementing other
479 languages. See the manual for details, or
480 http://en.wikipedia.org/wiki/Brainfuck for more information about the
481 Brainfuck language itself.
483 ** New language: Elisp
485 Guile now has an experimental Emacs Lisp compiler and runtime. You can
486 now switch to Elisp at the repl: `,language elisp'. All kudos to Daniel
487 Kraft, and all bugs to bug-guile@gnu.org.
489 ** Better documentation infrastructure for macros
491 It is now possible to introspect on the type of a macro, e.g.
492 syntax-rules, identifier-syntax, etc, and extract information about that
493 macro, such as the syntax-rules patterns or the defmacro arguments.
494 `(texinfo reflection)' takes advantage of this to give better macro
497 ** Support for arbitrary procedure metadata
499 Building on its support for docstrings, Guile now supports multiple
500 docstrings, adding them to the tail of a compiled procedure's
501 properties. For example:
507 (procedure-properties foo)
508 => ((name . foo) (documentation . "one") (documentation . "two"))
510 Also, vectors of pairs are now treated as additional metadata entries:
513 #((quz . #f) (docstring . "xyzzy"))
515 (procedure-properties bar)
516 => ((name . bar) (quz . #f) (docstring . "xyzzy"))
518 This allows arbitrary literals to be embedded as metadata in a compiled
521 ** The psyntax expander now knows how to interpret the @ and @@ special
524 ** The psyntax expander is now hygienic with respect to modules.
526 Free variables in a macro are scoped in the module that the macro was
527 defined in, not in the module the macro is used in. For example, code
530 (define-module (foo) #:export (bar))
531 (define (helper x) ...)
533 (syntax-rules () ((_ x) (helper x))))
535 (define-module (baz) #:use-module (foo))
538 It used to be you had to export `helper' from `(foo)' as well.
539 Thankfully, this has been fixed.
541 ** Complete support for version information in Guile's `module' form
543 Guile modules now have a `#:version' field. They may be loaded by
544 version as well. See "R6RS Version References", "General Information
545 about Modules", "Using Guile Modules", and "Creating Guile Modules" in
546 the manual for more information.
548 ** Support for renaming bindings on module export
550 Wherever Guile accepts a symbol as an argument to specify a binding to
551 export, it now also accepts a pair of symbols, indicating that a binding
552 should be renamed on export. See "Creating Guile Modules" in the manual
553 for more information.
555 ** `eval-case' has been deprecated, and replaced by `eval-when'.
557 The semantics of `eval-when' are easier to understand. See "Eval When"
558 in the manual, for more information.
560 ** Guile is now more strict about prohibiting definitions in expression
563 Although previous versions of Guile accepted it, the following
564 expression is not valid, in R5RS or R6RS:
566 (if test (define foo 'bar) (define foo 'baz))
568 In this specific case, it would be better to do:
570 (define foo (if test 'bar 'baz))
572 It is certainly possible to circumvent this resriction with e.g.
573 `(module-define! (current-module) 'foo 'baz)'. We would appreciate
574 feedback about this change (a consequence of using psyntax as the
575 default expander), and may choose to revisit this situation before 2.0
576 in response to user feedback.
578 ** Defmacros must now produce valid Scheme expressions.
580 It used to be that defmacros could unquote in Scheme values, as a way of
581 supporting partial evaluation, and avoiding some hygiene issues. For
584 (define (helper x) ...)
585 (define-macro (foo bar)
588 Assuming this macro is in the `(baz)' module, the direct translation of
591 (define (helper x) ...)
592 (define-macro (foo bar)
593 `((@@ (baz) helper) ,bar))
595 Of course, one could just use a hygienic macro instead:
599 ((_ bar) (helper bar))))
601 ** Guile's psyntax now supports docstrings and internal definitions.
603 The following Scheme is not strictly legal:
610 However its intent is fairly clear. Guile interprets "bar" to be the
611 docstring of `foo', and the definition of `baz' is still in definition
614 ** Macros need to be defined before their first use.
616 It used to be that with lazy memoization, this might work:
620 (define-macro (ref x) x)
623 But now, the body of `foo' is interpreted to mean a call to the toplevel
624 `ref' function, instead of a macro expansion. The solution is to define
625 macros before code that uses them.
627 ** Functions needed by macros at expand-time need to be present at
630 For example, this code will work at the REPL:
632 (define (double-helper x) (* x x))
633 (define-macro (double-literal x) (double-helper x))
634 (double-literal 2) => 4
636 But it will not work when a file is compiled, because the definition of
637 `double-helper' is not present at expand-time. The solution is to wrap
638 the definition of `double-helper' in `eval-when':
640 (eval-when (load compile eval)
641 (define (double-helper x) (* x x)))
642 (define-macro (double-literal x) (double-helper x))
643 (double-literal 2) => 4
645 See the documentation for eval-when for more information.
647 ** `macroexpand' produces structures, not S-expressions.
649 Given the need to maintain referential transparency, both lexically and
650 modular, the result of expanding Scheme expressions is no longer itself
651 an s-expression. If you want a human-readable approximation of the
652 result of `macroexpand', call `tree-il->scheme' from `(language
655 ** Removed function: `macroexpand-1'
657 It is unclear how to implement `macroexpand-1' with syntax-case, though
658 PLT Scheme does prove that it is possible.
660 ** New reader macros: #' #` #, #,@
662 These macros translate, respectively, to `syntax', `quasisyntax',
663 `unsyntax', and `unsyntax-splicing'. See the R6RS for more information.
664 These reader macros may be overridden by `read-hash-extend'.
666 ** Incompatible change to #'
668 Guile did have a #' hash-extension, by default, which just returned the
669 subsequent datum: #'foo => foo. In the unlikely event that anyone
670 actually used this, this behavior may be reinstated via the
671 `read-hash-extend' mechanism.
673 ** Scheme expresssions may be commented out with #;
675 #; comments out an entire expression. See SRFI-62 or the R6RS for more
678 ** Prompts: Delimited, composable continuations
680 Guile now has prompts as part of its primitive language. See "Prompts"
681 in the manual, for more information.
683 Expressions entered in at the REPL, or from the command line, are
684 surrounded by a prompt with the default prompt tag.
686 ** `make-stack' with a tail-called procedural narrowing argument no longer
687 works (with compiled procedures)
689 It used to be the case that a captured stack could be narrowed to select
690 calls only up to or from a certain procedure, even if that procedure
691 already tail-called another procedure. This was because the debug
692 information from the original procedure was kept on the stack.
694 Now with the new compiler, the stack only contains active frames from
695 the current continuation. A narrow to a procedure that is not in the
696 stack will result in an empty stack. To fix this, narrow to a procedure
697 that is active in the current continuation, or narrow to a specific
698 number of stack frames.
700 ** Backtraces through compiled procedures only show procedures that are
701 active in the current continuation
703 Similarly to the previous issue, backtraces in compiled code may be
704 different from backtraces in interpreted code. There are no semantic
705 differences, however. Please mail bug-guile@gnu.org if you see any
706 deficiencies with Guile's backtraces.
708 ** syntax-rules and syntax-case macros now propagate source information
709 through to the expanded code
711 This should result in better backtraces.
713 ** The currying behavior of `define' has been removed.
715 Before, `(define ((f a) b) (* a b))' would translate to
717 (define f (lambda (a) (lambda (b) (* a b))))
719 Now a syntax error is signaled, as this syntax is not supported by
720 default. Use the `(ice-9 curried-definitions)' module to get back the
723 ** New procedure, `define!'
725 `define!' is a procedure that takes two arguments, a symbol and a value,
726 and binds the value to the symbol in the current module. It's useful to
727 programmatically make definitions in the current module, and is slightly
728 less verbose than `module-define!'.
730 ** All modules have names now
732 Before, you could have anonymous modules: modules without names. Now,
733 because of hygiene and macros, all modules have names. If a module was
734 created without a name, the first time `module-name' is called on it, a
735 fresh name will be lazily generated for it.
737 ** Many syntax errors have different texts now
739 Syntax errors still throw to the `syntax-error' key, but the arguments
740 are often different now. Perhaps in the future, Guile will switch to
741 using standard SRFI-35 conditions.
743 ** Returning multiple values to compiled code will silently truncate the
744 values to the expected number
746 For example, the interpreter would raise an error evaluating the form,
747 `(+ (values 1 2) (values 3 4))', because it would see the operands as
748 being two compound "values" objects, to which `+' does not apply.
750 The compiler, on the other hand, receives multiple values on the stack,
751 not as a compound object. Given that it must check the number of values
752 anyway, if too many values are provided for a continuation, it chooses
753 to truncate those values, effectively evaluating `(+ 1 3)' instead.
755 The idea is that the semantics that the compiler implements is more
756 intuitive, and the use of the interpreter will fade out with time.
757 This behavior is allowed both by the R5RS and the R6RS.
759 ** Multiple values in compiled code are not represented by compound
762 This change may manifest itself in the following situation:
764 (let ((val (foo))) (do-something) val)
766 In the interpreter, if `foo' returns multiple values, multiple values
767 are produced from the `let' expression. In the compiler, those values
768 are truncated to the first value, and that first value is returned. In
769 the compiler, if `foo' returns no values, an error will be raised, while
770 the interpreter would proceed.
772 Both of these behaviors are allowed by R5RS and R6RS. The compiler's
773 behavior is more correct, however. If you wish to preserve a potentially
774 multiply-valued return, you will need to set up a multiple-value
775 continuation, using `call-with-values'.
777 ** Defmacros are now implemented in terms of syntax-case.
779 The practical ramification of this is that the `defmacro?' predicate has
780 been removed, along with `defmacro-transformer', `macro-table',
781 `xformer-table', `assert-defmacro?!', `set-defmacro-transformer!' and
782 `defmacro:transformer'. This is because defmacros are simply macros. If
783 any of these procedures provided useful facilities to you, we encourage
784 you to contact the Guile developers.
786 ** Hygienic macros documented as the primary syntactic extension mechanism.
788 The macro documentation was finally fleshed out with some documentation
789 on `syntax-rules' and `syntax-case' macros, and other parts of the macro
790 expansion process. See "Macros" in the manual, for details.
792 ** psyntax is now the default expander
794 Scheme code is now expanded by default by the psyntax hygienic macro
795 expander. Expansion is performed completely before compilation or
798 Notably, syntax errors will be signalled before interpretation begins.
799 In the past, many syntax errors were only detected at runtime if the
800 code in question was memoized.
802 As part of its expansion, psyntax renames all lexically-bound
803 identifiers. Original identifier names are preserved and given to the
804 compiler, but the interpreter will see the renamed variables, e.g.,
805 `x432' instead of `x'.
807 Note that the psyntax that Guile uses is a fork, as Guile already had
808 modules before incompatible modules were added to psyntax -- about 10
809 years ago! Thus there are surely a number of bugs that have been fixed
810 in psyntax since then. If you find one, please notify bug-guile@gnu.org.
812 ** syntax-rules and syntax-case are available by default.
814 There is no longer any need to import the `(ice-9 syncase)' module
815 (which is now deprecated). The expander may be invoked directly via
816 `macroexpand', though it is normally searched for via the current module
819 Also, the helper routines for syntax-case are available in the default
820 environment as well: `syntax->datum', `datum->syntax',
821 `bound-identifier=?', `free-identifier=?', `generate-temporaries',
822 `identifier?', and `syntax-violation'. See the R6RS for documentation.
824 ** Tail patterns in syntax-case
826 Guile has pulled in some more recent changes from the psyntax portable
827 syntax expander, to implement support for "tail patterns". Such patterns
828 are supported by syntax-rules and syntax-case. This allows a syntax-case
829 match clause to have ellipses, then a pattern at the end. For example:
833 ((_ val match-clause ... (else e e* ...))
836 Note how there is MATCH-CLAUSE, which is ellipsized, then there is a
837 tail pattern for the else clause. Thanks to Andreas Rottmann for the
838 patch, and Kent Dybvig for the code.
840 ** Lexical bindings introduced by hygienic macros may not be referenced
841 by nonhygienic macros.
843 If a lexical binding is introduced by a hygienic macro, it may not be
844 referenced by a nonhygienic macro. For example, this works:
847 (define-macro (bind-x val body)
848 `(let ((x ,val)) ,body))
849 (define-macro (ref x)
856 (define-syntax bind-x
858 ((_ val body) (let ((x val)) body))))
859 (define-macro (ref x)
863 It is not normal to run into this situation with existing code. However,
864 as code is ported over from defmacros to syntax-case, it is possible to
865 run into situations like this. In the future, Guile will probably port
866 its `while' macro to syntax-case, which makes this issue one to know
869 ** Macros may no longer be referenced as first-class values.
871 In the past, you could evaluate e.g. `if', and get its macro value. Now,
872 expanding this form raises a syntax error.
874 Macros still /exist/ as first-class values, but they must be
875 /referenced/ via the module system, e.g. `(module-ref (current-module)
878 ** Macros may now have docstrings.
880 `object-documentation' from `(ice-9 documentation)' may be used to
881 retrieve the docstring, once you have a macro value -- but see the above
882 note about first-class macros. Docstrings are associated with the syntax
883 transformer procedures.
885 ** `case-lambda' is now available in the default environment.
887 The binding in the default environment is equivalent to the one from the
888 `(srfi srfi-16)' module. Use the srfi-16 module explicitly if you wish
889 to maintain compatibility with Guile 1.8 and earlier.
891 ** Procedures may now have more than one arity.
893 This can be the case, for example, in case-lambda procedures. The
894 arities of compiled procedures may be accessed via procedures from the
895 `(system vm program)' module; see "Compiled Procedures", "Optional
896 Arguments", and "Case-lambda" in the manual.
898 ** `lambda*' and `define*' are now available in the default environment
900 As with `case-lambda', `(ice-9 optargs)' continues to be supported, for
901 compatibility purposes. No semantic change has been made (we hope).
902 Optional and keyword arguments now dispatch via special VM operations,
903 without the need to cons rest arguments, making them very fast.
905 ** New function, `truncated-print', with `format' support
907 `(ice-9 pretty-print)' now exports `truncated-print', a printer that
908 will ensure that the output stays within a certain width, truncating the
909 output in what is hopefully an intelligent manner. See the manual for
912 There is a new `format' specifier, `~@y', for doing a truncated
913 print (as opposed to `~y', which does a pretty-print). See the `format'
914 documentation for more details.
916 ** SRFI-4 vectors reimplemented in terms of R6RS bytevectors
918 Guile now implements SRFI-4 vectors using bytevectors. Often when you
919 have a numeric vector, you end up wanting to write its bytes somewhere,
920 or have access to the underlying bytes, or read in bytes from somewhere
921 else. Bytevectors are very good at this sort of thing. But the SRFI-4
922 APIs are nicer to use when doing number-crunching, because they are
923 addressed by element and not by byte.
925 So as a compromise, Guile allows all bytevector functions to operate on
926 numeric vectors. They address the underlying bytes in the native
927 endianness, as one would expect.
929 Following the same reasoning, that it's just bytes underneath, Guile
930 also allows uniform vectors of a given type to be accessed as if they
931 were of any type. One can fill a u32vector, and access its elements with
932 u8vector-ref. One can use f64vector-ref on bytevectors. It's all the
935 In this way, uniform numeric vectors may be written to and read from
936 input/output ports using the procedures that operate on bytevectors.
938 Calls to SRFI-4 accessors (ref and set functions) from Scheme are now
939 inlined to the VM instructions for bytevector access.
941 See "SRFI-4" in the manual, for more information.
943 ** Nonstandard SRFI-4 procedures now available from `(srfi srfi-4 gnu)'
945 Guile's `(srfi srfi-4)' now only exports those srfi-4 procedures that
946 are part of the standard. Complex uniform vectors and the
947 `any->FOOvector' family are now available only from `(srfi srfi-4 gnu)'.
949 Guile's default environment imports `(srfi srfi-4)', and probably should
950 import `(srfi srfi-4 gnu)' as well.
952 See "SRFI-4 Extensions" in the manual, for more information.
954 ** New syntax: include-from-path.
956 `include-from-path' is like `include', except it looks for its file in
957 the load path. It can be used to compile other files into a file.
959 ** New syntax: quasisyntax.
961 `quasisyntax' is to `syntax' as `quasiquote' is to `quote'. See the R6RS
962 documentation for more information. Thanks to Andre van Tonder for the
965 ** Unicode characters
967 Unicode characters may be entered in octal format via e.g. `#\454', or
968 created via (integer->char 300). A hex external representation will
969 probably be introduced at some point.
973 Internally, strings are now represented either in the `latin-1'
974 encoding, one byte per character, or in UTF-32, with four bytes per
975 character. Strings manage their own allocation, switching if needed.
977 Extended characters may be written in a literal string using the
978 hexadecimal escapes `\xXX', `\uXXXX', or `\UXXXXXX', for 8-bit, 16-bit,
979 or 24-bit codepoints, respectively, or entered directly in the native
980 encoding of the port on which the string is read.
984 One may now use U+03BB (GREEK SMALL LETTER LAMBDA) as an identifier.
986 ** Support for non-ASCII source code files
988 The default reader now handles source code files for some of the
989 non-ASCII character encodings, such as UTF-8. A non-ASCII source file
990 should have an encoding declaration near the top of the file. Also,
991 there is a new function, `file-encoding', that scans a port for a coding
992 declaration. See the section of the manual entitled, "Character Encoding
995 The pre-1.9.3 reader handled 8-bit clean but otherwise unspecified source
996 code. This use is now discouraged.
998 ** Support for locale transcoding when reading from and writing to ports
1000 Ports now have an associated character encoding, and port read and write
1001 operations do conversion to and from locales automatically. Ports also
1002 have an associated strategy for how to deal with locale conversion
1005 See the documentation in the manual for the four new support functions,
1006 `set-port-encoding!', `port-encoding', `set-port-conversion-strategy!',
1007 and `port-conversion-strategy'.
1009 ** String and SRFI-13 functions can operate on Unicode strings
1011 ** Unicode support for SRFI-14 character sets
1013 The default character sets are no longer locale dependent and contain
1014 characters from the whole Unicode range. There is a new predefined
1015 character set, `char-set:designated', which contains all assigned
1016 Unicode characters. There is a new debugging function, `%char-set-dump'.
1018 ** Character functions operate on Unicode characters
1020 `char-upcase' and `char-downcase' use default Unicode casing rules.
1021 Character comparisons such as `char<?' and `char-ci<?' now sort based on
1022 Unicode code points.
1024 ** Global variables `scm_charnames' and `scm_charnums' are removed
1026 These variables contained the names of control characters and were
1027 used when writing characters. While these were global, they were
1028 never intended to be public API. They have been replaced with private
1031 ** EBCDIC support is removed
1033 There was an EBCDIC compile flag that altered some of the character
1034 processing. It appeared that full EBCDIC support was never completed
1035 and was unmaintained.
1037 ** Compile-time warnings
1039 Guile can warn about potentially unbound free variables. Pass the
1040 -Wunbound-variable on the `guile-tools compile' command line, or add
1041 `#:warnings '(unbound-variable)' to your `compile' or `compile-file'
1044 Guile can also warn when you pass the wrong number of arguments to a
1045 procedure, with -Warity-mismatch, or `arity-mismatch' in the
1046 `#:warnings' as above.
1048 Other warnings include `-Wunused-variable' and `-Wunused-toplevel', to
1049 warn about unused local or global (top-level) variables.
1051 ** A new `memoize-symbol' evaluator trap has been added.
1053 This trap can be used for efficiently implementing a Scheme code
1056 ** Duplicate bindings among used modules are resolved lazily.
1058 This slightly improves program startup times.
1060 ** New thread cancellation and thread cleanup API
1062 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
1064 ** GOOPS dispatch in scheme
1066 As an implementation detail, GOOPS dispatch is no longer implemented by
1067 special evaluator bytecodes, but rather directly via a Scheme function
1068 associated with an applicable struct. There is some VM support for the
1069 underlying primitives, like `class-of'.
1071 This change will in the future allow users to customize generic function
1072 dispatch without incurring a performance penalty, and allow us to
1073 implement method combinations.
1075 ** Applicable struct support
1077 One may now make structs from Scheme that may be applied as procedures.
1078 To do so, make a struct whose vtable is `<applicable-struct-vtable>'.
1079 That struct will be the vtable of your applicable structs; instances of
1080 that new struct are assumed to have the procedure in their first slot.
1081 `<applicable-struct-vtable>' is like Common Lisp's
1082 `funcallable-standard-class'. Likewise there is
1083 `<applicable-struct-with-setter-vtable>', which looks for the setter in
1084 the second slot. This needs to be better documented.
1088 GOOPS had a number of concepts that were relevant to the days of Tcl,
1089 but not any more: operators and entities, mainly. These objects were
1090 never documented, and it is unlikely that they were ever used. Operators
1091 were a kind of generic specific to the Tcl support. Entities were
1092 replaced by applicable structs, mentioned above.
1094 ** New struct slot allocation: "hidden"
1096 A hidden slot is readable and writable, but will not be initialized by a
1097 call to make-struct. For example in your layout you would say "ph"
1098 instead of "pw". Hidden slots are useful for adding new slots to a
1099 vtable without breaking existing invocations to make-struct.
1101 ** eqv? not a generic
1103 One used to be able to extend `eqv?' as a primitive-generic, but no
1104 more. Because `eqv?' is in the expansion of `case' (via `memv'), which
1105 should be able to compile to static dispatch tables, it doesn't make
1106 sense to allow extensions that would subvert this optimization.
1108 ** `inet-ntop' and `inet-pton' are always available.
1110 Guile now use a portable implementation of `inet_pton'/`inet_ntop', so
1111 there is no more need to use `inet-aton'/`inet-ntoa'. The latter
1112 functions are deprecated.
1114 ** Fast bit operations.
1116 The bit-twiddling operations `ash', `logand', `logior', and `logxor' now
1117 have dedicated bytecodes. Guile is not just for symbolic computation,
1118 it's for number crunching too.
1120 ** Faster SRFI-9 record access
1122 SRFI-9 records are now implemented directly on top of Guile's structs,
1123 and their accessors are defined in such a way that normal call-sites
1124 inline to special VM opcodes, while still allowing for the general case
1125 (e.g. passing a record accessor to `apply').
1127 ** R6RS block comment support
1129 Guile now supports R6RS nested block comments. The start of a comment is
1130 marked with `#|', and the end with `|#'.
1132 ** `guile-2' cond-expand feature
1134 To test if your code is running under Guile 2.0 (or its alpha releases),
1135 test for the `guile-2' cond-expand feature. Like this:
1137 (cond-expand (guile-2 (eval-when (compile)
1138 ;; This must be evaluated at compile time.
1139 (fluid-set! current-reader my-reader)))
1141 ;; Earlier versions of Guile do not have a
1142 ;; separate compilation phase.
1143 (fluid-set! current-reader my-reader)))
1145 ** Fix bad interaction between `false-if-exception' and stack-call.
1147 Exceptions thrown by `false-if-exception' were erronously causing the
1148 stack to be saved, causing later errors to show the incorrectly-saved
1149 backtrace. This has been fixed.
1151 ** New global variables: %load-compiled-path, %load-compiled-extensions
1153 These are analogous to %load-path and %load-extensions.
1155 ** New procedure, `make-promise'
1157 `(make-promise (lambda () foo))' is equivalent to `(delay foo)'.
1159 ** `defined?' may accept a module as its second argument
1161 Previously it only accepted internal structures from the evaluator.
1163 ** New entry into %guile-build-info: `ccachedir'
1165 ** Fix bug in `module-bound?'.
1167 `module-bound?' was returning true if a module did have a local
1168 variable, but one that was unbound, but another imported module bound
1169 the variable. This was an error, and was fixed.
1171 ** `(ice-9 syncase)' has been deprecated.
1173 As syntax-case is available by default, importing `(ice-9 syncase)' has
1174 no effect, and will trigger a deprecation warning.
1176 ** New readline history functions
1178 The (ice-9 readline) module now provides add-history, read-history,
1179 write-history and clear-history, which wrap the corresponding GNU
1180 History library functions.
1182 ** Removed deprecated uniform array procedures:
1183 dimensions->uniform-array, list->uniform-array, array-prototype
1185 Instead, use make-typed-array, list->typed-array, or array-type,
1188 ** Deprecated: `lazy-catch'
1190 `lazy-catch' was a form that captured the stack at the point of a
1191 `throw', but the dynamic state at the point of the `catch'. It was a bit
1192 crazy. Please change to use `catch', possibly with a throw-handler, or
1193 `with-throw-handler'.
1195 ** Last but not least, the `λ' macro can be used in lieu of `lambda'
1197 * Changes to the C interface
1199 ** Guile now uses libgc, the Boehm-Demers-Weiser garbage collector
1201 The semantics of `scm_gc_malloc ()' have been changed, in a
1202 backward-compatible way. A new allocation routine,
1203 `scm_gc_malloc_pointerless ()', was added.
1205 Libgc is a conservative GC, which we hope will make interaction with C
1206 code easier and less error-prone.
1208 ** New type definitions for `scm_t_intptr' and friends.
1210 `SCM_T_UINTPTR_MAX', `SCM_T_INTPTR_MIN', `SCM_T_INTPTR_MAX',
1211 `SIZEOF_SCM_T_BITS', `scm_t_intptr' and `scm_t_uintptr' are now
1212 available to C. Have fun!
1214 ** The GH interface (deprecated in version 1.6, 2001) was removed.
1216 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
1218 This makes these internal functions technically not callable from
1221 ** Functions for handling `scm_option' now no longer require an argument
1222 indicating length of the `scm_t_option' array.
1224 ** Procedures-with-setters are now implemented using applicable structs
1226 From a user's perspective this doesn't mean very much. But if, for some
1227 odd reason, you used the SCM_PROCEDURE_WITH_SETTER_P, SCM_PROCEDURE, or
1228 SCM_SETTER macros, know that they're deprecated now. Also, scm_tc7_pws
1231 ** Remove old evaluator closures
1233 There used to be ranges of typecodes allocated to interpreted data
1234 structures, but that it no longer the case, given that interpreted
1235 procedure are now just regular VM closures. As a result, there is a
1236 newly free tc3, and a number of removed macros. See the ChangeLog for
1239 ** Primitive procedures are now VM trampoline procedures
1241 It used to be that there were something like 12 different typecodes
1242 allocated to primitive procedures, each with its own calling convention.
1243 Now there is only one, the gsubr. This may affect user code if you were
1244 defining a procedure using scm_c_make_subr rather scm_c_make_gsubr. The
1245 solution is to switch to use scm_c_make_gsubr. This solution works well
1246 both with the old 1.8 and and with the current 1.9 branch.
1248 Guile's old evaluator used to have special cases for applying "gsubrs",
1249 primitive procedures with specified numbers of required, optional, and
1250 rest arguments. Now, however, Guile represents gsubrs as normal VM
1251 procedures, with appropriate bytecode to parse out the correct number of
1252 arguments, including optional and rest arguments, and then with a
1253 special bytecode to apply the gsubr.
1255 This allows primitive procedures to appear on the VM stack, allowing
1256 them to be accurately counted in profiles. Also they now have more
1257 debugging information attached to them -- their number of arguments, for
1258 example. In addition, the VM can completely inline the application
1259 mechanics, allowing for faster primitive calls.
1261 However there are some changes on the C level. There is no more
1262 `scm_tc7_gsubr' or `scm_tcs_subrs' typecode for primitive procedures, as
1263 they are just VM procedures. Likewise the macros `SCM_GSUBR_TYPE',
1264 `SCM_GSUBR_MAKTYPE', `SCM_GSUBR_REQ', `SCM_GSUBR_OPT', and
1265 `SCM_GSUBR_REST' are gone, as are `SCM_SUBR_META_INFO', `SCM_SUBR_PROPS'
1266 `SCM_SET_SUBR_GENERIC_LOC', and `SCM_SUBR_ARITY_TO_TYPE'.
1268 Perhaps more significantly, `scm_c_make_subr',
1269 `scm_c_make_subr_with_generic', `scm_c_define_subr', and
1270 `scm_c_define_subr_with_generic'. They all operated on subr typecodes,
1271 and there are no more subr typecodes. Use the scm_c_make_gsubr family
1274 Normal users of gsubrs should not be affected, though, as the
1275 scm_c_make_gsubr family still is the correct way to create primitive
1278 ** Remove deprecated array C interfaces
1280 Removed the deprecated array functions `scm_i_arrayp',
1281 `scm_i_array_ndim', `scm_i_array_mem', `scm_i_array_v',
1282 `scm_i_array_base', `scm_i_array_dims', and the deprecated macros
1283 `SCM_ARRAYP', `SCM_ARRAY_NDIM', `SCM_ARRAY_CONTP', `SCM_ARRAY_MEM',
1284 `SCM_ARRAY_V', `SCM_ARRAY_BASE', and `SCM_ARRAY_DIMS'.
1286 ** Remove unused snarf macros
1288 `SCM_DEFINE1', `SCM_PRIMITIVE_GENERIC_1', `SCM_PROC1, and `SCM_GPROC1'
1289 are no more. Use SCM_DEFINE or SCM_PRIMITIVE_GENERIC instead.
1291 ** New functions: `scm_call_n', `scm_c_run_hookn'
1293 `scm_call_n' applies to apply a function to an array of arguments.
1294 `scm_c_run_hookn' runs a hook with an array of arguments.
1296 ** Some SMOB types changed to have static typecodes
1298 Fluids, dynamic states, and hash tables used to be SMOB objects, but now
1299 they have statically allocated tc7 typecodes.
1301 ** Preparations for changing SMOB representation
1303 If things go right, we'll be changing the SMOB representation soon. To
1304 that end, we did a lot of cleanups to calls to e.g. SCM_CELL_WORD_2(x) when
1305 the code meant SCM_SMOB_DATA_2(x); user code will need similar changes
1306 in the future. Code accessing SMOBs using SCM_CELL macros was never
1307 correct, but until now things still worked. Users should be aware of
1310 ** Changed invocation mechanics of applicable SMOBs
1312 Guile's old evaluator used to have special cases for applying SMOB
1313 objects. Now, with the VM, when Guile sees a SMOB, it looks up a VM
1314 trampoline procedure for it, and use the normal mechanics to apply the
1315 trampoline. This simplifies procedure application in the normal,
1318 The upshot is that the mechanics used to apply a SMOB are different from
1319 1.8. Descriptors no longer have `apply_0', `apply_1', `apply_2', and
1320 `apply_3' functions, and the macros SCM_SMOB_APPLY_0 and friends are now
1321 deprecated. Just use the scm_call_0 family of procedures.
1323 ** New C function: scm_module_public_interface
1325 This procedure corresponds to Scheme's `module-public-interface'.
1327 ** Undeprecate `scm_the_root_module ()'
1329 It's useful to be able to get the root module from C without doing a
1332 ** Inline vector allocation
1334 Instead of having vectors point out into the heap for their data, their
1335 data is now allocated inline to the vector object itself. The same is
1336 true for bytevectors, by default, though there is an indirection
1337 available which should allow for making a bytevector from an existing
1340 ** New struct constructors that don't involve making lists
1342 `scm_c_make_struct' and `scm_c_make_structv' are new varargs and array
1343 constructors, respectively, for structs. You might find them useful.
1347 In Guile 1.8, there were debugging frames on the C stack. Now there is
1348 no more need to explicitly mark the stack in this way, because Guile has
1349 a VM stack that it knows how to walk, which simplifies the C API
1350 considerably. See the ChangeLog for details; the relevant interface is
1351 in libguile/stacks.h. The Scheme API has not been changed significantly.
1353 ** Removal of Guile's primitive object system.
1355 There were a number of pieces in `objects.[ch]' that tried to be a
1356 minimal object system, but were never documented, and were quickly
1357 obseleted by GOOPS' merge into Guile proper. So `scm_make_class_object',
1358 `scm_make_subclass_object', `scm_metaclass_standard', and like symbols
1359 from objects.h are no more. In the very unlikely case in which these
1360 were useful to you, we urge you to contact guile-devel.
1364 Actually the future is still in the state that it was, is, and ever
1365 shall be, Amen, except that `futures.c' and `futures.h' are no longer a
1366 part of it. These files were experimental, never compiled, and would be
1367 better implemented in Scheme anyway. In the future, that is.
1369 ** Deprecate trampolines
1371 There used to be C functions `scm_trampoline_0', `scm_trampoline_1', and
1372 so on. The point was to do some precomputation on the type of the
1373 procedure, then return a specialized "call" procedure. However this
1374 optimization wasn't actually an optimization, so it is now deprecated.
1375 Just use `scm_call_0', etc instead.
1377 ** Better support for Lisp `nil'.
1379 The bit representation of `nil' has been tweaked so that it is now very
1380 efficient to check e.g. if a value is equal to Scheme's end-of-list or
1381 Lisp's nil. Additionally there are a heap of new, specific predicates
1382 like scm_is_null_or_nil.
1384 ** Better integration of Lisp `nil'.
1386 `scm_is_boolean', `scm_is_false', and `scm_is_null' all return true now
1387 for Lisp's `nil'. This shouldn't affect any Scheme code at this point,
1388 but when we start to integrate more with Emacs, it is possible that we
1389 break code that assumes that, for example, `(not x)' implies that `x' is
1390 `eq?' to `#f'. This is not a common assumption. Refactoring affected
1391 code to rely on properties instead of identities will improve code
1392 correctness. See "Nil" in the manual, for more details.
1394 ** Support for static allocation of strings, symbols, and subrs.
1396 Calls to snarfing CPP macros like SCM_DEFINE macro will now allocate
1397 much of their associated data as static variables, reducing Guile's
1400 ** `scm_stat' has an additional argument, `exception_on_error'
1401 ** `scm_primitive_load_path' has an additional argument `exception_on_not_found'
1403 ** `scm_set_port_seek' and `scm_set_port_truncate' use the `scm_t_off' type
1405 Previously they would use the `off_t' type, which is fragile since its
1406 definition depends on the application's value for `_FILE_OFFSET_BITS'.
1408 ** The `long_long' C type, deprecated in 1.8, has been removed
1410 ** Removed deprecated uniform array procedures: scm_make_uve,
1411 scm_array_prototype, scm_list_to_uniform_array,
1412 scm_dimensions_to_uniform_array, scm_make_ra, scm_shap2ra, scm_cvref,
1413 scm_ra_set_contp, scm_aind, scm_raprin1
1415 These functions have been deprecated since early 2005.
1417 * Changes to the distribution
1419 ** Guile's license is now LGPLv3+
1421 In other words the GNU Lesser General Public License, version 3 or
1422 later (at the discretion of each person that chooses to redistribute
1425 ** GOOPS documentation folded into Guile reference manual
1427 GOOPS, Guile's object system, used to be documented in separate manuals.
1428 This content is now included in Guile's manual directly.
1430 ** `guile-config' will be deprecated in favor of `pkg-config'
1432 `guile-config' has been rewritten to get its information from
1433 `pkg-config', so this should be a transparent change. Note however that
1434 guile.m4 has yet to be modified to call pkg-config instead of
1437 ** Guile now provides `guile-2.0.pc' instead of `guile-1.8.pc'
1439 Programs that use `pkg-config' to find Guile or one of its Autoconf
1440 macros should now require `guile-2.0' instead of `guile-1.8'.
1442 ** New installation directory: $(pkglibdir)/1.9/ccache
1444 If $(libdir) is /usr/lib, for example, Guile will install its .go files
1445 to /usr/lib/guile/1.9/ccache. These files are architecture-specific.
1447 ** Parallel installability fixes
1449 Guile now installs its header files to a effective-version-specific
1450 directory, and includes the effective version (e.g. 2.0) in the library
1451 name (e.g. libguile-2.0.so).
1453 This change should be transparent to users, who should detect Guile via
1454 the guile.m4 macro, or the guile-2.0.pc pkg-config file. It will allow
1455 parallel installs for multiple versions of Guile development
1458 ** Dynamically loadable extensions may be placed in a Guile-specific path
1460 Before, Guile only searched the system library paths for extensions
1461 (e.g. /usr/lib), which meant that the names of Guile extensions had to
1462 be globally unique. Installing them to a Guile-specific extensions
1463 directory is cleaner. Use `pkg-config --variable=extensionsdir
1464 guile-2.0' to get the location of the extensions directory.
1466 ** New dependency: libgc
1468 See http://www.hpl.hp.com/personal/Hans_Boehm/gc/, for more information.
1470 ** New dependency: GNU libunistring
1472 See http://www.gnu.org/software/libunistring/, for more information. Our
1473 Unicode support uses routines from libunistring.
1475 ** New dependency: libffi
1477 See http://sourceware.org/libffi/, for more information.
1481 Changes in 1.8.8 (since 1.8.7)
1485 ** Fix possible buffer overruns when parsing numbers
1486 ** Avoid clash with system setjmp/longjmp on IA64
1487 ** Fix `wrong type arg' exceptions with IPv6 addresses
1490 Changes in 1.8.7 (since 1.8.6)
1492 * New modules (see the manual for details)
1494 ** `(srfi srfi-98)', an interface to access environment variables
1498 ** Fix compilation with `--disable-deprecated'
1499 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
1500 ** Fix MinGW build problem caused by HAVE_STRUCT_TIMESPEC confusion
1501 ** Fix build problem when scm_t_timespec is different from struct timespec
1502 ** Fix build when compiled with -Wundef -Werror
1503 ** More build fixes for `alphaev56-dec-osf5.1b' (Tru64)
1504 ** Build fixes for `powerpc-ibm-aix5.3.0.0' (AIX 5.3)
1505 ** With GCC, always compile with `-mieee' on `alpha*' and `sh*'
1506 ** Better diagnose broken `(strftime "%z" ...)' in `time.test' (bug #24130)
1507 ** Fix parsing of SRFI-88/postfix keywords longer than 128 characters
1508 ** Fix reading of complex numbers where both parts are inexact decimals
1510 ** Allow @ macro to work with (ice-9 syncase)
1512 Previously, use of the @ macro in a module whose code is being
1513 transformed by (ice-9 syncase) would cause an "Invalid syntax" error.
1514 Now it works as you would expect (giving the value of the specified
1517 ** Have `scm_take_locale_symbol ()' return an interned symbol (bug #25865)
1520 Changes in 1.8.6 (since 1.8.5)
1522 * New features (see the manual for details)
1524 ** New convenience function `scm_c_symbol_length ()'
1526 ** Single stepping through code from Emacs
1528 When you use GDS to evaluate Scheme code from Emacs, you can now use
1529 `C-u' to indicate that you want to single step through that code. See
1530 `Evaluating Scheme Code' in the manual for more details.
1532 ** New "guile(1)" man page!
1534 * Changes to the distribution
1536 ** Automake's `AM_MAINTAINER_MODE' is no longer used
1538 Thus, the `--enable-maintainer-mode' configure option is no longer
1539 available: Guile is now always configured in "maintainer mode".
1541 ** `ChangeLog' files are no longer updated
1543 Instead, changes are detailed in the version control system's logs. See
1544 the top-level `ChangeLog' files for details.
1549 ** `symbol->string' now returns a read-only string, as per R5RS
1550 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
1551 ** `guile-config link' now prints `-L$libdir' before `-lguile'
1552 ** Fix memory corruption involving GOOPS' `class-redefinition'
1553 ** Fix possible deadlock in `mutex-lock'
1554 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
1555 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
1556 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
1557 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
1558 ** Fix misleading output from `(help rationalize)'
1559 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
1560 ** Fix `gcd' when called with a single, negative argument.
1561 ** Fix `Stack overflow' errors seen when building on some platforms
1562 ** Fix bug when `scm_with_guile ()' was called several times from the
1564 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
1565 dynamic environment of the call to `raise'
1566 ** Fix potential deadlock in `make-struct'
1567 ** Fix compilation problem with libltdl from Libtool 2.2.x
1568 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
1571 Changes in 1.8.5 (since 1.8.4)
1573 * Infrastructure changes
1575 ** Guile repository switched from CVS to Git
1577 The new repository can be accessed using
1578 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
1579 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
1581 ** Add support for `pkg-config'
1583 See "Autoconf Support" in the manual for details.
1585 * New modules (see the manual for details)
1589 * New features (see the manual for details)
1591 ** New `postfix' read option, for SRFI-88 keyword syntax
1592 ** Some I/O primitives have been inlined, which improves I/O performance
1593 ** New object-based traps infrastructure
1595 This is a GOOPS-based infrastructure that builds on Guile's low-level
1596 evaluator trap calls and facilitates the development of debugging
1597 features like single-stepping, breakpoints, tracing and profiling.
1598 See the `Traps' node of the manual for details.
1600 ** New support for working on Guile code from within Emacs
1602 Guile now incorporates the `GDS' library (previously distributed
1603 separately) for working on Guile code from within Emacs. See the
1604 `Using Guile In Emacs' node of the manual for details.
1608 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
1609 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
1611 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
1612 would trigger an unbound variable error for `match:andmap'.
1614 ** `(oop goops describe)' now properly provides the `describe' feature
1615 ** Fixed `args-fold' from `(srfi srfi-37)'
1617 Previously, parsing short option names of argument-less options would
1618 lead to a stack overflow.
1620 ** `(srfi srfi-35)' is now visible through `cond-expand'
1621 ** Fixed type-checking for the second argument of `eval'
1622 ** Fixed type-checking for SRFI-1 `partition'
1623 ** Fixed `struct-ref' and `struct-set!' on "light structs"
1624 ** Honor struct field access rights in GOOPS
1625 ** Changed the storage strategy of source properties, which fixes a deadlock
1626 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
1627 ** Fixed build issue for GNU/Linux on IA64
1628 ** Fixed build issues on NetBSD 1.6
1629 ** Fixed build issue on Solaris 2.10 x86_64
1630 ** Fixed build issue with DEC/Compaq/HP's compiler
1631 ** Fixed `scm_from_complex_double' build issue on FreeBSD
1632 ** Fixed `alloca' build issue on FreeBSD 6
1633 ** Removed use of non-portable makefile constructs
1634 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
1635 ** Make sure all tests honor `$TMPDIR'
1638 Changes in 1.8.4 (since 1.8.3)
1642 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
1643 ** Fixed a segmentation fault which occurred when displaying the
1644 backtrace of a stack with a promise object (made by `delay') in it.
1645 ** Make `accept' leave guile mode while blocking
1646 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
1647 ** Fixed a build problem on AIX (use of func_data identifier)
1648 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
1649 called with an associator proc that returns neither a pair nor #f.
1650 ** Secondary threads now always return a valid module for (current-module).
1651 ** Avoid MacOS build problems caused by incorrect combination of "64"
1652 system and library calls.
1653 ** `guile-snarf' now honors `$TMPDIR'
1654 ** `guile-config compile' now reports CPPFLAGS used at compile-time
1655 ** Fixed build with Sun Studio (Solaris 9)
1656 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
1657 uniform vectors on AIX.
1658 ** Fixed a deadlock that occurs upon GC with multiple threads.
1659 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
1660 ** Fixed autotool-derived build problems on AIX 6.1.
1661 ** Fixed NetBSD/alpha support
1662 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
1664 * New modules (see the manual for details)
1668 * Documentation fixes and improvements
1670 ** Removed premature breakpoint documentation
1672 The features described are not available in the series of 1.8.x
1673 releases, so the documentation was misleading and has been removed.
1675 ** More about Guile's default *random-state* variable
1677 ** GOOPS: more about how to use `next-method'
1679 * Changes to the distribution
1681 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
1683 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
1684 General Public License, and the few incorrect files have now been
1685 fixed to agree with the rest of the Guile distribution.
1687 ** Removed unnecessary extra copies of COPYING*
1689 The distribution now contains a single COPYING.LESSER at its top level.
1692 Changes in 1.8.3 (since 1.8.2)
1694 * New modules (see the manual for details)
1701 ** The `(ice-9 slib)' module now works as expected
1702 ** Expressions like "(set! 'x #t)" no longer yield a crash
1703 ** Warnings about duplicate bindings now go to stderr
1704 ** A memory leak in `make-socket-address' was fixed
1705 ** Alignment issues (e.g., on SPARC) in network routines were fixed
1706 ** A threading issue that showed up at least on NetBSD was fixed
1707 ** Build problems on Solaris and IRIX fixed
1709 * Implementation improvements
1711 ** The reader is now faster, which reduces startup time
1712 ** Procedures returned by `record-accessor' and `record-modifier' are faster
1715 Changes in 1.8.2 (since 1.8.1):
1717 * New procedures (see the manual for details)
1719 ** set-program-arguments
1722 * Incompatible changes
1724 ** The body of a top-level `define' no longer sees the binding being created
1726 In a top-level `define', the binding being created is no longer visible
1727 from the `define' body. This breaks code like
1728 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
1729 unbound in the body. However, such code was not R5RS-compliant anyway,
1734 ** Fractions were not `equal?' if stored in unreduced form.
1735 (A subtle problem, since printing a value reduced it, making it work.)
1736 ** srfi-60 `copy-bit' failed on 64-bit systems
1737 ** "guile --use-srfi" option at the REPL can replace core functions
1738 (Programs run with that option were ok, but in the interactive REPL
1739 the core bindings got priority, preventing SRFI replacements or
1741 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
1742 ** `kill' on mingw throws an error for a PID other than oneself
1743 ** Procedure names are attached to procedure-with-setters
1744 ** Array read syntax works with negative lower bound
1745 ** `array-in-bounds?' fix if an array has different lower bounds on each index
1746 ** `*' returns exact 0 for "(* inexact 0)"
1747 This follows what it always did for "(* 0 inexact)".
1748 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
1749 ** SRFI-19: `date->julian-day' did not account for timezone offset
1750 ** `ttyname' no longer crashes when passed a non-tty argument
1751 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
1752 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
1753 ** GOOPS: Fixed a bug in `method-more-specific?'
1754 ** Build problems on Solaris fixed
1755 ** Build problems on HP-UX IA64 fixed
1756 ** Build problems on MinGW fixed
1759 Changes in 1.8.1 (since 1.8.0):
1761 * LFS functions are now used to access 64-bit files on 32-bit systems.
1763 * New procedures (see the manual for details)
1765 ** primitive-_exit - [Scheme] the-root-module
1766 ** scm_primitive__exit - [C]
1767 ** make-completion-function - [Scheme] (ice-9 readline)
1768 ** scm_c_locale_stringn_to_number - [C]
1769 ** scm_srfi1_append_reverse [C]
1770 ** scm_srfi1_append_reverse_x [C]
1778 ** Build problems have been fixed on MacOS, SunOS, and QNX.
1780 ** `strftime' fix sign of %z timezone offset.
1782 ** A one-dimensional array can now be 'equal?' to a vector.
1784 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
1786 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
1788 ** `record-accessor' and `record-modifier' now have strict type checks.
1790 Record accessor and modifier procedures now throw an error if the
1791 record type of the record they're given is not the type expected.
1792 (Previously accessors returned #f and modifiers silently did nothing).
1794 ** It is now OK to use both autoload and use-modules on a given module.
1796 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
1798 Previously there was no checking on primatives like make-vector that
1799 accept "one or two" arguments. Now there is.
1801 ** The srfi-1 assoc function now calls its equality predicate properly.
1803 Previously srfi-1 assoc would call the equality predicate with the key
1804 last. According to the SRFI, the key should be first.
1806 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
1808 ** The array-set! procedure no longer segfaults when given a bit vector.
1810 ** Bugs in make-shared-array have been fixed.
1812 ** string<? and friends now follow char<? etc order on 8-bit chars.
1814 ** The format procedure now handles inf and nan values for ~f correctly.
1816 ** exact->inexact should no longer overflow when given certain large fractions.
1818 ** srfi-9 accessor and modifier procedures now have strict record type checks.
1820 This matches the srfi-9 specification.
1822 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
1824 Previously the (ice-9 ftw) procedures would ignore any file that had
1825 the same inode number as a file they had already seen, even if that
1826 file was on a different device.
1829 Changes in 1.8.0 (changes since the 1.6.x series):
1831 * Changes to the distribution
1833 ** Guile is now licensed with the GNU Lesser General Public License.
1835 ** The manual is now licensed with the GNU Free Documentation License.
1837 ** Guile now requires GNU MP (http://swox.com/gmp).
1839 Guile now uses the GNU MP library for arbitrary precision arithmetic.
1841 ** Guile now has separate private and public configuration headers.
1843 That is, things like HAVE_STRING_H no longer leak from Guile's
1846 ** Guile now provides and uses an "effective" version number.
1848 Guile now provides scm_effective_version and effective-version
1849 functions which return the "effective" version number. This is just
1850 the normal full version string without the final micro-version number,
1851 so the current effective-version is "1.8". The effective version
1852 should remain unchanged during a stable series, and should be used for
1853 items like the versioned share directory name
1854 i.e. /usr/share/guile/1.8.
1856 Providing an unchanging version number during a stable release for
1857 things like the versioned share directory can be particularly
1858 important for Guile "add-on" packages, since it provides a directory
1859 that they can install to that won't be changed out from under them
1860 with each micro release during a stable series.
1862 ** Thread implementation has changed.
1864 When you configure "--with-threads=null", you will get the usual
1865 threading API (call-with-new-thread, make-mutex, etc), but you can't
1866 actually create new threads. Also, "--with-threads=no" is now
1867 equivalent to "--with-threads=null". This means that the thread API
1868 is always present, although you might not be able to create new
1871 When you configure "--with-threads=pthreads" or "--with-threads=yes",
1872 you will get threads that are implemented with the portable POSIX
1873 threads. These threads can run concurrently (unlike the previous
1874 "coop" thread implementation), but need to cooperate for things like
1877 The default is "pthreads", unless your platform doesn't have pthreads,
1878 in which case "null" threads are used.
1880 See the manual for details, nodes "Initialization", "Multi-Threading",
1881 "Blocking", and others.
1883 ** There is the new notion of 'discouraged' features.
1885 This is a milder form of deprecation.
1887 Things that are discouraged should not be used in new code, but it is
1888 OK to leave them in old code for now. When a discouraged feature is
1889 used, no warning message is printed like there is for 'deprecated'
1890 features. Also, things that are merely discouraged are nevertheless
1891 implemented efficiently, while deprecated features can be very slow.
1893 You can omit discouraged features from libguile by configuring it with
1894 the '--disable-discouraged' option.
1896 ** Deprecation warnings can be controlled at run-time.
1898 (debug-enable 'warn-deprecated) switches them on and (debug-disable
1899 'warn-deprecated) switches them off.
1901 ** Support for SRFI 61, extended cond syntax for multiple values has
1904 This SRFI is always available.
1906 ** Support for require-extension, SRFI-55, has been added.
1908 The SRFI-55 special form `require-extension' has been added. It is
1909 available at startup, and provides a portable way to load Scheme
1910 extensions. SRFI-55 only requires support for one type of extension,
1911 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
1914 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
1916 The (srfi srfi-26) module is an implementation of SRFI-26 which
1917 provides the `cut' and `cute' syntax. These may be used to specialize
1918 parameters without currying.
1920 ** New module (srfi srfi-31)
1922 This is an implementation of SRFI-31 which provides a special form
1923 `rec' for recursive evaluation.
1925 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
1926 been merged with the core, making their functionality always
1929 The modules are still available, tho, and you could use them together
1930 with a renaming import, for example.
1932 ** Guile no longer includes its own version of libltdl.
1934 The official version is good enough now.
1936 ** The --enable-htmldoc option has been removed from 'configure'.
1938 Support for translating the documentation into HTML is now always
1939 provided. Use 'make html'.
1941 ** New module (ice-9 serialize):
1943 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
1944 don't trust the thread safety of most of your program, but where you
1945 have some section(s) of code which you consider can run in parallel to
1946 other sections. See ice-9/serialize.scm for more information.
1948 ** The configure option '--disable-arrays' has been removed.
1950 Support for arrays and uniform numeric arrays is now always included
1953 * Changes to the stand-alone interpreter
1955 ** New command line option `-L'.
1957 This option adds a directory to the front of the load path.
1959 ** New command line option `--no-debug'.
1961 Specifying `--no-debug' on the command line will keep the debugging
1962 evaluator turned off, even for interactive sessions.
1964 ** User-init file ~/.guile is now loaded with the debugging evaluator.
1966 Previously, the normal evaluator would have been used. Using the
1967 debugging evaluator gives better error messages.
1969 ** The '-e' option now 'read's its argument.
1971 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
1972 be used with '-e'. For example, you can now write a script like
1975 exec guile -e '(@ (demo) main)' -s "$0" "$@"
1978 (define-module (demo)
1982 (format #t "Demo: ~a~%" args))
1985 * Changes to Scheme functions and syntax
1987 ** Guardians have changed back to their original semantics
1989 Guardians now behave like described in the paper by Dybvig et al. In
1990 particular, they no longer make guarantees about the order in which
1991 they return objects, and they can no longer be greedy.
1993 They no longer drop cyclic data structures.
1995 The C function scm_make_guardian has been changed incompatibly and no
1996 longer takes the 'greedy_p' argument.
1998 ** New function hashx-remove!
2000 This function completes the set of 'hashx' functions.
2002 ** The concept of dynamic roots has been factored into continuation
2003 barriers and dynamic states.
2005 Each thread has a current dynamic state that carries the values of the
2006 fluids. You can create and copy dynamic states and use them as the
2007 second argument for 'eval'. See "Fluids and Dynamic States" in the
2010 To restrict the influence that captured continuations can have on the
2011 control flow, you can errect continuation barriers. See "Continuation
2012 Barriers" in the manual.
2014 The function call-with-dynamic-root now essentially temporarily
2015 installs a new dynamic state and errects a continuation barrier.
2017 ** The default load path no longer includes "." at the end.
2019 Automatically loading modules from the current directory should not
2020 happen by default. If you want to allow it in a more controlled
2021 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
2022 variable %load-path.
2024 ** The uniform vector and array support has been overhauled.
2026 It now complies with SRFI-4 and the weird prototype based uniform
2027 array creation has been deprecated. See the manual for more details.
2029 Some non-compatible changes have been made:
2030 - characters can no longer be stored into byte arrays.
2031 - strings and bit vectors are no longer considered to be uniform numeric
2033 - array-rank throws an error for non-arrays instead of returning zero.
2034 - array-ref does no longer accept non-arrays when no indices are given.
2036 There is the new notion of 'generalized vectors' and corresponding
2037 procedures like 'generalized-vector-ref'. Generalized vectors include
2038 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
2040 Arrays use generalized vectors as their storage, so that you still
2041 have arrays of characters, bits, etc. However, uniform-array-read!
2042 and uniform-array-write can no longer read/write strings and
2045 ** There is now support for copy-on-write substrings, mutation-sharing
2046 substrings and read-only strings.
2048 Three new procedures are related to this: substring/shared,
2049 substring/copy, and substring/read-only. See the manual for more
2052 ** Backtraces will now highlight the value that caused the error.
2054 By default, these values are enclosed in "{...}", such as in this
2063 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
2064 <unnamed port>:1:1: Wrong type (expecting pair): a
2065 ABORT: (wrong-type-arg)
2067 The prefix and suffix used for highlighting can be set via the two new
2068 printer options 'highlight-prefix' and 'highlight-suffix'. For
2069 example, putting this into ~/.guile will output the bad value in bold
2070 on an ANSI terminal:
2072 (print-set! highlight-prefix "\x1b[1m")
2073 (print-set! highlight-suffix "\x1b[22m")
2076 ** 'gettext' support for internationalization has been added.
2078 See the manual for details.
2080 ** New syntax '@' and '@@':
2082 You can now directly refer to variables exported from a module by
2085 (@ MODULE-NAME VARIABLE-NAME)
2087 For example (@ (ice-9 pretty-print) pretty-print) will directly access
2088 the pretty-print variable exported from the (ice-9 pretty-print)
2089 module. You don't need to 'use' that module first. You can also use
2090 '@' as a target of 'set!', as in (set! (@ mod var) val).
2092 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
2093 but it can also access variables that have not been exported. It is
2094 intended only for kluges and temporary fixes and for debugging, not
2097 ** Keyword syntax has been made more disciplined.
2099 Previously, the name of a keyword was read as a 'token' but printed as
2100 a symbol. Now, it is read as a general Scheme datum which must be a
2111 ERROR: In expression (a b c):
2115 ERROR: Unbound variable: foo
2120 ERROR: Wrong type (expecting symbol): 12
2124 ERROR: Wrong type (expecting symbol): (a b c)
2128 ** The printing of symbols that might look like keywords can be
2131 The new printer option 'quote-keywordish-symbols' controls how symbols
2132 are printed that have a colon as their first or last character. The
2133 default now is to only quote a symbol with #{...}# when the read
2134 option 'keywords' is not '#f'. Thus:
2136 guile> (define foo (string->symbol ":foo"))
2137 guile> (read-set! keywords #f)
2140 guile> (read-set! keywords 'prefix)
2143 guile> (print-set! quote-keywordish-symbols #f)
2147 ** 'while' now provides 'break' and 'continue'
2149 break and continue were previously bound in a while loop, but not
2150 documented, and continue didn't quite work properly. The undocumented
2151 parameter to break which gave a return value for the while has been
2154 ** 'call-with-current-continuation' is now also available under the name
2157 ** The module system now checks for duplicate bindings.
2159 The module system now can check for name conflicts among imported
2162 The behavior can be controlled by specifying one or more 'duplicates'
2163 handlers. For example, to make Guile return an error for every name
2166 (define-module (foo)
2171 The new default behavior of the module system when a name collision
2172 has been detected is to
2174 1. Give priority to bindings marked as a replacement.
2175 2. Issue a warning (different warning if overriding core binding).
2176 3. Give priority to the last encountered binding (this corresponds to
2179 If you want the old behavior back without replacements or warnings you
2182 (default-duplicate-binding-handler 'last)
2184 to your .guile init file.
2186 ** New define-module option: :replace
2188 :replace works as :export, but, in addition, marks the binding as a
2191 A typical example is `format' in (ice-9 format) which is a replacement
2192 for the core binding `format'.
2194 ** Adding prefixes to imported bindings in the module system
2196 There is now a new :use-module option :prefix. It can be used to add
2197 a prefix to all imported bindings.
2199 (define-module (foo)
2200 :use-module ((bar) :prefix bar:))
2202 will import all bindings exported from bar, but rename them by adding
2205 ** Conflicting generic functions can be automatically merged.
2207 When two imported bindings conflict and they are both generic
2208 functions, the two functions can now be merged automatically. This is
2209 activated with the 'duplicates' handler 'merge-generics'.
2211 ** New function: effective-version
2213 Returns the "effective" version number. This is just the normal full
2214 version string without the final micro-version number. See "Changes
2215 to the distribution" above.
2217 ** New threading functions: parallel, letpar, par-map, and friends
2219 These are convenient ways to run calculations in parallel in new
2220 threads. See "Parallel forms" in the manual for details.
2222 ** New function 'try-mutex'.
2224 This function will attempt to lock a mutex but will return immediately
2225 instead of blocking and indicate failure.
2227 ** Waiting on a condition variable can have a timeout.
2229 The function 'wait-condition-variable' now takes a third, optional
2230 argument that specifies the point in time where the waiting should be
2233 ** New function 'broadcast-condition-variable'.
2235 ** New functions 'all-threads' and 'current-thread'.
2237 ** Signals and system asyncs work better with threads.
2239 The function 'sigaction' now takes a fourth, optional, argument that
2240 specifies the thread that the handler should run in. When the
2241 argument is omitted, the handler will run in the thread that called
2244 Likewise, 'system-async-mark' takes a second, optional, argument that
2245 specifies the thread that the async should run in. When it is
2246 omitted, the async will run in the thread that called
2247 'system-async-mark'.
2249 C code can use the new functions scm_sigaction_for_thread and
2250 scm_system_async_mark_for_thread to pass the new thread argument.
2252 When a thread blocks on a mutex, a condition variable or is waiting
2253 for IO to be possible, it will still execute system asyncs. This can
2254 be used to interrupt such a thread by making it execute a 'throw', for
2257 ** The function 'system-async' is deprecated.
2259 You can now pass any zero-argument procedure to 'system-async-mark'.
2260 The function 'system-async' will just return its argument unchanged
2263 ** New functions 'call-with-blocked-asyncs' and
2264 'call-with-unblocked-asyncs'
2266 The expression (call-with-blocked-asyncs PROC) will call PROC and will
2267 block execution of system asyncs for the current thread by one level
2268 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
2269 procedure and will unblock the execution of system asyncs by one
2270 level for the current thread.
2272 Only system asyncs are affected by these functions.
2274 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
2276 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
2277 instead. Those functions are easier to use correctly and can be
2280 ** New function 'unsetenv'.
2282 ** New macro 'define-syntax-public'.
2284 It works like 'define-syntax' and also exports the defined macro (but
2287 ** There is support for Infinity and NaNs.
2289 Following PLT Scheme, Guile can now work with infinite numbers, and
2292 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
2293 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
2294 "+nan.0"). These numbers are inexact and have no exact counterpart.
2296 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
2297 sign of the dividend. The infinities are integers, and they answer #t
2298 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
2299 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
2310 ERROR: Numerical overflow
2312 Two new predicates 'inf?' and 'nan?' can be used to test for the
2315 ** Inexact zero can have a sign.
2317 Guile can now distinguish between plus and minus inexact zero, if your
2318 platform supports this, too. The two zeros are equal according to
2319 '=', but not according to 'eqv?'. For example
2330 ** Guile now has exact rationals.
2332 Guile can now represent fractions such as 1/3 exactly. Computing with
2333 them is also done exactly, of course:
2338 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
2339 for exact arguments.
2341 For example: (floor 2) now returns an exact 2 where in the past it
2342 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
2344 ** inexact->exact no longer returns only integers.
2346 Without exact rationals, the closest exact number was always an
2347 integer, but now inexact->exact returns the fraction that is exactly
2348 equal to a floating point number. For example:
2350 (inexact->exact 1.234)
2351 => 694680242521899/562949953421312
2353 When you want the old behavior, use 'round' explicitly:
2355 (inexact->exact (round 1.234))
2358 ** New function 'rationalize'.
2360 This function finds a simple fraction that is close to a given real
2361 number. For example (and compare with inexact->exact above):
2363 (rationalize (inexact->exact 1.234) 1/2000)
2366 Note that, as required by R5RS, rationalize returns only then an exact
2367 result when both its arguments are exact.
2369 ** 'odd?' and 'even?' work also for inexact integers.
2371 Previously, (odd? 1.0) would signal an error since only exact integers
2372 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
2373 returns #f and (odd? 1.5) signals an error.
2375 ** Guile now has uninterned symbols.
2377 The new function 'make-symbol' will return an uninterned symbol. This
2378 is a symbol that is unique and is guaranteed to remain unique.
2379 However, uninterned symbols can not yet be read back in.
2381 Use the new function 'symbol-interned?' to check whether a symbol is
2384 ** pretty-print has more options.
2386 The function pretty-print from the (ice-9 pretty-print) module can now
2387 also be invoked with keyword arguments that control things like
2388 maximum output width. See the manual for details.
2390 ** Variables have no longer a special behavior for `equal?'.
2392 Previously, comparing two variables with `equal?' would recursivly
2393 compare their values. This is no longer done. Variables are now only
2394 `equal?' if they are `eq?'.
2396 ** `(begin)' is now valid.
2398 You can now use an empty `begin' form. It will yield #<unspecified>
2399 when evaluated and simply be ignored in a definition context.
2401 ** Deprecated: procedure->macro
2403 Change your code to use 'define-macro' or r5rs macros. Also, be aware
2404 that macro expansion will not be done during evaluation, but prior to
2407 ** Soft ports now allow a `char-ready?' procedure
2409 The vector argument to `make-soft-port' can now have a length of
2410 either 5 or 6. (Previously the length had to be 5.) The optional 6th
2411 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
2412 that returns the number of characters that can be read immediately
2413 without the soft port blocking.
2415 ** Deprecated: undefine
2417 There is no replacement for undefine.
2419 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
2420 have been discouraged.
2422 They are relics from a time where a keyword like #:foo was used
2423 directly as a Tcl option "-foo" and thus keywords were internally
2424 stored as a symbol with a starting dash. We now store a symbol
2427 Use symbol->keyword and keyword->symbol instead.
2429 ** The `cheap' debug option is now obsolete
2431 Evaluator trap calls are now unconditionally "cheap" - in other words,
2432 they pass a debug object to the trap handler rather than a full
2433 continuation. The trap handler code can capture a full continuation
2434 by using `call-with-current-continuation' in the usual way, if it so
2437 The `cheap' option is retained for now so as not to break existing
2438 code which gets or sets it, but setting it now has no effect. It will
2439 be removed in the next major Guile release.
2441 ** Evaluator trap calls now support `tweaking'
2443 `Tweaking' means that the trap handler code can modify the Scheme
2444 expression that is about to be evaluated (in the case of an
2445 enter-frame trap) or the value that is being returned (in the case of
2446 an exit-frame trap). The trap handler code indicates that it wants to
2447 do this by returning a pair whose car is the symbol 'instead and whose
2448 cdr is the modified expression or return value.
2450 * Changes to the C interface
2452 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
2453 take a 'delete' function argument.
2455 This argument makes no sense since the delete function is used to
2456 remove a pair from an alist, and this must not be configurable.
2458 This is an incompatible change.
2460 ** The GH interface is now subject to the deprecation mechanism
2462 The GH interface has been deprecated for quite some time but now it is
2463 actually removed from Guile when it is configured with
2464 --disable-deprecated.
2466 See the manual "Transitioning away from GH" for more information.
2468 ** A new family of functions for converting between C values and
2469 Scheme values has been added.
2471 These functions follow a common naming scheme and are designed to be
2472 easier to use, thread-safe and more future-proof than the older
2475 - int scm_is_* (...)
2477 These are predicates that return a C boolean: 1 or 0. Instead of
2478 SCM_NFALSEP, you can now use scm_is_true, for example.
2480 - <type> scm_to_<type> (SCM val, ...)
2482 These are functions that convert a Scheme value into an appropriate
2483 C value. For example, you can use scm_to_int to safely convert from
2486 - SCM scm_from_<type> (<type> val, ...)
2488 These functions convert from a C type to a SCM value; for example,
2489 scm_from_int for ints.
2491 There is a huge number of these functions, for numbers, strings,
2492 symbols, vectors, etc. They are documented in the reference manual in
2493 the API section together with the types that they apply to.
2495 ** New functions for dealing with complex numbers in C have been added.
2497 The new functions are scm_c_make_rectangular, scm_c_make_polar,
2498 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
2499 They work like scm_make_rectangular etc but take or return doubles
2502 ** The function scm_make_complex has been discouraged.
2504 Use scm_c_make_rectangular instead.
2506 ** The INUM macros have been deprecated.
2508 A lot of code uses these macros to do general integer conversions,
2509 although the macros only work correctly with fixnums. Use the
2510 following alternatives.
2512 SCM_INUMP -> scm_is_integer or similar
2513 SCM_NINUMP -> !scm_is_integer or similar
2514 SCM_MAKINUM -> scm_from_int or similar
2515 SCM_INUM -> scm_to_int or similar
2517 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
2518 do the validating for you.
2520 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
2521 have been discouraged.
2523 Use the newer scm_to_<type> and scm_from_<type> functions instead for
2524 new code. The functions have been discouraged since they don't fit
2527 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
2529 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
2530 evaluates its argument twice. Use scm_is_true, etc. instead for new
2533 ** The macro SCM_EQ_P has been discouraged.
2535 Use scm_is_eq for new code, which fits better into the naming
2538 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
2541 Use the function scm_is_pair or scm_is_null instead.
2543 ** The functions scm_round and scm_truncate have been deprecated and
2544 are now available as scm_c_round and scm_c_truncate, respectively.
2546 These functions occupy the names that scm_round_number and
2547 scm_truncate_number should have.
2549 ** The functions scm_c_string2str, scm_c_substring2str, and
2550 scm_c_symbol2str have been deprecated.
2552 Use scm_to_locale_stringbuf or similar instead, maybe together with
2555 ** New functions scm_c_make_string, scm_c_string_length,
2556 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
2557 scm_c_substring_shared, scm_c_substring_copy.
2559 These are like scm_make_string, scm_length, etc. but are slightly
2560 easier to use from C.
2562 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
2563 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
2565 They export too many assumptions about the implementation of strings
2566 and symbols that are no longer true in the presence of
2567 mutation-sharing substrings and when Guile switches to some form of
2570 When working with strings, it is often best to use the normal string
2571 functions provided by Guile, such as scm_c_string_ref,
2572 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
2573 manual since many more such functions are now provided than
2576 When you want to convert a SCM string to a C string, use the
2577 scm_to_locale_string function or similar instead. For symbols, use
2578 scm_symbol_to_string and then work with that string. Because of the
2579 new string representation, scm_symbol_to_string does not need to copy
2580 and is thus quite efficient.
2582 ** Some string, symbol and keyword functions have been discouraged.
2584 They don't fit into the uniform naming scheme and are not explicit
2585 about the character encoding.
2587 Replace according to the following table:
2589 scm_allocate_string -> scm_c_make_string
2590 scm_take_str -> scm_take_locale_stringn
2591 scm_take0str -> scm_take_locale_string
2592 scm_mem2string -> scm_from_locale_stringn
2593 scm_str2string -> scm_from_locale_string
2594 scm_makfrom0str -> scm_from_locale_string
2595 scm_mem2symbol -> scm_from_locale_symboln
2596 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
2597 scm_str2symbol -> scm_from_locale_symbol
2599 SCM_SYMBOL_HASH -> scm_hashq
2600 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
2602 scm_c_make_keyword -> scm_from_locale_keyword
2604 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
2605 now also available to C code.
2607 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
2609 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
2610 the latter returns the true name of the keyword, not the 'dash name',
2611 as SCM_KEYWORDSYM used to do.
2613 ** A new way to access arrays in a thread-safe and efficient way has
2616 See the manual, node "Accessing Arrays From C".
2618 ** The old uniform vector and bitvector implementations have been
2619 unceremoniously removed.
2621 This implementation exposed the details of the tagging system of
2622 Guile. Use the new C API explained in the manual in node "Uniform
2623 Numeric Vectors" and "Bit Vectors", respectively.
2625 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
2626 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
2627 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
2628 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
2629 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
2630 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
2633 ** The macros dealing with vectors have been deprecated.
2635 Use the new functions scm_is_vector, scm_vector_elements,
2636 scm_vector_writable_elements, etc, or scm_is_simple_vector,
2637 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
2638 manual for more details.
2640 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
2641 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
2643 The following macros have been removed: SCM_VECTOR_BASE,
2644 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
2645 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
2647 ** Some C functions and macros related to arrays have been deprecated.
2649 Migrate according to the following table:
2651 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
2652 scm_make_ra -> scm_make_array
2653 scm_shap2ra -> scm_make_array
2654 scm_cvref -> scm_c_generalized_vector_ref
2655 scm_ra_set_contp -> do not use
2656 scm_aind -> scm_array_handle_pos
2657 scm_raprin1 -> scm_display or scm_write
2659 SCM_ARRAYP -> scm_is_array
2660 SCM_ARRAY_NDIM -> scm_c_array_rank
2661 SCM_ARRAY_DIMS -> scm_array_handle_dims
2662 SCM_ARRAY_CONTP -> do not use
2663 SCM_ARRAY_MEM -> do not use
2664 SCM_ARRAY_V -> scm_array_handle_elements or similar
2665 SCM_ARRAY_BASE -> do not use
2667 ** SCM_CELL_WORD_LOC has been deprecated.
2669 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
2670 to a SCM, as opposed to a pointer to a scm_t_bits.
2672 This was done to allow the correct use of pointers into the Scheme
2673 heap. Previously, the heap words were of type scm_t_bits and local
2674 variables and function arguments were of type SCM, making it
2675 non-standards-conformant to have a pointer that can point to both.
2677 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
2679 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
2680 second and third words of double smobs. Likewise for
2681 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
2683 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
2684 used to get and set the 16 exra bits in the zeroth word of a smob.
2686 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
2687 accesing the first immediate word of a smob as a SCM value, and there
2688 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
2689 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
2691 ** New way to deal with non-local exits and re-entries.
2693 There is a new set of functions that essentially do what
2694 scm_internal_dynamic_wind does, but in a way that is more convenient
2695 for C code in some situations. Here is a quick example of how to
2696 prevent a potential memory leak:
2703 scm_dynwind_begin (0);
2705 mem = scm_malloc (100);
2706 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
2708 /* MEM would leak if BAR throws an error.
2709 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
2716 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
2717 SCM_DYNWIND_END as well.
2721 For full documentation, see the node "Dynamic Wind" in the manual.
2723 ** New function scm_dynwind_free
2725 This function calls 'free' on a given pointer when a dynwind context
2726 is left. Thus the call to scm_dynwind_unwind_handler above could be
2727 replaced with simply scm_dynwind_free (mem).
2729 ** New functions scm_c_call_with_blocked_asyncs and
2730 scm_c_call_with_unblocked_asyncs
2732 Like scm_call_with_blocked_asyncs etc. but for C functions.
2734 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
2736 In addition to scm_c_call_with_blocked_asyncs you can now also use
2737 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
2738 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
2740 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
2741 SCM_REALLOW_INTS have been deprecated.
2743 They do no longer fulfill their original role of blocking signal
2744 delivery. Depending on what you want to achieve, replace a pair of
2745 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
2746 mutex, blocks asyncs, or both. See node "Critical Sections" in the
2749 ** The value 'scm_mask_ints' is no longer writable.
2751 Previously, you could set scm_mask_ints directly. This is no longer
2752 possible. Use scm_c_call_with_blocked_asyncs and
2753 scm_c_call_with_unblocked_asyncs instead.
2755 ** New way to temporarily set the current input, output or error ports
2757 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
2758 context' (see above). <foo> is one of "input", "output" or "error".
2760 ** New way to temporarily set fluids
2762 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
2763 above) to temporarily set the value of a fluid.
2765 ** New types scm_t_intmax and scm_t_uintmax.
2767 On platforms that have them, these types are identical to intmax_t and
2768 uintmax_t, respectively. On other platforms, they are identical to
2769 the largest integer types that Guile knows about.
2771 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
2773 You should not have used them.
2775 ** Many public #defines with generic names have been made private.
2777 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
2778 private or renamed with a more suitable public name.
2780 ** The macro SCM_TYP16S has been deprecated.
2782 This macro is not intended for public use.
2784 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
2786 Use scm_is_true (scm_inexact_p (...)) instead.
2788 ** The macro SCM_SLOPPY_REALP has been deprecated.
2790 Use scm_is_real instead.
2792 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
2794 Use scm_is_complex instead.
2796 ** Some preprocessor defines have been deprecated.
2798 These defines indicated whether a certain feature was present in Guile
2799 or not. Going forward, assume that the features are always present.
2801 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
2802 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
2804 The following macros have been removed completely: MEMOIZE_LOCALS,
2805 SCM_RECKLESS, SCM_CAUTIOUS.
2807 ** The preprocessor define STACK_DIRECTION has been deprecated.
2809 There should be no need to know about the stack direction for ordinary
2812 ** New function: scm_effective_version
2814 Returns the "effective" version number. This is just the normal full
2815 version string without the final micro-version number. See "Changes
2816 to the distribution" above.
2818 ** The function scm_call_with_new_thread has a new prototype.
2820 Instead of taking a list with the thunk and handler, these two
2821 arguments are now passed directly:
2823 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
2825 This is an incompatible change.
2827 ** New snarfer macro SCM_DEFINE_PUBLIC.
2829 This is like SCM_DEFINE, but also calls scm_c_export for the defined
2830 function in the init section.
2832 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
2834 ** Garbage collector rewrite.
2836 The garbage collector is cleaned up a lot, and now uses lazy
2837 sweeping. This is reflected in the output of (gc-stats); since cells
2838 are being freed when they are allocated, the cells-allocated field
2839 stays roughly constant.
2841 For malloc related triggers, the behavior is changed. It uses the same
2842 heuristic as the cell-triggered collections. It may be tuned with the
2843 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
2844 for minimum yield of malloc related triggers. The default is 40.
2845 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
2848 Debugging operations for the freelist have been deprecated, along with
2849 the C variables that control garbage collection. The environment
2850 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
2851 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
2853 For understanding the memory usage of a GUILE program, the routine
2854 gc-live-object-stats returns an alist containing the number of live
2855 objects for every type.
2858 ** The function scm_definedp has been renamed to scm_defined_p
2860 The name scm_definedp is deprecated.
2862 ** The struct scm_cell type has been renamed to scm_t_cell
2864 This is in accordance to Guile's naming scheme for types. Note that
2865 the name scm_cell is now used for a function that allocates and
2866 initializes a new cell (see below).
2868 ** New functions for memory management
2870 A new set of functions for memory management has been added since the
2871 old way (scm_must_malloc, scm_must_free, etc) was error prone and
2872 indeed, Guile itself contained some long standing bugs that could
2873 cause aborts in long running programs.
2875 The new functions are more symmetrical and do not need cooperation
2876 from smob free routines, among other improvements.
2878 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
2879 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
2880 scm_gc_free, scm_gc_register_collectable_memory, and
2881 scm_gc_unregister_collectable_memory. Refer to the manual for more
2882 details and for upgrading instructions.
2884 The old functions for memory management have been deprecated. They
2885 are: scm_must_malloc, scm_must_realloc, scm_must_free,
2886 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
2888 ** Declarations of exported features are marked with SCM_API.
2890 Every declaration of a feature that belongs to the exported Guile API
2891 has been marked by adding the macro "SCM_API" to the start of the
2892 declaration. This macro can expand into different things, the most
2893 common of which is just "extern" for Unix platforms. On Win32, it can
2894 be used to control which symbols are exported from a DLL.
2896 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
2897 will expand into "__declspec (dllimport) extern", which is needed for
2898 linking to the Guile DLL in Windows.
2900 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
2901 SCM_SRFI4_IMPORT, for the corresponding libraries.
2903 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
2905 Use the new functions scm_cell and scm_double_cell instead. The old
2906 macros had problems because with them allocation and initialization
2907 was separated and the GC could sometimes observe half initialized
2908 cells. Only careful coding by the user of SCM_NEWCELL and
2909 SCM_NEWCELL2 could make this safe and efficient.
2911 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
2913 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
2916 ** SRCBRKP has been deprecated.
2918 Use scm_c_source_property_breakpoint_p instead.
2920 ** Deprecated: scm_makmacro
2922 Change your code to use either scm_makmmacro or to define macros in
2923 Scheme, using 'define-macro'.
2925 ** New function scm_c_port_for_each.
2927 This function is like scm_port_for_each but takes a pointer to a C
2928 function as the callback instead of a SCM value.
2930 ** The names scm_internal_select, scm_thread_sleep, and
2931 scm_thread_usleep have been discouraged.
2933 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
2935 ** The GC can no longer be blocked.
2937 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
2938 The GC can now run (partially) concurrently with other code and thus
2939 blocking it is not well defined.
2941 ** Many definitions have been removed that were previously deprecated.
2943 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
2944 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
2945 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
2946 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
2947 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
2948 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
2949 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
2950 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
2951 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
2952 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
2953 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
2954 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
2955 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
2956 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
2957 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
2958 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
2959 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
2960 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
2961 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
2962 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
2963 scm_vector_set_length_x, scm_contregs, scm_debug_info,
2964 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
2965 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
2966 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
2967 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
2968 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
2969 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
2970 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
2971 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
2972 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
2973 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
2975 * Changes to bundled modules
2979 Using the (ice-9 debug) module no longer automatically switches Guile
2980 to use the debugging evaluator. If you want to switch to the
2981 debugging evaluator (which is needed for backtrace information if you
2982 hit an error), please add an explicit "(debug-enable 'debug)" to your
2983 code just after the code to use (ice-9 debug).
2986 Changes since Guile 1.4:
2988 * Changes to the distribution
2990 ** A top-level TODO file is included.
2992 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
2994 Guile now always uses three numbers to represent the version,
2995 i.e. "1.6.5". The first number, 1, is the major version number, the
2996 second number, 6, is the minor version number, and the third number,
2997 5, is the micro version number. Changes in major version number
2998 indicate major changes in Guile.
3000 Minor version numbers that are even denote stable releases, and odd
3001 minor version numbers denote development versions (which may be
3002 unstable). The micro version number indicates a minor sub-revision of
3003 a given MAJOR.MINOR release.
3005 In keeping with the new scheme, (minor-version) and scm_minor_version
3006 no longer return everything but the major version number. They now
3007 just return the minor version number. Two new functions
3008 (micro-version) and scm_micro_version have been added to report the
3009 micro version number.
3011 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
3013 ** New preprocessor definitions are available for checking versions.
3015 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
3016 SCM_MICRO_VERSION to the appropriate integer values.
3018 ** Guile now actively warns about deprecated features.
3020 The new configure option `--enable-deprecated=LEVEL' and the
3021 environment variable GUILE_WARN_DEPRECATED control this mechanism.
3022 See INSTALL and README for more information.
3024 ** Guile is much more likely to work on 64-bit architectures.
3026 Guile now compiles and passes "make check" with only two UNRESOLVED GC
3027 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
3028 for the use of a test machine, and thanks to Stefan Jahn for ia64
3031 ** New functions: setitimer and getitimer.
3033 These implement a fairly direct interface to the libc functions of the
3036 ** The #. reader extension is now disabled by default.
3038 For safety reasons, #. evaluation is disabled by default. To
3039 re-enable it, set the fluid read-eval? to #t. For example:
3041 (fluid-set! read-eval? #t)
3043 but make sure you realize the potential security risks involved. With
3044 read-eval? enabled, reading a data file from an untrusted source can
3047 ** New SRFI modules have been added:
3049 SRFI-0 `cond-expand' is now supported in Guile, without requiring
3052 (srfi srfi-1) is a library containing many useful pair- and list-processing
3055 (srfi srfi-2) exports and-let*.
3057 (srfi srfi-4) implements homogeneous numeric vector datatypes.
3059 (srfi srfi-6) is a dummy module for now, since guile already provides
3060 all of the srfi-6 procedures by default: open-input-string,
3061 open-output-string, get-output-string.
3063 (srfi srfi-8) exports receive.
3065 (srfi srfi-9) exports define-record-type.
3067 (srfi srfi-10) exports define-reader-ctor and implements the reader
3070 (srfi srfi-11) exports let-values and let*-values.
3072 (srfi srfi-13) implements the SRFI String Library.
3074 (srfi srfi-14) implements the SRFI Character-Set Library.
3076 (srfi srfi-17) implements setter and getter-with-setter and redefines
3077 some accessor procedures as procedures with getters. (such as car,
3078 cdr, vector-ref etc.)
3080 (srfi srfi-19) implements the SRFI Time/Date Library.
3082 ** New scripts / "executable modules"
3084 Subdirectory "scripts" contains Scheme modules that are packaged to
3085 also be executable as scripts. At this time, these scripts are available:
3094 See README there for more info.
3096 These scripts can be invoked from the shell with the new program
3097 "guile-tools", which keeps track of installation directory for you.
3100 $ guile-tools display-commentary srfi/*.scm
3102 guile-tools is copied to the standard $bindir on "make install".
3104 ** New module (ice-9 stack-catch):
3106 stack-catch is like catch, but saves the current state of the stack in
3107 the fluid the-last-stack. This fluid can be useful when using the
3108 debugger and when re-throwing an error.
3110 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
3112 This has been done to prevent problems on lesser operating systems
3113 that can't tolerate `*'s in file names. The exported macro continues
3114 to be named `and-let*', of course.
3116 On systems that support it, there is also a compatibility module named
3117 (ice-9 and-let*). It will go away in the next release.
3119 ** New modules (oop goops) etc.:
3122 (oop goops describe)
3124 (oop goops active-slot)
3125 (oop goops composite-slot)
3127 The Guile Object Oriented Programming System (GOOPS) has been
3128 integrated into Guile. For further information, consult the GOOPS
3129 manual and tutorial in the `doc' directory.
3131 ** New module (ice-9 rdelim).
3133 This exports the following procedures which were previously defined
3134 in the default environment:
3136 read-line read-line! read-delimited read-delimited! %read-delimited!
3137 %read-line write-line
3139 For backwards compatibility the definitions are still imported into the
3140 default environment in this version of Guile. However you should add:
3142 (use-modules (ice-9 rdelim))
3144 to any program which uses the definitions, since this may change in
3147 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
3148 can be used for similar functionality.
3150 ** New module (ice-9 rw)
3152 This is a subset of the (scsh rw) module from guile-scsh. Currently
3153 it defines two procedures:
3155 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
3157 Read characters from a port or file descriptor into a string STR.
3158 A port must have an underlying file descriptor -- a so-called
3159 fport. This procedure is scsh-compatible and can efficiently read
3162 *** New function: write-string/partial str [port_or_fdes [start [end]]]
3164 Write characters from a string STR to a port or file descriptor.
3165 A port must have an underlying file descriptor -- a so-called
3166 fport. This procedure is mostly compatible and can efficiently
3167 write large strings.
3169 ** New module (ice-9 match)
3171 This module includes Andrew K. Wright's pattern matcher. See
3172 ice-9/match.scm for brief description or
3174 http://www.star-lab.com/wright/code.html
3176 for complete documentation.
3178 ** New module (ice-9 buffered-input)
3180 This module provides procedures to construct an input port from an
3181 underlying source of input that reads and returns its input in chunks.
3182 The underlying input source is a Scheme procedure, specified by the
3183 caller, which the port invokes whenever it needs more input.
3185 This is useful when building an input port whose back end is Readline
3186 or a UI element such as the GtkEntry widget.
3190 The reference and tutorial documentation that was previously
3191 distributed separately, as `guile-doc', is now included in the core
3192 Guile distribution. The documentation consists of the following
3195 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
3198 - The Guile Reference Manual (guile.texi) contains (or is intended to
3199 contain) reference documentation on all aspects of Guile.
3201 - The GOOPS Manual (goops.texi) contains both tutorial-style and
3202 reference documentation for using GOOPS, Guile's Object Oriented
3205 - The Revised^5 Report on the Algorithmic Language Scheme
3208 See the README file in the `doc' directory for more details.
3210 ** There are a couple of examples in the examples/ directory now.
3212 * Changes to the stand-alone interpreter
3214 ** New command line option `--use-srfi'
3216 Using this option, SRFI modules can be loaded on startup and be
3217 available right from the beginning. This makes programming portable
3218 Scheme programs easier.
3220 The option `--use-srfi' expects a comma-separated list of numbers,
3221 each representing a SRFI number to be loaded into the interpreter
3222 before starting evaluating a script file or the REPL. Additionally,
3223 the feature identifier for the loaded SRFIs is recognized by
3224 `cond-expand' when using this option.
3227 $ guile --use-srfi=8,13
3228 guile> (receive (x z) (values 1 2) (+ 1 2))
3230 guile> (string-pad "bla" 20)
3233 ** Guile now always starts up in the `(guile-user)' module.
3235 Previously, scripts executed via the `-s' option would run in the
3236 `(guile)' module and the repl would run in the `(guile-user)' module.
3237 Now every user action takes place in the `(guile-user)' module by
3240 * Changes to Scheme functions and syntax
3242 ** Character classifiers work for non-ASCII characters.
3244 The predicates `char-alphabetic?', `char-numeric?',
3245 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
3246 no longer check whether their arguments are ASCII characters.
3247 Previously, a character would only be considered alphabetic when it
3248 was also ASCII, for example.
3250 ** Previously deprecated Scheme functions have been removed:
3252 tag - no replacement.
3253 fseek - replaced by seek.
3254 list* - replaced by cons*.
3256 ** It's now possible to create modules with controlled environments
3260 (use-modules (ice-9 safe))
3261 (define m (make-safe-module))
3262 ;;; m will now be a module containing only a safe subset of R5RS
3263 (eval '(+ 1 2) m) --> 3
3264 (eval 'load m) --> ERROR: Unbound variable: load
3266 ** Evaluation of "()", the empty list, is now an error.
3268 Previously, the expression "()" evaluated to the empty list. This has
3269 been changed to signal a "missing expression" error. The correct way
3270 to write the empty list as a literal constant is to use quote: "'()".
3272 ** New concept of `Guile Extensions'.
3274 A Guile Extension is just a ordinary shared library that can be linked
3275 at run-time. We found it advantageous to give this simple concept a
3276 dedicated name to distinguish the issues related to shared libraries
3277 from the issues related to the module system.
3279 *** New function: load-extension
3281 Executing (load-extension lib init) is mostly equivalent to
3283 (dynamic-call init (dynamic-link lib))
3285 except when scm_register_extension has been called previously.
3286 Whenever appropriate, you should use `load-extension' instead of
3287 dynamic-link and dynamic-call.
3289 *** New C function: scm_c_register_extension
3291 This function registers a initialization function for use by
3292 `load-extension'. Use it when you don't want specific extensions to
3293 be loaded as shared libraries (for example on platforms that don't
3294 support dynamic linking).
3296 ** Auto-loading of compiled-code modules is deprecated.
3298 Guile used to be able to automatically find and link a shared
3299 library to satisfy requests for a module. For example, the module
3300 `(foo bar)' could be implemented by placing a shared library named
3301 "foo/libbar.so" (or with a different extension) in a directory on the
3304 This has been found to be too tricky, and is no longer supported. The
3305 shared libraries are now called "extensions". You should now write a
3306 small Scheme file that calls `load-extension' to load the shared
3307 library and initialize it explicitly.
3309 The shared libraries themselves should be installed in the usual
3310 places for shared libraries, with names like "libguile-foo-bar".
3312 For example, place this into a file "foo/bar.scm"
3314 (define-module (foo bar))
3316 (load-extension "libguile-foo-bar" "foobar_init")
3318 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
3320 `eval' is now R5RS, that is it takes two arguments.
3321 The second argument is an environment specifier, i.e. either
3323 (scheme-report-environment 5)
3324 (null-environment 5)
3325 (interaction-environment)
3331 ** The module system has been made more disciplined.
3333 The function `eval' will save and restore the current module around
3334 the evaluation of the specified expression. While this expression is
3335 evaluated, `(current-module)' will now return the right module, which
3336 is the module specified as the second argument to `eval'.
3338 A consequence of this change is that `eval' is not particularly
3339 useful when you want allow the evaluated code to change what module is
3340 designated as the current module and have this change persist from one
3341 call to `eval' to the next. The read-eval-print-loop is an example
3342 where `eval' is now inadequate. To compensate, there is a new
3343 function `primitive-eval' that does not take a module specifier and
3344 that does not save/restore the current module. You should use this
3345 function together with `set-current-module', `current-module', etc
3346 when you want to have more control over the state that is carried from
3347 one eval to the next.
3349 Additionally, it has been made sure that forms that are evaluated at
3350 the top level are always evaluated with respect to the current module.
3351 Previously, subforms of top-level forms such as `begin', `case',
3352 etc. did not respect changes to the current module although these
3353 subforms are at the top-level as well.
3355 To prevent strange behavior, the forms `define-module',
3356 `use-modules', `use-syntax', and `export' have been restricted to only
3357 work on the top level. The forms `define-public' and
3358 `defmacro-public' only export the new binding on the top level. They
3359 behave just like `define' and `defmacro', respectively, when they are
3360 used in a lexical environment.
3362 Also, `export' will no longer silently re-export bindings imported
3363 from a used module. It will emit a `deprecation' warning and will
3364 cease to perform any re-export in the next version. If you actually
3365 want to re-export bindings, use the new `re-export' in place of
3366 `export'. The new `re-export' will not make copies of variables when
3367 rexporting them, as `export' did wrongly.
3369 ** Module system now allows selection and renaming of imported bindings
3371 Previously, when using `use-modules' or the `#:use-module' clause in
3372 the `define-module' form, all the bindings (association of symbols to
3373 values) for imported modules were added to the "current module" on an
3374 as-is basis. This has been changed to allow finer control through two
3375 new facilities: selection and renaming.
3377 You can now select which of the imported module's bindings are to be
3378 visible in the current module by using the `:select' clause. This
3379 clause also can be used to rename individual bindings. For example:
3381 ;; import all bindings no questions asked
3382 (use-modules (ice-9 common-list))
3384 ;; import four bindings, renaming two of them;
3385 ;; the current module sees: every some zonk-y zonk-n
3386 (use-modules ((ice-9 common-list)
3388 (remove-if . zonk-y)
3389 (remove-if-not . zonk-n))))
3391 You can also programmatically rename all selected bindings using the
3392 `:renamer' clause, which specifies a proc that takes a symbol and
3393 returns another symbol. Because it is common practice to use a prefix,
3394 we now provide the convenience procedure `symbol-prefix-proc'. For
3397 ;; import four bindings, renaming two of them specifically,
3398 ;; and all four w/ prefix "CL:";
3399 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
3400 (use-modules ((ice-9 common-list)
3402 (remove-if . zonk-y)
3403 (remove-if-not . zonk-n))
3404 :renamer (symbol-prefix-proc 'CL:)))
3406 ;; import four bindings, renaming two of them specifically,
3407 ;; and all four by upcasing.
3408 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
3409 (define (upcase-symbol sym)
3410 (string->symbol (string-upcase (symbol->string sym))))
3412 (use-modules ((ice-9 common-list)
3414 (remove-if . zonk-y)
3415 (remove-if-not . zonk-n))
3416 :renamer upcase-symbol))
3418 Note that programmatic renaming is done *after* individual renaming.
3419 Also, the above examples show `use-modules', but the same facilities are
3420 available for the `#:use-module' clause of `define-module'.
3422 See manual for more info.
3424 ** The semantics of guardians have changed.
3426 The changes are for the most part compatible. An important criterion
3427 was to keep the typical usage of guardians as simple as before, but to
3428 make the semantics safer and (as a result) more useful.
3430 *** All objects returned from guardians are now properly alive.
3432 It is now guaranteed that any object referenced by an object returned
3433 from a guardian is alive. It's now impossible for a guardian to
3434 return a "contained" object before its "containing" object.
3436 One incompatible (but probably not very important) change resulting
3437 from this is that it is no longer possible to guard objects that
3438 indirectly reference themselves (i.e. are parts of cycles). If you do
3439 so accidentally, you'll get a warning.
3441 *** There are now two types of guardians: greedy and sharing.
3443 If you call (make-guardian #t) or just (make-guardian), you'll get a
3444 greedy guardian, and for (make-guardian #f) a sharing guardian.
3446 Greedy guardians are the default because they are more "defensive".
3447 You can only greedily guard an object once. If you guard an object
3448 more than once, once in a greedy guardian and the rest of times in
3449 sharing guardians, then it is guaranteed that the object won't be
3450 returned from sharing guardians as long as it is greedily guarded
3453 Guardians returned by calls to `make-guardian' can now take one more
3454 optional parameter, which says whether to throw an error in case an
3455 attempt is made to greedily guard an object that is already greedily
3456 guarded. The default is true, i.e. throw an error. If the parameter
3457 is false, the guardian invocation returns #t if guarding was
3458 successful and #f if it wasn't.
3460 Also, since greedy guarding is, in effect, a side-effecting operation
3461 on objects, a new function is introduced: `destroy-guardian!'.
3462 Invoking this function on a guardian renders it unoperative and, if
3463 the guardian is greedy, clears the "greedily guarded" property of the
3464 objects that were guarded by it, thus undoing the side effect.
3466 Note that all this hair is hardly very important, since guardian
3467 objects are usually permanent.
3469 ** Continuations created by call-with-current-continuation now accept
3470 any number of arguments, as required by R5RS.
3472 ** New function `issue-deprecation-warning'
3474 This function is used to display the deprecation messages that are
3475 controlled by GUILE_WARN_DEPRECATION as explained in the README.
3478 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
3482 ;; `id' is deprecated. Use `identity' instead.
3487 ** New syntax `begin-deprecated'
3489 When deprecated features are included (as determined by the configure
3490 option --enable-deprecated), `begin-deprecated' is identical to
3491 `begin'. When deprecated features are excluded, it always evaluates
3492 to `#f', ignoring the body forms.
3494 ** New function `make-object-property'
3496 This function returns a new `procedure with setter' P that can be used
3497 to attach a property to objects. When calling P as
3501 where `obj' is any kind of object, it attaches `val' to `obj' in such
3502 a way that it can be retrieved by calling P as
3506 This function will replace procedure properties, symbol properties and
3507 source properties eventually.
3509 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
3511 Instead of #&optional, #&key, etc you should now use #:optional,
3512 #:key, etc. Since #:optional is a keyword, you can write it as just
3513 :optional when (read-set! keywords 'prefix) is active.
3515 The old reader syntax `#&' is still supported, but deprecated. It
3516 will be removed in the next release.
3518 ** New define-module option: pure
3520 Tells the module system not to include any bindings from the root
3525 (define-module (totally-empty-module)
3528 ** New define-module option: export NAME1 ...
3530 Export names NAME1 ...
3532 This option is required if you want to be able to export bindings from
3533 a module which doesn't import one of `define-public' or `export'.
3537 (define-module (foo)
3539 :use-module (ice-9 r5rs)
3542 ;;; Note that we're pure R5RS below this point!
3547 ** New function: object->string OBJ
3549 Return a Scheme string obtained by printing a given object.
3551 ** New function: port? X
3553 Returns a boolean indicating whether X is a port. Equivalent to
3554 `(or (input-port? X) (output-port? X))'.
3556 ** New function: file-port?
3558 Determines whether a given object is a port that is related to a file.
3560 ** New function: port-for-each proc
3562 Apply PROC to each port in the Guile port table in turn. The return
3563 value is unspecified. More specifically, PROC is applied exactly once
3564 to every port that exists in the system at the time PORT-FOR-EACH is
3565 invoked. Changes to the port table while PORT-FOR-EACH is running
3566 have no effect as far as PORT-FOR-EACH is concerned.
3568 ** New function: dup2 oldfd newfd
3570 A simple wrapper for the `dup2' system call. Copies the file
3571 descriptor OLDFD to descriptor number NEWFD, replacing the
3572 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
3573 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
3574 to move away ports which are using NEWFD. The return value is
3577 ** New function: close-fdes fd
3579 A simple wrapper for the `close' system call. Close file
3580 descriptor FD, which must be an integer. Unlike close (*note
3581 close: Ports and File Descriptors.), the file descriptor will be
3582 closed even if a port is using it. The return value is
3585 ** New function: crypt password salt
3587 Encrypts `password' using the standard unix password encryption
3590 ** New function: chroot path
3592 Change the root directory of the running process to `path'.
3594 ** New functions: getlogin, cuserid
3596 Return the login name or the user name of the current effective user
3599 ** New functions: getpriority which who, setpriority which who prio
3601 Get or set the priority of the running process.
3603 ** New function: getpass prompt
3605 Read a password from the terminal, first displaying `prompt' and
3608 ** New function: flock file operation
3610 Set/remove an advisory shared or exclusive lock on `file'.
3612 ** New functions: sethostname name, gethostname
3614 Set or get the hostname of the machine the current process is running
3617 ** New function: mkstemp! tmpl
3619 mkstemp creates a new unique file in the file system and returns a
3620 new buffered port open for reading and writing to the file. TMPL
3621 is a string specifying where the file should be created: it must
3622 end with `XXXXXX' and will be changed in place to return the name
3623 of the temporary file.
3625 ** New function: open-input-string string
3627 Return an input string port which delivers the characters from
3628 `string'. This procedure, together with `open-output-string' and
3629 `get-output-string' implements SRFI-6.
3631 ** New function: open-output-string
3633 Return an output string port which collects all data written to it.
3634 The data can then be retrieved by `get-output-string'.
3636 ** New function: get-output-string
3638 Return the contents of an output string port.
3640 ** New function: identity
3642 Return the argument.
3644 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
3645 are represented in Scheme as integers with normal host byte ordering.
3647 ** New function: inet-pton family address
3649 Convert a printable string network address into an integer. Note that
3650 unlike the C version of this function, the result is an integer with
3651 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
3654 (inet-pton AF_INET "127.0.0.1") => 2130706433
3655 (inet-pton AF_INET6 "::1") => 1
3657 ** New function: inet-ntop family address
3659 Convert an integer network address into a printable string. Note that
3660 unlike the C version of this function, the input is an integer with
3661 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
3664 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
3665 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
3666 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
3670 Use `identity' instead.
3676 ** Deprecated: return-it
3680 ** Deprecated: string-character-length
3682 Use `string-length' instead.
3684 ** Deprecated: flags
3686 Use `logior' instead.
3688 ** Deprecated: close-all-ports-except.
3690 This was intended for closing ports in a child process after a fork,
3691 but it has the undesirable side effect of flushing buffers.
3692 port-for-each is more flexible.
3694 ** The (ice-9 popen) module now attempts to set up file descriptors in
3695 the child process from the current Scheme ports, instead of using the
3696 current values of file descriptors 0, 1, and 2 in the parent process.
3698 ** Removed function: builtin-weak-bindings
3700 There is no such concept as a weak binding any more.
3702 ** Removed constants: bignum-radix, scm-line-incrementors
3704 ** define-method: New syntax mandatory.
3706 The new method syntax is now mandatory:
3708 (define-method (NAME ARG-SPEC ...) BODY ...)
3709 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
3711 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
3712 REST-ARG ::= ARG-NAME
3714 If you have old code using the old syntax, import
3715 (oop goops old-define-method) before (oop goops) as in:
3717 (use-modules (oop goops old-define-method) (oop goops))
3719 ** Deprecated function: builtin-variable
3720 Removed function: builtin-bindings
3722 There is no longer a distinction between builtin or other variables.
3723 Use module system operations for all variables.
3725 ** Lazy-catch handlers are no longer allowed to return.
3727 That is, a call to `throw', `error', etc is now guaranteed to not
3730 ** Bugfixes for (ice-9 getopt-long)
3732 This module is now tested using test-suite/tests/getopt-long.test.
3733 The following bugs have been fixed:
3735 *** Parsing for options that are specified to have `optional' args now checks
3736 if the next element is an option instead of unconditionally taking it as the
3739 *** An error is now thrown for `--opt=val' when the option description
3740 does not specify `(value #t)' or `(value optional)'. This condition used to
3741 be accepted w/o error, contrary to the documentation.
3743 *** The error message for unrecognized options is now more informative.
3744 It used to be "not a record", an artifact of the implementation.
3746 *** The error message for `--opt' terminating the arg list (no value), when
3747 `(value #t)' is specified, is now more informative. It used to be "not enough
3750 *** "Clumped" single-char args now preserve trailing string, use it as arg.
3751 The expansion used to be like so:
3753 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
3755 Note that the "5d" is dropped. Now it is like so:
3757 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
3759 This enables single-char options to have adjoining arguments as long as their
3760 constituent characters are not potential single-char options.
3762 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
3764 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
3765 property `arglist', which can be retrieved by `arity'. The result is that
3766 `arity' can give more detailed information than before:
3770 guile> (use-modules (ice-9 optargs))
3771 guile> (define* (foo #:optional a b c) a)
3773 0 or more arguments in `lambda*:G0'.
3778 3 optional arguments: `a', `b' and `c'.
3779 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
3781 2 required arguments: `a' and `b', 2 keyword arguments: `c'
3782 and `d', other keywords allowed.
3783 guile> (define* (baz a b #:optional c #:rest r) a)
3785 2 required arguments: `a' and `b', 1 optional argument: `c',
3788 * Changes to the C interface
3790 ** Types have been renamed from scm_*_t to scm_t_*.
3792 This has been done for POSIX sake. It reserves identifiers ending
3793 with "_t". What a concept.
3795 The old names are still available with status `deprecated'.
3797 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
3799 ** Deprecated features have been removed.
3803 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
3804 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
3806 *** C Functions removed
3808 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
3809 scm_fseek - replaced by scm_seek.
3810 gc-thunk - replaced by after-gc-hook.
3811 gh_int2scmb - replaced by gh_bool2scm.
3812 scm_tc_dblr - replaced by scm_tc16_real.
3813 scm_tc_dblc - replaced by scm_tc16_complex.
3814 scm_list_star - replaced by scm_cons_star.
3816 ** Deprecated: scm_makfromstr
3818 Use scm_mem2string instead.
3820 ** Deprecated: scm_make_shared_substring
3822 Explicit shared substrings will disappear from Guile.
3824 Instead, "normal" strings will be implemented using sharing
3825 internally, combined with a copy-on-write strategy.
3827 ** Deprecated: scm_read_only_string_p
3829 The concept of read-only strings will disappear in next release of
3832 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
3834 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
3836 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
3838 Call a procedure with the indicated number of arguments. See "Fly
3839 Evaluation" in the manual.
3841 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
3843 Call a procedure with the indicated number of arguments and a list of
3844 further arguments. See "Fly Evaluation" in the manual.
3846 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
3848 Create a list of the given number of elements. See "List
3849 Constructors" in the manual.
3851 ** Renamed function: scm_listify has been replaced by scm_list_n.
3853 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
3854 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
3856 Use functions scm_list_N instead.
3858 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
3860 Used by an application to read arbitrary number of bytes from a port.
3861 Same semantics as libc read, except that scm_c_read only returns less
3862 than SIZE bytes if at end-of-file.
3864 Warning: Doesn't update port line and column counts!
3866 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
3868 Used by an application to write arbitrary number of bytes to an SCM
3869 port. Similar semantics as libc write. However, unlike libc
3870 write, scm_c_write writes the requested number of bytes and has no
3873 Warning: Doesn't update port line and column counts!
3875 ** New function: scm_init_guile ()
3877 In contrast to scm_boot_guile, scm_init_guile will return normally
3878 after initializing Guile. It is not available on all systems, tho.
3880 ** New functions: scm_str2symbol, scm_mem2symbol
3882 The function scm_str2symbol takes a const char* pointing to a zero-terminated
3883 field of characters and creates a scheme symbol object from that C string.
3884 The function scm_mem2symbol takes a const char* and a number of characters and
3885 creates a symbol from the characters in that memory area.
3887 ** New functions: scm_primitive_make_property
3888 scm_primitive_property_ref
3889 scm_primitive_property_set_x
3890 scm_primitive_property_del_x
3892 These functions implement a new way to deal with object properties.
3893 See libguile/properties.c for their documentation.
3895 ** New function: scm_done_free (long size)
3897 This function is the inverse of scm_done_malloc. Use it to report the
3898 amount of smob memory you free. The previous method, which involved
3899 calling scm_done_malloc with negative argument, was somewhat
3900 unintuitive (and is still available, of course).
3902 ** New function: scm_c_memq (SCM obj, SCM list)
3904 This function provides a fast C level alternative for scm_memq for the case
3905 that the list parameter is known to be a proper list. The function is a
3906 replacement for scm_sloppy_memq, but is stricter in its requirements on its
3907 list input parameter, since for anything else but a proper list the function's
3908 behaviour is undefined - it may even crash or loop endlessly. Further, for
3909 the case that the object is not found in the list, scm_c_memq returns #f which
3910 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
3912 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
3913 scm_remember_upto_here
3915 These functions replace the function scm_remember.
3917 ** Deprecated function: scm_remember
3919 Use one of the new functions scm_remember_upto_here_1,
3920 scm_remember_upto_here_2 or scm_remember_upto_here instead.
3922 ** New function: scm_allocate_string
3924 This function replaces the function scm_makstr.
3926 ** Deprecated function: scm_makstr
3928 Use the new function scm_allocate_string instead.
3930 ** New global variable scm_gc_running_p introduced.
3932 Use this variable to find out if garbage collection is being executed. Up to
3933 now applications have used scm_gc_heap_lock to test if garbage collection was
3934 running, which also works because of the fact that up to know only the garbage
3935 collector has set this variable. But, this is an implementation detail that
3936 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
3937 of this variable is (and has been) not fully safe anyway.
3939 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
3941 Use these instead of SCM_LENGTH_MAX.
3943 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
3944 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
3945 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
3947 Use these instead of SCM_LENGTH.
3949 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
3950 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
3951 SCM_SET_BITVECTOR_LENGTH
3953 Use these instead of SCM_SETLENGTH
3955 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
3956 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
3959 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
3962 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
3963 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
3966 Use these instead of SCM_SETCHARS.
3968 ** New macro: SCM_BITVECTOR_P
3970 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
3972 Use instead of SCM_COERCE_SUBSTR.
3974 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
3976 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
3978 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
3979 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
3980 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
3981 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
3982 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
3983 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
3984 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
3985 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
3986 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
3987 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
3988 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
3989 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
3990 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
3991 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
3992 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
3994 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
3995 Use scm_memory_error instead of SCM_NALLOC.
3996 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
3997 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
3998 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
3999 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
4000 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
4001 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
4002 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
4003 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
4004 Use SCM_STRINGP instead of SCM_RWSTRINGP.
4005 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
4006 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
4007 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
4008 Use a type specific setter macro instead of SCM_SETLENGTH.
4009 Use a type specific setter macro instead of SCM_SETCHARS.
4010 Use a type specific length macro instead of SCM_LENGTH_MAX.
4011 Use SCM_GCMARKP instead of SCM_GC8MARKP.
4012 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
4013 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
4014 Use SCM_TYP16 instead of SCM_GCTYP16.
4015 Use SCM_CDR instead of SCM_GCCDR.
4016 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
4017 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
4018 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
4019 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
4020 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
4021 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
4022 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
4024 ** Removed function: scm_struct_init
4026 ** Removed variable: scm_symhash_dim
4028 ** Renamed function: scm_make_cont has been replaced by
4029 scm_make_continuation, which has a different interface.
4031 ** Deprecated function: scm_call_catching_errors
4033 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
4035 ** Deprecated function: scm_strhash
4037 Use scm_string_hash instead.
4039 ** Deprecated function: scm_vector_set_length_x
4041 Instead, create a fresh vector of the desired size and copy the contents.
4043 ** scm_gensym has changed prototype
4045 scm_gensym now only takes one argument.
4047 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
4050 There is now only a single symbol type scm_tc7_symbol.
4051 The tag scm_tc7_lvector was not used anyway.
4053 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
4055 Use scm_make_smob_type and scm_set_smob_XXX instead.
4057 ** New function scm_set_smob_apply.
4059 This can be used to set an apply function to a smob type.
4061 ** Deprecated function: scm_strprint_obj
4063 Use scm_object_to_string instead.
4065 ** Deprecated function: scm_wta
4067 Use scm_wrong_type_arg, or another appropriate error signalling function
4070 ** Explicit support for obarrays has been deprecated.
4072 Use `scm_str2symbol' and the generic hashtable functions instead.
4074 ** The concept of `vcells' has been deprecated.
4076 The data type `variable' is now used exclusively. `Vcells' have been
4077 a low-level concept so you are likely not affected by this change.
4079 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
4080 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
4082 Use scm_c_define or scm_c_lookup instead, as appropriate.
4084 *** New functions: scm_c_module_lookup, scm_c_lookup,
4085 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
4086 scm_module_define, scm_define.
4088 These functions work with variables instead of with vcells.
4090 ** New functions for creating and defining `subr's and `gsubr's.
4092 The new functions more clearly distinguish between creating a subr (or
4093 gsubr) object and adding it to the current module.
4095 These new functions are available: scm_c_make_subr, scm_c_define_subr,
4096 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
4097 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
4098 scm_c_define_gsubr_with_generic.
4100 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
4101 scm_make_subr_with_generic, scm_make_gsubr,
4102 scm_make_gsubr_with_generic.
4104 Use the new ones from above instead.
4106 ** C interface to the module system has changed.
4108 While we suggest that you avoid as many explicit module system
4109 operations from C as possible for the time being, the C interface has
4110 been made more similar to the high-level Scheme module system.
4112 *** New functions: scm_c_define_module, scm_c_use_module,
4113 scm_c_export, scm_c_resolve_module.
4115 They mostly work like their Scheme namesakes. scm_c_define_module
4116 takes a function that is called a context where the new module is
4119 *** Deprecated functions: scm_the_root_module, scm_make_module,
4120 scm_ensure_user_module, scm_load_scheme_module.
4122 Use the new functions instead.
4124 ** Renamed function: scm_internal_with_fluids becomes
4127 scm_internal_with_fluids is available as a deprecated function.
4129 ** New function: scm_c_with_fluid.
4131 Just like scm_c_with_fluids, but takes one fluid and one value instead
4134 ** Deprecated typedefs: long_long, ulong_long.
4136 They are of questionable utility and they pollute the global
4139 ** Deprecated typedef: scm_sizet
4141 It is of questionable utility now that Guile requires ANSI C, and is
4144 ** Deprecated typedefs: scm_port_rw_active, scm_port,
4145 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
4146 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
4148 Made more compliant with the naming policy by adding a _t at the end.
4150 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
4151 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
4153 With the exception of the mysterious scm_2ulong2big, they are still
4154 available under new names (scm_i_mkbig etc). These functions are not
4155 intended to be used in user code. You should avoid dealing with
4156 bignums directly, and should deal with numbers in general (which can
4159 ** Change in behavior: scm_num2long, scm_num2ulong
4161 The scm_num2[u]long functions don't any longer accept an inexact
4162 argument. This change in behavior is motivated by concordance with
4163 R5RS: It is more common that a primitive doesn't want to accept an
4164 inexact for an exact.
4166 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
4167 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
4168 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
4171 These are conversion functions between the various ANSI C integral
4172 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
4173 accept an inexact argument.
4175 ** New functions: scm_float2num, scm_double2num,
4176 scm_num2float, scm_num2double.
4178 These are conversion functions between the two ANSI C float types and
4181 ** New number validation macros:
4182 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
4186 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
4188 These are just nicer-named old scm_protect_object and
4189 scm_unprotect_object.
4191 ** Deprecated functions: scm_protect_object, scm_unprotect_object
4193 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
4195 These functions can be used to register pointers to locations that
4198 ** Deprecated function: scm_create_hook.
4200 Its sins are: misleading name, non-modularity and lack of general
4204 Changes since Guile 1.3.4:
4206 * Changes to the distribution
4208 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
4210 We've changed the way we handle generated files in the Guile source
4211 repository. As a result, the procedure for building trees obtained
4212 from the nightly FTP snapshots or via CVS has changed:
4213 - You must have appropriate versions of autoconf, automake, and
4214 libtool installed on your system. See README for info on how to
4215 obtain these programs.
4216 - Before configuring the tree, you must first run the script
4217 `autogen.sh' at the top of the source tree.
4219 The Guile repository used to contain not only source files, written by
4220 humans, but also some generated files, like configure scripts and
4221 Makefile.in files. Even though the contents of these files could be
4222 derived mechanically from other files present, we thought it would
4223 make the tree easier to build if we checked them into CVS.
4225 However, this approach means that minor differences between
4226 developer's installed tools and habits affected the whole team.
4227 So we have removed the generated files from the repository, and
4228 added the autogen.sh script, which will reconstruct them
4232 ** configure now has experimental options to remove support for certain
4235 --disable-arrays omit array and uniform array support
4236 --disable-posix omit posix interfaces
4237 --disable-networking omit networking interfaces
4238 --disable-regex omit regular expression interfaces
4240 These are likely to become separate modules some day.
4242 ** New configure option --enable-debug-freelist
4244 This enables a debugging version of SCM_NEWCELL(), and also registers
4245 an extra primitive, the setter `gc-set-debug-check-freelist!'.
4247 Configure with the --enable-debug-freelist option to enable
4248 the gc-set-debug-check-freelist! primitive, and then use:
4250 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
4251 (gc-set-debug-check-freelist! #f) # turn off checking
4253 Checking of the freelist forces a traversal of the freelist and
4254 a garbage collection before each allocation of a cell. This can
4255 slow down the interpreter dramatically, so the setter should be used to
4256 turn on this extra processing only when necessary.
4258 ** New configure option --enable-debug-malloc
4260 Include code for debugging of calls to scm_must_malloc/realloc/free.
4264 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
4265 2. objects reallocated by scm_must_realloc has been allocated by
4267 3. reallocated objects are reallocated with the same what string
4269 But, most importantly, it records the number of allocated objects of
4270 each kind. This is useful when searching for memory leaks.
4272 A Guile compiled with this option provides the primitive
4273 `malloc-stats' which returns an alist with pairs of kind and the
4274 number of objects of that kind.
4276 ** All includes are now referenced relative to the root directory
4278 Since some users have had problems with mixups between Guile and
4279 system headers, we have decided to always refer to Guile headers via
4280 their parent directories. This essentially creates a "private name
4281 space" for Guile headers. This means that the compiler only is given
4282 -I options for the root build and root source directory.
4284 ** Header files kw.h and genio.h have been removed.
4286 ** The module (ice-9 getopt-gnu-style) has been removed.
4288 ** New module (ice-9 documentation)
4290 Implements the interface to documentation strings associated with
4293 ** New module (ice-9 time)
4295 Provides a macro `time', which displays execution time of a given form.
4297 ** New module (ice-9 history)
4299 Loading this module enables value history in the repl.
4301 * Changes to the stand-alone interpreter
4303 ** New command line option --debug
4305 Start Guile with debugging evaluator and backtraces enabled.
4307 This is useful when debugging your .guile init file or scripts.
4309 ** New help facility
4311 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
4312 (help REGEXP) ditto for objects with names matching REGEXP (a string)
4313 (help 'NAME) gives documentation for NAME, even if it is not an object
4314 (help ,EXPR) gives documentation for object returned by EXPR
4315 (help (my module)) gives module commentary for `(my module)'
4316 (help) gives this text
4318 `help' searches among bindings exported from loaded modules, while
4319 `apropos' searches among bindings visible from the "current" module.
4321 Examples: (help help)
4323 (help "output-string")
4325 ** `help' and `apropos' now prints full module names
4327 ** Dynamic linking now uses libltdl from the libtool package.
4329 The old system dependent code for doing dynamic linking has been
4330 replaced with calls to the libltdl functions which do all the hairy
4333 The major improvement is that you can now directly pass libtool
4334 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
4335 will be able to do the best shared library job you can get, via
4338 The way dynamic libraries are found has changed and is not really
4339 portable across platforms, probably. It is therefore recommended to
4340 use absolute filenames when possible.
4342 If you pass a filename without an extension to `dynamic-link', it will
4343 try a few appropriate ones. Thus, the most platform ignorant way is
4344 to specify a name like "libfoo", without any directories and
4347 ** Guile COOP threads are now compatible with LinuxThreads
4349 Previously, COOP threading wasn't possible in applications linked with
4350 Linux POSIX threads due to their use of the stack pointer to find the
4351 thread context. This has now been fixed with a workaround which uses
4352 the pthreads to allocate the stack.
4354 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
4356 ** Positions of erring expression in scripts
4358 With version 1.3.4, the location of the erring expression in Guile
4359 scipts is no longer automatically reported. (This should have been
4360 documented before the 1.3.4 release.)
4362 You can get this information by enabling recording of positions of
4363 source expressions and running the debugging evaluator. Put this at
4364 the top of your script (or in your "site" file):
4366 (read-enable 'positions)
4367 (debug-enable 'debug)
4369 ** Backtraces in scripts
4371 It is now possible to get backtraces in scripts.
4375 (debug-enable 'debug 'backtrace)
4377 at the top of the script.
4379 (The first options enables the debugging evaluator.
4380 The second enables backtraces.)
4382 ** Part of module system symbol lookup now implemented in C
4384 The eval closure of most modules is now implemented in C. Since this
4385 was one of the bottlenecks for loading speed, Guile now loads code
4386 substantially faster than before.
4388 ** Attempting to get the value of an unbound variable now produces
4389 an exception with a key of 'unbound-variable instead of 'misc-error.
4391 ** The initial default output port is now unbuffered if it's using a
4392 tty device. Previously in this situation it was line-buffered.
4394 ** New hook: after-gc-hook
4396 after-gc-hook takes over the role of gc-thunk. This hook is run at
4397 the first SCM_TICK after a GC. (Thus, the code is run at the same
4398 point during evaluation as signal handlers.)
4400 Note that this hook should be used only for diagnostic and debugging
4401 purposes. It is not certain that it will continue to be well-defined
4402 when this hook is run in the future.
4404 C programmers: Note the new C level hooks scm_before_gc_c_hook,
4405 scm_before_sweep_c_hook, scm_after_gc_c_hook.
4407 ** Improvements to garbage collector
4409 Guile 1.4 has a new policy for triggering heap allocation and
4410 determining the sizes of heap segments. It fixes a number of problems
4413 1. The new policy can handle two separate pools of cells
4414 (2-word/4-word) better. (The old policy would run wild, allocating
4415 more and more memory for certain programs.)
4417 2. The old code would sometimes allocate far too much heap so that the
4418 Guile process became gigantic. The new code avoids this.
4420 3. The old code would sometimes allocate too little so that few cells
4421 were freed at GC so that, in turn, too much time was spent in GC.
4423 4. The old code would often trigger heap allocation several times in a
4424 row. (The new scheme predicts how large the segments needs to be
4425 in order not to need further allocation.)
4427 All in all, the new GC policy will make larger applications more
4430 The new GC scheme also is prepared for POSIX threading. Threads can
4431 allocate private pools of cells ("clusters") with just a single
4432 function call. Allocation of single cells from such a cluster can
4433 then proceed without any need of inter-thread synchronization.
4435 ** New environment variables controlling GC parameters
4437 GUILE_MAX_SEGMENT_SIZE Maximal segment size
4440 Allocation of 2-word cell heaps:
4442 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
4445 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
4446 GC in percent of total heap size
4449 Allocation of 4-word cell heaps
4450 (used for real numbers and misc other objects):
4452 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
4454 (See entry "Way for application to customize GC parameters" under
4455 section "Changes to the scm_ interface" below.)
4457 ** Guile now implements reals using 4-word cells
4459 This speeds up computation with reals. (They were earlier allocated
4460 with `malloc'.) There is still some room for optimizations, however.
4462 ** Some further steps toward POSIX thread support have been taken
4464 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
4465 don't have much effect any longer, and many of them will be removed in
4469 are only handled at the top of the evaluator loop, immediately after
4470 I/O, and in scm_equalp.
4472 *** The GC can allocate thread private pools of pairs.
4474 * Changes to Scheme functions and syntax
4476 ** close-input-port and close-output-port are now R5RS
4478 These procedures have been turned into primitives and have R5RS behaviour.
4480 ** New procedure: simple-format PORT MESSAGE ARG1 ...
4482 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
4483 extended by the more sophisticated version in (ice-9 format)
4485 (simple-format port message . args)
4486 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
4487 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
4488 the escapes are replaced with corresponding members of ARGS:
4489 ~A formats using `display' and ~S formats using `write'.
4490 If DESTINATION is #t, then use the `current-output-port',
4491 if DESTINATION is #f, then return a string containing the formatted text.
4492 Does not add a trailing newline."
4494 ** string-ref: the second argument is no longer optional.
4496 ** string, list->string: no longer accept strings in their arguments,
4497 only characters, for compatibility with R5RS.
4499 ** New procedure: port-closed? PORT
4500 Returns #t if PORT is closed or #f if it is open.
4502 ** Deprecated: list*
4504 The list* functionality is now provided by cons* (SRFI-1 compliant)
4506 ** New procedure: cons* ARG1 ARG2 ... ARGn
4508 Like `list', but the last arg provides the tail of the constructed list,
4509 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
4511 Requires at least one argument. If given one argument, that argument
4512 is returned as result.
4514 This function is called `list*' in some other Schemes and in Common LISP.
4516 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
4518 ** New procedure: object-documentation OBJECT
4520 Returns the documentation string associated with OBJECT. The
4521 procedure uses a caching mechanism so that subsequent lookups are
4524 Exported by (ice-9 documentation).
4526 ** module-name now returns full names of modules
4528 Previously, only the last part of the name was returned (`session' for
4529 `(ice-9 session)'). Ex: `(ice-9 session)'.
4531 * Changes to the gh_ interface
4533 ** Deprecated: gh_int2scmb
4535 Use gh_bool2scm instead.
4537 * Changes to the scm_ interface
4539 ** Guile primitives now carry docstrings!
4541 Thanks to Greg Badros!
4543 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
4545 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
4546 macros and must contain a docstring that is extracted into foo.doc using a new
4547 guile-doc-snarf script (that uses guile-doc-snarf.awk).
4549 However, a major overhaul of these macros is scheduled for the next release of
4552 ** Guile primitives use a new technique for validation of arguments
4554 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
4555 the readability of argument checking.
4557 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
4559 ** New macros: SCM_PACK, SCM_UNPACK
4561 Compose/decompose an SCM value.
4563 The SCM type is now treated as an abstract data type and may be defined as a
4564 long, a void* or as a struct, depending on the architecture and compile time
4565 options. This makes it easier to find several types of bugs, for example when
4566 SCM values are treated as integers without conversion. Values of the SCM type
4567 should be treated as "atomic" values. These macros are used when
4568 composing/decomposing an SCM value, either because you want to access
4569 individual bits, or because you want to treat it as an integer value.
4571 E.g., in order to set bit 7 in an SCM value x, use the expression
4573 SCM_PACK (SCM_UNPACK (x) | 0x80)
4575 ** The name property of hooks is deprecated.
4576 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
4578 You can emulate this feature by using object properties.
4580 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
4581 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
4584 These macros will be removed in a future release of Guile.
4586 ** The following types, functions and macros from numbers.h are deprecated:
4587 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
4588 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
4590 ** Port internals: the rw_random variable in the scm_port structure
4591 must be set to non-zero in any random access port. In recent Guile
4592 releases it was only set for bidirectional random-access ports.
4594 ** Port internals: the seek ptob procedure is now responsible for
4595 resetting the buffers if required. The change was made so that in the
4596 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
4597 the fport and strport ptobs can avoid resetting the buffers,
4598 in particular to avoid discarding unread chars. An existing port
4599 type can be fixed by adding something like the following to the
4600 beginning of the ptob seek procedure:
4602 if (pt->rw_active == SCM_PORT_READ)
4603 scm_end_input (object);
4604 else if (pt->rw_active == SCM_PORT_WRITE)
4605 ptob->flush (object);
4607 although to actually avoid resetting the buffers and discard unread
4608 chars requires further hacking that depends on the characteristics
4611 ** Deprecated functions: scm_fseek, scm_tag
4613 These functions are no longer used and will be removed in a future version.
4615 ** The scm_sysmissing procedure is no longer used in libguile.
4616 Unless it turns out to be unexpectedly useful to somebody, it will be
4617 removed in a future version.
4619 ** The format of error message strings has changed
4621 The two C procedures: scm_display_error and scm_error, as well as the
4622 primitive `scm-error', now use scm_simple_format to do their work.
4623 This means that the message strings of all code must be updated to use
4624 ~A where %s was used before, and ~S where %S was used before.
4626 During the period when there still are a lot of old Guiles out there,
4627 you might want to support both old and new versions of Guile.
4629 There are basically two methods to achieve this. Both methods use
4632 AC_CHECK_FUNCS(scm_simple_format)
4634 in your configure.in.
4636 Method 1: Use the string concatenation features of ANSI C's
4641 #ifdef HAVE_SCM_SIMPLE_FORMAT
4647 Then represent each of your error messages using a preprocessor macro:
4649 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
4653 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
4654 (define make-message string-append)
4656 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
4658 Method 2: Use the oldfmt function found in doc/oldfmt.c.
4662 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
4667 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
4671 ** Deprecated: coop_mutex_init, coop_condition_variable_init
4673 Don't use the functions coop_mutex_init and
4674 coop_condition_variable_init. They will change.
4676 Use scm_mutex_init and scm_cond_init instead.
4678 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
4679 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
4680 COND, as `scm_cond_wait' does, but it also bounds the duration
4681 of the wait. If COND has not been signaled before time ABSTIME,
4682 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
4683 returns the error code `ETIMEDOUT'.
4685 The ABSTIME parameter specifies an absolute time, with the same
4686 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
4687 to 00:00:00 GMT, January 1, 1970.
4689 ** New function: scm_cond_broadcast (scm_cond_t *COND)
4690 `scm_cond_broadcast' restarts all the threads that are waiting
4691 on the condition variable COND. Nothing happens if no threads are
4694 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
4695 `scm_key_create' allocates a new TSD key. The key is stored in
4696 the location pointed to by KEY. There is no limit on the number
4697 of keys allocated at a given time. The value initially associated
4698 with the returned key is `NULL' in all currently executing threads.
4700 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
4701 function associated with the key. When a thread terminates,
4702 DESTR_FUNCTION is called on the value associated with the key in
4703 that thread. The DESTR_FUNCTION is not called if a key is deleted
4704 with `scm_key_delete' or a value is changed with
4705 `scm_setspecific'. The order in which destructor functions are
4706 called at thread termination time is unspecified.
4708 Destructors are not yet implemented.
4710 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
4711 `scm_setspecific' changes the value associated with KEY in the
4712 calling thread, storing the given POINTER instead.
4714 ** New function: scm_getspecific (scm_key_t KEY)
4715 `scm_getspecific' returns the value currently associated with
4716 KEY in the calling thread.
4718 ** New function: scm_key_delete (scm_key_t KEY)
4719 `scm_key_delete' deallocates a TSD key. It does not check
4720 whether non-`NULL' values are associated with that key in the
4721 currently executing threads, nor call the destructor function
4722 associated with the key.
4724 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
4726 Initialize a C level hook HOOK with associated HOOK_DATA and type
4727 TYPE. (See scm_c_hook_run ().)
4729 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
4731 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
4732 is true, add it last, otherwise first. The same FUNC can be added
4733 multiple times if FUNC_DATA differ and vice versa.
4735 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
4737 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
4738 function is only removed if both FUNC and FUNC_DATA matches.
4740 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
4742 Run hook HOOK passing DATA to the hook functions.
4744 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
4745 returned is undefined.
4747 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
4748 returns a non-NULL value. This value is returned as the result of
4749 scm_c_hook_run. If all functions return NULL, NULL is returned.
4751 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
4752 returns a NULL value, and NULL is returned. If all functions returns
4753 a non-NULL value, the last value is returned.
4755 ** New C level GC hooks
4757 Five new C level hooks has been added to the garbage collector.
4759 scm_before_gc_c_hook
4762 are run before locking and after unlocking the heap. The system is
4763 thus in a mode where evaluation can take place. (Except that
4764 scm_before_gc_c_hook must not allocate new cells.)
4766 scm_before_mark_c_hook
4767 scm_before_sweep_c_hook
4768 scm_after_sweep_c_hook
4770 are run when the heap is locked. These are intended for extension of
4771 the GC in a modular fashion. Examples are the weaks and guardians
4774 ** Way for application to customize GC parameters
4776 The application can set up other default values for the GC heap
4777 allocation parameters
4779 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
4780 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
4781 GUILE_MAX_SEGMENT_SIZE,
4785 scm_default_init_heap_size_1, scm_default_min_yield_1,
4786 scm_default_init_heap_size_2, scm_default_min_yield_2,
4787 scm_default_max_segment_size
4789 respectively before callong scm_boot_guile.
4791 (See entry "New environment variables ..." in section
4792 "Changes to the stand-alone interpreter" above.)
4794 ** scm_protect_object/scm_unprotect_object now nest
4796 This means that you can call scm_protect_object multiple times on an
4797 object and count on the object being protected until
4798 scm_unprotect_object has been call the same number of times.
4800 The functions also have better time complexity.
4802 Still, it is usually possible to structure the application in a way
4803 that you don't need to use these functions. For example, if you use a
4804 protected standard Guile list to keep track of live objects rather
4805 than some custom data type, objects will die a natural death when they
4806 are no longer needed.
4808 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
4810 Guile does not provide the float representation for inexact real numbers any
4811 more. Now, only doubles are used to represent inexact real numbers. Further,
4812 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
4813 and scm_tc16_complex, respectively.
4815 ** Removed deprecated type scm_smobfuns
4817 ** Removed deprecated function scm_newsmob
4819 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
4821 There is an ongoing discussion among the developers whether to
4822 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
4823 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
4824 until this issue has been settled.
4826 ** Removed deprecated type tag scm_tc16_kw
4828 ** Added type tag scm_tc16_keyword
4830 (This was introduced already in release 1.3.4 but was not documented
4833 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
4835 * Changes to system call interfaces:
4837 ** The "select" procedure now tests port buffers for the ability to
4838 provide input or accept output. Previously only the underlying file
4839 descriptors were checked.
4841 ** New variable PIPE_BUF: the maximum number of bytes that can be
4842 atomically written to a pipe.
4844 ** If a facility is not available on the system when Guile is
4845 compiled, the corresponding primitive procedure will not be defined.
4846 Previously it would have been defined but would throw a system-error
4847 exception if called. Exception handlers which catch this case may
4848 need minor modification: an error will be thrown with key
4849 'unbound-variable instead of 'system-error. Alternatively it's
4850 now possible to use `defined?' to check whether the facility is
4853 ** Procedures which depend on the timezone should now give the correct
4854 result on systems which cache the TZ environment variable, even if TZ
4855 is changed without calling tzset.
4857 * Changes to the networking interfaces:
4859 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
4860 long integers between network and host format. For now, it's not
4861 particularly convenient to do this kind of thing, but consider:
4863 (define write-network-long
4864 (lambda (value port)
4865 (let ((v (make-uniform-vector 1 1 0)))
4866 (uniform-vector-set! v 0 (htonl value))
4867 (uniform-vector-write v port))))
4869 (define read-network-long
4871 (let ((v (make-uniform-vector 1 1 0)))
4872 (uniform-vector-read! v port)
4873 (ntohl (uniform-vector-ref v 0)))))
4875 ** If inet-aton fails, it now throws an error with key 'misc-error
4876 instead of 'system-error, since errno is not relevant.
4878 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
4879 specific keys instead of 'system-error. The latter is inappropriate
4880 since errno will not have been set. The keys are:
4881 'host-not-found, 'try-again, 'no-recovery and 'no-data.
4883 ** sethostent, setnetent, setprotoent, setservent: now take an
4884 optional argument STAYOPEN, which specifies whether the database
4885 remains open after a database entry is accessed randomly (e.g., using
4886 gethostbyname for the hosts database.) The default is #f. Previously
4890 Changes since Guile 1.3.2:
4892 * Changes to the stand-alone interpreter
4896 An initial version of the Guile debugger written by Chris Hanson has
4897 been added. The debugger is still under development but is included
4898 in the distribution anyway since it is already quite useful.
4904 after an error to enter the debugger. Type `help' inside the debugger
4905 for a description of available commands.
4907 If you prefer to have stack frames numbered and printed in
4908 anti-chronological order and prefer up in the stack to be down on the
4909 screen as is the case in gdb, you can put
4911 (debug-enable 'backwards)
4913 in your .guile startup file. (However, this means that Guile can't
4914 use indentation to indicate stack level.)
4916 The debugger is autoloaded into Guile at the first use.
4918 ** Further enhancements to backtraces
4920 There is a new debug option `width' which controls the maximum width
4921 on the screen of printed stack frames. Fancy printing parameters
4922 ("level" and "length" as in Common LISP) are adaptively adjusted for
4923 each stack frame to give maximum information while still fitting
4924 within the bounds. If the stack frame can't be made to fit by
4925 adjusting parameters, it is simply cut off at the end. This is marked
4928 ** Some modules are now only loaded when the repl is started
4930 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
4931 regex) are now loaded into (guile-user) only if the repl has been
4932 started. The effect is that the startup time for scripts has been
4933 reduced to 30% of what it was previously.
4935 Correctly written scripts load the modules they require at the top of
4936 the file and should not be affected by this change.
4938 ** Hooks are now represented as smobs
4940 * Changes to Scheme functions and syntax
4942 ** Readline support has changed again.
4944 The old (readline-activator) module is gone. Use (ice-9 readline)
4945 instead, which now contains all readline functionality. So the code
4946 to activate readline is now
4948 (use-modules (ice-9 readline))
4951 This should work at any time, including from the guile prompt.
4953 To avoid confusion about the terms of Guile's license, please only
4954 enable readline for your personal use; please don't make it the
4955 default for others. Here is why we make this rather odd-sounding
4958 Guile is normally licensed under a weakened form of the GNU General
4959 Public License, which allows you to link code with Guile without
4960 placing that code under the GPL. This exception is important to some
4963 However, since readline is distributed under the GNU General Public
4964 License, when you link Guile with readline, either statically or
4965 dynamically, you effectively change Guile's license to the strict GPL.
4966 Whenever you link any strictly GPL'd code into Guile, uses of Guile
4967 which are normally permitted become forbidden. This is a rather
4968 non-obvious consequence of the licensing terms.
4970 So, to make sure things remain clear, please let people choose for
4971 themselves whether to link GPL'd libraries like readline with Guile.
4973 ** regexp-substitute/global has changed slightly, but incompatibly.
4975 If you include a function in the item list, the string of the match
4976 object it receives is the same string passed to
4977 regexp-substitute/global, not some suffix of that string.
4978 Correspondingly, the match's positions are relative to the entire
4979 string, not the suffix.
4981 If the regexp can match the empty string, the way matches are chosen
4982 from the string has changed. regexp-substitute/global recognizes the
4983 same set of matches that list-matches does; see below.
4985 ** New function: list-matches REGEXP STRING [FLAGS]
4987 Return a list of match objects, one for every non-overlapping, maximal
4988 match of REGEXP in STRING. The matches appear in left-to-right order.
4989 list-matches only reports matches of the empty string if there are no
4990 other matches which begin on, end at, or include the empty match's
4993 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4995 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
4997 For each match of REGEXP in STRING, apply PROC to the match object,
4998 and the last value PROC returned, or INIT for the first call. Return
4999 the last value returned by PROC. We apply PROC to the matches as they
5000 appear from left to right.
5002 This function recognizes matches according to the same criteria as
5005 Thus, you could define list-matches like this:
5007 (define (list-matches regexp string . flags)
5008 (reverse! (apply fold-matches regexp string '() cons flags)))
5010 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
5014 *** New function: hook? OBJ
5016 Return #t if OBJ is a hook, otherwise #f.
5018 *** New function: make-hook-with-name NAME [ARITY]
5020 Return a hook with name NAME and arity ARITY. The default value for
5021 ARITY is 0. The only effect of NAME is that it will appear when the
5022 hook object is printed to ease debugging.
5024 *** New function: hook-empty? HOOK
5026 Return #t if HOOK doesn't contain any procedures, otherwise #f.
5028 *** New function: hook->list HOOK
5030 Return a list of the procedures that are called when run-hook is
5033 ** `map' signals an error if its argument lists are not all the same length.
5035 This is the behavior required by R5RS, so this change is really a bug
5036 fix. But it seems to affect a lot of people's code, so we're
5037 mentioning it here anyway.
5039 ** Print-state handling has been made more transparent
5041 Under certain circumstances, ports are represented as a port with an
5042 associated print state. Earlier, this pair was represented as a pair
5043 (see "Some magic has been added to the printer" below). It is now
5044 indistinguishable (almost; see `get-print-state') from a port on the
5047 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
5049 Return a new port with the associated print state PRINT-STATE.
5051 *** New function: get-print-state OUTPUT-PORT
5053 Return the print state associated with this port if it exists,
5054 otherwise return #f.
5056 *** New function: directory-stream? OBJECT
5058 Returns true iff OBJECT is a directory stream --- the sort of object
5059 returned by `opendir'.
5061 ** New function: using-readline?
5063 Return #t if readline is in use in the current repl.
5065 ** structs will be removed in 1.4
5067 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
5068 and use GOOPS objects as the fundamental record type.
5070 * Changes to the scm_ interface
5072 ** structs will be removed in 1.4
5074 The entire current struct interface (struct.c, struct.h) will be
5075 replaced in Guile 1.4. We will merge GOOPS into libguile and use
5076 GOOPS objects as the fundamental record type.
5078 ** The internal representation of subr's has changed
5080 Instead of giving a hint to the subr name, the CAR field of the subr
5081 now contains an index to a subr entry in scm_subr_table.
5083 *** New variable: scm_subr_table
5085 An array of subr entries. A subr entry contains the name, properties
5086 and documentation associated with the subr. The properties and
5087 documentation slots are not yet used.
5089 ** A new scheme for "forwarding" calls to a builtin to a generic function
5091 It is now possible to extend the functionality of some Guile
5092 primitives by letting them defer a call to a GOOPS generic function on
5093 argument mismatch. This means that there is no loss of efficiency in
5098 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
5099 (define-method + ((x <string>) (y <string>))
5100 (string-append x y))
5102 + will still be as efficient as usual in numerical calculations, but
5103 can also be used for concatenating strings.
5105 Who will be the first one to extend Guile's numerical tower to
5106 rationals? :) [OK, there a few other things to fix before this can
5107 be made in a clean way.]
5109 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
5111 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
5113 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
5115 These do the same job as SCM_PROC and SCM_PROC1, but they also define
5116 a variable GENERIC which can be used by the dispatch macros below.
5118 [This is experimental code which may change soon.]
5120 *** New macros for forwarding control to a generic on arg type error
5122 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
5124 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
5126 These correspond to the scm_wta function call, and have the same
5127 behaviour until the user has called the GOOPS primitive
5128 `enable-primitive-generic!'. After that, these macros will apply the
5129 generic function GENERIC to the argument(s) instead of calling
5132 [This is experimental code which may change soon.]
5134 *** New macros for argument testing with generic dispatch
5136 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
5138 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
5140 These correspond to the SCM_ASSERT macro, but will defer control to
5141 GENERIC on error after `enable-primitive-generic!' has been called.
5143 [This is experimental code which may change soon.]
5145 ** New function: SCM scm_eval_body (SCM body, SCM env)
5147 Evaluates the body of a special form.
5149 ** The internal representation of struct's has changed
5151 Previously, four slots were allocated for the procedure(s) of entities
5152 and operators. The motivation for this representation had to do with
5153 the structure of the evaluator, the wish to support tail-recursive
5154 generic functions, and efficiency. Since the generic function
5155 dispatch mechanism has changed, there is no longer a need for such an
5156 expensive representation, and the representation has been simplified.
5158 This should not make any difference for most users.
5160 ** GOOPS support has been cleaned up.
5162 Some code has been moved from eval.c to objects.c and code in both of
5163 these compilation units has been cleaned up and better structured.
5165 *** New functions for applying generic functions
5167 New function: SCM scm_apply_generic (GENERIC, ARGS)
5168 New function: SCM scm_call_generic_0 (GENERIC)
5169 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
5170 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
5171 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
5173 ** Deprecated function: scm_make_named_hook
5175 It is now replaced by:
5177 ** New function: SCM scm_create_hook (const char *name, int arity)
5179 Creates a hook in the same way as make-hook above but also
5180 binds a variable named NAME to it.
5182 This is the typical way of creating a hook from C code.
5184 Currently, the variable is created in the "current" module.
5185 This might change when we get the new module system.
5187 [The behaviour is identical to scm_make_named_hook.]
5191 Changes since Guile 1.3:
5193 * Changes to mailing lists
5195 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
5197 See the README file to find current addresses for all the Guile
5200 * Changes to the distribution
5202 ** Readline support is no longer included with Guile by default.
5204 Based on the different license terms of Guile and Readline, we
5205 concluded that Guile should not *by default* cause the linking of
5206 Readline into an application program. Readline support is now offered
5207 as a separate module, which is linked into an application only when
5208 you explicitly specify it.
5210 Although Guile is GNU software, its distribution terms add a special
5211 exception to the usual GNU General Public License (GPL). Guile's
5212 license includes a clause that allows you to link Guile with non-free
5213 programs. We add this exception so as not to put Guile at a
5214 disadvantage vis-a-vis other extensibility packages that support other
5217 In contrast, the GNU Readline library is distributed under the GNU
5218 General Public License pure and simple. This means that you may not
5219 link Readline, even dynamically, into an application unless it is
5220 distributed under a free software license that is compatible the GPL.
5222 Because of this difference in distribution terms, an application that
5223 can use Guile may not be able to use Readline. Now users will be
5224 explicitly offered two independent decisions about the use of these
5227 You can activate the readline support by issuing
5229 (use-modules (readline-activator))
5232 from your ".guile" file, for example.
5234 * Changes to the stand-alone interpreter
5236 ** All builtins now print as primitives.
5237 Previously builtin procedures not belonging to the fundamental subr
5238 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
5239 Now, they print as #<primitive-procedure NAME>.
5241 ** Backtraces slightly more intelligible.
5242 gsubr-apply and macro transformer application frames no longer appear
5245 * Changes to Scheme functions and syntax
5247 ** Guile now correctly handles internal defines by rewriting them into
5248 their equivalent letrec. Previously, internal defines would
5249 incrementally add to the innermost environment, without checking
5250 whether the restrictions specified in RnRS were met. This lead to the
5251 correct behaviour when these restriction actually were met, but didn't
5252 catch all illegal uses. Such an illegal use could lead to crashes of
5253 the Guile interpreter or or other unwanted results. An example of
5254 incorrect internal defines that made Guile behave erratically:
5266 The problem with this example is that the definition of `c' uses the
5267 value of `b' directly. This confuses the meoization machine of Guile
5268 so that the second call of `b' (this time in a larger environment that
5269 also contains bindings for `c' and `d') refers to the binding of `c'
5270 instead of `a'. You could also make Guile crash with a variation on
5275 (define (b flag) (if flag a 1))
5276 (define c (1+ (b flag)))
5284 From now on, Guile will issue an `Unbound variable: b' error message
5289 A hook contains a list of functions which should be called on
5290 particular occasions in an existing program. Hooks are used for
5293 A window manager might have a hook before-window-map-hook. The window
5294 manager uses the function run-hooks to call all functions stored in
5295 before-window-map-hook each time a window is mapped. The user can
5296 store functions in the hook using add-hook!.
5298 In Guile, hooks are first class objects.
5300 *** New function: make-hook [N_ARGS]
5302 Return a hook for hook functions which can take N_ARGS arguments.
5303 The default value for N_ARGS is 0.
5305 (See also scm_make_named_hook below.)
5307 *** New function: add-hook! HOOK PROC [APPEND_P]
5309 Put PROC at the beginning of the list of functions stored in HOOK.
5310 If APPEND_P is supplied, and non-false, put PROC at the end instead.
5312 PROC must be able to take the number of arguments specified when the
5315 If PROC already exists in HOOK, then remove it first.
5317 *** New function: remove-hook! HOOK PROC
5319 Remove PROC from the list of functions in HOOK.
5321 *** New function: reset-hook! HOOK
5323 Clear the list of hook functions stored in HOOK.
5325 *** New function: run-hook HOOK ARG1 ...
5327 Run all hook functions stored in HOOK with arguments ARG1 ... .
5328 The number of arguments supplied must correspond to the number given
5329 when the hook was created.
5331 ** The function `dynamic-link' now takes optional keyword arguments.
5332 The only keyword argument that is currently defined is `:global
5333 BOOL'. With it, you can control whether the shared library will be
5334 linked in global mode or not. In global mode, the symbols from the
5335 linked library can be used to resolve references from other
5336 dynamically linked libraries. In non-global mode, the linked
5337 library is essentially invisible and can only be accessed via
5338 `dynamic-func', etc. The default is now to link in global mode.
5339 Previously, the default has been non-global mode.
5341 The `#:global' keyword is only effective on platforms that support
5342 the dlopen family of functions.
5344 ** New function `provided?'
5346 - Function: provided? FEATURE
5347 Return true iff FEATURE is supported by this installation of
5348 Guile. FEATURE must be a symbol naming a feature; the global
5349 variable `*features*' is a list of available features.
5351 ** Changes to the module (ice-9 expect):
5353 *** The expect-strings macro now matches `$' in a regular expression
5354 only at a line-break or end-of-file by default. Previously it would
5355 match the end of the string accumulated so far. The old behaviour
5356 can be obtained by setting the variable `expect-strings-exec-flags'
5359 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
5360 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
5361 in a regular expression will still match before a line-break or
5362 end-of-file. The default is `regexp/noteol'.
5364 *** The expect-strings macro now uses a variable
5365 `expect-strings-compile-flags' for the flags to be supplied to
5366 `make-regexp'. The default is `regexp/newline', which was previously
5369 *** The expect macro now supplies two arguments to a match procedure:
5370 the current accumulated string and a flag to indicate whether
5371 end-of-file has been reached. Previously only the string was supplied.
5372 If end-of-file is reached, the match procedure will be called an
5373 additional time with the same accumulated string as the previous call
5374 but with the flag set.
5376 ** New module (ice-9 format), implementing the Common Lisp `format' function.
5378 This code, and the documentation for it that appears here, was
5379 borrowed from SLIB, with minor adaptations for Guile.
5381 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
5382 An almost complete implementation of Common LISP format description
5383 according to the CL reference book `Common LISP' from Guy L.
5384 Steele, Digital Press. Backward compatible to most of the
5385 available Scheme format implementations.
5387 Returns `#t', `#f' or a string; has side effect of printing
5388 according to FORMAT-STRING. If DESTINATION is `#t', the output is
5389 to the current output port and `#t' is returned. If DESTINATION
5390 is `#f', a formatted string is returned as the result of the call.
5391 NEW: If DESTINATION is a string, DESTINATION is regarded as the
5392 format string; FORMAT-STRING is then the first argument and the
5393 output is returned as a string. If DESTINATION is a number, the
5394 output is to the current error port if available by the
5395 implementation. Otherwise DESTINATION must be an output port and
5398 FORMAT-STRING must be a string. In case of a formatting error
5399 format returns `#f' and prints a message on the current output or
5400 error port. Characters are output as if the string were output by
5401 the `display' function with the exception of those prefixed by a
5402 tilde (~). For a detailed description of the FORMAT-STRING syntax
5403 please consult a Common LISP format reference manual. For a test
5404 suite to verify this format implementation load `formatst.scm'.
5405 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
5407 Note: `format' is not reentrant, i.e. only one `format'-call may
5408 be executed at a time.
5411 *** Format Specification (Format version 3.0)
5413 Please consult a Common LISP format reference manual for a detailed
5414 description of the format string syntax. For a demonstration of the
5415 implemented directives see `formatst.scm'.
5417 This implementation supports directive parameters and modifiers (`:'
5418 and `@' characters). Multiple parameters must be separated by a comma
5419 (`,'). Parameters can be numerical parameters (positive or negative),
5420 character parameters (prefixed by a quote character (`''), variable
5421 parameters (`v'), number of rest arguments parameter (`#'), empty and
5422 default parameters. Directive characters are case independent. The
5423 general form of a directive is:
5425 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
5427 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
5429 *** Implemented CL Format Control Directives
5431 Documentation syntax: Uppercase characters represent the
5432 corresponding control directive characters. Lowercase characters
5433 represent control directive parameter descriptions.
5436 Any (print as `display' does).
5440 `~MINCOL,COLINC,MINPAD,PADCHARA'
5444 S-expression (print as `write' does).
5448 `~MINCOL,COLINC,MINPAD,PADCHARS'
5454 print number sign always.
5457 print comma separated.
5459 `~MINCOL,PADCHAR,COMMACHARD'
5465 print number sign always.
5468 print comma separated.
5470 `~MINCOL,PADCHAR,COMMACHARX'
5476 print number sign always.
5479 print comma separated.
5481 `~MINCOL,PADCHAR,COMMACHARO'
5487 print number sign always.
5490 print comma separated.
5492 `~MINCOL,PADCHAR,COMMACHARB'
5497 `~N,MINCOL,PADCHAR,COMMACHARR'
5501 print a number as a Roman numeral.
5504 print a number as an "old fashioned" Roman numeral.
5507 print a number as an ordinal English number.
5510 print a number as a cardinal English number.
5515 prints `y' and `ies'.
5518 as `~P but jumps 1 argument backward.'
5521 as `~@P but jumps 1 argument backward.'
5526 prints a character as the reader can understand it (i.e. `#\'
5530 prints a character as emacs does (eg. `^C' for ASCII 03).
5533 Fixed-format floating-point (prints a flonum like MMM.NNN).
5534 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
5536 If the number is positive a plus sign is printed.
5539 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
5540 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
5542 If the number is positive a plus sign is printed.
5545 General floating-point (prints a flonum either fixed or
5547 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
5549 If the number is positive a plus sign is printed.
5552 Dollars floating-point (prints a flonum in fixed with signs
5554 `~DIGITS,SCALE,WIDTH,PADCHAR$'
5556 If the number is positive a plus sign is printed.
5559 A sign is always printed and appears before the padding.
5562 The sign appears before the padding.
5570 print newline if not at the beginning of the output line.
5572 prints `~&' and then N-1 newlines.
5577 print N page separators.
5587 newline is ignored, white space left.
5590 newline is left, white space ignored.
5595 relative tabulation.
5601 Indirection (expects indirect arguments as a list).
5603 extracts indirect arguments from format arguments.
5606 Case conversion (converts by `string-downcase').
5608 converts by `string-capitalize'.
5611 converts by `string-capitalize-first'.
5614 converts by `string-upcase'.
5617 Argument Jumping (jumps 1 argument forward).
5619 jumps N arguments forward.
5622 jumps 1 argument backward.
5625 jumps N arguments backward.
5628 jumps to the 0th argument.
5631 jumps to the Nth argument (beginning from 0)
5633 `~[STR0~;STR1~;...~;STRN~]'
5634 Conditional Expression (numerical clause conditional).
5636 take argument from N.
5639 true test conditional.
5642 if-else-then conditional.
5648 default clause follows.
5651 Iteration (args come from the next argument (a list)).
5653 at most N iterations.
5656 args from next arg (a list of lists).
5659 args from the rest of arguments.
5662 args from the rest args (lists).
5673 aborts if N <= M <= K
5675 *** Not Implemented CL Format Control Directives
5678 print `#f' as an empty list (see below).
5681 print `#f' as an empty list (see below).
5687 (sorry I don't understand its semantics completely)
5689 *** Extended, Replaced and Additional Control Directives
5691 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
5692 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
5693 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
5694 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
5695 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
5696 COMMAWIDTH is the number of characters between two comma
5700 print a R4RS complex number as `~F~@Fi' with passed parameters for
5704 Pretty print formatting of an argument for scheme code lists.
5710 Flushes the output if format DESTINATION is a port.
5713 Print a `#\space' character
5715 print N `#\space' characters.
5718 Print a `#\tab' character
5720 print N `#\tab' characters.
5723 Takes N as an integer representation for a character. No arguments
5724 are consumed. N is converted to a character by `integer->char'. N
5725 must be a positive decimal number.
5728 Print out readproof. Prints out internal objects represented as
5729 `#<...>' as strings `"#<...>"' so that the format output can always
5730 be processed by `read'.
5733 Print out readproof. Prints out internal objects represented as
5734 `#<...>' as strings `"#<...>"' so that the format output can always
5735 be processed by `read'.
5738 Prints information and a copyright notice on the format
5741 prints format version.
5744 may also print number strings, i.e. passing a number as a string
5745 and format it accordingly.
5747 *** Configuration Variables
5749 The format module exports some configuration variables to suit the
5750 systems and users needs. There should be no modification necessary for
5751 the configuration that comes with Guile. Format detects automatically
5752 if the running scheme system implements floating point numbers and
5755 format:symbol-case-conv
5756 Symbols are converted by `symbol->string' so the case type of the
5757 printed symbols is implementation dependent.
5758 `format:symbol-case-conv' is a one arg closure which is either
5759 `#f' (no conversion), `string-upcase', `string-downcase' or
5760 `string-capitalize'. (default `#f')
5762 format:iobj-case-conv
5763 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
5764 implementation internal objects. (default `#f')
5767 The character prefixing the exponent value in `~E' printing.
5770 *** Compatibility With Other Format Implementations
5776 Downward compatible except for padding support and `~A', `~S',
5777 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
5778 `printf' padding support which is completely replaced by the CL
5779 `format' padding style.
5782 Downward compatible except for `~', which is not documented
5783 (ignores all characters inside the format string up to a newline
5784 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
5785 numerical and variable parameters and `:/@' modifiers in the CL
5789 Downward compatible except for `~A' and `~S' which print in
5790 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
5791 directive parameters or modifiers)).
5794 Downward compatible except for an optional destination parameter:
5795 S2C accepts a format call without a destination which returns a
5796 formatted string. This is equivalent to a #f destination in S2C.
5797 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
5798 parameters or modifiers)).
5801 ** Changes to string-handling functions.
5803 These functions were added to support the (ice-9 format) module, above.
5805 *** New function: string-upcase STRING
5806 *** New function: string-downcase STRING
5808 These are non-destructive versions of the existing string-upcase! and
5809 string-downcase! functions.
5811 *** New function: string-capitalize! STRING
5812 *** New function: string-capitalize STRING
5814 These functions convert the first letter of each word in the string to
5817 (string-capitalize "howdy there")
5820 As with the other functions, string-capitalize! modifies the string in
5821 place, while string-capitalize returns a modified copy of its argument.
5823 *** New function: string-ci->symbol STRING
5825 Return a symbol whose name is STRING, but having the same case as if
5826 the symbol had be read by `read'.
5828 Guile can be configured to be sensitive or insensitive to case
5829 differences in Scheme identifiers. If Guile is case-insensitive, all
5830 symbols are converted to lower case on input. The `string-ci->symbol'
5831 function returns a symbol whose name in STRING, transformed as Guile
5832 would if STRING were input.
5834 *** New function: substring-move! STRING1 START END STRING2 START
5836 Copy the substring of STRING1 from START (inclusive) to END
5837 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
5838 string, and the source and destination areas may overlap; in all
5839 cases, the function behaves as if all the characters were copied
5842 *** Extended functions: substring-move-left! substring-move-right!
5844 These functions now correctly copy arbitrarily overlapping substrings;
5845 they are both synonyms for substring-move!.
5848 ** New module (ice-9 getopt-long), with the function `getopt-long'.
5850 getopt-long is a function for parsing command-line arguments in a
5851 manner consistent with other GNU programs.
5853 (getopt-long ARGS GRAMMAR)
5854 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
5856 ARGS should be a list of strings. Its first element should be the
5857 name of the program; subsequent elements should be the arguments
5858 that were passed to the program on the command line. The
5859 `program-arguments' procedure returns a list of this form.
5861 GRAMMAR is a list of the form:
5862 ((OPTION (PROPERTY VALUE) ...) ...)
5864 Each OPTION should be a symbol. `getopt-long' will accept a
5865 command-line option named `--OPTION'.
5866 Each option can have the following (PROPERTY VALUE) pairs:
5868 (single-char CHAR) --- Accept `-CHAR' as a single-character
5869 equivalent to `--OPTION'. This is how to specify traditional
5871 (required? BOOL) --- If BOOL is true, the option is required.
5872 getopt-long will raise an error if it is not found in ARGS.
5873 (value BOOL) --- If BOOL is #t, the option accepts a value; if
5874 it is #f, it does not; and if it is the symbol
5875 `optional', the option may appear in ARGS with or
5877 (predicate FUNC) --- If the option accepts a value (i.e. you
5878 specified `(value #t)' for this option), then getopt
5879 will apply FUNC to the value, and throw an exception
5880 if it returns #f. FUNC should be a procedure which
5881 accepts a string and returns a boolean value; you may
5882 need to use quasiquotes to get it into GRAMMAR.
5884 The (PROPERTY VALUE) pairs may occur in any order, but each
5885 property may occur only once. By default, options do not have
5886 single-character equivalents, are not required, and do not take
5889 In ARGS, single-character options may be combined, in the usual
5890 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
5891 accepts values, then it must be the last option in the
5892 combination; the value is the next argument. So, for example, using
5893 the following grammar:
5894 ((apples (single-char #\a))
5895 (blimps (single-char #\b) (value #t))
5896 (catalexis (single-char #\c) (value #t)))
5897 the following argument lists would be acceptable:
5898 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
5899 for "blimps" and "catalexis")
5900 ("-ab" "bang" "-c" "couth") (same)
5901 ("-ac" "couth" "-b" "bang") (same)
5902 ("-abc" "couth" "bang") (an error, since `-b' is not the
5903 last option in its combination)
5905 If an option's value is optional, then `getopt-long' decides
5906 whether it has a value by looking at what follows it in ARGS. If
5907 the next element is a string, and it does not appear to be an
5908 option itself, then that string is the option's value.
5910 The value of a long option can appear as the next element in ARGS,
5911 or it can follow the option name, separated by an `=' character.
5912 Thus, using the same grammar as above, the following argument lists
5914 ("--apples" "Braeburn" "--blimps" "Goodyear")
5915 ("--apples=Braeburn" "--blimps" "Goodyear")
5916 ("--blimps" "Goodyear" "--apples=Braeburn")
5918 If the option "--" appears in ARGS, argument parsing stops there;
5919 subsequent arguments are returned as ordinary arguments, even if
5920 they resemble options. So, in the argument list:
5921 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
5922 `getopt-long' will recognize the `apples' option as having the
5923 value "Granny Smith", but it will not recognize the `blimp'
5924 option; it will return the strings "--blimp" and "Goodyear" as
5925 ordinary argument strings.
5927 The `getopt-long' function returns the parsed argument list as an
5928 assocation list, mapping option names --- the symbols from GRAMMAR
5929 --- onto their values, or #t if the option does not accept a value.
5930 Unused options do not appear in the alist.
5932 All arguments that are not the value of any option are returned
5933 as a list, associated with the empty list.
5935 `getopt-long' throws an exception if:
5936 - it finds an unrecognized option in ARGS
5937 - a required option is omitted
5938 - an option that requires an argument doesn't get one
5939 - an option that doesn't accept an argument does get one (this can
5940 only happen using the long option `--opt=value' syntax)
5941 - an option predicate fails
5946 `((lockfile-dir (required? #t)
5949 (predicate ,file-is-directory?))
5950 (verbose (required? #f)
5953 (x-includes (single-char #\x))
5954 (rnet-server (single-char #\y)
5955 (predicate ,string?))))
5957 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
5958 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
5960 => ((() "foo1" "-fred" "foo2" "foo3")
5961 (rnet-server . "lamprod")
5962 (x-includes . "/usr/include")
5963 (lockfile-dir . "/tmp")
5966 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
5968 It will be removed in a few releases.
5970 ** New syntax: lambda*
5971 ** New syntax: define*
5972 ** New syntax: define*-public
5973 ** New syntax: defmacro*
5974 ** New syntax: defmacro*-public
5975 Guile now supports optional arguments.
5977 `lambda*', `define*', `define*-public', `defmacro*' and
5978 `defmacro*-public' are identical to the non-* versions except that
5979 they use an extended type of parameter list that has the following BNF
5980 syntax (parentheses are literal, square brackets indicate grouping,
5981 and `*', `+' and `?' have the usual meaning):
5983 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
5984 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
5985 [[#&rest identifier]|[. identifier]]? ) | [identifier]
5987 ext-var-decl ::= identifier | ( identifier expression )
5989 The semantics are best illustrated with the following documentation
5990 and examples for `lambda*':
5993 lambda extended for optional and keyword arguments
5995 lambda* creates a procedure that takes optional arguments. These
5996 are specified by putting them inside brackets at the end of the
5997 paramater list, but before any dotted rest argument. For example,
5998 (lambda* (a b #&optional c d . e) '())
5999 creates a procedure with fixed arguments a and b, optional arguments c
6000 and d, and rest argument e. If the optional arguments are omitted
6001 in a call, the variables for them are unbound in the procedure. This
6002 can be checked with the bound? macro.
6004 lambda* can also take keyword arguments. For example, a procedure
6006 (lambda* (#&key xyzzy larch) '())
6007 can be called with any of the argument lists (#:xyzzy 11)
6008 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
6009 are given as keywords are bound to values.
6011 Optional and keyword arguments can also be given default values
6012 which they take on when they are not present in a call, by giving a
6013 two-item list in place of an optional argument, for example in:
6014 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
6015 foo is a fixed argument, bar is an optional argument with default
6016 value 42, and baz is a keyword argument with default value 73.
6017 Default value expressions are not evaluated unless they are needed
6018 and until the procedure is called.
6020 lambda* now supports two more special parameter list keywords.
6022 lambda*-defined procedures now throw an error by default if a
6023 keyword other than one of those specified is found in the actual
6024 passed arguments. However, specifying #&allow-other-keys
6025 immediately after the kyword argument declarations restores the
6026 previous behavior of ignoring unknown keywords. lambda* also now
6027 guarantees that if the same keyword is passed more than once, the
6028 last one passed is the one that takes effect. For example,
6029 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
6030 #:heads 37 #:tails 42 #:heads 99)
6031 would result in (99 47) being displayed.
6033 #&rest is also now provided as a synonym for the dotted syntax rest
6034 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
6035 all respects to lambda*. This is provided for more similarity to DSSSL,
6036 MIT-Scheme and Kawa among others, as well as for refugees from other
6039 Further documentation may be found in the optargs.scm file itself.
6041 The optional argument module also exports the macros `let-optional',
6042 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
6043 are not documented here because they may be removed in the future, but
6044 full documentation is still available in optargs.scm.
6046 ** New syntax: and-let*
6047 Guile now supports the `and-let*' form, described in the draft SRFI-2.
6049 Syntax: (land* (<clause> ...) <body> ...)
6050 Each <clause> should have one of the following forms:
6051 (<variable> <expression>)
6054 Each <variable> or <bound-variable> should be an identifier. Each
6055 <expression> should be a valid expression. The <body> should be a
6056 possibly empty sequence of expressions, like the <body> of a
6059 Semantics: A LAND* expression is evaluated by evaluating the
6060 <expression> or <bound-variable> of each of the <clause>s from
6061 left to right. The value of the first <expression> or
6062 <bound-variable> that evaluates to a false value is returned; the
6063 remaining <expression>s and <bound-variable>s are not evaluated.
6064 The <body> forms are evaluated iff all the <expression>s and
6065 <bound-variable>s evaluate to true values.
6067 The <expression>s and the <body> are evaluated in an environment
6068 binding each <variable> of the preceding (<variable> <expression>)
6069 clauses to the value of the <expression>. Later bindings
6070 shadow earlier bindings.
6072 Guile's and-let* macro was contributed by Michael Livshin.
6074 ** New sorting functions
6076 *** New function: sorted? SEQUENCE LESS?
6077 Returns `#t' when the sequence argument is in non-decreasing order
6078 according to LESS? (that is, there is no adjacent pair `... x y
6079 ...' for which `(less? y x)').
6081 Returns `#f' when the sequence contains at least one out-of-order
6082 pair. It is an error if the sequence is neither a list nor a
6085 *** New function: merge LIST1 LIST2 LESS?
6086 LIST1 and LIST2 are sorted lists.
6087 Returns the sorted list of all elements in LIST1 and LIST2.
6089 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
6090 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
6091 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
6092 (Here "<" should read "comes before".)
6094 *** New procedure: merge! LIST1 LIST2 LESS?
6095 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
6096 the result. If the code is compiled, and LESS? constructs no new
6097 pairs, no pairs at all will be allocated. The first pair of the
6098 result will be either the first pair of LIST1 or the first pair of
6101 *** New function: sort SEQUENCE LESS?
6102 Accepts either a list or a vector, and returns a new sequence
6103 which is sorted. The new sequence is the same type as the input.
6104 Always `(sorted? (sort sequence less?) less?)'. The original
6105 sequence is not altered in any way. The new sequence shares its
6106 elements with the old one; no elements are copied.
6108 *** New procedure: sort! SEQUENCE LESS
6109 Returns its sorted result in the original boxes. No new storage is
6110 allocated at all. Proper usage: (set! slist (sort! slist <))
6112 *** New function: stable-sort SEQUENCE LESS?
6113 Similar to `sort' but stable. That is, if "equal" elements are
6114 ordered a < b in the original sequence, they will have the same order
6117 *** New function: stable-sort! SEQUENCE LESS?
6118 Similar to `sort!' but stable.
6119 Uses temporary storage when sorting vectors.
6121 *** New functions: sort-list, sort-list!
6122 Added for compatibility with scsh.
6124 ** New built-in random number support
6126 *** New function: random N [STATE]
6127 Accepts a positive integer or real N and returns a number of the
6128 same type between zero (inclusive) and N (exclusive). The values
6129 returned have a uniform distribution.
6131 The optional argument STATE must be of the type produced by
6132 `copy-random-state' or `seed->random-state'. It defaults to the value
6133 of the variable `*random-state*'. This object is used to maintain the
6134 state of the pseudo-random-number generator and is altered as a side
6135 effect of the `random' operation.
6137 *** New variable: *random-state*
6138 Holds a data structure that encodes the internal state of the
6139 random-number generator that `random' uses by default. The nature
6140 of this data structure is implementation-dependent. It may be
6141 printed out and successfully read back in, but may or may not
6142 function correctly as a random-number state object in another
6145 *** New function: copy-random-state [STATE]
6146 Returns a new object of type suitable for use as the value of the
6147 variable `*random-state*' and as a second argument to `random'.
6148 If argument STATE is given, a copy of it is returned. Otherwise a
6149 copy of `*random-state*' is returned.
6151 *** New function: seed->random-state SEED
6152 Returns a new object of type suitable for use as the value of the
6153 variable `*random-state*' and as a second argument to `random'.
6154 SEED is a string or a number. A new state is generated and
6155 initialized using SEED.
6157 *** New function: random:uniform [STATE]
6158 Returns an uniformly distributed inexact real random number in the
6159 range between 0 and 1.
6161 *** New procedure: random:solid-sphere! VECT [STATE]
6162 Fills VECT with inexact real random numbers the sum of whose
6163 squares is less than 1.0. Thinking of VECT as coordinates in
6164 space of dimension N = `(vector-length VECT)', the coordinates are
6165 uniformly distributed within the unit N-shere. The sum of the
6166 squares of the numbers is returned. VECT can be either a vector
6167 or a uniform vector of doubles.
6169 *** New procedure: random:hollow-sphere! VECT [STATE]
6170 Fills VECT with inexact real random numbers the sum of whose squares
6171 is equal to 1.0. Thinking of VECT as coordinates in space of
6172 dimension n = `(vector-length VECT)', the coordinates are uniformly
6173 distributed over the surface of the unit n-shere. VECT can be either
6174 a vector or a uniform vector of doubles.
6176 *** New function: random:normal [STATE]
6177 Returns an inexact real in a normal distribution with mean 0 and
6178 standard deviation 1. For a normal distribution with mean M and
6179 standard deviation D use `(+ M (* D (random:normal)))'.
6181 *** New procedure: random:normal-vector! VECT [STATE]
6182 Fills VECT with inexact real random numbers which are independent and
6183 standard normally distributed (i.e., with mean 0 and variance 1).
6184 VECT can be either a vector or a uniform vector of doubles.
6186 *** New function: random:exp STATE
6187 Returns an inexact real in an exponential distribution with mean 1.
6188 For an exponential distribution with mean U use (* U (random:exp)).
6190 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
6192 These functions now operate on numbers in the range of a C unsigned
6195 These functions used to operate on numbers in the range of a C signed
6196 long; however, this seems inappropriate, because Guile integers don't
6199 ** New function: make-guardian
6200 This is an implementation of guardians as described in
6201 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
6202 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
6203 Programming Language Design and Implementation, June 1993
6204 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
6206 ** New functions: delq1!, delv1!, delete1!
6207 These procedures behave similar to delq! and friends but delete only
6208 one object if at all.
6210 ** New function: unread-string STRING PORT
6211 Unread STRING to PORT, that is, push it back onto the port so that
6212 next read operation will work on the pushed back characters.
6214 ** unread-char can now be called multiple times
6215 If unread-char is called multiple times, the unread characters will be
6216 read again in last-in first-out order.
6218 ** the procedures uniform-array-read! and uniform-array-write! now
6219 work on any kind of port, not just ports which are open on a file.
6221 ** Now 'l' in a port mode requests line buffering.
6223 ** The procedure truncate-file now works on string ports as well
6224 as file ports. If the size argument is omitted, the current
6225 file position is used.
6227 ** new procedure: seek PORT/FDES OFFSET WHENCE
6228 The arguments are the same as for the old fseek procedure, but it
6229 works on string ports as well as random-access file ports.
6231 ** the fseek procedure now works on string ports, since it has been
6232 redefined using seek.
6234 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
6235 size is not supplied.
6237 ** the newline procedure no longer flushes the port if it's not
6238 line-buffered: previously it did if it was the current output port.
6240 ** open-pipe and close-pipe are no longer primitive procedures, but
6241 an emulation can be obtained using `(use-modules (ice-9 popen))'.
6243 ** the freopen procedure has been removed.
6245 ** new procedure: drain-input PORT
6246 Drains PORT's read buffers (including any pushed-back characters)
6247 and returns the contents as a single string.
6249 ** New function: map-in-order PROC LIST1 LIST2 ...
6250 Version of `map' which guarantees that the procedure is applied to the
6251 lists in serial order.
6253 ** Renamed `serial-array-copy!' and `serial-array-map!' to
6254 `array-copy-in-order!' and `array-map-in-order!'. The old names are
6255 now obsolete and will go away in release 1.5.
6257 ** New syntax: collect BODY1 ...
6258 Version of `begin' which returns a list of the results of the body
6259 forms instead of the result of the last body form. In contrast to
6260 `begin', `collect' allows an empty body.
6262 ** New functions: read-history FILENAME, write-history FILENAME
6263 Read/write command line history from/to file. Returns #t on success
6264 and #f if an error occured.
6266 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
6268 These procedures return a list of definitions available in the specified
6269 argument, a relative module reference. In the case of no argument,
6270 `(current-module)' is now consulted for definitions to return, instead
6271 of simply returning #f, the former behavior.
6273 ** The #/ syntax for lists is no longer supported.
6275 Earlier versions of Scheme accepted this syntax, but printed a
6278 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
6280 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
6283 * Changes to the gh_ interface
6287 Now takes a second argument which is the result array. If this
6288 pointer is NULL, a new array is malloced (the old behaviour).
6290 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
6291 gh_scm2shorts, gh_scm2longs, gh_scm2floats
6295 * Changes to the scm_ interface
6297 ** Function: scm_make_named_hook (char* name, int n_args)
6299 Creates a hook in the same way as make-hook above but also
6300 binds a variable named NAME to it.
6302 This is the typical way of creating a hook from C code.
6304 Currently, the variable is created in the "current" module. This
6305 might change when we get the new module system.
6307 ** The smob interface
6309 The interface for creating smobs has changed. For documentation, see
6310 data-rep.info (made from guile-core/doc/data-rep.texi).
6312 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
6314 >>> This function will be removed in 1.3.4. <<<
6318 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
6319 This function adds a new smob type, named NAME, with instance size
6320 SIZE to the system. The return value is a tag that is used in
6321 creating instances of the type. If SIZE is 0, then no memory will
6322 be allocated when instances of the smob are created, and nothing
6323 will be freed by the default free function.
6325 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
6326 This function sets the smob marking procedure for the smob type
6327 specified by the tag TC. TC is the tag returned by
6328 `scm_make_smob_type'.
6330 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
6331 This function sets the smob freeing procedure for the smob type
6332 specified by the tag TC. TC is the tag returned by
6333 `scm_make_smob_type'.
6335 *** Function: void scm_set_smob_print (tc, print)
6337 - Function: void scm_set_smob_print (long tc,
6338 scm_sizet (*print) (SCM,
6342 This function sets the smob printing procedure for the smob type
6343 specified by the tag TC. TC is the tag returned by
6344 `scm_make_smob_type'.
6346 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
6347 This function sets the smob equality-testing predicate for the
6348 smob type specified by the tag TC. TC is the tag returned by
6349 `scm_make_smob_type'.
6351 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
6352 Make VALUE contain a smob instance of the type with type code TC and
6353 smob data DATA. VALUE must be previously declared as C type `SCM'.
6355 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
6356 This macro expands to a block of code that creates a smob instance
6357 of the type with type code TC and smob data DATA, and returns that
6358 `SCM' value. It should be the last piece of code in a block.
6360 ** The interfaces for using I/O ports and implementing port types
6361 (ptobs) have changed significantly. The new interface is based on
6362 shared access to buffers and a new set of ptob procedures.
6364 *** scm_newptob has been removed
6368 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
6370 - Function: SCM scm_make_port_type (char *type_name,
6371 int (*fill_buffer) (SCM port),
6372 void (*write_flush) (SCM port));
6374 Similarly to the new smob interface, there is a set of function
6375 setters by which the user can customize the behaviour of his port
6376 type. See ports.h (scm_set_port_XXX).
6378 ** scm_strport_to_string: New function: creates a new string from
6379 a string port's buffer.
6381 ** Plug in interface for random number generators
6382 The variable `scm_the_rng' in random.c contains a value and three
6383 function pointers which together define the current random number
6384 generator being used by the Scheme level interface and the random
6385 number library functions.
6387 The user is free to replace the default generator with the generator
6390 *** Variable: size_t scm_the_rng.rstate_size
6391 The size of the random state type used by the current RNG
6394 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
6395 Given the random STATE, return 32 random bits.
6397 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
6398 Seed random state STATE using string S of length N.
6400 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
6401 Given random state STATE, return a malloced copy.
6404 The default RNG is the MWC (Multiply With Carry) random number
6405 generator described by George Marsaglia at the Department of
6406 Statistics and Supercomputer Computations Research Institute, The
6407 Florida State University (http://stat.fsu.edu/~geo).
6409 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
6410 passes all tests in the DIEHARD test suite
6411 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
6412 costs one multiply and one add on platforms which either supports long
6413 longs (gcc does this on most systems) or have 64 bit longs. The cost
6414 is four multiply on other systems but this can be optimized by writing
6415 scm_i_uniform32 in assembler.
6417 These functions are provided through the scm_the_rng interface for use
6418 by libguile and the application.
6420 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
6421 Given the random STATE, return 32 random bits.
6422 Don't use this function directly. Instead go through the plugin
6423 interface (see "Plug in interface" above).
6425 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
6426 Initialize STATE using SEED of length N.
6428 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
6429 Return a malloc:ed copy of STATE. This function can easily be re-used
6430 in the interfaces to other RNGs.
6432 ** Random number library functions
6433 These functions use the current RNG through the scm_the_rng interface.
6434 It might be a good idea to use these functions from your C code so
6435 that only one random generator is used by all code in your program.
6437 The default random state is stored in:
6439 *** Variable: SCM scm_var_random_state
6440 Contains the vcell of the Scheme variable "*random-state*" which is
6441 used as default state by all random number functions in the Scheme
6446 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
6448 *** Function: scm_rstate *scm_c_default_rstate (void)
6449 This is a convenience function which returns the value of
6450 scm_var_random_state. An error message is generated if this value
6451 isn't a random state.
6453 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
6454 Make a new random state from the string SEED of length LENGTH.
6456 It is generally not a good idea to use multiple random states in a
6457 program. While subsequent random numbers generated from one random
6458 state are guaranteed to be reasonably independent, there is no such
6459 guarantee for numbers generated from different random states.
6461 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
6462 Return 32 random bits.
6464 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
6465 Return a sample from the uniform(0,1) distribution.
6467 *** Function: double scm_c_normal01 (scm_rstate *STATE)
6468 Return a sample from the normal(0,1) distribution.
6470 *** Function: double scm_c_exp1 (scm_rstate *STATE)
6471 Return a sample from the exp(1) distribution.
6473 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
6474 Return a sample from the discrete uniform(0,M) distribution.
6476 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
6477 Return a sample from the discrete uniform(0,M) distribution.
6478 M must be a bignum object. The returned value may be an INUM.
6482 Changes in Guile 1.3 (released Monday, October 19, 1998):
6484 * Changes to the distribution
6486 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
6487 To avoid conflicts, programs should name environment variables after
6488 themselves, except when there's a common practice establishing some
6491 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
6492 giving the former precedence, and printing a warning message if the
6493 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
6495 ** The header files related to multi-byte characters have been removed.
6496 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
6497 which referred to these explicitly will probably need to be rewritten,
6498 since the support for the variant string types has been removed; see
6501 ** The header files append.h and sequences.h have been removed. These
6502 files implemented non-R4RS operations which would encourage
6503 non-portable programming style and less easy-to-read code.
6505 * Changes to the stand-alone interpreter
6507 ** New procedures have been added to implement a "batch mode":
6509 *** Function: batch-mode?
6511 Returns a boolean indicating whether the interpreter is in batch
6514 *** Function: set-batch-mode?! ARG
6516 If ARG is true, switches the interpreter to batch mode. The `#f'
6517 case has not been implemented.
6519 ** Guile now provides full command-line editing, when run interactively.
6520 To use this feature, you must have the readline library installed.
6521 The Guile build process will notice it, and automatically include
6524 The readline library is available via anonymous FTP from any GNU
6525 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
6527 ** the-last-stack is now a fluid.
6529 * Changes to the procedure for linking libguile with your programs
6531 ** You can now use the `guile-config' utility to build programs that use Guile.
6533 Guile now includes a command-line utility called `guile-config', which
6534 can provide information about how to compile and link programs that
6537 *** `guile-config compile' prints any C compiler flags needed to use Guile.
6538 You should include this command's output on the command line you use
6539 to compile C or C++ code that #includes the Guile header files. It's
6540 usually just a `-I' flag to help the compiler find the Guile headers.
6543 *** `guile-config link' prints any linker flags necessary to link with Guile.
6545 This command writes to its standard output a list of flags which you
6546 must pass to the linker to link your code against the Guile library.
6547 The flags include '-lguile' itself, any other libraries the Guile
6548 library depends upon, and any `-L' flags needed to help the linker
6549 find those libraries.
6551 For example, here is a Makefile rule that builds a program named 'foo'
6552 from the object files ${FOO_OBJECTS}, and links them against Guile:
6555 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
6557 Previous Guile releases recommended that you use autoconf to detect
6558 which of a predefined set of libraries were present on your system.
6559 It is more robust to use `guile-config', since it records exactly which
6560 libraries the installed Guile library requires.
6562 This was originally called `build-guile', but was renamed to
6563 `guile-config' before Guile 1.3 was released, to be consistent with
6564 the analogous script for the GTK+ GUI toolkit, which is called
6568 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
6570 If you are using the GNU autoconf package to configure your program,
6571 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
6572 (described above) and gather the necessary values for use in your
6575 The GUILE_FLAGS macro expands to configure script code which runs the
6576 `guile-config' script, to find out where Guile's header files and
6577 libraries are installed. It sets two variables, marked for
6578 substitution, as by AC_SUBST.
6580 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
6581 code that uses Guile header files. This is almost always just a
6584 GUILE_LDFLAGS --- flags to pass to the linker to link a
6585 program against Guile. This includes `-lguile' for the Guile
6586 library itself, any libraries that Guile itself requires (like
6587 -lqthreads), and so on. It may also include a -L flag to tell the
6588 compiler where to find the libraries.
6590 GUILE_FLAGS is defined in the file guile.m4, in the top-level
6591 directory of the Guile distribution. You can copy it into your
6592 package's aclocal.m4 file, and then use it in your configure.in file.
6594 If you are using the `aclocal' program, distributed with GNU automake,
6595 to maintain your aclocal.m4 file, the Guile installation process
6596 installs guile.m4 where aclocal will find it. All you need to do is
6597 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
6598 this will copy the definition of GUILE_FLAGS into your aclocal.m4
6602 * Changes to Scheme functions and syntax
6604 ** Multi-byte strings have been removed, as have multi-byte and wide
6605 ports. We felt that these were the wrong approach to
6606 internationalization support.
6608 ** New function: readline [PROMPT]
6609 Read a line from the terminal, and allow the user to edit it,
6610 prompting with PROMPT. READLINE provides a large set of Emacs-like
6611 editing commands, lets the user recall previously typed lines, and
6612 works on almost every kind of terminal, including dumb terminals.
6614 READLINE assumes that the cursor is at the beginning of the line when
6615 it is invoked. Thus, you can't print a prompt yourself, and then call
6616 READLINE; you need to package up your prompt as a string, pass it to
6617 the function, and let READLINE print the prompt itself. This is
6618 because READLINE needs to know the prompt's screen width.
6620 For Guile to provide this function, you must have the readline
6621 library, version 2.1 or later, installed on your system. Readline is
6622 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
6623 any GNU mirror site.
6625 See also ADD-HISTORY function.
6627 ** New function: add-history STRING
6628 Add STRING as the most recent line in the history used by the READLINE
6629 command. READLINE does not add lines to the history itself; you must
6630 call ADD-HISTORY to make previous input available to the user.
6632 ** The behavior of the read-line function has changed.
6634 This function now uses standard C library functions to read the line,
6635 for speed. This means that it doesn not respect the value of
6636 scm-line-incrementors; it assumes that lines are delimited with
6639 (Note that this is read-line, the function that reads a line of text
6640 from a port, not readline, the function that reads a line from a
6641 terminal, providing full editing capabilities.)
6643 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
6645 This module provides some simple argument parsing. It exports one
6648 Function: getopt-gnu-style ARG-LS
6649 Parse a list of program arguments into an alist of option
6652 Each item in the list of program arguments is examined to see if
6653 it meets the syntax of a GNU long-named option. An argument like
6654 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
6655 returned alist, where MUMBLE is a keyword object with the same
6656 name as the argument. An argument like `--MUMBLE=FROB' produces
6657 an element of the form (MUMBLE . FROB), where FROB is a string.
6659 As a special case, the returned alist also contains a pair whose
6660 car is the symbol `rest'. The cdr of this pair is a list
6661 containing all the items in the argument list that are not options
6662 of the form mentioned above.
6664 The argument `--' is treated specially: all items in the argument
6665 list appearing after such an argument are not examined, and are
6666 returned in the special `rest' list.
6668 This function does not parse normal single-character switches.
6669 You will need to parse them out of the `rest' list yourself.
6671 ** The read syntax for byte vectors and short vectors has changed.
6673 Instead of #bytes(...), write #y(...).
6675 Instead of #short(...), write #h(...).
6677 This may seem nutty, but, like the other uniform vectors, byte vectors
6678 and short vectors want to have the same print and read syntax (and,
6679 more basic, want to have read syntax!). Changing the read syntax to
6680 use multiple characters after the hash sign breaks with the
6681 conventions used in R5RS and the conventions used for the other
6682 uniform vectors. It also introduces complexity in the current reader,
6683 both on the C and Scheme levels. (The Right solution is probably to
6684 change the syntax and prototypes for uniform vectors entirely.)
6687 ** The new module (ice-9 session) provides useful interactive functions.
6689 *** New procedure: (apropos REGEXP OPTION ...)
6691 Display a list of top-level variables whose names match REGEXP, and
6692 the modules they are imported from. Each OPTION should be one of the
6695 value --- Show the value of each matching variable.
6696 shadow --- Show bindings shadowed by subsequently imported modules.
6697 full --- Same as both `shadow' and `value'.
6701 guile> (apropos "trace" 'full)
6702 debug: trace #<procedure trace args>
6703 debug: untrace #<procedure untrace args>
6704 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
6705 the-scm-module: before-backtrace-hook ()
6706 the-scm-module: backtrace #<primitive-procedure backtrace>
6707 the-scm-module: after-backtrace-hook ()
6708 the-scm-module: has-shown-backtrace-hint? #f
6711 ** There are new functions and syntax for working with macros.
6713 Guile implements macros as a special object type. Any variable whose
6714 top-level binding is a macro object acts as a macro. The macro object
6715 specifies how the expression should be transformed before evaluation.
6717 *** Macro objects now print in a reasonable way, resembling procedures.
6719 *** New function: (macro? OBJ)
6720 True iff OBJ is a macro object.
6722 *** New function: (primitive-macro? OBJ)
6723 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
6724 macro transformers, implemented in eval.c rather than Scheme code.
6726 Why do we have this function?
6727 - For symmetry with procedure? and primitive-procedure?,
6728 - to allow custom print procedures to tell whether a macro is
6729 primitive, and display it differently, and
6730 - to allow compilers and user-written evaluators to distinguish
6731 builtin special forms from user-defined ones, which could be
6734 *** New function: (macro-type OBJ)
6735 Return a value indicating what kind of macro OBJ is. Possible return
6738 The symbol `syntax' --- a macro created by procedure->syntax.
6739 The symbol `macro' --- a macro created by procedure->macro.
6740 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
6741 The boolean #f --- if OBJ is not a macro object.
6743 *** New function: (macro-name MACRO)
6744 Return the name of the macro object MACRO's procedure, as returned by
6747 *** New function: (macro-transformer MACRO)
6748 Return the transformer procedure for MACRO.
6750 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
6752 Specify a new macro expander to use in the current module. Each
6753 MODULE is a module name, with the same meaning as in the `use-modules'
6754 form; each named module's exported bindings are added to the current
6755 top-level environment. TRANSFORMER is an expression evaluated in the
6756 resulting environment which must yield a procedure to use as the
6757 module's eval transformer: every expression evaluated in this module
6758 is passed to this function, and the result passed to the Guile
6761 *** macro-eval! is removed. Use local-eval instead.
6763 ** Some magic has been added to the printer to better handle user
6764 written printing routines (like record printers, closure printers).
6766 The problem is that these user written routines must have access to
6767 the current `print-state' to be able to handle fancy things like
6768 detection of circular references. These print-states have to be
6769 passed to the builtin printing routines (display, write, etc) to
6770 properly continue the print chain.
6772 We didn't want to change all existing print code so that it
6773 explicitly passes thru a print state in addition to a port. Instead,
6774 we extented the possible values that the builtin printing routines
6775 accept as a `port'. In addition to a normal port, they now also take
6776 a pair of a normal port and a print-state. Printing will go to the
6777 port and the print-state will be used to control the detection of
6778 circular references, etc. If the builtin function does not care for a
6779 print-state, it is simply ignored.
6781 User written callbacks are now called with such a pair as their
6782 `port', but because every function now accepts this pair as a PORT
6783 argument, you don't have to worry about that. In fact, it is probably
6784 safest to not check for these pairs.
6786 However, it is sometimes necessary to continue a print chain on a
6787 different port, for example to get a intermediate string
6788 representation of the printed value, mangle that string somehow, and
6789 then to finally print the mangled string. Use the new function
6791 inherit-print-state OLD-PORT NEW-PORT
6793 for this. It constructs a new `port' that prints to NEW-PORT but
6794 inherits the print-state of OLD-PORT.
6796 ** struct-vtable-offset renamed to vtable-offset-user
6798 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
6800 ** There is now a third optional argument to make-vtable-vtable
6801 (and fourth to make-struct) when constructing new types (vtables).
6802 This argument initializes field vtable-index-printer of the vtable.
6804 ** The detection of circular references has been extended to structs.
6805 That is, a structure that -- in the process of being printed -- prints
6806 itself does not lead to infinite recursion.
6808 ** There is now some basic support for fluids. Please read
6809 "libguile/fluid.h" to find out more. It is accessible from Scheme with
6810 the following functions and macros:
6812 Function: make-fluid
6814 Create a new fluid object. Fluids are not special variables or
6815 some other extension to the semantics of Scheme, but rather
6816 ordinary Scheme objects. You can store them into variables (that
6817 are still lexically scoped, of course) or into any other place you
6818 like. Every fluid has a initial value of `#f'.
6820 Function: fluid? OBJ
6822 Test whether OBJ is a fluid.
6824 Function: fluid-ref FLUID
6825 Function: fluid-set! FLUID VAL
6827 Access/modify the fluid FLUID. Modifications are only visible
6828 within the current dynamic root (that includes threads).
6830 Function: with-fluids* FLUIDS VALUES THUNK
6832 FLUIDS is a list of fluids and VALUES a corresponding list of
6833 values for these fluids. Before THUNK gets called the values are
6834 installed in the fluids and the old values of the fluids are
6835 saved in the VALUES list. When the flow of control leaves THUNK
6836 or reenters it, the values get swapped again. You might think of
6837 this as a `safe-fluid-excursion'. Note that the VALUES list is
6838 modified by `with-fluids*'.
6840 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
6842 The same as `with-fluids*' but with a different syntax. It looks
6843 just like `let', but both FLUID and VALUE are evaluated. Remember,
6844 fluids are not special variables but ordinary objects. FLUID
6845 should evaluate to a fluid.
6847 ** Changes to system call interfaces:
6849 *** close-port, close-input-port and close-output-port now return a
6850 boolean instead of an `unspecified' object. #t means that the port
6851 was successfully closed, while #f means it was already closed. It is
6852 also now possible for these procedures to raise an exception if an
6853 error occurs (some errors from write can be delayed until close.)
6855 *** the first argument to chmod, fcntl, ftell and fseek can now be a
6858 *** the third argument to fcntl is now optional.
6860 *** the first argument to chown can now be a file descriptor or a port.
6862 *** the argument to stat can now be a port.
6864 *** The following new procedures have been added (most use scsh
6867 *** procedure: close PORT/FD
6868 Similar to close-port (*note close-port: Closing Ports.), but also
6869 works on file descriptors. A side effect of closing a file
6870 descriptor is that any ports using that file descriptor are moved
6871 to a different file descriptor and have their revealed counts set
6874 *** procedure: port->fdes PORT
6875 Returns the integer file descriptor underlying PORT. As a side
6876 effect the revealed count of PORT is incremented.
6878 *** procedure: fdes->ports FDES
6879 Returns a list of existing ports which have FDES as an underlying
6880 file descriptor, without changing their revealed counts.
6882 *** procedure: fdes->inport FDES
6883 Returns an existing input port which has FDES as its underlying
6884 file descriptor, if one exists, and increments its revealed count.
6885 Otherwise, returns a new input port with a revealed count of 1.
6887 *** procedure: fdes->outport FDES
6888 Returns an existing output port which has FDES as its underlying
6889 file descriptor, if one exists, and increments its revealed count.
6890 Otherwise, returns a new output port with a revealed count of 1.
6892 The next group of procedures perform a `dup2' system call, if NEWFD
6893 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
6894 duplicated can be supplied as an integer or contained in a port. The
6895 type of value returned varies depending on which procedure is used.
6897 All procedures also have the side effect when performing `dup2' that
6898 any ports using NEWFD are moved to a different file descriptor and have
6899 their revealed counts set to zero.
6901 *** procedure: dup->fdes PORT/FD [NEWFD]
6902 Returns an integer file descriptor.
6904 *** procedure: dup->inport PORT/FD [NEWFD]
6905 Returns a new input port using the new file descriptor.
6907 *** procedure: dup->outport PORT/FD [NEWFD]
6908 Returns a new output port using the new file descriptor.
6910 *** procedure: dup PORT/FD [NEWFD]
6911 Returns a new port if PORT/FD is a port, with the same mode as the
6912 supplied port, otherwise returns an integer file descriptor.
6914 *** procedure: dup->port PORT/FD MODE [NEWFD]
6915 Returns a new port using the new file descriptor. MODE supplies a
6916 mode string for the port (*note open-file: File Ports.).
6918 *** procedure: setenv NAME VALUE
6919 Modifies the environment of the current process, which is also the
6920 default environment inherited by child processes.
6922 If VALUE is `#f', then NAME is removed from the environment.
6923 Otherwise, the string NAME=VALUE is added to the environment,
6924 replacing any existing string with name matching NAME.
6926 The return value is unspecified.
6928 *** procedure: truncate-file OBJ SIZE
6929 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
6930 can be a string containing a file name or an integer file
6931 descriptor or port open for output on the file. The underlying
6932 system calls are `truncate' and `ftruncate'.
6934 The return value is unspecified.
6936 *** procedure: setvbuf PORT MODE [SIZE]
6937 Set the buffering mode for PORT. MODE can be:
6945 block buffered, using a newly allocated buffer of SIZE bytes.
6946 However if SIZE is zero or unspecified, the port will be made
6949 This procedure should not be used after I/O has been performed with
6952 Ports are usually block buffered by default, with a default buffer
6953 size. Procedures e.g., *Note open-file: File Ports, which accept a
6954 mode string allow `0' to be added to request an unbuffered port.
6956 *** procedure: fsync PORT/FD
6957 Copies any unwritten data for the specified output file descriptor
6958 to disk. If PORT/FD is a port, its buffer is flushed before the
6959 underlying file descriptor is fsync'd. The return value is
6962 *** procedure: open-fdes PATH FLAGS [MODES]
6963 Similar to `open' but returns a file descriptor instead of a port.
6965 *** procedure: execle PATH ENV [ARG] ...
6966 Similar to `execl', but the environment of the new process is
6967 specified by ENV, which must be a list of strings as returned by
6968 the `environ' procedure.
6970 This procedure is currently implemented using the `execve' system
6971 call, but we call it `execle' because of its Scheme calling
6974 *** procedure: strerror ERRNO
6975 Returns the Unix error message corresponding to ERRNO, an integer.
6977 *** procedure: primitive-exit [STATUS]
6978 Terminate the current process without unwinding the Scheme stack.
6979 This is would typically be useful after a fork. The exit status
6980 is STATUS if supplied, otherwise zero.
6982 *** procedure: times
6983 Returns an object with information about real and processor time.
6984 The following procedures accept such an object as an argument and
6985 return a selected component:
6988 The current real time, expressed as time units relative to an
6992 The CPU time units used by the calling process.
6995 The CPU time units used by the system on behalf of the
6999 The CPU time units used by terminated child processes of the
7000 calling process, whose status has been collected (e.g., using
7004 Similarly, the CPU times units used by the system on behalf of
7005 terminated child processes.
7007 ** Removed: list-length
7008 ** Removed: list-append, list-append!
7009 ** Removed: list-reverse, list-reverse!
7011 ** array-map renamed to array-map!
7013 ** serial-array-map renamed to serial-array-map!
7015 ** catch doesn't take #f as first argument any longer
7017 Previously, it was possible to pass #f instead of a key to `catch'.
7018 That would cause `catch' to pass a jump buffer object to the procedure
7019 passed as second argument. The procedure could then use this jump
7020 buffer objekt as an argument to throw.
7022 This mechanism has been removed since its utility doesn't motivate the
7023 extra complexity it introduces.
7025 ** The `#/' notation for lists now provokes a warning message from Guile.
7026 This syntax will be removed from Guile in the near future.
7028 To disable the warning message, set the GUILE_HUSH environment
7029 variable to any non-empty value.
7031 ** The newline character now prints as `#\newline', following the
7032 normal Scheme notation, not `#\nl'.
7034 * Changes to the gh_ interface
7036 ** The gh_enter function now takes care of loading the Guile startup files.
7037 gh_enter works by calling scm_boot_guile; see the remarks below.
7039 ** Function: void gh_write (SCM x)
7041 Write the printed representation of the scheme object x to the current
7042 output port. Corresponds to the scheme level `write'.
7044 ** gh_list_length renamed to gh_length.
7046 ** vector handling routines
7048 Several major changes. In particular, gh_vector() now resembles
7049 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
7050 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
7051 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
7052 vector-related gh_ functions have been implemented.
7054 ** pair and list routines
7056 Implemented several of the R4RS pair and list functions that were
7059 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
7061 New function. Converts double arrays back and forth between Scheme
7064 * Changes to the scm_ interface
7066 ** The function scm_boot_guile now takes care of loading the startup files.
7068 Guile's primary initialization function, scm_boot_guile, now takes
7069 care of loading `boot-9.scm', in the `ice-9' module, to initialize
7070 Guile, define the module system, and put together some standard
7071 bindings. It also loads `init.scm', which is intended to hold
7072 site-specific initialization code.
7074 Since Guile cannot operate properly until boot-9.scm is loaded, there
7075 is no reason to separate loading boot-9.scm from Guile's other
7076 initialization processes.
7078 This job used to be done by scm_compile_shell_switches, which didn't
7079 make much sense; in particular, it meant that people using Guile for
7080 non-shell-like applications had to jump through hoops to get Guile
7081 initialized properly.
7083 ** The function scm_compile_shell_switches no longer loads the startup files.
7084 Now, Guile always loads the startup files, whenever it is initialized;
7085 see the notes above for scm_boot_guile and scm_load_startup_files.
7087 ** Function: scm_load_startup_files
7088 This new function takes care of loading Guile's initialization file
7089 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
7090 this is always called by the Guile initialization process, it's
7091 probably not too useful to call this yourself, but it's there anyway.
7093 ** The semantics of smob marking have changed slightly.
7095 The smob marking function (the `mark' member of the scm_smobfuns
7096 structure) is no longer responsible for setting the mark bit on the
7097 smob. The generic smob handling code in the garbage collector will
7098 set this bit. The mark function need only ensure that any other
7099 objects the smob refers to get marked.
7101 Note that this change means that the smob's GC8MARK bit is typically
7102 already set upon entry to the mark function. Thus, marking functions
7103 which look like this:
7106 if (SCM_GC8MARKP (ptr))
7108 SCM_SETGC8MARK (ptr);
7109 ... mark objects to which the smob refers ...
7112 are now incorrect, since they will return early, and fail to mark any
7113 other objects the smob refers to. Some code in the Guile library used
7116 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
7118 If you have implemented your own I/O port type, by writing the
7119 functions required by the scm_ptobfuns and then calling scm_newptob,
7120 you will need to change your functions slightly.
7122 The functions in a scm_ptobfuns structure now expect the port itself
7123 as their argument; they used to expect the `stream' member of the
7124 port's scm_port_table structure. This allows functions in an
7125 scm_ptobfuns structure to easily access the port's cell (and any flags
7126 it its CAR), and the port's scm_port_table structure.
7128 Guile now passes the I/O port itself as the `port' argument in the
7129 following scm_ptobfuns functions:
7131 int (*free) (SCM port);
7132 int (*fputc) (int, SCM port);
7133 int (*fputs) (char *, SCM port);
7134 scm_sizet (*fwrite) SCM_P ((char *ptr,
7138 int (*fflush) (SCM port);
7139 int (*fgetc) (SCM port);
7140 int (*fclose) (SCM port);
7142 The interfaces to the `mark', `print', `equalp', and `fgets' methods
7145 If you have existing code which defines its own port types, it is easy
7146 to convert your code to the new interface; simply apply SCM_STREAM to
7147 the port argument to yield the value you code used to expect.
7149 Note that since both the port and the stream have the same type in the
7150 C code --- they are both SCM values --- the C compiler will not remind
7151 you if you forget to update your scm_ptobfuns functions.
7154 ** Function: int scm_internal_select (int fds,
7158 struct timeval *timeout);
7160 This is a replacement for the `select' function provided by the OS.
7161 It enables I/O blocking and sleeping to happen for one cooperative
7162 thread without blocking other threads. It also avoids busy-loops in
7163 these situations. It is intended that all I/O blocking and sleeping
7164 will finally go through this function. Currently, this function is
7165 only available on systems providing `gettimeofday' and `select'.
7167 ** Function: SCM scm_internal_stack_catch (SCM tag,
7168 scm_catch_body_t body,
7170 scm_catch_handler_t handler,
7173 A new sibling to the other two C level `catch' functions
7174 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
7175 the stack to be saved automatically into the variable `the-last-stack'
7176 (scm_the_last_stack_var) on error. This is necessary if you want to
7177 use advanced error reporting, such as calling scm_display_error and
7178 scm_display_backtrace. (They both take a stack object as argument.)
7180 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
7182 scm_catch_handler_t handler,
7185 Spawns a new thread. It does a job similar to
7186 scm_call_with_new_thread but takes arguments more suitable when
7187 spawning threads from application C code.
7189 ** The hook scm_error_callback has been removed. It was originally
7190 intended as a way for the user to install his own error handler. But
7191 that method works badly since it intervenes between throw and catch,
7192 thereby changing the semantics of expressions like (catch #t ...).
7193 The correct way to do it is to use one of the C level catch functions
7194 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
7196 ** Removed functions:
7198 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
7199 scm_list_reverse, scm_list_reverse_x
7201 ** New macros: SCM_LISTn where n is one of the integers 0-9.
7203 These can be used for pretty list creation from C. The idea is taken
7204 from Erick Gallesio's STk.
7206 ** scm_array_map renamed to scm_array_map_x
7208 ** mbstrings are now removed
7210 This means that the type codes scm_tc7_mb_string and
7211 scm_tc7_mb_substring has been removed.
7213 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
7215 Since we no longer support multi-byte strings, these I/O functions
7216 have been simplified, and renamed. Here are their old names, and
7217 their new names and arguments:
7219 scm_gen_putc -> void scm_putc (int c, SCM port);
7220 scm_gen_puts -> void scm_puts (char *s, SCM port);
7221 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
7222 scm_gen_getc -> void scm_getc (SCM port);
7225 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
7227 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
7229 SCM_TYP7S now masks away the bit which distinguishes substrings from
7232 ** scm_catch_body_t: Backward incompatible change!
7234 Body functions to scm_internal_catch and friends do not any longer
7235 take a second argument. This is because it is no longer possible to
7236 pass a #f arg to catch.
7238 ** Calls to scm_protect_object and scm_unprotect now nest properly.
7240 The function scm_protect_object protects its argument from being freed
7241 by the garbage collector. scm_unprotect_object removes that
7244 These functions now nest properly. That is, for every object O, there
7245 is a counter which scm_protect_object(O) increments and
7246 scm_unprotect_object(O) decrements, if the counter is greater than
7247 zero. Every object's counter is zero when it is first created. If an
7248 object's counter is greater than zero, the garbage collector will not
7249 reclaim its storage.
7251 This allows you to use scm_protect_object in your code without
7252 worrying that some other function you call will call
7253 scm_unprotect_object, and allow it to be freed. Assuming that the
7254 functions you call are well-behaved, and unprotect only those objects
7255 they protect, you can follow the same rule and have confidence that
7256 objects will be freed only at appropriate times.
7259 Changes in Guile 1.2 (released Tuesday, June 24 1997):
7261 * Changes to the distribution
7263 ** Nightly snapshots are now available from ftp.red-bean.com.
7264 The old server, ftp.cyclic.com, has been relinquished to its rightful
7267 Nightly snapshots of the Guile development sources are now available via
7268 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
7270 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
7271 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
7273 ** To run Guile without installing it, the procedure has changed a bit.
7275 If you used a separate build directory to compile Guile, you'll need
7276 to include the build directory in SCHEME_LOAD_PATH, as well as the
7277 source directory. See the `INSTALL' file for examples.
7279 * Changes to the procedure for linking libguile with your programs
7281 ** The standard Guile load path for Scheme code now includes
7282 $(datadir)/guile (usually /usr/local/share/guile). This means that
7283 you can install your own Scheme files there, and Guile will find them.
7284 (Previous versions of Guile only checked a directory whose name
7285 contained the Guile version number, so you had to re-install or move
7286 your Scheme sources each time you installed a fresh version of Guile.)
7288 The load path also includes $(datadir)/guile/site; we recommend
7289 putting individual Scheme files there. If you want to install a
7290 package with multiple source files, create a directory for them under
7293 ** Guile 1.2 will now use the Rx regular expression library, if it is
7294 installed on your system. When you are linking libguile into your own
7295 programs, this means you will have to link against -lguile, -lqt (if
7296 you configured Guile with thread support), and -lrx.
7298 If you are using autoconf to generate configuration scripts for your
7299 application, the following lines should suffice to add the appropriate
7300 libraries to your link command:
7302 ### Find Rx, quickthreads and libguile.
7303 AC_CHECK_LIB(rx, main)
7304 AC_CHECK_LIB(qt, main)
7305 AC_CHECK_LIB(guile, scm_shell)
7307 The Guile 1.2 distribution does not contain sources for the Rx
7308 library, as Guile 1.0 did. If you want to use Rx, you'll need to
7309 retrieve it from a GNU FTP site and install it separately.
7311 * Changes to Scheme functions and syntax
7313 ** The dynamic linking features of Guile are now enabled by default.
7314 You can disable them by giving the `--disable-dynamic-linking' option
7317 (dynamic-link FILENAME)
7319 Find the object file denoted by FILENAME (a string) and link it
7320 into the running Guile application. When everything works out,
7321 return a Scheme object suitable for representing the linked object
7322 file. Otherwise an error is thrown. How object files are
7323 searched is system dependent.
7325 (dynamic-object? VAL)
7327 Determine whether VAL represents a dynamically linked object file.
7329 (dynamic-unlink DYNOBJ)
7331 Unlink the indicated object file from the application. DYNOBJ
7332 should be one of the values returned by `dynamic-link'.
7334 (dynamic-func FUNCTION DYNOBJ)
7336 Search the C function indicated by FUNCTION (a string or symbol)
7337 in DYNOBJ and return some Scheme object that can later be used
7338 with `dynamic-call' to actually call this function. Right now,
7339 these Scheme objects are formed by casting the address of the
7340 function to `long' and converting this number to its Scheme
7343 (dynamic-call FUNCTION DYNOBJ)
7345 Call the C function indicated by FUNCTION and DYNOBJ. The
7346 function is passed no arguments and its return value is ignored.
7347 When FUNCTION is something returned by `dynamic-func', call that
7348 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
7349 etc.), look it up in DYNOBJ; this is equivalent to
7351 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
7353 Interrupts are deferred while the C function is executing (with
7354 SCM_DEFER_INTS/SCM_ALLOW_INTS).
7356 (dynamic-args-call FUNCTION DYNOBJ ARGS)
7358 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
7359 some arguments and return its return value. The C function is
7360 expected to take two arguments and return an `int', just like
7363 int c_func (int argc, char **argv);
7365 ARGS must be a list of strings and is converted into an array of
7366 `char *'. The array is passed in ARGV and its size in ARGC. The
7367 return value is converted to a Scheme number and returned from the
7368 call to `dynamic-args-call'.
7370 When dynamic linking is disabled or not supported on your system,
7371 the above functions throw errors, but they are still available.
7373 Here is a small example that works on GNU/Linux:
7375 (define libc-obj (dynamic-link "libc.so"))
7376 (dynamic-args-call 'rand libc-obj '())
7378 See the file `libguile/DYNAMIC-LINKING' for additional comments.
7380 ** The #/ syntax for module names is depreciated, and will be removed
7381 in a future version of Guile. Instead of
7389 The latter syntax is more consistent with existing Lisp practice.
7391 ** Guile now does fancier printing of structures. Structures are the
7392 underlying implementation for records, which in turn are used to
7393 implement modules, so all of these object now print differently and in
7394 a more informative way.
7396 The Scheme printer will examine the builtin variable *struct-printer*
7397 whenever it needs to print a structure object. When this variable is
7398 not `#f' it is deemed to be a procedure and will be applied to the
7399 structure object and the output port. When *struct-printer* is `#f'
7400 or the procedure return `#f' the structure object will be printed in
7401 the boring #<struct 80458270> form.
7403 This hook is used by some routines in ice-9/boot-9.scm to implement
7404 type specific printing routines. Please read the comments there about
7407 One of the more specific uses of structs are records. The printing
7408 procedure that could be passed to MAKE-RECORD-TYPE is now actually
7409 called. It should behave like a *struct-printer* procedure (described
7412 ** Guile now supports a new R4RS-compliant syntax for keywords. A
7413 token of the form #:NAME, where NAME has the same syntax as a Scheme
7414 symbol, is the external representation of the keyword named NAME.
7415 Keyword objects print using this syntax as well, so values containing
7416 keyword objects can be read back into Guile. When used in an
7417 expression, keywords are self-quoting objects.
7419 Guile suports this read syntax, and uses this print syntax, regardless
7420 of the current setting of the `keyword' read option. The `keyword'
7421 read option only controls whether Guile recognizes the `:NAME' syntax,
7422 which is incompatible with R4RS. (R4RS says such token represent
7425 ** Guile has regular expression support again. Guile 1.0 included
7426 functions for matching regular expressions, based on the Rx library.
7427 In Guile 1.1, the Guile/Rx interface was removed to simplify the
7428 distribution, and thus Guile had no regular expression support. Guile
7429 1.2 again supports the most commonly used functions, and supports all
7430 of SCSH's regular expression functions.
7432 If your system does not include a POSIX regular expression library,
7433 and you have not linked Guile with a third-party regexp library such as
7434 Rx, these functions will not be available. You can tell whether your
7435 Guile installation includes regular expression support by checking
7436 whether the `*features*' list includes the `regex' symbol.
7438 *** regexp functions
7440 By default, Guile supports POSIX extended regular expressions. That
7441 means that the characters `(', `)', `+' and `?' are special, and must
7442 be escaped if you wish to match the literal characters.
7444 This regular expression interface was modeled after that implemented
7445 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
7446 with SCSH regular expressions.
7448 **** Function: string-match PATTERN STR [START]
7449 Compile the string PATTERN into a regular expression and compare
7450 it with STR. The optional numeric argument START specifies the
7451 position of STR at which to begin matching.
7453 `string-match' returns a "match structure" which describes what,
7454 if anything, was matched by the regular expression. *Note Match
7455 Structures::. If STR does not match PATTERN at all,
7456 `string-match' returns `#f'.
7458 Each time `string-match' is called, it must compile its PATTERN
7459 argument into a regular expression structure. This operation is
7460 expensive, which makes `string-match' inefficient if the same regular
7461 expression is used several times (for example, in a loop). For better
7462 performance, you can compile a regular expression in advance and then
7463 match strings against the compiled regexp.
7465 **** Function: make-regexp STR [FLAGS]
7466 Compile the regular expression described by STR, and return the
7467 compiled regexp structure. If STR does not describe a legal
7468 regular expression, `make-regexp' throws a
7469 `regular-expression-syntax' error.
7471 FLAGS may be the bitwise-or of one or more of the following:
7473 **** Constant: regexp/extended
7474 Use POSIX Extended Regular Expression syntax when interpreting
7475 STR. If not set, POSIX Basic Regular Expression syntax is used.
7476 If the FLAGS argument is omitted, we assume regexp/extended.
7478 **** Constant: regexp/icase
7479 Do not differentiate case. Subsequent searches using the
7480 returned regular expression will be case insensitive.
7482 **** Constant: regexp/newline
7483 Match-any-character operators don't match a newline.
7485 A non-matching list ([^...]) not containing a newline matches a
7488 Match-beginning-of-line operator (^) matches the empty string
7489 immediately after a newline, regardless of whether the FLAGS
7490 passed to regexp-exec contain regexp/notbol.
7492 Match-end-of-line operator ($) matches the empty string
7493 immediately before a newline, regardless of whether the FLAGS
7494 passed to regexp-exec contain regexp/noteol.
7496 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
7497 Match the compiled regular expression REGEXP against `str'. If
7498 the optional integer START argument is provided, begin matching
7499 from that position in the string. Return a match structure
7500 describing the results of the match, or `#f' if no match could be
7503 FLAGS may be the bitwise-or of one or more of the following:
7505 **** Constant: regexp/notbol
7506 The match-beginning-of-line operator always fails to match (but
7507 see the compilation flag regexp/newline above) This flag may be
7508 used when different portions of a string are passed to
7509 regexp-exec and the beginning of the string should not be
7510 interpreted as the beginning of the line.
7512 **** Constant: regexp/noteol
7513 The match-end-of-line operator always fails to match (but see the
7514 compilation flag regexp/newline above)
7516 **** Function: regexp? OBJ
7517 Return `#t' if OBJ is a compiled regular expression, or `#f'
7520 Regular expressions are commonly used to find patterns in one string
7521 and replace them with the contents of another string.
7523 **** Function: regexp-substitute PORT MATCH [ITEM...]
7524 Write to the output port PORT selected contents of the match
7525 structure MATCH. Each ITEM specifies what should be written, and
7526 may be one of the following arguments:
7528 * A string. String arguments are written out verbatim.
7530 * An integer. The submatch with that number is written.
7532 * The symbol `pre'. The portion of the matched string preceding
7533 the regexp match is written.
7535 * The symbol `post'. The portion of the matched string
7536 following the regexp match is written.
7538 PORT may be `#f', in which case nothing is written; instead,
7539 `regexp-substitute' constructs a string from the specified ITEMs
7542 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
7543 Similar to `regexp-substitute', but can be used to perform global
7544 substitutions on STR. Instead of taking a match structure as an
7545 argument, `regexp-substitute/global' takes two string arguments: a
7546 REGEXP string describing a regular expression, and a TARGET string
7547 which should be matched against this regular expression.
7549 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
7552 * A function may be supplied. When this function is called, it
7553 will be passed one argument: a match structure for a given
7554 regular expression match. It should return a string to be
7555 written out to PORT.
7557 * The `post' symbol causes `regexp-substitute/global' to recurse
7558 on the unmatched portion of STR. This *must* be supplied in
7559 order to perform global search-and-replace on STR; if it is
7560 not present among the ITEMs, then `regexp-substitute/global'
7561 will return after processing a single match.
7563 *** Match Structures
7565 A "match structure" is the object returned by `string-match' and
7566 `regexp-exec'. It describes which portion of a string, if any, matched
7567 the given regular expression. Match structures include: a reference to
7568 the string that was checked for matches; the starting and ending
7569 positions of the regexp match; and, if the regexp included any
7570 parenthesized subexpressions, the starting and ending positions of each
7573 In each of the regexp match functions described below, the `match'
7574 argument must be a match structure returned by a previous call to
7575 `string-match' or `regexp-exec'. Most of these functions return some
7576 information about the original target string that was matched against a
7577 regular expression; we will call that string TARGET for easy reference.
7579 **** Function: regexp-match? OBJ
7580 Return `#t' if OBJ is a match structure returned by a previous
7581 call to `regexp-exec', or `#f' otherwise.
7583 **** Function: match:substring MATCH [N]
7584 Return the portion of TARGET matched by subexpression number N.
7585 Submatch 0 (the default) represents the entire regexp match. If
7586 the regular expression as a whole matched, but the subexpression
7587 number N did not match, return `#f'.
7589 **** Function: match:start MATCH [N]
7590 Return the starting position of submatch number N.
7592 **** Function: match:end MATCH [N]
7593 Return the ending position of submatch number N.
7595 **** Function: match:prefix MATCH
7596 Return the unmatched portion of TARGET preceding the regexp match.
7598 **** Function: match:suffix MATCH
7599 Return the unmatched portion of TARGET following the regexp match.
7601 **** Function: match:count MATCH
7602 Return the number of parenthesized subexpressions from MATCH.
7603 Note that the entire regular expression match itself counts as a
7604 subexpression, and failed submatches are included in the count.
7606 **** Function: match:string MATCH
7607 Return the original TARGET string.
7609 *** Backslash Escapes
7611 Sometimes you will want a regexp to match characters like `*' or `$'
7612 exactly. For example, to check whether a particular string represents
7613 a menu entry from an Info node, it would be useful to match it against
7614 a regexp like `^* [^:]*::'. However, this won't work; because the
7615 asterisk is a metacharacter, it won't match the `*' at the beginning of
7616 the string. In this case, we want to make the first asterisk un-magic.
7618 You can do this by preceding the metacharacter with a backslash
7619 character `\'. (This is also called "quoting" the metacharacter, and
7620 is known as a "backslash escape".) When Guile sees a backslash in a
7621 regular expression, it considers the following glyph to be an ordinary
7622 character, no matter what special meaning it would ordinarily have.
7623 Therefore, we can make the above example work by changing the regexp to
7624 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
7625 to match only a single asterisk in the target string.
7627 Since the backslash is itself a metacharacter, you may force a
7628 regexp to match a backslash in the target string by preceding the
7629 backslash with itself. For example, to find variable references in a
7630 TeX program, you might want to find occurrences of the string `\let\'
7631 followed by any number of alphabetic characters. The regular expression
7632 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
7633 each match a single backslash in the target string.
7635 **** Function: regexp-quote STR
7636 Quote each special character found in STR with a backslash, and
7637 return the resulting string.
7639 *Very important:* Using backslash escapes in Guile source code (as
7640 in Emacs Lisp or C) can be tricky, because the backslash character has
7641 special meaning for the Guile reader. For example, if Guile encounters
7642 the character sequence `\n' in the middle of a string while processing
7643 Scheme code, it replaces those characters with a newline character.
7644 Similarly, the character sequence `\t' is replaced by a horizontal tab.
7645 Several of these "escape sequences" are processed by the Guile reader
7646 before your code is executed. Unrecognized escape sequences are
7647 ignored: if the characters `\*' appear in a string, they will be
7648 translated to the single character `*'.
7650 This translation is obviously undesirable for regular expressions,
7651 since we want to be able to include backslashes in a string in order to
7652 escape regexp metacharacters. Therefore, to make sure that a backslash
7653 is preserved in a string in your Guile program, you must use *two*
7654 consecutive backslashes:
7656 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
7658 The string in this example is preprocessed by the Guile reader before
7659 any code is executed. The resulting argument to `make-regexp' is the
7660 string `^\* [^:]*', which is what we really want.
7662 This also means that in order to write a regular expression that
7663 matches a single backslash character, the regular expression string in
7664 the source code must include *four* backslashes. Each consecutive pair
7665 of backslashes gets translated by the Guile reader to a single
7666 backslash, and the resulting double-backslash is interpreted by the
7667 regexp engine as matching a single backslash character. Hence:
7669 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
7671 The reason for the unwieldiness of this syntax is historical. Both
7672 regular expression pattern matchers and Unix string processing systems
7673 have traditionally used backslashes with the special meanings described
7674 above. The POSIX regular expression specification and ANSI C standard
7675 both require these semantics. Attempting to abandon either convention
7676 would cause other kinds of compatibility problems, possibly more severe
7677 ones. Therefore, without extending the Scheme reader to support
7678 strings with different quoting conventions (an ungainly and confusing
7679 extension when implemented in other languages), we must adhere to this
7680 cumbersome escape syntax.
7682 * Changes to the gh_ interface
7684 * Changes to the scm_ interface
7686 * Changes to system call interfaces:
7688 ** The value returned by `raise' is now unspecified. It throws an exception
7691 *** A new procedure `sigaction' can be used to install signal handlers
7693 (sigaction signum [action] [flags])
7695 signum is the signal number, which can be specified using the value
7698 If action is omitted, sigaction returns a pair: the CAR is the current
7699 signal hander, which will be either an integer with the value SIG_DFL
7700 (default action) or SIG_IGN (ignore), or the Scheme procedure which
7701 handles the signal, or #f if a non-Scheme procedure handles the
7702 signal. The CDR contains the current sigaction flags for the handler.
7704 If action is provided, it is installed as the new handler for signum.
7705 action can be a Scheme procedure taking one argument, or the value of
7706 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
7707 whatever signal handler was installed before sigaction was first used.
7708 Flags can optionally be specified for the new handler (SA_RESTART is
7709 always used if the system provides it, so need not be specified.) The
7710 return value is a pair with information about the old handler as
7713 This interface does not provide access to the "signal blocking"
7714 facility. Maybe this is not needed, since the thread support may
7715 provide solutions to the problem of consistent access to data
7718 *** A new procedure `flush-all-ports' is equivalent to running
7719 `force-output' on every port open for output.
7721 ** Guile now provides information on how it was built, via the new
7722 global variable, %guile-build-info. This variable records the values
7723 of the standard GNU makefile directory variables as an assocation
7724 list, mapping variable names (symbols) onto directory paths (strings).
7725 For example, to find out where the Guile link libraries were
7726 installed, you can say:
7728 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
7731 * Changes to the scm_ interface
7733 ** The new function scm_handle_by_message_noexit is just like the
7734 existing scm_handle_by_message function, except that it doesn't call
7735 exit to terminate the process. Instead, it prints a message and just
7736 returns #f. This might be a more appropriate catch-all handler for
7737 new dynamic roots and threads.
7740 Changes in Guile 1.1 (released Friday, May 16 1997):
7742 * Changes to the distribution.
7744 The Guile 1.0 distribution has been split up into several smaller
7746 guile-core --- the Guile interpreter itself.
7747 guile-tcltk --- the interface between the Guile interpreter and
7748 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
7749 is a toolkit for building graphical user interfaces.
7750 guile-rgx-ctax --- the interface between Guile and the Rx regular
7751 expression matcher, and the translator for the Ctax
7752 programming language. These are packaged together because the
7753 Ctax translator uses Rx to parse Ctax source code.
7755 This NEWS file describes the changes made to guile-core since the 1.0
7758 We no longer distribute the documentation, since it was either out of
7759 date, or incomplete. As soon as we have current documentation, we
7764 * Changes to the stand-alone interpreter
7766 ** guile now accepts command-line arguments compatible with SCSH, Olin
7767 Shivers' Scheme Shell.
7769 In general, arguments are evaluated from left to right, but there are
7770 exceptions. The following switches stop argument processing, and
7771 stash all remaining command-line arguments as the value returned by
7772 the (command-line) function.
7773 -s SCRIPT load Scheme source code from FILE, and exit
7774 -c EXPR evalute Scheme expression EXPR, and exit
7775 -- stop scanning arguments; run interactively
7777 The switches below are processed as they are encountered.
7778 -l FILE load Scheme source code from FILE
7779 -e FUNCTION after reading script, apply FUNCTION to
7780 command line arguments
7781 -ds do -s script at this point
7782 --emacs enable Emacs protocol (experimental)
7783 -h, --help display this help and exit
7784 -v, --version display version information and exit
7785 \ read arguments from following script lines
7787 So, for example, here is a Guile script named `ekko' (thanks, Olin)
7788 which re-implements the traditional "echo" command:
7790 #!/usr/local/bin/guile -s
7793 (map (lambda (arg) (display arg) (display " "))
7797 (main (command-line))
7799 Suppose we invoke this script as follows:
7801 ekko a speckled gecko
7803 Through the magic of Unix script processing (triggered by the `#!'
7804 token at the top of the file), /usr/local/bin/guile receives the
7805 following list of command-line arguments:
7807 ("-s" "./ekko" "a" "speckled" "gecko")
7809 Unix inserts the name of the script after the argument specified on
7810 the first line of the file (in this case, "-s"), and then follows that
7811 with the arguments given to the script. Guile loads the script, which
7812 defines the `main' function, and then applies it to the list of
7813 remaining command-line arguments, ("a" "speckled" "gecko").
7815 In Unix, the first line of a script file must take the following form:
7817 #!INTERPRETER ARGUMENT
7819 where INTERPRETER is the absolute filename of the interpreter
7820 executable, and ARGUMENT is a single command-line argument to pass to
7823 You may only pass one argument to the interpreter, and its length is
7824 limited. These restrictions can be annoying to work around, so Guile
7825 provides a general mechanism (borrowed from, and compatible with,
7826 SCSH) for circumventing them.
7828 If the ARGUMENT in a Guile script is a single backslash character,
7829 `\', Guile will open the script file, parse arguments from its second
7830 and subsequent lines, and replace the `\' with them. So, for example,
7831 here is another implementation of the `ekko' script:
7833 #!/usr/local/bin/guile \
7837 (for-each (lambda (arg) (display arg) (display " "))
7841 If the user invokes this script as follows:
7843 ekko a speckled gecko
7845 Unix expands this into
7847 /usr/local/bin/guile \ ekko a speckled gecko
7849 When Guile sees the `\' argument, it replaces it with the arguments
7850 read from the second line of the script, producing:
7852 /usr/local/bin/guile -e main -s ekko a speckled gecko
7854 This tells Guile to load the `ekko' script, and apply the function
7855 `main' to the argument list ("a" "speckled" "gecko").
7857 Here is how Guile parses the command-line arguments:
7858 - Each space character terminates an argument. This means that two
7859 spaces in a row introduce an empty-string argument.
7860 - The tab character is not permitted (unless you quote it with the
7861 backslash character, as described below), to avoid confusion.
7862 - The newline character terminates the sequence of arguments, and will
7863 also terminate a final non-empty argument. (However, a newline
7864 following a space will not introduce a final empty-string argument;
7865 it only terminates the argument list.)
7866 - The backslash character is the escape character. It escapes
7867 backslash, space, tab, and newline. The ANSI C escape sequences
7868 like \n and \t are also supported. These produce argument
7869 constituents; the two-character combination \n doesn't act like a
7870 terminating newline. The escape sequence \NNN for exactly three
7871 octal digits reads as the character whose ASCII code is NNN. As
7872 above, characters produced this way are argument constituents.
7873 Backslash followed by other characters is not allowed.
7875 * Changes to the procedure for linking libguile with your programs
7877 ** Guile now builds and installs a shared guile library, if your
7878 system support shared libraries. (It still builds a static library on
7879 all systems.) Guile automatically detects whether your system
7880 supports shared libraries. To prevent Guile from buildisg shared
7881 libraries, pass the `--disable-shared' flag to the configure script.
7883 Guile takes longer to compile when it builds shared libraries, because
7884 it must compile every file twice --- once to produce position-
7885 independent object code, and once to produce normal object code.
7887 ** The libthreads library has been merged into libguile.
7889 To link a program against Guile, you now need only link against
7890 -lguile and -lqt; -lthreads is no longer needed. If you are using
7891 autoconf to generate configuration scripts for your application, the
7892 following lines should suffice to add the appropriate libraries to
7895 ### Find quickthreads and libguile.
7896 AC_CHECK_LIB(qt, main)
7897 AC_CHECK_LIB(guile, scm_shell)
7899 * Changes to Scheme functions
7901 ** Guile Scheme's special syntax for keyword objects is now optional,
7902 and disabled by default.
7904 The syntax variation from R4RS made it difficult to port some
7905 interesting packages to Guile. The routines which accepted keyword
7906 arguments (mostly in the module system) have been modified to also
7907 accept symbols whose names begin with `:'.
7909 To change the keyword syntax, you must first import the (ice-9 debug)
7911 (use-modules (ice-9 debug))
7913 Then you can enable the keyword syntax as follows:
7914 (read-set! keywords 'prefix)
7916 To disable keyword syntax, do this:
7917 (read-set! keywords #f)
7919 ** Many more primitive functions accept shared substrings as
7920 arguments. In the past, these functions required normal, mutable
7921 strings as arguments, although they never made use of this
7924 ** The uniform array functions now operate on byte vectors. These
7925 functions are `array-fill!', `serial-array-copy!', `array-copy!',
7926 `serial-array-map', `array-map', `array-for-each', and
7929 ** The new functions `trace' and `untrace' implement simple debugging
7930 support for Scheme functions.
7932 The `trace' function accepts any number of procedures as arguments,
7933 and tells the Guile interpreter to display each procedure's name and
7934 arguments each time the procedure is invoked. When invoked with no
7935 arguments, `trace' returns the list of procedures currently being
7938 The `untrace' function accepts any number of procedures as arguments,
7939 and tells the Guile interpreter not to trace them any more. When
7940 invoked with no arguments, `untrace' untraces all curretly traced
7943 The tracing in Guile has an advantage over most other systems: we
7944 don't create new procedure objects, but mark the procedure objects
7945 themselves. This means that anonymous and internal procedures can be
7948 ** The function `assert-repl-prompt' has been renamed to
7949 `set-repl-prompt!'. It takes one argument, PROMPT.
7950 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
7951 - If PROMPT is a string, we use it as a prompt.
7952 - If PROMPT is a procedure accepting no arguments, we call it, and
7953 display the result as a prompt.
7954 - Otherwise, we display "> ".
7956 ** The new function `eval-string' reads Scheme expressions from a
7957 string and evaluates them, returning the value of the last expression
7958 in the string. If the string contains no expressions, it returns an
7961 ** The new function `thunk?' returns true iff its argument is a
7962 procedure of zero arguments.
7964 ** `defined?' is now a builtin function, instead of syntax. This
7965 means that its argument should be quoted. It returns #t iff its
7966 argument is bound in the current module.
7968 ** The new syntax `use-modules' allows you to add new modules to your
7969 environment without re-typing a complete `define-module' form. It
7970 accepts any number of module names as arguments, and imports their
7971 public bindings into the current module.
7973 ** The new function (module-defined? NAME MODULE) returns true iff
7974 NAME, a symbol, is defined in MODULE, a module object.
7976 ** The new function `builtin-bindings' creates and returns a hash
7977 table containing copies of all the root module's bindings.
7979 ** The new function `builtin-weak-bindings' does the same as
7980 `builtin-bindings', but creates a doubly-weak hash table.
7982 ** The `equal?' function now considers variable objects to be
7983 equivalent if they have the same name and the same value.
7985 ** The new function `command-line' returns the command-line arguments
7986 given to Guile, as a list of strings.
7988 When using guile as a script interpreter, `command-line' returns the
7989 script's arguments; those processed by the interpreter (like `-s' or
7990 `-c') are omitted. (In other words, you get the normal, expected
7991 behavior.) Any application that uses scm_shell to process its
7992 command-line arguments gets this behavior as well.
7994 ** The new function `load-user-init' looks for a file called `.guile'
7995 in the user's home directory, and loads it if it exists. This is
7996 mostly for use by the code generated by scm_compile_shell_switches,
7997 but we thought it might also be useful in other circumstances.
7999 ** The new function `log10' returns the base-10 logarithm of its
8002 ** Changes to I/O functions
8004 *** The functions `read', `primitive-load', `read-and-eval!', and
8005 `primitive-load-path' no longer take optional arguments controlling
8006 case insensitivity and a `#' parser.
8008 Case sensitivity is now controlled by a read option called
8009 `case-insensitive'. The user can add new `#' syntaxes with the
8010 `read-hash-extend' function (see below).
8012 *** The new function `read-hash-extend' allows the user to change the
8013 syntax of Guile Scheme in a somewhat controlled way.
8015 (read-hash-extend CHAR PROC)
8016 When parsing S-expressions, if we read a `#' character followed by
8017 the character CHAR, use PROC to parse an object from the stream.
8018 If PROC is #f, remove any parsing procedure registered for CHAR.
8020 The reader applies PROC to two arguments: CHAR and an input port.
8022 *** The new functions read-delimited and read-delimited! provide a
8023 general mechanism for doing delimited input on streams.
8025 (read-delimited DELIMS [PORT HANDLE-DELIM])
8026 Read until we encounter one of the characters in DELIMS (a string),
8027 or end-of-file. PORT is the input port to read from; it defaults to
8028 the current input port. The HANDLE-DELIM parameter determines how
8029 the terminating character is handled; it should be one of the
8032 'trim omit delimiter from result
8033 'peek leave delimiter character in input stream
8034 'concat append delimiter character to returned value
8035 'split return a pair: (RESULT . TERMINATOR)
8037 HANDLE-DELIM defaults to 'peek.
8039 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
8040 A side-effecting variant of `read-delimited'.
8042 The data is written into the string BUF at the indices in the
8043 half-open interval [START, END); the default interval is the whole
8044 string: START = 0 and END = (string-length BUF). The values of
8045 START and END must specify a well-defined interval in BUF, i.e.
8046 0 <= START <= END <= (string-length BUF).
8048 It returns NBYTES, the number of bytes read. If the buffer filled
8049 up without a delimiter character being found, it returns #f. If the
8050 port is at EOF when the read starts, it returns the EOF object.
8052 If an integer is returned (i.e., the read is successfully terminated
8053 by reading a delimiter character), then the HANDLE-DELIM parameter
8054 determines how to handle the terminating character. It is described
8055 above, and defaults to 'peek.
8057 (The descriptions of these functions were borrowed from the SCSH
8058 manual, by Olin Shivers and Brian Carlstrom.)
8060 *** The `%read-delimited!' function is the primitive used to implement
8061 `read-delimited' and `read-delimited!'.
8063 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
8065 This returns a pair of values: (TERMINATOR . NUM-READ).
8066 - TERMINATOR describes why the read was terminated. If it is a
8067 character or the eof object, then that is the value that terminated
8068 the read. If it is #f, the function filled the buffer without finding
8069 a delimiting character.
8070 - NUM-READ is the number of characters read into BUF.
8072 If the read is successfully terminated by reading a delimiter
8073 character, then the gobble? parameter determines what to do with the
8074 terminating character. If true, the character is removed from the
8075 input stream; if false, the character is left in the input stream
8076 where a subsequent read operation will retrieve it. In either case,
8077 the character is also the first value returned by the procedure call.
8079 (The descriptions of this function was borrowed from the SCSH manual,
8080 by Olin Shivers and Brian Carlstrom.)
8082 *** The `read-line' and `read-line!' functions have changed; they now
8083 trim the terminator by default; previously they appended it to the
8084 returned string. For the old behavior, use (read-line PORT 'concat).
8086 *** The functions `uniform-array-read!' and `uniform-array-write!' now
8087 take new optional START and END arguments, specifying the region of
8088 the array to read and write.
8090 *** The `ungetc-char-ready?' function has been removed. We feel it's
8091 inappropriate for an interface to expose implementation details this
8094 ** Changes to the Unix library and system call interface
8096 *** The new fcntl function provides access to the Unix `fcntl' system
8099 (fcntl PORT COMMAND VALUE)
8100 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
8101 Values for COMMAND are:
8103 F_DUPFD duplicate a file descriptor
8104 F_GETFD read the descriptor's close-on-exec flag
8105 F_SETFD set the descriptor's close-on-exec flag to VALUE
8106 F_GETFL read the descriptor's flags, as set on open
8107 F_SETFL set the descriptor's flags, as set on open to VALUE
8108 F_GETOWN return the process ID of a socket's owner, for SIGIO
8109 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
8110 FD_CLOEXEC not sure what this is
8112 For details, see the documentation for the fcntl system call.
8114 *** The arguments to `select' have changed, for compatibility with
8115 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
8116 expected behavior. The MILLISECONDS parameter has been changed to
8117 MICROSECONDS, to more closely resemble the underlying system call.
8118 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
8119 corresponding return set will be the same.
8121 *** The arguments to the `mknod' system call have changed. They are
8124 (mknod PATH TYPE PERMS DEV)
8125 Create a new file (`node') in the file system. PATH is the name of
8126 the file to create. TYPE is the kind of file to create; it should
8127 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
8128 permission bits to give the newly created file. If TYPE is
8129 'block-special or 'char-special, DEV specifies which device the
8130 special file refers to; its interpretation depends on the kind of
8131 special file being created.
8133 *** The `fork' function has been renamed to `primitive-fork', to avoid
8134 clashing with various SCSH forks.
8136 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
8137 and `recvfrom!'. They no longer accept a size for a second argument;
8138 you must pass a string to hold the received value. They no longer
8139 return the buffer. Instead, `recv' returns the length of the message
8140 received, and `recvfrom' returns a pair containing the packet's length
8141 and originating address.
8143 *** The file descriptor datatype has been removed, as have the
8144 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
8145 We plan to replace these functions with a SCSH-compatible interface.
8147 *** The `create' function has been removed; it's just a special case
8150 *** There are new functions to break down process termination status
8151 values. In the descriptions below, STATUS is a value returned by
8154 (status:exit-val STATUS)
8155 If the child process exited normally, this function returns the exit
8156 code for the child process (i.e., the value passed to exit, or
8157 returned from main). If the child process did not exit normally,
8158 this function returns #f.
8160 (status:stop-sig STATUS)
8161 If the child process was suspended by a signal, this function
8162 returns the signal that suspended the child. Otherwise, it returns
8165 (status:term-sig STATUS)
8166 If the child process terminated abnormally, this function returns
8167 the signal that terminated the child. Otherwise, this function
8170 POSIX promises that exactly one of these functions will return true on
8171 a valid STATUS value.
8173 These functions are compatible with SCSH.
8175 *** There are new accessors and setters for the broken-out time vectors
8176 returned by `localtime', `gmtime', and that ilk. They are:
8178 Component Accessor Setter
8179 ========================= ============ ============
8180 seconds tm:sec set-tm:sec
8181 minutes tm:min set-tm:min
8182 hours tm:hour set-tm:hour
8183 day of the month tm:mday set-tm:mday
8184 month tm:mon set-tm:mon
8185 year tm:year set-tm:year
8186 day of the week tm:wday set-tm:wday
8187 day in the year tm:yday set-tm:yday
8188 daylight saving time tm:isdst set-tm:isdst
8189 GMT offset, seconds tm:gmtoff set-tm:gmtoff
8190 name of time zone tm:zone set-tm:zone
8192 *** There are new accessors for the vectors returned by `uname',
8193 describing the host system:
8196 ============================================== ================
8197 name of the operating system implementation utsname:sysname
8198 network name of this machine utsname:nodename
8199 release level of the operating system utsname:release
8200 version level of the operating system utsname:version
8201 machine hardware platform utsname:machine
8203 *** There are new accessors for the vectors returned by `getpw',
8204 `getpwnam', `getpwuid', and `getpwent', describing entries from the
8205 system's user database:
8208 ====================== =================
8209 user name passwd:name
8210 user password passwd:passwd
8213 real name passwd:gecos
8214 home directory passwd:dir
8215 shell program passwd:shell
8217 *** There are new accessors for the vectors returned by `getgr',
8218 `getgrnam', `getgrgid', and `getgrent', describing entries from the
8219 system's group database:
8222 ======================= ============
8223 group name group:name
8224 group password group:passwd
8226 group members group:mem
8228 *** There are new accessors for the vectors returned by `gethost',
8229 `gethostbyaddr', `gethostbyname', and `gethostent', describing
8233 ========================= ===============
8234 official name of host hostent:name
8235 alias list hostent:aliases
8236 host address type hostent:addrtype
8237 length of address hostent:length
8238 list of addresses hostent:addr-list
8240 *** There are new accessors for the vectors returned by `getnet',
8241 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
8245 ========================= ===============
8246 official name of net netent:name
8247 alias list netent:aliases
8248 net number type netent:addrtype
8249 net number netent:net
8251 *** There are new accessors for the vectors returned by `getproto',
8252 `getprotobyname', `getprotobynumber', and `getprotoent', describing
8256 ========================= ===============
8257 official protocol name protoent:name
8258 alias list protoent:aliases
8259 protocol number protoent:proto
8261 *** There are new accessors for the vectors returned by `getserv',
8262 `getservbyname', `getservbyport', and `getservent', describing
8266 ========================= ===============
8267 official service name servent:name
8268 alias list servent:aliases
8269 port number servent:port
8270 protocol to use servent:proto
8272 *** There are new accessors for the sockaddr structures returned by
8273 `accept', `getsockname', `getpeername', `recvfrom!':
8276 ======================================== ===============
8277 address format (`family') sockaddr:fam
8278 path, for file domain addresses sockaddr:path
8279 address, for internet domain addresses sockaddr:addr
8280 TCP or UDP port, for internet sockaddr:port
8282 *** The `getpwent', `getgrent', `gethostent', `getnetent',
8283 `getprotoent', and `getservent' functions now return #f at the end of
8284 the user database. (They used to throw an exception.)
8286 Note that calling MUMBLEent function is equivalent to calling the
8287 corresponding MUMBLE function with no arguments.
8289 *** The `setpwent', `setgrent', `sethostent', `setnetent',
8290 `setprotoent', and `setservent' routines now take no arguments.
8292 *** The `gethost', `getproto', `getnet', and `getserv' functions now
8293 provide more useful information when they throw an exception.
8295 *** The `lnaof' function has been renamed to `inet-lnaof'.
8297 *** Guile now claims to have the `current-time' feature.
8299 *** The `mktime' function now takes an optional second argument ZONE,
8300 giving the time zone to use for the conversion. ZONE should be a
8301 string, in the same format as expected for the "TZ" environment variable.
8303 *** The `strptime' function now returns a pair (TIME . COUNT), where
8304 TIME is the parsed time as a vector, and COUNT is the number of
8305 characters from the string left unparsed. This function used to
8306 return the remaining characters as a string.
8308 *** The `gettimeofday' function has replaced the old `time+ticks' function.
8309 The return value is now (SECONDS . MICROSECONDS); the fractional
8310 component is no longer expressed in "ticks".
8312 *** The `ticks/sec' constant has been removed, in light of the above change.
8314 * Changes to the gh_ interface
8316 ** gh_eval_str() now returns an SCM object which is the result of the
8319 ** gh_scm2str() now copies the Scheme data to a caller-provided C
8322 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
8323 and returns the array
8325 ** gh_scm2str0() is gone: there is no need to distinguish
8326 null-terminated from non-null-terminated, since gh_scm2newstr() allows
8327 the user to interpret the data both ways.
8329 * Changes to the scm_ interface
8331 ** The new function scm_symbol_value0 provides an easy way to get a
8332 symbol's value from C code:
8334 SCM scm_symbol_value0 (char *NAME)
8335 Return the value of the symbol named by the null-terminated string
8336 NAME in the current module. If the symbol named NAME is unbound in
8337 the current module, return SCM_UNDEFINED.
8339 ** The new function scm_sysintern0 creates new top-level variables,
8340 without assigning them a value.
8342 SCM scm_sysintern0 (char *NAME)
8343 Create a new Scheme top-level variable named NAME. NAME is a
8344 null-terminated string. Return the variable's value cell.
8346 ** The function scm_internal_catch is the guts of catch. It handles
8347 all the mechanics of setting up a catch target, invoking the catch
8348 body, and perhaps invoking the handler if the body does a throw.
8350 The function is designed to be usable from C code, but is general
8351 enough to implement all the semantics Guile Scheme expects from throw.
8353 TAG is the catch tag. Typically, this is a symbol, but this function
8354 doesn't actually care about that.
8356 BODY is a pointer to a C function which runs the body of the catch;
8357 this is the code you can throw from. We call it like this:
8358 BODY (BODY_DATA, JMPBUF)
8360 BODY_DATA is just the BODY_DATA argument we received; we pass it
8361 through to BODY as its first argument. The caller can make
8362 BODY_DATA point to anything useful that BODY might need.
8363 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
8364 which we have just created and initialized.
8366 HANDLER is a pointer to a C function to deal with a throw to TAG,
8367 should one occur. We call it like this:
8368 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
8370 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
8371 same idea as BODY_DATA above.
8372 THROWN_TAG is the tag that the user threw to; usually this is
8373 TAG, but it could be something else if TAG was #t (i.e., a
8374 catch-all), or the user threw to a jmpbuf.
8375 THROW_ARGS is the list of arguments the user passed to the THROW
8378 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
8379 is just a pointer we pass through to HANDLER. We don't actually
8380 use either of those pointers otherwise ourselves. The idea is
8381 that, if our caller wants to communicate something to BODY or
8382 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
8383 HANDLER can then use. Think of it as a way to make BODY and
8384 HANDLER closures, not just functions; MUMBLE_DATA points to the
8387 Of course, it's up to the caller to make sure that any data a
8388 MUMBLE_DATA needs is protected from GC. A common way to do this is
8389 to make MUMBLE_DATA a pointer to data stored in an automatic
8390 structure variable; since the collector must scan the stack for
8391 references anyway, this assures that any references in MUMBLE_DATA
8394 ** The new function scm_internal_lazy_catch is exactly like
8395 scm_internal_catch, except:
8397 - It does not unwind the stack (this is the major difference).
8398 - If handler returns, its value is returned from the throw.
8399 - BODY always receives #f as its JMPBUF argument (since there's no
8400 jmpbuf associated with a lazy catch, because we don't unwind the
8403 ** scm_body_thunk is a new body function you can pass to
8404 scm_internal_catch if you want the body to be like Scheme's `catch'
8405 --- a thunk, or a function of one argument if the tag is #f.
8407 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
8408 contains the Scheme procedure to invoke as the body, and the tag
8409 we're catching. If the tag is #f, then we pass JMPBUF (created by
8410 scm_internal_catch) to the body procedure; otherwise, the body gets
8413 ** scm_handle_by_proc is a new handler function you can pass to
8414 scm_internal_catch if you want the handler to act like Scheme's catch
8415 --- call a procedure with the tag and the throw arguments.
8417 If the user does a throw to this catch, this function runs a handler
8418 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
8419 variable holding the Scheme procedure object to invoke. It ought to
8420 be a pointer to an automatic variable (i.e., one living on the stack),
8421 or the procedure object should be otherwise protected from GC.
8423 ** scm_handle_by_message is a new handler function to use with
8424 `scm_internal_catch' if you want Guile to print a message and die.
8425 It's useful for dealing with throws to uncaught keys at the top level.
8427 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
8428 message header to print; if zero, we use "guile" instead. That
8429 text is followed by a colon, then the message described by ARGS.
8431 ** The return type of scm_boot_guile is now void; the function does
8432 not return a value, and indeed, never returns at all.
8434 ** The new function scm_shell makes it easy for user applications to
8435 process command-line arguments in a way that is compatible with the
8436 stand-alone guile interpreter (which is in turn compatible with SCSH,
8439 To use the scm_shell function, first initialize any guile modules
8440 linked into your application, and then call scm_shell with the values
8441 of ARGC and ARGV your `main' function received. scm_shell will add
8442 any SCSH-style meta-arguments from the top of the script file to the
8443 argument vector, and then process the command-line arguments. This
8444 generally means loading a script file or starting up an interactive
8445 command interpreter. For details, see "Changes to the stand-alone
8448 ** The new functions scm_get_meta_args and scm_count_argv help you
8449 implement the SCSH-style meta-argument, `\'.
8451 char **scm_get_meta_args (int ARGC, char **ARGV)
8452 If the second element of ARGV is a string consisting of a single
8453 backslash character (i.e. "\\" in Scheme notation), open the file
8454 named by the following argument, parse arguments from it, and return
8455 the spliced command line. The returned array is terminated by a
8458 For details of argument parsing, see above, under "guile now accepts
8459 command-line arguments compatible with SCSH..."
8461 int scm_count_argv (char **ARGV)
8462 Count the arguments in ARGV, assuming it is terminated by a null
8465 For an example of how these functions might be used, see the source
8466 code for the function scm_shell in libguile/script.c.
8468 You will usually want to use scm_shell instead of calling this
8471 ** The new function scm_compile_shell_switches turns an array of
8472 command-line arguments into Scheme code to carry out the actions they
8473 describe. Given ARGC and ARGV, it returns a Scheme expression to
8474 evaluate, and calls scm_set_program_arguments to make any remaining
8475 command-line arguments available to the Scheme code. For example,
8476 given the following arguments:
8478 -e main -s ekko a speckled gecko
8480 scm_set_program_arguments will return the following expression:
8482 (begin (load "ekko") (main (command-line)) (quit))
8484 You will usually want to use scm_shell instead of calling this
8487 ** The function scm_shell_usage prints a usage message appropriate for
8488 an interpreter that uses scm_compile_shell_switches to handle its
8489 command-line arguments.
8491 void scm_shell_usage (int FATAL, char *MESSAGE)
8492 Print a usage message to the standard error output. If MESSAGE is
8493 non-zero, write it before the usage message, followed by a newline.
8494 If FATAL is non-zero, exit the process, using FATAL as the
8495 termination status. (If you want to be compatible with Guile,
8496 always use 1 as the exit status when terminating due to command-line
8499 You will usually want to use scm_shell instead of calling this
8502 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
8503 expressions. It used to return SCM_EOL. Earth-shattering.
8505 ** The macros for declaring scheme objects in C code have been
8506 rearranged slightly. They are now:
8508 SCM_SYMBOL (C_NAME, SCHEME_NAME)
8509 Declare a static SCM variable named C_NAME, and initialize it to
8510 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
8511 be a C identifier, and SCHEME_NAME should be a C string.
8513 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
8514 Just like SCM_SYMBOL, but make C_NAME globally visible.
8516 SCM_VCELL (C_NAME, SCHEME_NAME)
8517 Create a global variable at the Scheme level named SCHEME_NAME.
8518 Declare a static SCM variable named C_NAME, and initialize it to
8519 point to the Scheme variable's value cell.
8521 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
8522 Just like SCM_VCELL, but make C_NAME globally visible.
8524 The `guile-snarf' script writes initialization code for these macros
8525 to its standard output, given C source code as input.
8527 The SCM_GLOBAL macro is gone.
8529 ** The scm_read_line and scm_read_line_x functions have been replaced
8530 by Scheme code based on the %read-delimited! procedure (known to C
8531 code as scm_read_delimited_x). See its description above for more
8534 ** The function scm_sys_open has been renamed to scm_open. It now
8535 returns a port instead of an FD object.
8537 * The dynamic linking support has changed. For more information, see
8538 libguile/DYNAMIC-LINKING.
8543 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
8546 * Changes to the 'guile' program:
8548 ** Guile now loads some new files when it starts up. Guile first
8549 searches the load path for init.scm, and loads it if found. Then, if
8550 Guile is not being used to execute a script, and the user's home
8551 directory contains a file named `.guile', Guile loads that.
8553 ** You can now use Guile as a shell script interpreter.
8555 To paraphrase the SCSH manual:
8557 When Unix tries to execute an executable file whose first two
8558 characters are the `#!', it treats the file not as machine code to
8559 be directly executed by the native processor, but as source code
8560 to be executed by some interpreter. The interpreter to use is
8561 specified immediately after the #! sequence on the first line of
8562 the source file. The kernel reads in the name of the interpreter,
8563 and executes that instead. It passes the interpreter the source
8564 filename as its first argument, with the original arguments
8565 following. Consult the Unix man page for the `exec' system call
8566 for more information.
8568 Now you can use Guile as an interpreter, using a mechanism which is a
8569 compatible subset of that provided by SCSH.
8571 Guile now recognizes a '-s' command line switch, whose argument is the
8572 name of a file of Scheme code to load. It also treats the two
8573 characters `#!' as the start of a comment, terminated by `!#'. Thus,
8574 to make a file of Scheme code directly executable by Unix, insert the
8575 following two lines at the top of the file:
8577 #!/usr/local/bin/guile -s
8580 Guile treats the argument of the `-s' command-line switch as the name
8581 of a file of Scheme code to load, and treats the sequence `#!' as the
8582 start of a block comment, terminated by `!#'.
8584 For example, here's a version of 'echo' written in Scheme:
8586 #!/usr/local/bin/guile -s
8588 (let loop ((args (cdr (program-arguments))))
8591 (display (car args))
8592 (if (pair? (cdr args))
8594 (loop (cdr args)))))
8597 Why does `#!' start a block comment terminated by `!#', instead of the
8598 end of the line? That is the notation SCSH uses, and although we
8599 don't yet support the other SCSH features that motivate that choice,
8600 we would like to be backward-compatible with any existing Guile
8601 scripts once we do. Furthermore, if the path to Guile on your system
8602 is too long for your kernel, you can start the script with this
8606 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
8609 Note that some very old Unix systems don't support the `#!' syntax.
8612 ** You can now run Guile without installing it.
8614 Previous versions of the interactive Guile interpreter (`guile')
8615 couldn't start up unless Guile's Scheme library had been installed;
8616 they used the value of the environment variable `SCHEME_LOAD_PATH'
8617 later on in the startup process, but not to find the startup code
8618 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
8621 To run Guile without installing it, build it in the normal way, and
8622 then set the environment variable `SCHEME_LOAD_PATH' to a
8623 colon-separated list of directories, including the top-level directory
8624 of the Guile sources. For example, if you unpacked Guile so that the
8625 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
8628 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
8631 ** Guile's read-eval-print loop no longer prints #<unspecified>
8632 results. If the user wants to see this, she can evaluate the
8633 expression (assert-repl-print-unspecified #t), perhaps in her startup
8636 ** Guile no longer shows backtraces by default when an error occurs;
8637 however, it does display a message saying how to get one, and how to
8638 request that they be displayed by default. After an error, evaluate
8640 to see a backtrace, and
8641 (debug-enable 'backtrace)
8642 to see them by default.
8646 * Changes to Guile Scheme:
8648 ** Guile now distinguishes between #f and the empty list.
8650 This is for compatibility with the IEEE standard, the (possibly)
8651 upcoming Revised^5 Report on Scheme, and many extant Scheme
8654 Guile used to have #f and '() denote the same object, to make Scheme's
8655 type system more compatible with Emacs Lisp's. However, the change
8656 caused too much trouble for Scheme programmers, and we found another
8657 way to reconcile Emacs Lisp with Scheme that didn't require this.
8660 ** Guile's delq, delv, delete functions, and their destructive
8661 counterparts, delq!, delv!, and delete!, now remove all matching
8662 elements from the list, not just the first. This matches the behavior
8663 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
8664 functions which inspired them.
8666 I recognize that this change may break code in subtle ways, but it
8667 seems best to make the change before the FSF's first Guile release,
8671 ** The compiled-library-path function has been deleted from libguile.
8673 ** The facilities for loading Scheme source files have changed.
8675 *** The variable %load-path now tells Guile which directories to search
8676 for Scheme code. Its value is a list of strings, each of which names
8679 *** The variable %load-extensions now tells Guile which extensions to
8680 try appending to a filename when searching the load path. Its value
8681 is a list of strings. Its default value is ("" ".scm").
8683 *** (%search-load-path FILENAME) searches the directories listed in the
8684 value of the %load-path variable for a Scheme file named FILENAME,
8685 with all the extensions listed in %load-extensions. If it finds a
8686 match, then it returns its full filename. If FILENAME is absolute, it
8687 returns it unchanged. Otherwise, it returns #f.
8689 %search-load-path will not return matches that refer to directories.
8691 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
8692 uses %seach-load-path to find a file named FILENAME, and loads it if
8693 it finds it. If it can't read FILENAME for any reason, it throws an
8696 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
8699 *** load uses the same searching semantics as primitive-load.
8701 *** The functions %try-load, try-load-with-path, %load, load-with-path,
8702 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
8703 path, and load-module-with-path have been deleted. The functions
8704 above should serve their purposes.
8706 *** If the value of the variable %load-hook is a procedure,
8707 `primitive-load' applies its value to the name of the file being
8708 loaded (without the load path directory name prepended). If its value
8709 is #f, it is ignored. Otherwise, an error occurs.
8711 This is mostly useful for printing load notification messages.
8714 ** The function `eval!' is no longer accessible from the scheme level.
8715 We can't allow operations which introduce glocs into the scheme level,
8716 because Guile's type system can't handle these as data. Use `eval' or
8717 `read-and-eval!' (see below) as replacement.
8719 ** The new function read-and-eval! reads an expression from PORT,
8720 evaluates it, and returns the result. This is more efficient than
8721 simply calling `read' and `eval', since it is not necessary to make a
8722 copy of the expression for the evaluator to munge.
8724 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
8725 for the `read' function.
8728 ** The function `int?' has been removed; its definition was identical
8729 to that of `integer?'.
8731 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
8732 use the R4RS names for these functions.
8734 ** The function object-properties no longer returns the hash handle;
8735 it simply returns the object's property list.
8737 ** Many functions have been changed to throw errors, instead of
8738 returning #f on failure. The point of providing exception handling in
8739 the language is to simplify the logic of user code, but this is less
8740 useful if Guile's primitives don't throw exceptions.
8742 ** The function `fileno' has been renamed from `%fileno'.
8744 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
8747 * Changes to Guile's C interface:
8749 ** The library's initialization procedure has been simplified.
8750 scm_boot_guile now has the prototype:
8752 void scm_boot_guile (int ARGC,
8754 void (*main_func) (),
8757 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
8758 MAIN_FUNC should do all the work of the program (initializing other
8759 packages, reading user input, etc.) before returning. When MAIN_FUNC
8760 returns, call exit (0); this function never returns. If you want some
8761 other exit value, MAIN_FUNC may call exit itself.
8763 scm_boot_guile arranges for program-arguments to return the strings
8764 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
8765 scm_set_program_arguments with the final list, so Scheme code will
8766 know which arguments have been processed.
8768 scm_boot_guile establishes a catch-all catch handler which prints an
8769 error message and exits the process. This means that Guile exits in a
8770 coherent way when system errors occur and the user isn't prepared to
8771 handle it. If the user doesn't like this behavior, they can establish
8772 their own universal catcher in MAIN_FUNC to shadow this one.
8774 Why must the caller do all the real work from MAIN_FUNC? The garbage
8775 collector assumes that all local variables of type SCM will be above
8776 scm_boot_guile's stack frame on the stack. If you try to manipulate
8777 SCM values after this function returns, it's the luck of the draw
8778 whether the GC will be able to find the objects you allocate. So,
8779 scm_boot_guile function exits, rather than returning, to discourage
8780 people from making that mistake.
8782 The IN, OUT, and ERR arguments were removed; there are other
8783 convenient ways to override these when desired.
8785 The RESULT argument was deleted; this function should never return.
8787 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
8791 ** Guile's header files should no longer conflict with your system's
8794 In order to compile code which #included <libguile.h>, previous
8795 versions of Guile required you to add a directory containing all the
8796 Guile header files to your #include path. This was a problem, since
8797 Guile's header files have names which conflict with many systems'
8800 Now only <libguile.h> need appear in your #include path; you must
8801 refer to all Guile's other header files as <libguile/mumble.h>.
8802 Guile's installation procedure puts libguile.h in $(includedir), and
8803 the rest in $(includedir)/libguile.
8806 ** Two new C functions, scm_protect_object and scm_unprotect_object,
8807 have been added to the Guile library.
8809 scm_protect_object (OBJ) protects OBJ from the garbage collector.
8810 OBJ will not be freed, even if all other references are dropped,
8811 until someone does scm_unprotect_object (OBJ). Both functions
8814 Note that calls to scm_protect_object do not nest. You can call
8815 scm_protect_object any number of times on a given object, and the
8816 next call to scm_unprotect_object will unprotect it completely.
8818 Basically, scm_protect_object and scm_unprotect_object just
8819 maintain a list of references to things. Since the GC knows about
8820 this list, all objects it mentions stay alive. scm_protect_object
8821 adds its argument to the list; scm_unprotect_object remove its
8822 argument from the list.
8825 ** scm_eval_0str now returns the value of the last expression
8828 ** The new function scm_read_0str reads an s-expression from a
8829 null-terminated string, and returns it.
8831 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
8832 to a Scheme port object.
8834 ** The new function `scm_set_program_arguments' allows C code to set
8835 the value returned by the Scheme `program-arguments' function.
8840 * Guile no longer includes sophisticated Tcl/Tk support.
8842 The old Tcl/Tk support was unsatisfying to us, because it required the
8843 user to link against the Tcl library, as well as Tk and Guile. The
8844 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
8845 referring to widgets by names, rather than exporting widgets to Scheme
8846 code as a special datatype.
8848 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
8849 maintainers described some very interesting changes in progress to the
8850 Tcl/Tk internals, which would facilitate clean interfaces between lone
8851 Tk and other interpreters --- even for garbage-collected languages
8852 like Scheme. They expected the new Tk to be publicly available in the
8855 Since it seems that Guile might soon have a new, cleaner interface to
8856 lone Tk, and that the old Guile/Tk glue code would probably need to be
8857 completely rewritten, we (Jim Blandy and Richard Stallman) have
8858 decided not to support the old code. We'll spend the time instead on
8859 a good interface to the newer Tk, as soon as it is available.
8861 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
8864 Copyright information:
8866 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
8868 Permission is granted to anyone to make or distribute verbatim copies
8869 of this document as received, in any medium, provided that the
8870 copyright notice and this permission notice are preserved,
8871 thus giving the recipient permission to redistribute in turn.
8873 Permission is granted to distribute modified versions
8874 of this document, or of portions of it,
8875 under the above conditions, provided also that they
8876 carry prominent notices stating who last changed them.
8881 paragraph-separate: "[
\f]*$"