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 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 1.9.1 (since the 1.9.0 prerelease):
13 ** `scm_set_port_seek' and `scm_set_port_truncate' use the `scm_t_off' type
15 Previously they would use the `off_t' type, which is fragile since its
16 definition depends on the application's value for `_FILE_OFFSET_BITS'.
18 ** Automatically compiled files will be placed in ~/.cache, not ~/.guile-ccache.
20 Actually, they will be placed in $XDG_CACHE_HOME/guile/ccache/1.9,
21 defaulting to XDG_CACHE_HOME=~/.cache. Users may remove their
22 ~/.guile-ccache directories.
24 ** New language: Brainfuck.
26 Brainfuck is a toy language that closely models Turing machines. Guile's
27 brainfuck compiler is meant to be an example of implementing other
28 languages. See the manual for details, or
29 http://en.wikipedia.org/wiki/Brainfuck for more information about the
30 Brainfuck language itself.
32 ** A number of Scheme files were corrected to be LGPLv3+.
34 Some Scheme files imported for the compiler were erroneously labeled as
35 being LGPLv2+ or GPLv2+. This oversight has been fixed.
37 ** Bytevectors may now be accessed with a C-friendly API.
39 New functions: `scm_is_bytevector ()', `scm_c_bytevector_length ()',
40 `scm_c_bytevector_length ()', and `scm_c_bytevector_set_x ()'. See the
43 ** Bytevectors are now accessible using the generalized-vector API.
45 As a side effect, this change allows compilation of literal bytevectors
48 ** Meta-commands to the REPL work better with strange languages.
50 Specifically, meta-commands that take expressions as arguments will use
51 the current language's reader to read those expressions, which may span
52 multiple lines, with readline integration if the user has that enabled.
54 ** The object code file format has changed.
56 The objcode loader will complain about a "bad header cookie" if it
57 happens to find an old file. The workaround for that is currently to
58 find all stale .go files and remove them. This is likely to affect users
59 who have checked out Guile's git repository, not those that build from
62 ** Vector access has been sped up considerably.
64 Guile's virtual machine now has vector and bytevector operations. Using
65 Guile to process large amounts of data is now easier. This is because
66 `vector-ref' and `vector-set!' now have fast opcodes. In addition, there
67 are opcodes for `ref' and `set' operations on bytevectors for everything
68 from 8-bit integers to 64-bit floating-point values.
70 In the next release, we hope to extend this speedup to other kinds of
73 ** The `long_long' C type, deprecated in 1.8, has been removed.
75 ** And of course, the usual collection of bugfixes.
77 Interested users should see the ChangeLog for more information.
79 Changes in 1.9.x (since the 1.8.x series):
81 * New modules (see the manual for details)
83 ** `(srfi srfi-18)', more sophisticated multithreading support
84 ** `(ice-9 i18n)', internationalization support
85 ** `(rnrs bytevector)', the R6RS bytevector API
86 ** `(rnrs io ports)', a subset of the R6RS I/O port API
87 ** `(system xref)', a cross-referencing facility (FIXME undocumented)
89 * Changes to the stand-alone interpreter
91 ** Guile now can compile Scheme to bytecode for a custom virtual machine.
93 Compiled code loads much faster than Scheme source code, and runs around
94 3 or 4 times as fast, generating much less garbage in the process.
96 ** The stack limit is now initialized from the environment.
98 If getrlimit(2) is available and a stack limit is set, Guile will set
99 its stack limit to 80% of the rlimit. Otherwise the limit is 160000
100 words, a four-fold increase from the earlier default limit.
102 ** New environment variables: GUILE_LOAD_COMPILED_PATH,
103 GUILE_SYSTEM_LOAD_COMPILED_PATH
105 GUILE_LOAD_COMPILED_PATH is for compiled files what GUILE_LOAD_PATH is
106 for source files. It is a different path, however, because compiled
107 files are architecture-specific. GUILE_SYSTEM_LOAD_COMPILED_PATH is like
110 ** New read-eval-print loop (REPL) implementation
112 Running Guile with no arguments drops the user into the new REPL. While
113 it is self-documenting to an extent, the new REPL has not yet been
114 documented in the manual. This will be fixed before 2.0.
116 ** New `guile-tools' commands: `compile', `disassemble'
118 Pass the `--help' command-line option to these commands for more
121 * Changes to Scheme functions and syntax
123 ** Procedure removed: `the-environment'
125 This procedure was part of the interpreter's execution model, and does
126 not apply to the compiler.
128 ** Files loaded with `primitive-load-path' will now be compiled
131 If a compiled .go file corresponding to a .scm file is not found or is
132 not fresh, the .scm file will be compiled on the fly, and the resulting
133 .go file stored away. An advisory note will be printed on the console.
135 Note that this mechanism depends on preservation of the .scm and .go
136 modification times; if the .scm or .go files are moved after
137 installation, care should be taken to preserve their original
140 Autocompiled files will be stored in the $XDG_CACHE_HOME/guile/ccache
141 directory, where $XDG_CACHE_HOME defaults to ~/.cache. This directory
142 will be created if needed.
144 To inhibit autocompilation, set the GUILE_AUTO_COMPILE environment
145 variable to 0, or pass --no-autocompile on the Guile command line.
147 Note that there is currently a bug here: automatic compilation will
148 sometimes be attempted when it shouldn't.
150 For example, the old (lang elisp) modules are meant to be interpreted,
151 not compiled. This bug will be fixed before 2.0. FIXME 2.0: Should say
152 something here about module-transformer called for compile.
154 ** New POSIX procedures: `getrlimit' and `setrlimit'
156 Note however that the interface of these functions is likely to change
157 in the next prerelease.
159 ** New procedure in `(oops goops)': `method-formals'
161 ** BUG: (procedure-property func 'arity) does not work on compiled
164 This will be fixed one way or another before 2.0.
166 ** New procedures in (ice-9 session): `add-value-help-handler!',
167 `remove-value-help-handler!', `add-name-help-handler!'
168 `remove-name-help-handler!', `procedure-arguments',
170 The value and name help handlers provide some minimal extensibility to
171 the help interface. Guile-lib's `(texinfo reflection)' uses them, for
172 example, to make stexinfo help documentation available. See those
173 procedures' docstrings for more information.
175 `procedure-arguments' describes the arguments that a procedure can take,
176 combining arity and formals. For example:
178 (procedure-arguments resolve-interface)
179 => ((required . (name)) (rest . args))
181 Additionally, `module-commentary' is now publically exported from
184 ** Deprecated: `procedure->memoizing-macro', `procedure->syntax'
186 These procedures will not work with syncase expansion, and indeed are
187 not used in the normal course of Guile. They are still used by the old
188 Emacs Lisp support, however.
190 ** New language: ECMAScript
192 Guile now ships with one other high-level language supported,
193 ECMAScript. The goal is to support all of version 3.1 of the standard,
194 but not all of the libraries are there yet. This support is not yet
195 documented; ask on the mailing list if you are interested.
197 ** New language: Brainfuck
199 Brainfuck is a toy language that closely models Turing machines. Guile's
200 brainfuck compiler is meant to be an example of implementing other
201 languages. See the manual for details, or
202 http://en.wikipedia.org/wiki/Brainfuck for more information about the
203 Brainfuck language itself.
205 ** Defmacros may now have docstrings.
207 Indeed, any macro may have a docstring. `object-documentation' from
208 `(ice-9 documentation)' may be used to retrieve the docstring, once you
209 have a macro value -- but see the above note about first-class macros.
210 Docstrings are associated with the syntax transformer procedures.
212 ** The psyntax expander now knows how to interpret the @ and @@ special
215 ** The psyntax expander is now hygienic with respect to modules.
217 Free variables in a macro are scoped in the module that the macro was
218 defined in, not in the module the macro is used in. For example, code
221 (define-module (foo) #:export (bar))
222 (define (helper x) ...)
224 (syntax-rules () ((_ x) (helper x))))
226 (define-module (baz) #:use-module (foo))
229 It used to be you had to export `helper' from `(foo)' as well.
230 Thankfully, this has been fixed.
232 ** New function, `procedure-module'
234 While useful on its own, `procedure-module' is used by psyntax on syntax
235 transformers to determine the module in which to scope introduced
238 ** `eval-case' has been deprecated, and replaced by `eval-when'.
240 The semantics of `eval-when' are easier to understand. It is still
241 missing documentation, however.
243 ** Guile is now more strict about prohibiting definitions in expression
246 Although previous versions of Guile accepted it, the following
247 expression is not valid, in R5RS or R6RS:
249 (if test (define foo 'bar) (define foo 'baz))
251 In this specific case, it would be better to do:
253 (define foo (if test 'bar 'baz))
255 It is certainly possible to circumvent this resriction with e.g.
256 `(module-define! (current-module) 'foo 'baz)'. We would appreciate
257 feedback about this change (a consequence of using psyntax as the
258 default expander), and may choose to revisit this situation before 2.0
259 in response to user feedback.
261 ** Defmacros must now produce valid Scheme expressions.
263 It used to be that defmacros could unquote in Scheme values, as a way of
264 supporting partial evaluation, and avoiding some hygiene issues. For
267 (define (helper x) ...)
268 (define-macro (foo bar)
271 Assuming this macro is in the `(baz)' module, the direct translation of
274 (define (helper x) ...)
275 (define-macro (foo bar)
276 `((@@ (baz) helper) ,bar))
278 Of course, one could just use a hygienic macro instead:
282 ((_ bar) (helper bar))))
284 ** Guile's psyntax now supports docstrings and internal definitions.
286 The following Scheme is not strictly legal:
293 However its intent is fairly clear. Guile interprets "bar" to be the
294 docstring of `foo', and the definition of `baz' is still in definition
297 ** Macros need to be defined before their first use.
299 It used to be that with lazy memoization, this might work:
303 (define-macro (ref x) x)
306 But now, the body of `foo' is interpreted to mean a call to the toplevel
307 `ref' function, instead of a macro expansion. The solution is to define
308 macros before code that uses them.
310 ** Functions needed by macros at expand-time need to be present at
313 For example, this code will work at the REPL:
315 (define (double-helper x) (* x x))
316 (define-macro (double-literal x) (double-helper x))
317 (double-literal 2) => 4
319 But it will not work when a file is compiled, because the definition of
320 `double-helper' is not present at expand-time. The solution is to wrap
321 the definition of `double-helper' in `eval-when':
323 (eval-when (load compile eval)
324 (define (double-helper x) (* x x)))
325 (define-macro (double-literal x) (double-helper x))
326 (double-literal 2) => 4
328 See the (currently missing) documentation for eval-when for more
331 ** New variable, %pre-modules-transformer
333 Need to document this one some more.
335 ** Temporarily removed functions: `macroexpand', `macroexpand-1'
337 `macroexpand' will be added back before 2.0. It is unclear how to
338 implement `macroexpand-1' with syntax-case, though PLT Scheme does prove
341 ** New reader macros: #' #` #, #,@
343 These macros translate, respectively, to `syntax', `quasisyntax',
344 `unsyntax', and `unsyntax-splicing'. See the R6RS for more information.
345 These reader macros may be overridden by `read-hash-extend'.
347 ** Incompatible change to #'
349 Guile did have a #' hash-extension, by default, which just returned the
350 subsequent datum: #'foo => foo. In the unlikely event that anyone
351 actually used this, this behavior may be reinstated via the
352 `read-hash-extend' mechanism.
354 ** Scheme expresssions may be commented out with #;
356 #; comments out an entire expression. See SRFI-62 or the R6RS for more
359 ** `make-stack' with a tail-called procedural narrowing argument no longer
360 works (with compiled procedures)
362 It used to be the case that a captured stack could be narrowed to select
363 calls only up to or from a certain procedure, even if that procedure
364 already tail-called another procedure. This was because the debug
365 information from the original procedure was kept on the stack.
367 Now with the new compiler, the stack only contains active frames from
368 the current continuation. A narrow to a procedure that is not in the
369 stack will result in an empty stack. To fix this, narrow to a procedure
370 that is active in the current continuation, or narrow to a specific
371 number of stack frames.
373 ** backtraces through compiled procedures only show procedures that are
374 active in the current continuation
376 Similarly to the previous issue, backtraces in compiled code may be
377 different from backtraces in interpreted code. There are no semantic
378 differences, however. Please mail bug-guile@gnu.org if you see any
379 deficiencies with Guile's backtraces.
381 ** syntax-rules and syntax-case macros now propagate source information
382 through to the expanded code
384 This should result in better backtraces.
386 ** The currying behavior of `define' has been removed.
388 Before, `(define ((f a) b) (* a b))' would translate to
390 (define f (lambda (a) (lambda (b) (* a b))))
392 Now a syntax error is signaled, as this syntax is not supported by
393 default. If there is sufficient demand, this syntax can be supported
396 ** All modules have names now
398 Before, you could have anonymous modules: modules without names. Now,
399 because of hygiene and macros, all modules have names. If a module was
400 created without a name, the first time `module-name' is called on it, a
401 fresh name will be lazily generated for it.
403 ** Many syntax errors have different texts now
405 Syntax errors still throw to the `syntax-error' key, but the arguments
406 are often different now. Perhaps in the future, Guile will switch to
407 using standard SRFI-35 conditions.
409 ** Returning multiple values to compiled code will silently truncate the
410 values to the expected number
412 For example, the interpreter would raise an error evaluating the form,
413 `(+ (values 1 2) (values 3 4))', because it would see the operands as
414 being two compound "values" objects, to which `+' does not apply.
416 The compiler, on the other hand, receives multiple values on the stack,
417 not as a compound object. Given that it must check the number of values
418 anyway, if too many values are provided for a continuation, it chooses
419 to truncate those values, effectively evaluating `(+ 1 3)' instead.
421 The idea is that the semantics that the compiler implements is more
422 intuitive, and the use of the interpreter will fade out with time.
423 This behavior is allowed both by the R5RS and the R6RS.
425 ** Multiple values in compiled code are not represented by compound
428 This change may manifest itself in the following situation:
430 (let ((val (foo))) (do-something) val)
432 In the interpreter, if `foo' returns multiple values, multiple values
433 are produced from the `let' expression. In the compiler, those values
434 are truncated to the first value, and that first value is returned. In
435 the compiler, if `foo' returns no values, an error will be raised, while
436 the interpreter would proceed.
438 Both of these behaviors are allowed by R5RS and R6RS. The compiler's
439 behavior is more correct, however. If you wish to preserve a potentially
440 multiply-valued return, you will need to set up a multiple-value
441 continuation, using `call-with-values'.
443 ** Defmacros are now implemented in terms of syntax-case.
445 The practical ramification of this is that the `defmacro?' predicate has
446 been removed, along with `defmacro-transformer', `macro-table',
447 `xformer-table', `assert-defmacro?!', `set-defmacro-transformer!' and
448 `defmacro:transformer'. This is because defmacros are simply macros. If
449 any of these procedures provided useful facilities to you, we encourage
450 you to contact the Guile developers.
452 ** psyntax is now the default expander
454 Scheme code is now expanded by default by the psyntax hygienic macro
455 expander. Expansion is performed completely before compilation or
458 Notably, syntax errors will be signalled before interpretation begins.
459 In the past, many syntax errors were only detected at runtime if the
460 code in question was memoized.
462 As part of its expansion, psyntax renames all lexically-bound
463 identifiers. Original identifier names are preserved and given to the
464 compiler, but the interpreter will see the renamed variables, e.g.,
465 `x432' instead of `x'.
467 Note that the psyntax that Guile uses is a fork, as Guile already had
468 modules before incompatible modules were added to psyntax -- about 10
469 years ago! Thus there are surely a number of bugs that have been fixed
470 in psyntax since then. If you find one, please notify bug-guile@gnu.org.
472 ** syntax-rules and syntax-case are available by default.
474 There is no longer any need to import the `(ice-9 syncase)' module
475 (which is now deprecated). The expander may be invoked directly via
476 `sc-expand', though it is normally searched for via the current module
479 Also, the helper routines for syntax-case are available in the default
480 environment as well: `syntax->datum', `datum->syntax',
481 `bound-identifier=?', `free-identifier=?', `generate-temporaries',
482 `identifier?', and `syntax-violation'. See the R6RS for documentation.
484 ** Lexical bindings introduced by hygienic macros may not be referenced
485 by nonhygienic macros.
487 If a lexical binding is introduced by a hygienic macro, it may not be
488 referenced by a nonhygienic macro. For example, this works:
491 (define-macro (bind-x val body)
492 `(let ((x ,val)) ,body))
493 (define-macro (ref x)
500 (define-syntax bind-x
502 ((_ val body) (let ((x val)) body))))
503 (define-macro (ref x)
507 It is not normal to run into this situation with existing code. However,
508 as code is ported over from defmacros to syntax-case, it is possible to
509 run into situations like this. In the future, Guile will probably port
510 its `while' macro to syntax-case, which makes this issue one to know
513 ** Macros may no longer be referenced as first-class values.
515 In the past, you could evaluate e.g. `if', and get its macro value. Now,
516 expanding this form raises a syntax error.
518 Macros still /exist/ as first-class values, but they must be
519 /referenced/ via the module system, e.g. `(module-ref (current-module)
522 This decision may be revisited before the 2.0 release. Feedback welcome
523 to guile-devel@gnu.org (subscription required) or bug-guile@gnu.org (no
524 subscription required).
526 ** New macro type: syncase-macro
528 XXX Need to decide whether to document this for 2.0, probably should:
529 make-syncase-macro, make-extended-syncase-macro, macro-type,
530 syncase-macro-type, syncase-macro-binding
532 ** A new `memoize-symbol' evaluator trap has been added.
534 This trap can be used for efficiently implementing a Scheme code
537 ** Duplicate bindings among used modules are resolved lazily.
539 This slightly improves program startup times.
541 ** New thread cancellation and thread cleanup API
543 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
545 ** Fix bad interaction between `false-if-exception' and stack-call.
547 Exceptions thrown by `false-if-exception' were erronously causing the
548 stack to be saved, causing later errors to show the incorrectly-saved
549 backtrace. This has been fixed.
551 ** New global variables: %load-compiled-path, %load-compiled-extensions
553 These are analogous to %load-path and %load-extensions.
555 ** New procedure, `make-promise'
557 `(make-promise (lambda () foo))' is equivalent to `(delay foo)'.
559 ** New entry into %guile-build-info: `ccachedir'
561 ** Fix bug in `module-bound?'.
563 `module-bound?' was returning true if a module did have a local
564 variable, but one that was unbound, but another imported module bound
565 the variable. This was an error, and was fixed.
567 ** `(ice-9 syncase)' has been deprecated.
569 As syntax-case is available by default, importing `(ice-9 syncase)' has
570 no effect, and will trigger a deprecation warning.
572 * Changes to the C interface
574 ** The GH interface (deprecated in version 1.6, 2001) was removed.
576 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
578 This makes these internal functions technically not callable from
581 ** Functions for handling `scm_option' now no longer require an argument
582 indicating length of the `scm_t_option' array.
584 ** scm_primitive_load_path has additional argument, exception_on_error
586 ** New C function: scm_module_public_interface
588 This procedure corresponds to Scheme's `module-public-interface'.
590 ** `scm_stat' has an additional argument, `exception_on_error'
591 ** `scm_primitive_load_path' has an additional argument `exception_on_not_found'
593 ** `scm_set_port_seek' and `scm_set_port_truncate' use the `scm_t_off' type
595 Previously they would use the `off_t' type, which is fragile since its
596 definition depends on the application's value for `_FILE_OFFSET_BITS'.
598 ** The `long_long' C type, deprecated in 1.8, has been removed
600 * Changes to the distribution
602 ** Guile's license is now LGPLv3+
604 In other words the GNU Lesser General Public License, version 3 or
605 later (at the discretion of each person that chooses to redistribute
608 ** `guile-config' will be deprecated in favor of `pkg-config'
610 `guile-config' has been rewritten to get its information from
611 `pkg-config', so this should be a transparent change. Note however that
612 guile.m4 has yet to be modified to call pkg-config instead of
615 ** New installation directory: $(pkglibdir)/1.9/ccache
617 If $(libdir) is /usr/lib, for example, Guile will install its .go files
618 to /usr/lib/guile/1.9/ccache. These files are architecture-specific.
620 ** New dependency: GNU libunistring.
622 See http://www.gnu.org/software/libunistring/. We hope to merge in
623 Unicode support in the next prerelease.
627 Changes in 1.8.7 (since 1.8.6)
629 * New modules (see the manual for details)
631 ** `(srfi srfi-98)', an interface to access environment variables
635 ** Fix compilation with `--disable-deprecated'
636 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
637 ** Fix MinGW build problem caused by HAVE_STRUCT_TIMESPEC confusion
638 ** Fix build problem when scm_t_timespec is different from struct timespec
639 ** Fix build when compiled with -Wundef -Werror
640 ** More build fixes for `alphaev56-dec-osf5.1b' (Tru64)
641 ** Build fixes for `powerpc-ibm-aix5.3.0.0' (AIX 5.3)
642 ** With GCC, always compile with `-mieee' on `alpha*' and `sh*'
643 ** Better diagnose broken `(strftime "%z" ...)' in `time.test' (bug #24130)
644 ** Fix parsing of SRFI-88/postfix keywords longer than 128 characters
645 ** Fix reading of complex numbers where both parts are inexact decimals
647 ** Allow @ macro to work with (ice-9 syncase)
649 Previously, use of the @ macro in a module whose code is being
650 transformed by (ice-9 syncase) would cause an "Invalid syntax" error.
651 Now it works as you would expect (giving the value of the specified
654 ** Have `scm_take_locale_symbol ()' return an interned symbol (bug #25865)
657 Changes in 1.8.6 (since 1.8.5)
659 * New features (see the manual for details)
661 ** New convenience function `scm_c_symbol_length ()'
663 ** Single stepping through code from Emacs
665 When you use GDS to evaluate Scheme code from Emacs, you can now use
666 `C-u' to indicate that you want to single step through that code. See
667 `Evaluating Scheme Code' in the manual for more details.
669 ** New "guile(1)" man page!
671 * Changes to the distribution
673 ** Automake's `AM_MAINTAINER_MODE' is no longer used
675 Thus, the `--enable-maintainer-mode' configure option is no longer
676 available: Guile is now always configured in "maintainer mode".
678 ** `ChangeLog' files are no longer updated
680 Instead, changes are detailed in the version control system's logs. See
681 the top-level `ChangeLog' files for details.
686 ** `symbol->string' now returns a read-only string, as per R5RS
687 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
688 ** `guile-config link' now prints `-L$libdir' before `-lguile'
689 ** Fix memory corruption involving GOOPS' `class-redefinition'
690 ** Fix possible deadlock in `mutex-lock'
691 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
692 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
693 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
694 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
695 ** Fix misleading output from `(help rationalize)'
696 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
697 ** Fix `gcd' when called with a single, negative argument.
698 ** Fix `Stack overflow' errors seen when building on some platforms
699 ** Fix bug when `scm_with_guile ()' was called several times from the
701 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
702 dynamic environment of the call to `raise'
703 ** Fix potential deadlock in `make-struct'
704 ** Fix compilation problem with libltdl from Libtool 2.2.x
705 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
708 Changes in 1.8.5 (since 1.8.4)
710 * Infrastructure changes
712 ** Guile repository switched from CVS to Git
714 The new repository can be accessed using
715 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
716 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
718 ** Add support for `pkg-config'
720 See "Autoconf Support" in the manual for details.
722 * New modules (see the manual for details)
726 * New features (see the manual for details)
728 ** New `postfix' read option, for SRFI-88 keyword syntax
729 ** Some I/O primitives have been inlined, which improves I/O performance
730 ** New object-based traps infrastructure
732 This is a GOOPS-based infrastructure that builds on Guile's low-level
733 evaluator trap calls and facilitates the development of debugging
734 features like single-stepping, breakpoints, tracing and profiling.
735 See the `Traps' node of the manual for details.
737 ** New support for working on Guile code from within Emacs
739 Guile now incorporates the `GDS' library (previously distributed
740 separately) for working on Guile code from within Emacs. See the
741 `Using Guile In Emacs' node of the manual for details.
745 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
746 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
748 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
749 would trigger an unbound variable error for `match:andmap'.
751 ** `(oop goops describe)' now properly provides the `describe' feature
752 ** Fixed `args-fold' from `(srfi srfi-37)'
754 Previously, parsing short option names of argument-less options would
755 lead to a stack overflow.
757 ** `(srfi srfi-35)' is now visible through `cond-expand'
758 ** Fixed type-checking for the second argument of `eval'
759 ** Fixed type-checking for SRFI-1 `partition'
760 ** Fixed `struct-ref' and `struct-set!' on "light structs"
761 ** Honor struct field access rights in GOOPS
762 ** Changed the storage strategy of source properties, which fixes a deadlock
763 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
764 ** Fixed build issue for GNU/Linux on IA64
765 ** Fixed build issues on NetBSD 1.6
766 ** Fixed build issue on Solaris 2.10 x86_64
767 ** Fixed build issue with DEC/Compaq/HP's compiler
768 ** Fixed `scm_from_complex_double' build issue on FreeBSD
769 ** Fixed `alloca' build issue on FreeBSD 6
770 ** Removed use of non-portable makefile constructs
771 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
772 ** Make sure all tests honor `$TMPDIR'
775 Changes in 1.8.4 (since 1.8.3)
779 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
780 ** Fixed a segmentation fault which occurred when displaying the
781 backtrace of a stack with a promise object (made by `delay') in it.
782 ** Make `accept' leave guile mode while blocking
783 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
784 ** Fixed a build problem on AIX (use of func_data identifier)
785 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
786 called with an associator proc that returns neither a pair nor #f.
787 ** Secondary threads now always return a valid module for (current-module).
788 ** Avoid MacOS build problems caused by incorrect combination of "64"
789 system and library calls.
790 ** `guile-snarf' now honors `$TMPDIR'
791 ** `guile-config compile' now reports CPPFLAGS used at compile-time
792 ** Fixed build with Sun Studio (Solaris 9)
793 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
794 uniform vectors on AIX.
795 ** Fixed a deadlock that occurs upon GC with multiple threads.
796 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
797 ** Fixed autotool-derived build problems on AIX 6.1.
798 ** Fixed NetBSD/alpha support
799 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
801 * New modules (see the manual for details)
805 * Documentation fixes and improvements
807 ** Removed premature breakpoint documentation
809 The features described are not available in the series of 1.8.x
810 releases, so the documentation was misleading and has been removed.
812 ** More about Guile's default *random-state* variable
814 ** GOOPS: more about how to use `next-method'
816 * Changes to the distribution
818 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
820 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
821 General Public License, and the few incorrect files have now been
822 fixed to agree with the rest of the Guile distribution.
824 ** Removed unnecessary extra copies of COPYING*
826 The distribution now contains a single COPYING.LESSER at its top level.
829 Changes in 1.8.3 (since 1.8.2)
831 * New modules (see the manual for details)
838 ** The `(ice-9 slib)' module now works as expected
839 ** Expressions like "(set! 'x #t)" no longer yield a crash
840 ** Warnings about duplicate bindings now go to stderr
841 ** A memory leak in `make-socket-address' was fixed
842 ** Alignment issues (e.g., on SPARC) in network routines were fixed
843 ** A threading issue that showed up at least on NetBSD was fixed
844 ** Build problems on Solaris and IRIX fixed
846 * Implementation improvements
848 ** The reader is now faster, which reduces startup time
849 ** Procedures returned by `record-accessor' and `record-modifier' are faster
852 Changes in 1.8.2 (since 1.8.1):
854 * New procedures (see the manual for details)
856 ** set-program-arguments
859 * Incompatible changes
861 ** The body of a top-level `define' no longer sees the binding being created
863 In a top-level `define', the binding being created is no longer visible
864 from the `define' body. This breaks code like
865 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
866 unbound in the body. However, such code was not R5RS-compliant anyway,
871 ** Fractions were not `equal?' if stored in unreduced form.
872 (A subtle problem, since printing a value reduced it, making it work.)
873 ** srfi-60 `copy-bit' failed on 64-bit systems
874 ** "guile --use-srfi" option at the REPL can replace core functions
875 (Programs run with that option were ok, but in the interactive REPL
876 the core bindings got priority, preventing SRFI replacements or
878 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
879 ** `kill' on mingw throws an error for a PID other than oneself
880 ** Procedure names are attached to procedure-with-setters
881 ** Array read syntax works with negative lower bound
882 ** `array-in-bounds?' fix if an array has different lower bounds on each index
883 ** `*' returns exact 0 for "(* inexact 0)"
884 This follows what it always did for "(* 0 inexact)".
885 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
886 ** SRFI-19: `date->julian-day' did not account for timezone offset
887 ** `ttyname' no longer crashes when passed a non-tty argument
888 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
889 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
890 ** GOOPS: Fixed a bug in `method-more-specific?'
891 ** Build problems on Solaris fixed
892 ** Build problems on HP-UX IA64 fixed
893 ** Build problems on MinGW fixed
896 Changes in 1.8.1 (since 1.8.0):
898 * LFS functions are now used to access 64-bit files on 32-bit systems.
900 * New procedures (see the manual for details)
902 ** primitive-_exit - [Scheme] the-root-module
903 ** scm_primitive__exit - [C]
904 ** make-completion-function - [Scheme] (ice-9 readline)
905 ** scm_c_locale_stringn_to_number - [C]
906 ** scm_srfi1_append_reverse [C]
907 ** scm_srfi1_append_reverse_x [C]
915 ** Build problems have been fixed on MacOS, SunOS, and QNX.
917 ** `strftime' fix sign of %z timezone offset.
919 ** A one-dimensional array can now be 'equal?' to a vector.
921 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
923 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
925 ** `record-accessor' and `record-modifier' now have strict type checks.
927 Record accessor and modifier procedures now throw an error if the
928 record type of the record they're given is not the type expected.
929 (Previously accessors returned #f and modifiers silently did nothing).
931 ** It is now OK to use both autoload and use-modules on a given module.
933 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
935 Previously there was no checking on primatives like make-vector that
936 accept "one or two" arguments. Now there is.
938 ** The srfi-1 assoc function now calls its equality predicate properly.
940 Previously srfi-1 assoc would call the equality predicate with the key
941 last. According to the SRFI, the key should be first.
943 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
945 ** The array-set! procedure no longer segfaults when given a bit vector.
947 ** Bugs in make-shared-array have been fixed.
949 ** string<? and friends now follow char<? etc order on 8-bit chars.
951 ** The format procedure now handles inf and nan values for ~f correctly.
953 ** exact->inexact should no longer overflow when given certain large fractions.
955 ** srfi-9 accessor and modifier procedures now have strict record type checks.
957 This matches the srfi-9 specification.
959 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
961 Previously the (ice-9 ftw) procedures would ignore any file that had
962 the same inode number as a file they had already seen, even if that
963 file was on a different device.
966 Changes in 1.8.0 (changes since the 1.6.x series):
968 * Changes to the distribution
970 ** Guile is now licensed with the GNU Lesser General Public License.
972 ** The manual is now licensed with the GNU Free Documentation License.
974 ** Guile now requires GNU MP (http://swox.com/gmp).
976 Guile now uses the GNU MP library for arbitrary precision arithmetic.
978 ** Guile now has separate private and public configuration headers.
980 That is, things like HAVE_STRING_H no longer leak from Guile's
983 ** Guile now provides and uses an "effective" version number.
985 Guile now provides scm_effective_version and effective-version
986 functions which return the "effective" version number. This is just
987 the normal full version string without the final micro-version number,
988 so the current effective-version is "1.8". The effective version
989 should remain unchanged during a stable series, and should be used for
990 items like the versioned share directory name
991 i.e. /usr/share/guile/1.8.
993 Providing an unchanging version number during a stable release for
994 things like the versioned share directory can be particularly
995 important for Guile "add-on" packages, since it provides a directory
996 that they can install to that won't be changed out from under them
997 with each micro release during a stable series.
999 ** Thread implementation has changed.
1001 When you configure "--with-threads=null", you will get the usual
1002 threading API (call-with-new-thread, make-mutex, etc), but you can't
1003 actually create new threads. Also, "--with-threads=no" is now
1004 equivalent to "--with-threads=null". This means that the thread API
1005 is always present, although you might not be able to create new
1008 When you configure "--with-threads=pthreads" or "--with-threads=yes",
1009 you will get threads that are implemented with the portable POSIX
1010 threads. These threads can run concurrently (unlike the previous
1011 "coop" thread implementation), but need to cooperate for things like
1014 The default is "pthreads", unless your platform doesn't have pthreads,
1015 in which case "null" threads are used.
1017 See the manual for details, nodes "Initialization", "Multi-Threading",
1018 "Blocking", and others.
1020 ** There is the new notion of 'discouraged' features.
1022 This is a milder form of deprecation.
1024 Things that are discouraged should not be used in new code, but it is
1025 OK to leave them in old code for now. When a discouraged feature is
1026 used, no warning message is printed like there is for 'deprecated'
1027 features. Also, things that are merely discouraged are nevertheless
1028 implemented efficiently, while deprecated features can be very slow.
1030 You can omit discouraged features from libguile by configuring it with
1031 the '--disable-discouraged' option.
1033 ** Deprecation warnings can be controlled at run-time.
1035 (debug-enable 'warn-deprecated) switches them on and (debug-disable
1036 'warn-deprecated) switches them off.
1038 ** Support for SRFI 61, extended cond syntax for multiple values has
1041 This SRFI is always available.
1043 ** Support for require-extension, SRFI-55, has been added.
1045 The SRFI-55 special form `require-extension' has been added. It is
1046 available at startup, and provides a portable way to load Scheme
1047 extensions. SRFI-55 only requires support for one type of extension,
1048 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
1051 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
1053 The (srfi srfi-26) module is an implementation of SRFI-26 which
1054 provides the `cut' and `cute' syntax. These may be used to specialize
1055 parameters without currying.
1057 ** New module (srfi srfi-31)
1059 This is an implementation of SRFI-31 which provides a special form
1060 `rec' for recursive evaluation.
1062 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
1063 been merged with the core, making their functionality always
1066 The modules are still available, tho, and you could use them together
1067 with a renaming import, for example.
1069 ** Guile no longer includes its own version of libltdl.
1071 The official version is good enough now.
1073 ** The --enable-htmldoc option has been removed from 'configure'.
1075 Support for translating the documentation into HTML is now always
1076 provided. Use 'make html'.
1078 ** New module (ice-9 serialize):
1080 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
1081 don't trust the thread safety of most of your program, but where you
1082 have some section(s) of code which you consider can run in parallel to
1083 other sections. See ice-9/serialize.scm for more information.
1085 ** The configure option '--disable-arrays' has been removed.
1087 Support for arrays and uniform numeric arrays is now always included
1090 * Changes to the stand-alone interpreter
1092 ** New command line option `-L'.
1094 This option adds a directory to the front of the load path.
1096 ** New command line option `--no-debug'.
1098 Specifying `--no-debug' on the command line will keep the debugging
1099 evaluator turned off, even for interactive sessions.
1101 ** User-init file ~/.guile is now loaded with the debugging evaluator.
1103 Previously, the normal evaluator would have been used. Using the
1104 debugging evaluator gives better error messages.
1106 ** The '-e' option now 'read's its argument.
1108 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
1109 be used with '-e'. For example, you can now write a script like
1112 exec guile -e '(@ (demo) main)' -s "$0" "$@"
1115 (define-module (demo)
1119 (format #t "Demo: ~a~%" args))
1122 * Changes to Scheme functions and syntax
1124 ** Guardians have changed back to their original semantics
1126 Guardians now behave like described in the paper by Dybvig et al. In
1127 particular, they no longer make guarantees about the order in which
1128 they return objects, and they can no longer be greedy.
1130 They no longer drop cyclic data structures.
1132 The C function scm_make_guardian has been changed incompatibly and no
1133 longer takes the 'greedy_p' argument.
1135 ** New function hashx-remove!
1137 This function completes the set of 'hashx' functions.
1139 ** The concept of dynamic roots has been factored into continuation
1140 barriers and dynamic states.
1142 Each thread has a current dynamic state that carries the values of the
1143 fluids. You can create and copy dynamic states and use them as the
1144 second argument for 'eval'. See "Fluids and Dynamic States" in the
1147 To restrict the influence that captured continuations can have on the
1148 control flow, you can errect continuation barriers. See "Continuation
1149 Barriers" in the manual.
1151 The function call-with-dynamic-root now essentially temporarily
1152 installs a new dynamic state and errects a continuation barrier.
1154 ** The default load path no longer includes "." at the end.
1156 Automatically loading modules from the current directory should not
1157 happen by default. If you want to allow it in a more controlled
1158 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
1159 variable %load-path.
1161 ** The uniform vector and array support has been overhauled.
1163 It now complies with SRFI-4 and the weird prototype based uniform
1164 array creation has been deprecated. See the manual for more details.
1166 Some non-compatible changes have been made:
1167 - characters can no longer be stored into byte arrays.
1168 - strings and bit vectors are no longer considered to be uniform numeric
1170 - array-rank throws an error for non-arrays instead of returning zero.
1171 - array-ref does no longer accept non-arrays when no indices are given.
1173 There is the new notion of 'generalized vectors' and corresponding
1174 procedures like 'generalized-vector-ref'. Generalized vectors include
1175 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
1177 Arrays use generalized vectors as their storage, so that you still
1178 have arrays of characters, bits, etc. However, uniform-array-read!
1179 and uniform-array-write can no longer read/write strings and
1182 ** There is now support for copy-on-write substrings, mutation-sharing
1183 substrings and read-only strings.
1185 Three new procedures are related to this: substring/shared,
1186 substring/copy, and substring/read-only. See the manual for more
1189 ** Backtraces will now highlight the value that caused the error.
1191 By default, these values are enclosed in "{...}", such as in this
1200 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
1201 <unnamed port>:1:1: Wrong type (expecting pair): a
1202 ABORT: (wrong-type-arg)
1204 The prefix and suffix used for highlighting can be set via the two new
1205 printer options 'highlight-prefix' and 'highlight-suffix'. For
1206 example, putting this into ~/.guile will output the bad value in bold
1207 on an ANSI terminal:
1209 (print-set! highlight-prefix "\x1b[1m")
1210 (print-set! highlight-suffix "\x1b[22m")
1213 ** 'gettext' support for internationalization has been added.
1215 See the manual for details.
1217 ** New syntax '@' and '@@':
1219 You can now directly refer to variables exported from a module by
1222 (@ MODULE-NAME VARIABLE-NAME)
1224 For example (@ (ice-9 pretty-print) pretty-print) will directly access
1225 the pretty-print variable exported from the (ice-9 pretty-print)
1226 module. You don't need to 'use' that module first. You can also use
1227 '@' as a target of 'set!', as in (set! (@ mod var) val).
1229 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
1230 but it can also access variables that have not been exported. It is
1231 intended only for kluges and temporary fixes and for debugging, not
1234 ** Keyword syntax has been made more disciplined.
1236 Previously, the name of a keyword was read as a 'token' but printed as
1237 a symbol. Now, it is read as a general Scheme datum which must be a
1248 ERROR: In expression (a b c):
1252 ERROR: Unbound variable: foo
1257 ERROR: Wrong type (expecting symbol): 12
1261 ERROR: Wrong type (expecting symbol): (a b c)
1265 ** The printing of symbols that might look like keywords can be
1268 The new printer option 'quote-keywordish-symbols' controls how symbols
1269 are printed that have a colon as their first or last character. The
1270 default now is to only quote a symbol with #{...}# when the read
1271 option 'keywords' is not '#f'. Thus:
1273 guile> (define foo (string->symbol ":foo"))
1274 guile> (read-set! keywords #f)
1277 guile> (read-set! keywords 'prefix)
1280 guile> (print-set! quote-keywordish-symbols #f)
1284 ** 'while' now provides 'break' and 'continue'
1286 break and continue were previously bound in a while loop, but not
1287 documented, and continue didn't quite work properly. The undocumented
1288 parameter to break which gave a return value for the while has been
1291 ** 'call-with-current-continuation' is now also available under the name
1294 ** The module system now checks for duplicate bindings.
1296 The module system now can check for name conflicts among imported
1299 The behavior can be controlled by specifying one or more 'duplicates'
1300 handlers. For example, to make Guile return an error for every name
1303 (define-module (foo)
1308 The new default behavior of the module system when a name collision
1309 has been detected is to
1311 1. Give priority to bindings marked as a replacement.
1312 2. Issue a warning (different warning if overriding core binding).
1313 3. Give priority to the last encountered binding (this corresponds to
1316 If you want the old behavior back without replacements or warnings you
1319 (default-duplicate-binding-handler 'last)
1321 to your .guile init file.
1323 ** New define-module option: :replace
1325 :replace works as :export, but, in addition, marks the binding as a
1328 A typical example is `format' in (ice-9 format) which is a replacement
1329 for the core binding `format'.
1331 ** Adding prefixes to imported bindings in the module system
1333 There is now a new :use-module option :prefix. It can be used to add
1334 a prefix to all imported bindings.
1336 (define-module (foo)
1337 :use-module ((bar) :prefix bar:))
1339 will import all bindings exported from bar, but rename them by adding
1342 ** Conflicting generic functions can be automatically merged.
1344 When two imported bindings conflict and they are both generic
1345 functions, the two functions can now be merged automatically. This is
1346 activated with the 'duplicates' handler 'merge-generics'.
1348 ** New function: effective-version
1350 Returns the "effective" version number. This is just the normal full
1351 version string without the final micro-version number. See "Changes
1352 to the distribution" above.
1354 ** New threading functions: parallel, letpar, par-map, and friends
1356 These are convenient ways to run calculations in parallel in new
1357 threads. See "Parallel forms" in the manual for details.
1359 ** New function 'try-mutex'.
1361 This function will attempt to lock a mutex but will return immediately
1362 instead of blocking and indicate failure.
1364 ** Waiting on a condition variable can have a timeout.
1366 The function 'wait-condition-variable' now takes a third, optional
1367 argument that specifies the point in time where the waiting should be
1370 ** New function 'broadcast-condition-variable'.
1372 ** New functions 'all-threads' and 'current-thread'.
1374 ** Signals and system asyncs work better with threads.
1376 The function 'sigaction' now takes a fourth, optional, argument that
1377 specifies the thread that the handler should run in. When the
1378 argument is omitted, the handler will run in the thread that called
1381 Likewise, 'system-async-mark' takes a second, optional, argument that
1382 specifies the thread that the async should run in. When it is
1383 omitted, the async will run in the thread that called
1384 'system-async-mark'.
1386 C code can use the new functions scm_sigaction_for_thread and
1387 scm_system_async_mark_for_thread to pass the new thread argument.
1389 When a thread blocks on a mutex, a condition variable or is waiting
1390 for IO to be possible, it will still execute system asyncs. This can
1391 be used to interrupt such a thread by making it execute a 'throw', for
1394 ** The function 'system-async' is deprecated.
1396 You can now pass any zero-argument procedure to 'system-async-mark'.
1397 The function 'system-async' will just return its argument unchanged
1400 ** New functions 'call-with-blocked-asyncs' and
1401 'call-with-unblocked-asyncs'
1403 The expression (call-with-blocked-asyncs PROC) will call PROC and will
1404 block execution of system asyncs for the current thread by one level
1405 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
1406 procedure and will unblock the execution of system asyncs by one
1407 level for the current thread.
1409 Only system asyncs are affected by these functions.
1411 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
1413 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
1414 instead. Those functions are easier to use correctly and can be
1417 ** New function 'unsetenv'.
1419 ** New macro 'define-syntax-public'.
1421 It works like 'define-syntax' and also exports the defined macro (but
1424 ** There is support for Infinity and NaNs.
1426 Following PLT Scheme, Guile can now work with infinite numbers, and
1429 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
1430 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
1431 "+nan.0"). These numbers are inexact and have no exact counterpart.
1433 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
1434 sign of the dividend. The infinities are integers, and they answer #t
1435 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
1436 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
1447 ERROR: Numerical overflow
1449 Two new predicates 'inf?' and 'nan?' can be used to test for the
1452 ** Inexact zero can have a sign.
1454 Guile can now distinguish between plus and minus inexact zero, if your
1455 platform supports this, too. The two zeros are equal according to
1456 '=', but not according to 'eqv?'. For example
1467 ** Guile now has exact rationals.
1469 Guile can now represent fractions such as 1/3 exactly. Computing with
1470 them is also done exactly, of course:
1475 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
1476 for exact arguments.
1478 For example: (floor 2) now returns an exact 2 where in the past it
1479 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
1481 ** inexact->exact no longer returns only integers.
1483 Without exact rationals, the closest exact number was always an
1484 integer, but now inexact->exact returns the fraction that is exactly
1485 equal to a floating point number. For example:
1487 (inexact->exact 1.234)
1488 => 694680242521899/562949953421312
1490 When you want the old behavior, use 'round' explicitly:
1492 (inexact->exact (round 1.234))
1495 ** New function 'rationalize'.
1497 This function finds a simple fraction that is close to a given real
1498 number. For example (and compare with inexact->exact above):
1500 (rationalize (inexact->exact 1.234) 1/2000)
1503 Note that, as required by R5RS, rationalize returns only then an exact
1504 result when both its arguments are exact.
1506 ** 'odd?' and 'even?' work also for inexact integers.
1508 Previously, (odd? 1.0) would signal an error since only exact integers
1509 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
1510 returns #f and (odd? 1.5) signals an error.
1512 ** Guile now has uninterned symbols.
1514 The new function 'make-symbol' will return an uninterned symbol. This
1515 is a symbol that is unique and is guaranteed to remain unique.
1516 However, uninterned symbols can not yet be read back in.
1518 Use the new function 'symbol-interned?' to check whether a symbol is
1521 ** pretty-print has more options.
1523 The function pretty-print from the (ice-9 pretty-print) module can now
1524 also be invoked with keyword arguments that control things like
1525 maximum output width. See the manual for details.
1527 ** Variables have no longer a special behavior for `equal?'.
1529 Previously, comparing two variables with `equal?' would recursivly
1530 compare their values. This is no longer done. Variables are now only
1531 `equal?' if they are `eq?'.
1533 ** `(begin)' is now valid.
1535 You can now use an empty `begin' form. It will yield #<unspecified>
1536 when evaluated and simply be ignored in a definition context.
1538 ** Deprecated: procedure->macro
1540 Change your code to use 'define-macro' or r5rs macros. Also, be aware
1541 that macro expansion will not be done during evaluation, but prior to
1544 ** Soft ports now allow a `char-ready?' procedure
1546 The vector argument to `make-soft-port' can now have a length of
1547 either 5 or 6. (Previously the length had to be 5.) The optional 6th
1548 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
1549 that returns the number of characters that can be read immediately
1550 without the soft port blocking.
1552 ** Deprecated: undefine
1554 There is no replacement for undefine.
1556 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
1557 have been discouraged.
1559 They are relics from a time where a keyword like #:foo was used
1560 directly as a Tcl option "-foo" and thus keywords were internally
1561 stored as a symbol with a starting dash. We now store a symbol
1564 Use symbol->keyword and keyword->symbol instead.
1566 ** The `cheap' debug option is now obsolete
1568 Evaluator trap calls are now unconditionally "cheap" - in other words,
1569 they pass a debug object to the trap handler rather than a full
1570 continuation. The trap handler code can capture a full continuation
1571 by using `call-with-current-continuation' in the usual way, if it so
1574 The `cheap' option is retained for now so as not to break existing
1575 code which gets or sets it, but setting it now has no effect. It will
1576 be removed in the next major Guile release.
1578 ** Evaluator trap calls now support `tweaking'
1580 `Tweaking' means that the trap handler code can modify the Scheme
1581 expression that is about to be evaluated (in the case of an
1582 enter-frame trap) or the value that is being returned (in the case of
1583 an exit-frame trap). The trap handler code indicates that it wants to
1584 do this by returning a pair whose car is the symbol 'instead and whose
1585 cdr is the modified expression or return value.
1587 * Changes to the C interface
1589 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
1590 take a 'delete' function argument.
1592 This argument makes no sense since the delete function is used to
1593 remove a pair from an alist, and this must not be configurable.
1595 This is an incompatible change.
1597 ** The GH interface is now subject to the deprecation mechanism
1599 The GH interface has been deprecated for quite some time but now it is
1600 actually removed from Guile when it is configured with
1601 --disable-deprecated.
1603 See the manual "Transitioning away from GH" for more information.
1605 ** A new family of functions for converting between C values and
1606 Scheme values has been added.
1608 These functions follow a common naming scheme and are designed to be
1609 easier to use, thread-safe and more future-proof than the older
1612 - int scm_is_* (...)
1614 These are predicates that return a C boolean: 1 or 0. Instead of
1615 SCM_NFALSEP, you can now use scm_is_true, for example.
1617 - <type> scm_to_<type> (SCM val, ...)
1619 These are functions that convert a Scheme value into an appropriate
1620 C value. For example, you can use scm_to_int to safely convert from
1623 - SCM scm_from_<type> (<type> val, ...)
1625 These functions convert from a C type to a SCM value; for example,
1626 scm_from_int for ints.
1628 There is a huge number of these functions, for numbers, strings,
1629 symbols, vectors, etc. They are documented in the reference manual in
1630 the API section together with the types that they apply to.
1632 ** New functions for dealing with complex numbers in C have been added.
1634 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1635 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1636 They work like scm_make_rectangular etc but take or return doubles
1639 ** The function scm_make_complex has been discouraged.
1641 Use scm_c_make_rectangular instead.
1643 ** The INUM macros have been deprecated.
1645 A lot of code uses these macros to do general integer conversions,
1646 although the macros only work correctly with fixnums. Use the
1647 following alternatives.
1649 SCM_INUMP -> scm_is_integer or similar
1650 SCM_NINUMP -> !scm_is_integer or similar
1651 SCM_MAKINUM -> scm_from_int or similar
1652 SCM_INUM -> scm_to_int or similar
1654 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1655 do the validating for you.
1657 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1658 have been discouraged.
1660 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1661 new code. The functions have been discouraged since they don't fit
1664 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1666 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1667 evaluates its argument twice. Use scm_is_true, etc. instead for new
1670 ** The macro SCM_EQ_P has been discouraged.
1672 Use scm_is_eq for new code, which fits better into the naming
1675 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1678 Use the function scm_is_pair or scm_is_null instead.
1680 ** The functions scm_round and scm_truncate have been deprecated and
1681 are now available as scm_c_round and scm_c_truncate, respectively.
1683 These functions occupy the names that scm_round_number and
1684 scm_truncate_number should have.
1686 ** The functions scm_c_string2str, scm_c_substring2str, and
1687 scm_c_symbol2str have been deprecated.
1689 Use scm_to_locale_stringbuf or similar instead, maybe together with
1692 ** New functions scm_c_make_string, scm_c_string_length,
1693 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1694 scm_c_substring_shared, scm_c_substring_copy.
1696 These are like scm_make_string, scm_length, etc. but are slightly
1697 easier to use from C.
1699 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1700 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1702 They export too many assumptions about the implementation of strings
1703 and symbols that are no longer true in the presence of
1704 mutation-sharing substrings and when Guile switches to some form of
1707 When working with strings, it is often best to use the normal string
1708 functions provided by Guile, such as scm_c_string_ref,
1709 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1710 manual since many more such functions are now provided than
1713 When you want to convert a SCM string to a C string, use the
1714 scm_to_locale_string function or similar instead. For symbols, use
1715 scm_symbol_to_string and then work with that string. Because of the
1716 new string representation, scm_symbol_to_string does not need to copy
1717 and is thus quite efficient.
1719 ** Some string, symbol and keyword functions have been discouraged.
1721 They don't fit into the uniform naming scheme and are not explicit
1722 about the character encoding.
1724 Replace according to the following table:
1726 scm_allocate_string -> scm_c_make_string
1727 scm_take_str -> scm_take_locale_stringn
1728 scm_take0str -> scm_take_locale_string
1729 scm_mem2string -> scm_from_locale_stringn
1730 scm_str2string -> scm_from_locale_string
1731 scm_makfrom0str -> scm_from_locale_string
1732 scm_mem2symbol -> scm_from_locale_symboln
1733 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1734 scm_str2symbol -> scm_from_locale_symbol
1736 SCM_SYMBOL_HASH -> scm_hashq
1737 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1739 scm_c_make_keyword -> scm_from_locale_keyword
1741 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1742 now also available to C code.
1744 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1746 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1747 the latter returns the true name of the keyword, not the 'dash name',
1748 as SCM_KEYWORDSYM used to do.
1750 ** A new way to access arrays in a thread-safe and efficient way has
1753 See the manual, node "Accessing Arrays From C".
1755 ** The old uniform vector and bitvector implementations have been
1756 unceremoniously removed.
1758 This implementation exposed the details of the tagging system of
1759 Guile. Use the new C API explained in the manual in node "Uniform
1760 Numeric Vectors" and "Bit Vectors", respectively.
1762 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1763 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1764 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1765 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1766 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1767 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1770 ** The macros dealing with vectors have been deprecated.
1772 Use the new functions scm_is_vector, scm_vector_elements,
1773 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1774 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1775 manual for more details.
1777 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1778 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1780 The following macros have been removed: SCM_VECTOR_BASE,
1781 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1782 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1784 ** Some C functions and macros related to arrays have been deprecated.
1786 Migrate according to the following table:
1788 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1789 scm_make_ra -> scm_make_array
1790 scm_shap2ra -> scm_make_array
1791 scm_cvref -> scm_c_generalized_vector_ref
1792 scm_ra_set_contp -> do not use
1793 scm_aind -> scm_array_handle_pos
1794 scm_raprin1 -> scm_display or scm_write
1796 SCM_ARRAYP -> scm_is_array
1797 SCM_ARRAY_NDIM -> scm_c_array_rank
1798 SCM_ARRAY_DIMS -> scm_array_handle_dims
1799 SCM_ARRAY_CONTP -> do not use
1800 SCM_ARRAY_MEM -> do not use
1801 SCM_ARRAY_V -> scm_array_handle_elements or similar
1802 SCM_ARRAY_BASE -> do not use
1804 ** SCM_CELL_WORD_LOC has been deprecated.
1806 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1807 to a SCM, as opposed to a pointer to a scm_t_bits.
1809 This was done to allow the correct use of pointers into the Scheme
1810 heap. Previously, the heap words were of type scm_t_bits and local
1811 variables and function arguments were of type SCM, making it
1812 non-standards-conformant to have a pointer that can point to both.
1814 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1816 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1817 second and third words of double smobs. Likewise for
1818 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1820 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1821 used to get and set the 16 exra bits in the zeroth word of a smob.
1823 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1824 accesing the first immediate word of a smob as a SCM value, and there
1825 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1826 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1828 ** New way to deal with non-local exits and re-entries.
1830 There is a new set of functions that essentially do what
1831 scm_internal_dynamic_wind does, but in a way that is more convenient
1832 for C code in some situations. Here is a quick example of how to
1833 prevent a potential memory leak:
1840 scm_dynwind_begin (0);
1842 mem = scm_malloc (100);
1843 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1845 /* MEM would leak if BAR throws an error.
1846 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1853 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1854 SCM_DYNWIND_END as well.
1858 For full documentation, see the node "Dynamic Wind" in the manual.
1860 ** New function scm_dynwind_free
1862 This function calls 'free' on a given pointer when a dynwind context
1863 is left. Thus the call to scm_dynwind_unwind_handler above could be
1864 replaced with simply scm_dynwind_free (mem).
1866 ** New functions scm_c_call_with_blocked_asyncs and
1867 scm_c_call_with_unblocked_asyncs
1869 Like scm_call_with_blocked_asyncs etc. but for C functions.
1871 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1873 In addition to scm_c_call_with_blocked_asyncs you can now also use
1874 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1875 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1877 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1878 SCM_REALLOW_INTS have been deprecated.
1880 They do no longer fulfill their original role of blocking signal
1881 delivery. Depending on what you want to achieve, replace a pair of
1882 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1883 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1886 ** The value 'scm_mask_ints' is no longer writable.
1888 Previously, you could set scm_mask_ints directly. This is no longer
1889 possible. Use scm_c_call_with_blocked_asyncs and
1890 scm_c_call_with_unblocked_asyncs instead.
1892 ** New way to temporarily set the current input, output or error ports
1894 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1895 context' (see above). <foo> is one of "input", "output" or "error".
1897 ** New way to temporarily set fluids
1899 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1900 above) to temporarily set the value of a fluid.
1902 ** New types scm_t_intmax and scm_t_uintmax.
1904 On platforms that have them, these types are identical to intmax_t and
1905 uintmax_t, respectively. On other platforms, they are identical to
1906 the largest integer types that Guile knows about.
1908 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1910 You should not have used them.
1912 ** Many public #defines with generic names have been made private.
1914 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1915 private or renamed with a more suitable public name.
1917 ** The macro SCM_TYP16S has been deprecated.
1919 This macro is not intended for public use.
1921 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1923 Use scm_is_true (scm_inexact_p (...)) instead.
1925 ** The macro SCM_SLOPPY_REALP has been deprecated.
1927 Use scm_is_real instead.
1929 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1931 Use scm_is_complex instead.
1933 ** Some preprocessor defines have been deprecated.
1935 These defines indicated whether a certain feature was present in Guile
1936 or not. Going forward, assume that the features are always present.
1938 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1939 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1941 The following macros have been removed completely: MEMOIZE_LOCALS,
1942 SCM_RECKLESS, SCM_CAUTIOUS.
1944 ** The preprocessor define STACK_DIRECTION has been deprecated.
1946 There should be no need to know about the stack direction for ordinary
1949 ** New function: scm_effective_version
1951 Returns the "effective" version number. This is just the normal full
1952 version string without the final micro-version number. See "Changes
1953 to the distribution" above.
1955 ** The function scm_call_with_new_thread has a new prototype.
1957 Instead of taking a list with the thunk and handler, these two
1958 arguments are now passed directly:
1960 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1962 This is an incompatible change.
1964 ** New snarfer macro SCM_DEFINE_PUBLIC.
1966 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1967 function in the init section.
1969 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1971 ** Garbage collector rewrite.
1973 The garbage collector is cleaned up a lot, and now uses lazy
1974 sweeping. This is reflected in the output of (gc-stats); since cells
1975 are being freed when they are allocated, the cells-allocated field
1976 stays roughly constant.
1978 For malloc related triggers, the behavior is changed. It uses the same
1979 heuristic as the cell-triggered collections. It may be tuned with the
1980 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1981 for minimum yield of malloc related triggers. The default is 40.
1982 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1985 Debugging operations for the freelist have been deprecated, along with
1986 the C variables that control garbage collection. The environment
1987 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1988 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1990 For understanding the memory usage of a GUILE program, the routine
1991 gc-live-object-stats returns an alist containing the number of live
1992 objects for every type.
1995 ** The function scm_definedp has been renamed to scm_defined_p
1997 The name scm_definedp is deprecated.
1999 ** The struct scm_cell type has been renamed to scm_t_cell
2001 This is in accordance to Guile's naming scheme for types. Note that
2002 the name scm_cell is now used for a function that allocates and
2003 initializes a new cell (see below).
2005 ** New functions for memory management
2007 A new set of functions for memory management has been added since the
2008 old way (scm_must_malloc, scm_must_free, etc) was error prone and
2009 indeed, Guile itself contained some long standing bugs that could
2010 cause aborts in long running programs.
2012 The new functions are more symmetrical and do not need cooperation
2013 from smob free routines, among other improvements.
2015 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
2016 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
2017 scm_gc_free, scm_gc_register_collectable_memory, and
2018 scm_gc_unregister_collectable_memory. Refer to the manual for more
2019 details and for upgrading instructions.
2021 The old functions for memory management have been deprecated. They
2022 are: scm_must_malloc, scm_must_realloc, scm_must_free,
2023 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
2025 ** Declarations of exported features are marked with SCM_API.
2027 Every declaration of a feature that belongs to the exported Guile API
2028 has been marked by adding the macro "SCM_API" to the start of the
2029 declaration. This macro can expand into different things, the most
2030 common of which is just "extern" for Unix platforms. On Win32, it can
2031 be used to control which symbols are exported from a DLL.
2033 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
2034 will expand into "__declspec (dllimport) extern", which is needed for
2035 linking to the Guile DLL in Windows.
2037 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
2038 SCM_SRFI4_IMPORT, for the corresponding libraries.
2040 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
2042 Use the new functions scm_cell and scm_double_cell instead. The old
2043 macros had problems because with them allocation and initialization
2044 was separated and the GC could sometimes observe half initialized
2045 cells. Only careful coding by the user of SCM_NEWCELL and
2046 SCM_NEWCELL2 could make this safe and efficient.
2048 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
2050 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
2053 ** SRCBRKP has been deprecated.
2055 Use scm_c_source_property_breakpoint_p instead.
2057 ** Deprecated: scm_makmacro
2059 Change your code to use either scm_makmmacro or to define macros in
2060 Scheme, using 'define-macro'.
2062 ** New function scm_c_port_for_each.
2064 This function is like scm_port_for_each but takes a pointer to a C
2065 function as the callback instead of a SCM value.
2067 ** The names scm_internal_select, scm_thread_sleep, and
2068 scm_thread_usleep have been discouraged.
2070 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
2072 ** The GC can no longer be blocked.
2074 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
2075 The GC can now run (partially) concurrently with other code and thus
2076 blocking it is not well defined.
2078 ** Many definitions have been removed that were previously deprecated.
2080 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
2081 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
2082 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
2083 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
2084 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
2085 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
2086 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
2087 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
2088 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
2089 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
2090 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
2091 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
2092 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
2093 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
2094 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
2095 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
2096 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
2097 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
2098 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
2099 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
2100 scm_vector_set_length_x, scm_contregs, scm_debug_info,
2101 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
2102 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
2103 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
2104 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
2105 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
2106 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
2107 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
2108 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
2109 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
2110 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
2112 * Changes to bundled modules
2116 Using the (ice-9 debug) module no longer automatically switches Guile
2117 to use the debugging evaluator. If you want to switch to the
2118 debugging evaluator (which is needed for backtrace information if you
2119 hit an error), please add an explicit "(debug-enable 'debug)" to your
2120 code just after the code to use (ice-9 debug).
2123 Changes since Guile 1.4:
2125 * Changes to the distribution
2127 ** A top-level TODO file is included.
2129 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
2131 Guile now always uses three numbers to represent the version,
2132 i.e. "1.6.5". The first number, 1, is the major version number, the
2133 second number, 6, is the minor version number, and the third number,
2134 5, is the micro version number. Changes in major version number
2135 indicate major changes in Guile.
2137 Minor version numbers that are even denote stable releases, and odd
2138 minor version numbers denote development versions (which may be
2139 unstable). The micro version number indicates a minor sub-revision of
2140 a given MAJOR.MINOR release.
2142 In keeping with the new scheme, (minor-version) and scm_minor_version
2143 no longer return everything but the major version number. They now
2144 just return the minor version number. Two new functions
2145 (micro-version) and scm_micro_version have been added to report the
2146 micro version number.
2148 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
2150 ** New preprocessor definitions are available for checking versions.
2152 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
2153 SCM_MICRO_VERSION to the appropriate integer values.
2155 ** Guile now actively warns about deprecated features.
2157 The new configure option `--enable-deprecated=LEVEL' and the
2158 environment variable GUILE_WARN_DEPRECATED control this mechanism.
2159 See INSTALL and README for more information.
2161 ** Guile is much more likely to work on 64-bit architectures.
2163 Guile now compiles and passes "make check" with only two UNRESOLVED GC
2164 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
2165 for the use of a test machine, and thanks to Stefan Jahn for ia64
2168 ** New functions: setitimer and getitimer.
2170 These implement a fairly direct interface to the libc functions of the
2173 ** The #. reader extension is now disabled by default.
2175 For safety reasons, #. evaluation is disabled by default. To
2176 re-enable it, set the fluid read-eval? to #t. For example:
2178 (fluid-set! read-eval? #t)
2180 but make sure you realize the potential security risks involved. With
2181 read-eval? enabled, reading a data file from an untrusted source can
2184 ** New SRFI modules have been added:
2186 SRFI-0 `cond-expand' is now supported in Guile, without requiring
2189 (srfi srfi-1) is a library containing many useful pair- and list-processing
2192 (srfi srfi-2) exports and-let*.
2194 (srfi srfi-4) implements homogeneous numeric vector datatypes.
2196 (srfi srfi-6) is a dummy module for now, since guile already provides
2197 all of the srfi-6 procedures by default: open-input-string,
2198 open-output-string, get-output-string.
2200 (srfi srfi-8) exports receive.
2202 (srfi srfi-9) exports define-record-type.
2204 (srfi srfi-10) exports define-reader-ctor and implements the reader
2207 (srfi srfi-11) exports let-values and let*-values.
2209 (srfi srfi-13) implements the SRFI String Library.
2211 (srfi srfi-14) implements the SRFI Character-Set Library.
2213 (srfi srfi-17) implements setter and getter-with-setter and redefines
2214 some accessor procedures as procedures with getters. (such as car,
2215 cdr, vector-ref etc.)
2217 (srfi srfi-19) implements the SRFI Time/Date Library.
2219 ** New scripts / "executable modules"
2221 Subdirectory "scripts" contains Scheme modules that are packaged to
2222 also be executable as scripts. At this time, these scripts are available:
2231 See README there for more info.
2233 These scripts can be invoked from the shell with the new program
2234 "guile-tools", which keeps track of installation directory for you.
2237 $ guile-tools display-commentary srfi/*.scm
2239 guile-tools is copied to the standard $bindir on "make install".
2241 ** New module (ice-9 stack-catch):
2243 stack-catch is like catch, but saves the current state of the stack in
2244 the fluid the-last-stack. This fluid can be useful when using the
2245 debugger and when re-throwing an error.
2247 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
2249 This has been done to prevent problems on lesser operating systems
2250 that can't tolerate `*'s in file names. The exported macro continues
2251 to be named `and-let*', of course.
2253 On systems that support it, there is also a compatibility module named
2254 (ice-9 and-let*). It will go away in the next release.
2256 ** New modules (oop goops) etc.:
2259 (oop goops describe)
2261 (oop goops active-slot)
2262 (oop goops composite-slot)
2264 The Guile Object Oriented Programming System (GOOPS) has been
2265 integrated into Guile. For further information, consult the GOOPS
2266 manual and tutorial in the `doc' directory.
2268 ** New module (ice-9 rdelim).
2270 This exports the following procedures which were previously defined
2271 in the default environment:
2273 read-line read-line! read-delimited read-delimited! %read-delimited!
2274 %read-line write-line
2276 For backwards compatibility the definitions are still imported into the
2277 default environment in this version of Guile. However you should add:
2279 (use-modules (ice-9 rdelim))
2281 to any program which uses the definitions, since this may change in
2284 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
2285 can be used for similar functionality.
2287 ** New module (ice-9 rw)
2289 This is a subset of the (scsh rw) module from guile-scsh. Currently
2290 it defines two procedures:
2292 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
2294 Read characters from a port or file descriptor into a string STR.
2295 A port must have an underlying file descriptor -- a so-called
2296 fport. This procedure is scsh-compatible and can efficiently read
2299 *** New function: write-string/partial str [port_or_fdes [start [end]]]
2301 Write characters from a string STR to a port or file descriptor.
2302 A port must have an underlying file descriptor -- a so-called
2303 fport. This procedure is mostly compatible and can efficiently
2304 write large strings.
2306 ** New module (ice-9 match)
2308 This module includes Andrew K. Wright's pattern matcher. See
2309 ice-9/match.scm for brief description or
2311 http://www.star-lab.com/wright/code.html
2313 for complete documentation.
2315 ** New module (ice-9 buffered-input)
2317 This module provides procedures to construct an input port from an
2318 underlying source of input that reads and returns its input in chunks.
2319 The underlying input source is a Scheme procedure, specified by the
2320 caller, which the port invokes whenever it needs more input.
2322 This is useful when building an input port whose back end is Readline
2323 or a UI element such as the GtkEntry widget.
2327 The reference and tutorial documentation that was previously
2328 distributed separately, as `guile-doc', is now included in the core
2329 Guile distribution. The documentation consists of the following
2332 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
2335 - The Guile Reference Manual (guile.texi) contains (or is intended to
2336 contain) reference documentation on all aspects of Guile.
2338 - The GOOPS Manual (goops.texi) contains both tutorial-style and
2339 reference documentation for using GOOPS, Guile's Object Oriented
2342 - The Revised^5 Report on the Algorithmic Language Scheme
2345 See the README file in the `doc' directory for more details.
2347 ** There are a couple of examples in the examples/ directory now.
2349 * Changes to the stand-alone interpreter
2351 ** New command line option `--use-srfi'
2353 Using this option, SRFI modules can be loaded on startup and be
2354 available right from the beginning. This makes programming portable
2355 Scheme programs easier.
2357 The option `--use-srfi' expects a comma-separated list of numbers,
2358 each representing a SRFI number to be loaded into the interpreter
2359 before starting evaluating a script file or the REPL. Additionally,
2360 the feature identifier for the loaded SRFIs is recognized by
2361 `cond-expand' when using this option.
2364 $ guile --use-srfi=8,13
2365 guile> (receive (x z) (values 1 2) (+ 1 2))
2367 guile> (string-pad "bla" 20)
2370 ** Guile now always starts up in the `(guile-user)' module.
2372 Previously, scripts executed via the `-s' option would run in the
2373 `(guile)' module and the repl would run in the `(guile-user)' module.
2374 Now every user action takes place in the `(guile-user)' module by
2377 * Changes to Scheme functions and syntax
2379 ** Character classifiers work for non-ASCII characters.
2381 The predicates `char-alphabetic?', `char-numeric?',
2382 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
2383 no longer check whether their arguments are ASCII characters.
2384 Previously, a character would only be considered alphabetic when it
2385 was also ASCII, for example.
2387 ** Previously deprecated Scheme functions have been removed:
2389 tag - no replacement.
2390 fseek - replaced by seek.
2391 list* - replaced by cons*.
2393 ** It's now possible to create modules with controlled environments
2397 (use-modules (ice-9 safe))
2398 (define m (make-safe-module))
2399 ;;; m will now be a module containing only a safe subset of R5RS
2400 (eval '(+ 1 2) m) --> 3
2401 (eval 'load m) --> ERROR: Unbound variable: load
2403 ** Evaluation of "()", the empty list, is now an error.
2405 Previously, the expression "()" evaluated to the empty list. This has
2406 been changed to signal a "missing expression" error. The correct way
2407 to write the empty list as a literal constant is to use quote: "'()".
2409 ** New concept of `Guile Extensions'.
2411 A Guile Extension is just a ordinary shared library that can be linked
2412 at run-time. We found it advantageous to give this simple concept a
2413 dedicated name to distinguish the issues related to shared libraries
2414 from the issues related to the module system.
2416 *** New function: load-extension
2418 Executing (load-extension lib init) is mostly equivalent to
2420 (dynamic-call init (dynamic-link lib))
2422 except when scm_register_extension has been called previously.
2423 Whenever appropriate, you should use `load-extension' instead of
2424 dynamic-link and dynamic-call.
2426 *** New C function: scm_c_register_extension
2428 This function registers a initialization function for use by
2429 `load-extension'. Use it when you don't want specific extensions to
2430 be loaded as shared libraries (for example on platforms that don't
2431 support dynamic linking).
2433 ** Auto-loading of compiled-code modules is deprecated.
2435 Guile used to be able to automatically find and link a shared
2436 library to satisfy requests for a module. For example, the module
2437 `(foo bar)' could be implemented by placing a shared library named
2438 "foo/libbar.so" (or with a different extension) in a directory on the
2441 This has been found to be too tricky, and is no longer supported. The
2442 shared libraries are now called "extensions". You should now write a
2443 small Scheme file that calls `load-extension' to load the shared
2444 library and initialize it explicitly.
2446 The shared libraries themselves should be installed in the usual
2447 places for shared libraries, with names like "libguile-foo-bar".
2449 For example, place this into a file "foo/bar.scm"
2451 (define-module (foo bar))
2453 (load-extension "libguile-foo-bar" "foobar_init")
2455 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
2457 `eval' is now R5RS, that is it takes two arguments.
2458 The second argument is an environment specifier, i.e. either
2460 (scheme-report-environment 5)
2461 (null-environment 5)
2462 (interaction-environment)
2468 ** The module system has been made more disciplined.
2470 The function `eval' will save and restore the current module around
2471 the evaluation of the specified expression. While this expression is
2472 evaluated, `(current-module)' will now return the right module, which
2473 is the module specified as the second argument to `eval'.
2475 A consequence of this change is that `eval' is not particularly
2476 useful when you want allow the evaluated code to change what module is
2477 designated as the current module and have this change persist from one
2478 call to `eval' to the next. The read-eval-print-loop is an example
2479 where `eval' is now inadequate. To compensate, there is a new
2480 function `primitive-eval' that does not take a module specifier and
2481 that does not save/restore the current module. You should use this
2482 function together with `set-current-module', `current-module', etc
2483 when you want to have more control over the state that is carried from
2484 one eval to the next.
2486 Additionally, it has been made sure that forms that are evaluated at
2487 the top level are always evaluated with respect to the current module.
2488 Previously, subforms of top-level forms such as `begin', `case',
2489 etc. did not respect changes to the current module although these
2490 subforms are at the top-level as well.
2492 To prevent strange behavior, the forms `define-module',
2493 `use-modules', `use-syntax', and `export' have been restricted to only
2494 work on the top level. The forms `define-public' and
2495 `defmacro-public' only export the new binding on the top level. They
2496 behave just like `define' and `defmacro', respectively, when they are
2497 used in a lexical environment.
2499 Also, `export' will no longer silently re-export bindings imported
2500 from a used module. It will emit a `deprecation' warning and will
2501 cease to perform any re-export in the next version. If you actually
2502 want to re-export bindings, use the new `re-export' in place of
2503 `export'. The new `re-export' will not make copies of variables when
2504 rexporting them, as `export' did wrongly.
2506 ** Module system now allows selection and renaming of imported bindings
2508 Previously, when using `use-modules' or the `#:use-module' clause in
2509 the `define-module' form, all the bindings (association of symbols to
2510 values) for imported modules were added to the "current module" on an
2511 as-is basis. This has been changed to allow finer control through two
2512 new facilities: selection and renaming.
2514 You can now select which of the imported module's bindings are to be
2515 visible in the current module by using the `:select' clause. This
2516 clause also can be used to rename individual bindings. For example:
2518 ;; import all bindings no questions asked
2519 (use-modules (ice-9 common-list))
2521 ;; import four bindings, renaming two of them;
2522 ;; the current module sees: every some zonk-y zonk-n
2523 (use-modules ((ice-9 common-list)
2525 (remove-if . zonk-y)
2526 (remove-if-not . zonk-n))))
2528 You can also programmatically rename all selected bindings using the
2529 `:renamer' clause, which specifies a proc that takes a symbol and
2530 returns another symbol. Because it is common practice to use a prefix,
2531 we now provide the convenience procedure `symbol-prefix-proc'. For
2534 ;; import four bindings, renaming two of them specifically,
2535 ;; and all four w/ prefix "CL:";
2536 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
2537 (use-modules ((ice-9 common-list)
2539 (remove-if . zonk-y)
2540 (remove-if-not . zonk-n))
2541 :renamer (symbol-prefix-proc 'CL:)))
2543 ;; import four bindings, renaming two of them specifically,
2544 ;; and all four by upcasing.
2545 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
2546 (define (upcase-symbol sym)
2547 (string->symbol (string-upcase (symbol->string sym))))
2549 (use-modules ((ice-9 common-list)
2551 (remove-if . zonk-y)
2552 (remove-if-not . zonk-n))
2553 :renamer upcase-symbol))
2555 Note that programmatic renaming is done *after* individual renaming.
2556 Also, the above examples show `use-modules', but the same facilities are
2557 available for the `#:use-module' clause of `define-module'.
2559 See manual for more info.
2561 ** The semantics of guardians have changed.
2563 The changes are for the most part compatible. An important criterion
2564 was to keep the typical usage of guardians as simple as before, but to
2565 make the semantics safer and (as a result) more useful.
2567 *** All objects returned from guardians are now properly alive.
2569 It is now guaranteed that any object referenced by an object returned
2570 from a guardian is alive. It's now impossible for a guardian to
2571 return a "contained" object before its "containing" object.
2573 One incompatible (but probably not very important) change resulting
2574 from this is that it is no longer possible to guard objects that
2575 indirectly reference themselves (i.e. are parts of cycles). If you do
2576 so accidentally, you'll get a warning.
2578 *** There are now two types of guardians: greedy and sharing.
2580 If you call (make-guardian #t) or just (make-guardian), you'll get a
2581 greedy guardian, and for (make-guardian #f) a sharing guardian.
2583 Greedy guardians are the default because they are more "defensive".
2584 You can only greedily guard an object once. If you guard an object
2585 more than once, once in a greedy guardian and the rest of times in
2586 sharing guardians, then it is guaranteed that the object won't be
2587 returned from sharing guardians as long as it is greedily guarded
2590 Guardians returned by calls to `make-guardian' can now take one more
2591 optional parameter, which says whether to throw an error in case an
2592 attempt is made to greedily guard an object that is already greedily
2593 guarded. The default is true, i.e. throw an error. If the parameter
2594 is false, the guardian invocation returns #t if guarding was
2595 successful and #f if it wasn't.
2597 Also, since greedy guarding is, in effect, a side-effecting operation
2598 on objects, a new function is introduced: `destroy-guardian!'.
2599 Invoking this function on a guardian renders it unoperative and, if
2600 the guardian is greedy, clears the "greedily guarded" property of the
2601 objects that were guarded by it, thus undoing the side effect.
2603 Note that all this hair is hardly very important, since guardian
2604 objects are usually permanent.
2606 ** Continuations created by call-with-current-continuation now accept
2607 any number of arguments, as required by R5RS.
2609 ** New function `issue-deprecation-warning'
2611 This function is used to display the deprecation messages that are
2612 controlled by GUILE_WARN_DEPRECATION as explained in the README.
2615 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
2619 ;; `id' is deprecated. Use `identity' instead.
2624 ** New syntax `begin-deprecated'
2626 When deprecated features are included (as determined by the configure
2627 option --enable-deprecated), `begin-deprecated' is identical to
2628 `begin'. When deprecated features are excluded, it always evaluates
2629 to `#f', ignoring the body forms.
2631 ** New function `make-object-property'
2633 This function returns a new `procedure with setter' P that can be used
2634 to attach a property to objects. When calling P as
2638 where `obj' is any kind of object, it attaches `val' to `obj' in such
2639 a way that it can be retrieved by calling P as
2643 This function will replace procedure properties, symbol properties and
2644 source properties eventually.
2646 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2648 Instead of #&optional, #&key, etc you should now use #:optional,
2649 #:key, etc. Since #:optional is a keyword, you can write it as just
2650 :optional when (read-set! keywords 'prefix) is active.
2652 The old reader syntax `#&' is still supported, but deprecated. It
2653 will be removed in the next release.
2655 ** New define-module option: pure
2657 Tells the module system not to include any bindings from the root
2662 (define-module (totally-empty-module)
2665 ** New define-module option: export NAME1 ...
2667 Export names NAME1 ...
2669 This option is required if you want to be able to export bindings from
2670 a module which doesn't import one of `define-public' or `export'.
2674 (define-module (foo)
2676 :use-module (ice-9 r5rs)
2679 ;;; Note that we're pure R5RS below this point!
2684 ** New function: object->string OBJ
2686 Return a Scheme string obtained by printing a given object.
2688 ** New function: port? X
2690 Returns a boolean indicating whether X is a port. Equivalent to
2691 `(or (input-port? X) (output-port? X))'.
2693 ** New function: file-port?
2695 Determines whether a given object is a port that is related to a file.
2697 ** New function: port-for-each proc
2699 Apply PROC to each port in the Guile port table in turn. The return
2700 value is unspecified. More specifically, PROC is applied exactly once
2701 to every port that exists in the system at the time PORT-FOR-EACH is
2702 invoked. Changes to the port table while PORT-FOR-EACH is running
2703 have no effect as far as PORT-FOR-EACH is concerned.
2705 ** New function: dup2 oldfd newfd
2707 A simple wrapper for the `dup2' system call. Copies the file
2708 descriptor OLDFD to descriptor number NEWFD, replacing the
2709 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2710 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2711 to move away ports which are using NEWFD. The return value is
2714 ** New function: close-fdes fd
2716 A simple wrapper for the `close' system call. Close file
2717 descriptor FD, which must be an integer. Unlike close (*note
2718 close: Ports and File Descriptors.), the file descriptor will be
2719 closed even if a port is using it. The return value is
2722 ** New function: crypt password salt
2724 Encrypts `password' using the standard unix password encryption
2727 ** New function: chroot path
2729 Change the root directory of the running process to `path'.
2731 ** New functions: getlogin, cuserid
2733 Return the login name or the user name of the current effective user
2736 ** New functions: getpriority which who, setpriority which who prio
2738 Get or set the priority of the running process.
2740 ** New function: getpass prompt
2742 Read a password from the terminal, first displaying `prompt' and
2745 ** New function: flock file operation
2747 Set/remove an advisory shared or exclusive lock on `file'.
2749 ** New functions: sethostname name, gethostname
2751 Set or get the hostname of the machine the current process is running
2754 ** New function: mkstemp! tmpl
2756 mkstemp creates a new unique file in the file system and returns a
2757 new buffered port open for reading and writing to the file. TMPL
2758 is a string specifying where the file should be created: it must
2759 end with `XXXXXX' and will be changed in place to return the name
2760 of the temporary file.
2762 ** New function: open-input-string string
2764 Return an input string port which delivers the characters from
2765 `string'. This procedure, together with `open-output-string' and
2766 `get-output-string' implements SRFI-6.
2768 ** New function: open-output-string
2770 Return an output string port which collects all data written to it.
2771 The data can then be retrieved by `get-output-string'.
2773 ** New function: get-output-string
2775 Return the contents of an output string port.
2777 ** New function: identity
2779 Return the argument.
2781 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2782 are represented in Scheme as integers with normal host byte ordering.
2784 ** New function: inet-pton family address
2786 Convert a printable string network address into an integer. Note that
2787 unlike the C version of this function, the result is an integer with
2788 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2791 (inet-pton AF_INET "127.0.0.1") => 2130706433
2792 (inet-pton AF_INET6 "::1") => 1
2794 ** New function: inet-ntop family address
2796 Convert an integer network address into a printable string. Note that
2797 unlike the C version of this function, the input is an integer with
2798 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2801 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2802 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2803 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2807 Use `identity' instead.
2813 ** Deprecated: return-it
2817 ** Deprecated: string-character-length
2819 Use `string-length' instead.
2821 ** Deprecated: flags
2823 Use `logior' instead.
2825 ** Deprecated: close-all-ports-except.
2827 This was intended for closing ports in a child process after a fork,
2828 but it has the undesirable side effect of flushing buffers.
2829 port-for-each is more flexible.
2831 ** The (ice-9 popen) module now attempts to set up file descriptors in
2832 the child process from the current Scheme ports, instead of using the
2833 current values of file descriptors 0, 1, and 2 in the parent process.
2835 ** Removed function: builtin-weak-bindings
2837 There is no such concept as a weak binding any more.
2839 ** Removed constants: bignum-radix, scm-line-incrementors
2841 ** define-method: New syntax mandatory.
2843 The new method syntax is now mandatory:
2845 (define-method (NAME ARG-SPEC ...) BODY ...)
2846 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2848 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2849 REST-ARG ::= ARG-NAME
2851 If you have old code using the old syntax, import
2852 (oop goops old-define-method) before (oop goops) as in:
2854 (use-modules (oop goops old-define-method) (oop goops))
2856 ** Deprecated function: builtin-variable
2857 Removed function: builtin-bindings
2859 There is no longer a distinction between builtin or other variables.
2860 Use module system operations for all variables.
2862 ** Lazy-catch handlers are no longer allowed to return.
2864 That is, a call to `throw', `error', etc is now guaranteed to not
2867 ** Bugfixes for (ice-9 getopt-long)
2869 This module is now tested using test-suite/tests/getopt-long.test.
2870 The following bugs have been fixed:
2872 *** Parsing for options that are specified to have `optional' args now checks
2873 if the next element is an option instead of unconditionally taking it as the
2876 *** An error is now thrown for `--opt=val' when the option description
2877 does not specify `(value #t)' or `(value optional)'. This condition used to
2878 be accepted w/o error, contrary to the documentation.
2880 *** The error message for unrecognized options is now more informative.
2881 It used to be "not a record", an artifact of the implementation.
2883 *** The error message for `--opt' terminating the arg list (no value), when
2884 `(value #t)' is specified, is now more informative. It used to be "not enough
2887 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2888 The expansion used to be like so:
2890 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2892 Note that the "5d" is dropped. Now it is like so:
2894 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2896 This enables single-char options to have adjoining arguments as long as their
2897 constituent characters are not potential single-char options.
2899 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2901 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2902 property `arglist', which can be retrieved by `arity'. The result is that
2903 `arity' can give more detailed information than before:
2907 guile> (use-modules (ice-9 optargs))
2908 guile> (define* (foo #:optional a b c) a)
2910 0 or more arguments in `lambda*:G0'.
2915 3 optional arguments: `a', `b' and `c'.
2916 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2918 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2919 and `d', other keywords allowed.
2920 guile> (define* (baz a b #:optional c #:rest r) a)
2922 2 required arguments: `a' and `b', 1 optional argument: `c',
2925 * Changes to the C interface
2927 ** Types have been renamed from scm_*_t to scm_t_*.
2929 This has been done for POSIX sake. It reserves identifiers ending
2930 with "_t". What a concept.
2932 The old names are still available with status `deprecated'.
2934 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2936 ** Deprecated features have been removed.
2940 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2941 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2943 *** C Functions removed
2945 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2946 scm_fseek - replaced by scm_seek.
2947 gc-thunk - replaced by after-gc-hook.
2948 gh_int2scmb - replaced by gh_bool2scm.
2949 scm_tc_dblr - replaced by scm_tc16_real.
2950 scm_tc_dblc - replaced by scm_tc16_complex.
2951 scm_list_star - replaced by scm_cons_star.
2953 ** Deprecated: scm_makfromstr
2955 Use scm_mem2string instead.
2957 ** Deprecated: scm_make_shared_substring
2959 Explicit shared substrings will disappear from Guile.
2961 Instead, "normal" strings will be implemented using sharing
2962 internally, combined with a copy-on-write strategy.
2964 ** Deprecated: scm_read_only_string_p
2966 The concept of read-only strings will disappear in next release of
2969 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2971 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2973 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2975 Call a procedure with the indicated number of arguments. See "Fly
2976 Evaluation" in the manual.
2978 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2980 Call a procedure with the indicated number of arguments and a list of
2981 further arguments. See "Fly Evaluation" in the manual.
2983 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2985 Create a list of the given number of elements. See "List
2986 Constructors" in the manual.
2988 ** Renamed function: scm_listify has been replaced by scm_list_n.
2990 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2991 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2993 Use functions scm_list_N instead.
2995 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2997 Used by an application to read arbitrary number of bytes from a port.
2998 Same semantics as libc read, except that scm_c_read only returns less
2999 than SIZE bytes if at end-of-file.
3001 Warning: Doesn't update port line and column counts!
3003 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
3005 Used by an application to write arbitrary number of bytes to an SCM
3006 port. Similar semantics as libc write. However, unlike libc
3007 write, scm_c_write writes the requested number of bytes and has no
3010 Warning: Doesn't update port line and column counts!
3012 ** New function: scm_init_guile ()
3014 In contrast to scm_boot_guile, scm_init_guile will return normally
3015 after initializing Guile. It is not available on all systems, tho.
3017 ** New functions: scm_str2symbol, scm_mem2symbol
3019 The function scm_str2symbol takes a const char* pointing to a zero-terminated
3020 field of characters and creates a scheme symbol object from that C string.
3021 The function scm_mem2symbol takes a const char* and a number of characters and
3022 creates a symbol from the characters in that memory area.
3024 ** New functions: scm_primitive_make_property
3025 scm_primitive_property_ref
3026 scm_primitive_property_set_x
3027 scm_primitive_property_del_x
3029 These functions implement a new way to deal with object properties.
3030 See libguile/properties.c for their documentation.
3032 ** New function: scm_done_free (long size)
3034 This function is the inverse of scm_done_malloc. Use it to report the
3035 amount of smob memory you free. The previous method, which involved
3036 calling scm_done_malloc with negative argument, was somewhat
3037 unintuitive (and is still available, of course).
3039 ** New function: scm_c_memq (SCM obj, SCM list)
3041 This function provides a fast C level alternative for scm_memq for the case
3042 that the list parameter is known to be a proper list. The function is a
3043 replacement for scm_sloppy_memq, but is stricter in its requirements on its
3044 list input parameter, since for anything else but a proper list the function's
3045 behaviour is undefined - it may even crash or loop endlessly. Further, for
3046 the case that the object is not found in the list, scm_c_memq returns #f which
3047 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
3049 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
3050 scm_remember_upto_here
3052 These functions replace the function scm_remember.
3054 ** Deprecated function: scm_remember
3056 Use one of the new functions scm_remember_upto_here_1,
3057 scm_remember_upto_here_2 or scm_remember_upto_here instead.
3059 ** New function: scm_allocate_string
3061 This function replaces the function scm_makstr.
3063 ** Deprecated function: scm_makstr
3065 Use the new function scm_allocate_string instead.
3067 ** New global variable scm_gc_running_p introduced.
3069 Use this variable to find out if garbage collection is being executed. Up to
3070 now applications have used scm_gc_heap_lock to test if garbage collection was
3071 running, which also works because of the fact that up to know only the garbage
3072 collector has set this variable. But, this is an implementation detail that
3073 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
3074 of this variable is (and has been) not fully safe anyway.
3076 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
3078 Use these instead of SCM_LENGTH_MAX.
3080 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
3081 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
3082 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
3084 Use these instead of SCM_LENGTH.
3086 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
3087 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
3088 SCM_SET_BITVECTOR_LENGTH
3090 Use these instead of SCM_SETLENGTH
3092 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
3093 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
3096 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
3099 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
3100 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
3103 Use these instead of SCM_SETCHARS.
3105 ** New macro: SCM_BITVECTOR_P
3107 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
3109 Use instead of SCM_COERCE_SUBSTR.
3111 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
3113 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
3115 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
3116 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
3117 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
3118 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
3119 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
3120 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
3121 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
3122 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
3123 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
3124 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
3125 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
3126 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
3127 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
3128 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
3129 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
3131 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
3132 Use scm_memory_error instead of SCM_NALLOC.
3133 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
3134 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
3135 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
3136 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
3137 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
3138 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
3139 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
3140 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
3141 Use SCM_STRINGP instead of SCM_RWSTRINGP.
3142 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
3143 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
3144 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
3145 Use a type specific setter macro instead of SCM_SETLENGTH.
3146 Use a type specific setter macro instead of SCM_SETCHARS.
3147 Use a type specific length macro instead of SCM_LENGTH_MAX.
3148 Use SCM_GCMARKP instead of SCM_GC8MARKP.
3149 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
3150 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
3151 Use SCM_TYP16 instead of SCM_GCTYP16.
3152 Use SCM_CDR instead of SCM_GCCDR.
3153 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
3154 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
3155 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
3156 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
3157 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
3158 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
3159 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
3161 ** Removed function: scm_struct_init
3163 ** Removed variable: scm_symhash_dim
3165 ** Renamed function: scm_make_cont has been replaced by
3166 scm_make_continuation, which has a different interface.
3168 ** Deprecated function: scm_call_catching_errors
3170 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
3172 ** Deprecated function: scm_strhash
3174 Use scm_string_hash instead.
3176 ** Deprecated function: scm_vector_set_length_x
3178 Instead, create a fresh vector of the desired size and copy the contents.
3180 ** scm_gensym has changed prototype
3182 scm_gensym now only takes one argument.
3184 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
3187 There is now only a single symbol type scm_tc7_symbol.
3188 The tag scm_tc7_lvector was not used anyway.
3190 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
3192 Use scm_make_smob_type and scm_set_smob_XXX instead.
3194 ** New function scm_set_smob_apply.
3196 This can be used to set an apply function to a smob type.
3198 ** Deprecated function: scm_strprint_obj
3200 Use scm_object_to_string instead.
3202 ** Deprecated function: scm_wta
3204 Use scm_wrong_type_arg, or another appropriate error signalling function
3207 ** Explicit support for obarrays has been deprecated.
3209 Use `scm_str2symbol' and the generic hashtable functions instead.
3211 ** The concept of `vcells' has been deprecated.
3213 The data type `variable' is now used exclusively. `Vcells' have been
3214 a low-level concept so you are likely not affected by this change.
3216 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
3217 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
3219 Use scm_c_define or scm_c_lookup instead, as appropriate.
3221 *** New functions: scm_c_module_lookup, scm_c_lookup,
3222 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
3223 scm_module_define, scm_define.
3225 These functions work with variables instead of with vcells.
3227 ** New functions for creating and defining `subr's and `gsubr's.
3229 The new functions more clearly distinguish between creating a subr (or
3230 gsubr) object and adding it to the current module.
3232 These new functions are available: scm_c_make_subr, scm_c_define_subr,
3233 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
3234 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
3235 scm_c_define_gsubr_with_generic.
3237 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
3238 scm_make_subr_with_generic, scm_make_gsubr,
3239 scm_make_gsubr_with_generic.
3241 Use the new ones from above instead.
3243 ** C interface to the module system has changed.
3245 While we suggest that you avoid as many explicit module system
3246 operations from C as possible for the time being, the C interface has
3247 been made more similar to the high-level Scheme module system.
3249 *** New functions: scm_c_define_module, scm_c_use_module,
3250 scm_c_export, scm_c_resolve_module.
3252 They mostly work like their Scheme namesakes. scm_c_define_module
3253 takes a function that is called a context where the new module is
3256 *** Deprecated functions: scm_the_root_module, scm_make_module,
3257 scm_ensure_user_module, scm_load_scheme_module.
3259 Use the new functions instead.
3261 ** Renamed function: scm_internal_with_fluids becomes
3264 scm_internal_with_fluids is available as a deprecated function.
3266 ** New function: scm_c_with_fluid.
3268 Just like scm_c_with_fluids, but takes one fluid and one value instead
3271 ** Deprecated typedefs: long_long, ulong_long.
3273 They are of questionable utility and they pollute the global
3276 ** Deprecated typedef: scm_sizet
3278 It is of questionable utility now that Guile requires ANSI C, and is
3281 ** Deprecated typedefs: scm_port_rw_active, scm_port,
3282 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
3283 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
3285 Made more compliant with the naming policy by adding a _t at the end.
3287 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
3288 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
3290 With the exception of the mysterious scm_2ulong2big, they are still
3291 available under new names (scm_i_mkbig etc). These functions are not
3292 intended to be used in user code. You should avoid dealing with
3293 bignums directly, and should deal with numbers in general (which can
3296 ** Change in behavior: scm_num2long, scm_num2ulong
3298 The scm_num2[u]long functions don't any longer accept an inexact
3299 argument. This change in behavior is motivated by concordance with
3300 R5RS: It is more common that a primitive doesn't want to accept an
3301 inexact for an exact.
3303 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
3304 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
3305 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
3308 These are conversion functions between the various ANSI C integral
3309 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
3310 accept an inexact argument.
3312 ** New functions: scm_float2num, scm_double2num,
3313 scm_num2float, scm_num2double.
3315 These are conversion functions between the two ANSI C float types and
3318 ** New number validation macros:
3319 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
3323 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
3325 These are just nicer-named old scm_protect_object and
3326 scm_unprotect_object.
3328 ** Deprecated functions: scm_protect_object, scm_unprotect_object
3330 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
3332 These functions can be used to register pointers to locations that
3335 ** Deprecated function: scm_create_hook.
3337 Its sins are: misleading name, non-modularity and lack of general
3341 Changes since Guile 1.3.4:
3343 * Changes to the distribution
3345 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
3347 We've changed the way we handle generated files in the Guile source
3348 repository. As a result, the procedure for building trees obtained
3349 from the nightly FTP snapshots or via CVS has changed:
3350 - You must have appropriate versions of autoconf, automake, and
3351 libtool installed on your system. See README for info on how to
3352 obtain these programs.
3353 - Before configuring the tree, you must first run the script
3354 `autogen.sh' at the top of the source tree.
3356 The Guile repository used to contain not only source files, written by
3357 humans, but also some generated files, like configure scripts and
3358 Makefile.in files. Even though the contents of these files could be
3359 derived mechanically from other files present, we thought it would
3360 make the tree easier to build if we checked them into CVS.
3362 However, this approach means that minor differences between
3363 developer's installed tools and habits affected the whole team.
3364 So we have removed the generated files from the repository, and
3365 added the autogen.sh script, which will reconstruct them
3369 ** configure now has experimental options to remove support for certain
3372 --disable-arrays omit array and uniform array support
3373 --disable-posix omit posix interfaces
3374 --disable-networking omit networking interfaces
3375 --disable-regex omit regular expression interfaces
3377 These are likely to become separate modules some day.
3379 ** New configure option --enable-debug-freelist
3381 This enables a debugging version of SCM_NEWCELL(), and also registers
3382 an extra primitive, the setter `gc-set-debug-check-freelist!'.
3384 Configure with the --enable-debug-freelist option to enable
3385 the gc-set-debug-check-freelist! primitive, and then use:
3387 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
3388 (gc-set-debug-check-freelist! #f) # turn off checking
3390 Checking of the freelist forces a traversal of the freelist and
3391 a garbage collection before each allocation of a cell. This can
3392 slow down the interpreter dramatically, so the setter should be used to
3393 turn on this extra processing only when necessary.
3395 ** New configure option --enable-debug-malloc
3397 Include code for debugging of calls to scm_must_malloc/realloc/free.
3401 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
3402 2. objects reallocated by scm_must_realloc has been allocated by
3404 3. reallocated objects are reallocated with the same what string
3406 But, most importantly, it records the number of allocated objects of
3407 each kind. This is useful when searching for memory leaks.
3409 A Guile compiled with this option provides the primitive
3410 `malloc-stats' which returns an alist with pairs of kind and the
3411 number of objects of that kind.
3413 ** All includes are now referenced relative to the root directory
3415 Since some users have had problems with mixups between Guile and
3416 system headers, we have decided to always refer to Guile headers via
3417 their parent directories. This essentially creates a "private name
3418 space" for Guile headers. This means that the compiler only is given
3419 -I options for the root build and root source directory.
3421 ** Header files kw.h and genio.h have been removed.
3423 ** The module (ice-9 getopt-gnu-style) has been removed.
3425 ** New module (ice-9 documentation)
3427 Implements the interface to documentation strings associated with
3430 ** New module (ice-9 time)
3432 Provides a macro `time', which displays execution time of a given form.
3434 ** New module (ice-9 history)
3436 Loading this module enables value history in the repl.
3438 * Changes to the stand-alone interpreter
3440 ** New command line option --debug
3442 Start Guile with debugging evaluator and backtraces enabled.
3444 This is useful when debugging your .guile init file or scripts.
3446 ** New help facility
3448 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
3449 (help REGEXP) ditto for objects with names matching REGEXP (a string)
3450 (help 'NAME) gives documentation for NAME, even if it is not an object
3451 (help ,EXPR) gives documentation for object returned by EXPR
3452 (help (my module)) gives module commentary for `(my module)'
3453 (help) gives this text
3455 `help' searches among bindings exported from loaded modules, while
3456 `apropos' searches among bindings visible from the "current" module.
3458 Examples: (help help)
3460 (help "output-string")
3462 ** `help' and `apropos' now prints full module names
3464 ** Dynamic linking now uses libltdl from the libtool package.
3466 The old system dependent code for doing dynamic linking has been
3467 replaced with calls to the libltdl functions which do all the hairy
3470 The major improvement is that you can now directly pass libtool
3471 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
3472 will be able to do the best shared library job you can get, via
3475 The way dynamic libraries are found has changed and is not really
3476 portable across platforms, probably. It is therefore recommended to
3477 use absolute filenames when possible.
3479 If you pass a filename without an extension to `dynamic-link', it will
3480 try a few appropriate ones. Thus, the most platform ignorant way is
3481 to specify a name like "libfoo", without any directories and
3484 ** Guile COOP threads are now compatible with LinuxThreads
3486 Previously, COOP threading wasn't possible in applications linked with
3487 Linux POSIX threads due to their use of the stack pointer to find the
3488 thread context. This has now been fixed with a workaround which uses
3489 the pthreads to allocate the stack.
3491 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
3493 ** Positions of erring expression in scripts
3495 With version 1.3.4, the location of the erring expression in Guile
3496 scipts is no longer automatically reported. (This should have been
3497 documented before the 1.3.4 release.)
3499 You can get this information by enabling recording of positions of
3500 source expressions and running the debugging evaluator. Put this at
3501 the top of your script (or in your "site" file):
3503 (read-enable 'positions)
3504 (debug-enable 'debug)
3506 ** Backtraces in scripts
3508 It is now possible to get backtraces in scripts.
3512 (debug-enable 'debug 'backtrace)
3514 at the top of the script.
3516 (The first options enables the debugging evaluator.
3517 The second enables backtraces.)
3519 ** Part of module system symbol lookup now implemented in C
3521 The eval closure of most modules is now implemented in C. Since this
3522 was one of the bottlenecks for loading speed, Guile now loads code
3523 substantially faster than before.
3525 ** Attempting to get the value of an unbound variable now produces
3526 an exception with a key of 'unbound-variable instead of 'misc-error.
3528 ** The initial default output port is now unbuffered if it's using a
3529 tty device. Previously in this situation it was line-buffered.
3531 ** New hook: after-gc-hook
3533 after-gc-hook takes over the role of gc-thunk. This hook is run at
3534 the first SCM_TICK after a GC. (Thus, the code is run at the same
3535 point during evaluation as signal handlers.)
3537 Note that this hook should be used only for diagnostic and debugging
3538 purposes. It is not certain that it will continue to be well-defined
3539 when this hook is run in the future.
3541 C programmers: Note the new C level hooks scm_before_gc_c_hook,
3542 scm_before_sweep_c_hook, scm_after_gc_c_hook.
3544 ** Improvements to garbage collector
3546 Guile 1.4 has a new policy for triggering heap allocation and
3547 determining the sizes of heap segments. It fixes a number of problems
3550 1. The new policy can handle two separate pools of cells
3551 (2-word/4-word) better. (The old policy would run wild, allocating
3552 more and more memory for certain programs.)
3554 2. The old code would sometimes allocate far too much heap so that the
3555 Guile process became gigantic. The new code avoids this.
3557 3. The old code would sometimes allocate too little so that few cells
3558 were freed at GC so that, in turn, too much time was spent in GC.
3560 4. The old code would often trigger heap allocation several times in a
3561 row. (The new scheme predicts how large the segments needs to be
3562 in order not to need further allocation.)
3564 All in all, the new GC policy will make larger applications more
3567 The new GC scheme also is prepared for POSIX threading. Threads can
3568 allocate private pools of cells ("clusters") with just a single
3569 function call. Allocation of single cells from such a cluster can
3570 then proceed without any need of inter-thread synchronization.
3572 ** New environment variables controlling GC parameters
3574 GUILE_MAX_SEGMENT_SIZE Maximal segment size
3577 Allocation of 2-word cell heaps:
3579 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
3582 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
3583 GC in percent of total heap size
3586 Allocation of 4-word cell heaps
3587 (used for real numbers and misc other objects):
3589 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
3591 (See entry "Way for application to customize GC parameters" under
3592 section "Changes to the scm_ interface" below.)
3594 ** Guile now implements reals using 4-word cells
3596 This speeds up computation with reals. (They were earlier allocated
3597 with `malloc'.) There is still some room for optimizations, however.
3599 ** Some further steps toward POSIX thread support have been taken
3601 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
3602 don't have much effect any longer, and many of them will be removed in
3606 are only handled at the top of the evaluator loop, immediately after
3607 I/O, and in scm_equalp.
3609 *** The GC can allocate thread private pools of pairs.
3611 * Changes to Scheme functions and syntax
3613 ** close-input-port and close-output-port are now R5RS
3615 These procedures have been turned into primitives and have R5RS behaviour.
3617 ** New procedure: simple-format PORT MESSAGE ARG1 ...
3619 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
3620 extended by the more sophisticated version in (ice-9 format)
3622 (simple-format port message . args)
3623 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
3624 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
3625 the escapes are replaced with corresponding members of ARGS:
3626 ~A formats using `display' and ~S formats using `write'.
3627 If DESTINATION is #t, then use the `current-output-port',
3628 if DESTINATION is #f, then return a string containing the formatted text.
3629 Does not add a trailing newline."
3631 ** string-ref: the second argument is no longer optional.
3633 ** string, list->string: no longer accept strings in their arguments,
3634 only characters, for compatibility with R5RS.
3636 ** New procedure: port-closed? PORT
3637 Returns #t if PORT is closed or #f if it is open.
3639 ** Deprecated: list*
3641 The list* functionality is now provided by cons* (SRFI-1 compliant)
3643 ** New procedure: cons* ARG1 ARG2 ... ARGn
3645 Like `list', but the last arg provides the tail of the constructed list,
3646 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3648 Requires at least one argument. If given one argument, that argument
3649 is returned as result.
3651 This function is called `list*' in some other Schemes and in Common LISP.
3653 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3655 ** New procedure: object-documentation OBJECT
3657 Returns the documentation string associated with OBJECT. The
3658 procedure uses a caching mechanism so that subsequent lookups are
3661 Exported by (ice-9 documentation).
3663 ** module-name now returns full names of modules
3665 Previously, only the last part of the name was returned (`session' for
3666 `(ice-9 session)'). Ex: `(ice-9 session)'.
3668 * Changes to the gh_ interface
3670 ** Deprecated: gh_int2scmb
3672 Use gh_bool2scm instead.
3674 * Changes to the scm_ interface
3676 ** Guile primitives now carry docstrings!
3678 Thanks to Greg Badros!
3680 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3682 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3683 macros and must contain a docstring that is extracted into foo.doc using a new
3684 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3686 However, a major overhaul of these macros is scheduled for the next release of
3689 ** Guile primitives use a new technique for validation of arguments
3691 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3692 the readability of argument checking.
3694 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3696 ** New macros: SCM_PACK, SCM_UNPACK
3698 Compose/decompose an SCM value.
3700 The SCM type is now treated as an abstract data type and may be defined as a
3701 long, a void* or as a struct, depending on the architecture and compile time
3702 options. This makes it easier to find several types of bugs, for example when
3703 SCM values are treated as integers without conversion. Values of the SCM type
3704 should be treated as "atomic" values. These macros are used when
3705 composing/decomposing an SCM value, either because you want to access
3706 individual bits, or because you want to treat it as an integer value.
3708 E.g., in order to set bit 7 in an SCM value x, use the expression
3710 SCM_PACK (SCM_UNPACK (x) | 0x80)
3712 ** The name property of hooks is deprecated.
3713 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3715 You can emulate this feature by using object properties.
3717 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3718 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3721 These macros will be removed in a future release of Guile.
3723 ** The following types, functions and macros from numbers.h are deprecated:
3724 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3725 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3727 ** Port internals: the rw_random variable in the scm_port structure
3728 must be set to non-zero in any random access port. In recent Guile
3729 releases it was only set for bidirectional random-access ports.
3731 ** Port internals: the seek ptob procedure is now responsible for
3732 resetting the buffers if required. The change was made so that in the
3733 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3734 the fport and strport ptobs can avoid resetting the buffers,
3735 in particular to avoid discarding unread chars. An existing port
3736 type can be fixed by adding something like the following to the
3737 beginning of the ptob seek procedure:
3739 if (pt->rw_active == SCM_PORT_READ)
3740 scm_end_input (object);
3741 else if (pt->rw_active == SCM_PORT_WRITE)
3742 ptob->flush (object);
3744 although to actually avoid resetting the buffers and discard unread
3745 chars requires further hacking that depends on the characteristics
3748 ** Deprecated functions: scm_fseek, scm_tag
3750 These functions are no longer used and will be removed in a future version.
3752 ** The scm_sysmissing procedure is no longer used in libguile.
3753 Unless it turns out to be unexpectedly useful to somebody, it will be
3754 removed in a future version.
3756 ** The format of error message strings has changed
3758 The two C procedures: scm_display_error and scm_error, as well as the
3759 primitive `scm-error', now use scm_simple_format to do their work.
3760 This means that the message strings of all code must be updated to use
3761 ~A where %s was used before, and ~S where %S was used before.
3763 During the period when there still are a lot of old Guiles out there,
3764 you might want to support both old and new versions of Guile.
3766 There are basically two methods to achieve this. Both methods use
3769 AC_CHECK_FUNCS(scm_simple_format)
3771 in your configure.in.
3773 Method 1: Use the string concatenation features of ANSI C's
3778 #ifdef HAVE_SCM_SIMPLE_FORMAT
3784 Then represent each of your error messages using a preprocessor macro:
3786 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3790 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3791 (define make-message string-append)
3793 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3795 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3799 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3804 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3808 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3810 Don't use the functions coop_mutex_init and
3811 coop_condition_variable_init. They will change.
3813 Use scm_mutex_init and scm_cond_init instead.
3815 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3816 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3817 COND, as `scm_cond_wait' does, but it also bounds the duration
3818 of the wait. If COND has not been signaled before time ABSTIME,
3819 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3820 returns the error code `ETIMEDOUT'.
3822 The ABSTIME parameter specifies an absolute time, with the same
3823 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3824 to 00:00:00 GMT, January 1, 1970.
3826 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3827 `scm_cond_broadcast' restarts all the threads that are waiting
3828 on the condition variable COND. Nothing happens if no threads are
3831 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3832 `scm_key_create' allocates a new TSD key. The key is stored in
3833 the location pointed to by KEY. There is no limit on the number
3834 of keys allocated at a given time. The value initially associated
3835 with the returned key is `NULL' in all currently executing threads.
3837 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3838 function associated with the key. When a thread terminates,
3839 DESTR_FUNCTION is called on the value associated with the key in
3840 that thread. The DESTR_FUNCTION is not called if a key is deleted
3841 with `scm_key_delete' or a value is changed with
3842 `scm_setspecific'. The order in which destructor functions are
3843 called at thread termination time is unspecified.
3845 Destructors are not yet implemented.
3847 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3848 `scm_setspecific' changes the value associated with KEY in the
3849 calling thread, storing the given POINTER instead.
3851 ** New function: scm_getspecific (scm_key_t KEY)
3852 `scm_getspecific' returns the value currently associated with
3853 KEY in the calling thread.
3855 ** New function: scm_key_delete (scm_key_t KEY)
3856 `scm_key_delete' deallocates a TSD key. It does not check
3857 whether non-`NULL' values are associated with that key in the
3858 currently executing threads, nor call the destructor function
3859 associated with the key.
3861 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3863 Initialize a C level hook HOOK with associated HOOK_DATA and type
3864 TYPE. (See scm_c_hook_run ().)
3866 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3868 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3869 is true, add it last, otherwise first. The same FUNC can be added
3870 multiple times if FUNC_DATA differ and vice versa.
3872 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3874 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3875 function is only removed if both FUNC and FUNC_DATA matches.
3877 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3879 Run hook HOOK passing DATA to the hook functions.
3881 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3882 returned is undefined.
3884 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3885 returns a non-NULL value. This value is returned as the result of
3886 scm_c_hook_run. If all functions return NULL, NULL is returned.
3888 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3889 returns a NULL value, and NULL is returned. If all functions returns
3890 a non-NULL value, the last value is returned.
3892 ** New C level GC hooks
3894 Five new C level hooks has been added to the garbage collector.
3896 scm_before_gc_c_hook
3899 are run before locking and after unlocking the heap. The system is
3900 thus in a mode where evaluation can take place. (Except that
3901 scm_before_gc_c_hook must not allocate new cells.)
3903 scm_before_mark_c_hook
3904 scm_before_sweep_c_hook
3905 scm_after_sweep_c_hook
3907 are run when the heap is locked. These are intended for extension of
3908 the GC in a modular fashion. Examples are the weaks and guardians
3911 ** Way for application to customize GC parameters
3913 The application can set up other default values for the GC heap
3914 allocation parameters
3916 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3917 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3918 GUILE_MAX_SEGMENT_SIZE,
3922 scm_default_init_heap_size_1, scm_default_min_yield_1,
3923 scm_default_init_heap_size_2, scm_default_min_yield_2,
3924 scm_default_max_segment_size
3926 respectively before callong scm_boot_guile.
3928 (See entry "New environment variables ..." in section
3929 "Changes to the stand-alone interpreter" above.)
3931 ** scm_protect_object/scm_unprotect_object now nest
3933 This means that you can call scm_protect_object multiple times on an
3934 object and count on the object being protected until
3935 scm_unprotect_object has been call the same number of times.
3937 The functions also have better time complexity.
3939 Still, it is usually possible to structure the application in a way
3940 that you don't need to use these functions. For example, if you use a
3941 protected standard Guile list to keep track of live objects rather
3942 than some custom data type, objects will die a natural death when they
3943 are no longer needed.
3945 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3947 Guile does not provide the float representation for inexact real numbers any
3948 more. Now, only doubles are used to represent inexact real numbers. Further,
3949 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3950 and scm_tc16_complex, respectively.
3952 ** Removed deprecated type scm_smobfuns
3954 ** Removed deprecated function scm_newsmob
3956 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3958 There is an ongoing discussion among the developers whether to
3959 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3960 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3961 until this issue has been settled.
3963 ** Removed deprecated type tag scm_tc16_kw
3965 ** Added type tag scm_tc16_keyword
3967 (This was introduced already in release 1.3.4 but was not documented
3970 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3972 * Changes to system call interfaces:
3974 ** The "select" procedure now tests port buffers for the ability to
3975 provide input or accept output. Previously only the underlying file
3976 descriptors were checked.
3978 ** New variable PIPE_BUF: the maximum number of bytes that can be
3979 atomically written to a pipe.
3981 ** If a facility is not available on the system when Guile is
3982 compiled, the corresponding primitive procedure will not be defined.
3983 Previously it would have been defined but would throw a system-error
3984 exception if called. Exception handlers which catch this case may
3985 need minor modification: an error will be thrown with key
3986 'unbound-variable instead of 'system-error. Alternatively it's
3987 now possible to use `defined?' to check whether the facility is
3990 ** Procedures which depend on the timezone should now give the correct
3991 result on systems which cache the TZ environment variable, even if TZ
3992 is changed without calling tzset.
3994 * Changes to the networking interfaces:
3996 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3997 long integers between network and host format. For now, it's not
3998 particularly convenient to do this kind of thing, but consider:
4000 (define write-network-long
4001 (lambda (value port)
4002 (let ((v (make-uniform-vector 1 1 0)))
4003 (uniform-vector-set! v 0 (htonl value))
4004 (uniform-vector-write v port))))
4006 (define read-network-long
4008 (let ((v (make-uniform-vector 1 1 0)))
4009 (uniform-vector-read! v port)
4010 (ntohl (uniform-vector-ref v 0)))))
4012 ** If inet-aton fails, it now throws an error with key 'misc-error
4013 instead of 'system-error, since errno is not relevant.
4015 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
4016 specific keys instead of 'system-error. The latter is inappropriate
4017 since errno will not have been set. The keys are:
4018 'host-not-found, 'try-again, 'no-recovery and 'no-data.
4020 ** sethostent, setnetent, setprotoent, setservent: now take an
4021 optional argument STAYOPEN, which specifies whether the database
4022 remains open after a database entry is accessed randomly (e.g., using
4023 gethostbyname for the hosts database.) The default is #f. Previously
4027 Changes since Guile 1.3.2:
4029 * Changes to the stand-alone interpreter
4033 An initial version of the Guile debugger written by Chris Hanson has
4034 been added. The debugger is still under development but is included
4035 in the distribution anyway since it is already quite useful.
4041 after an error to enter the debugger. Type `help' inside the debugger
4042 for a description of available commands.
4044 If you prefer to have stack frames numbered and printed in
4045 anti-chronological order and prefer up in the stack to be down on the
4046 screen as is the case in gdb, you can put
4048 (debug-enable 'backwards)
4050 in your .guile startup file. (However, this means that Guile can't
4051 use indentation to indicate stack level.)
4053 The debugger is autoloaded into Guile at the first use.
4055 ** Further enhancements to backtraces
4057 There is a new debug option `width' which controls the maximum width
4058 on the screen of printed stack frames. Fancy printing parameters
4059 ("level" and "length" as in Common LISP) are adaptively adjusted for
4060 each stack frame to give maximum information while still fitting
4061 within the bounds. If the stack frame can't be made to fit by
4062 adjusting parameters, it is simply cut off at the end. This is marked
4065 ** Some modules are now only loaded when the repl is started
4067 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
4068 regex) are now loaded into (guile-user) only if the repl has been
4069 started. The effect is that the startup time for scripts has been
4070 reduced to 30% of what it was previously.
4072 Correctly written scripts load the modules they require at the top of
4073 the file and should not be affected by this change.
4075 ** Hooks are now represented as smobs
4077 * Changes to Scheme functions and syntax
4079 ** Readline support has changed again.
4081 The old (readline-activator) module is gone. Use (ice-9 readline)
4082 instead, which now contains all readline functionality. So the code
4083 to activate readline is now
4085 (use-modules (ice-9 readline))
4088 This should work at any time, including from the guile prompt.
4090 To avoid confusion about the terms of Guile's license, please only
4091 enable readline for your personal use; please don't make it the
4092 default for others. Here is why we make this rather odd-sounding
4095 Guile is normally licensed under a weakened form of the GNU General
4096 Public License, which allows you to link code with Guile without
4097 placing that code under the GPL. This exception is important to some
4100 However, since readline is distributed under the GNU General Public
4101 License, when you link Guile with readline, either statically or
4102 dynamically, you effectively change Guile's license to the strict GPL.
4103 Whenever you link any strictly GPL'd code into Guile, uses of Guile
4104 which are normally permitted become forbidden. This is a rather
4105 non-obvious consequence of the licensing terms.
4107 So, to make sure things remain clear, please let people choose for
4108 themselves whether to link GPL'd libraries like readline with Guile.
4110 ** regexp-substitute/global has changed slightly, but incompatibly.
4112 If you include a function in the item list, the string of the match
4113 object it receives is the same string passed to
4114 regexp-substitute/global, not some suffix of that string.
4115 Correspondingly, the match's positions are relative to the entire
4116 string, not the suffix.
4118 If the regexp can match the empty string, the way matches are chosen
4119 from the string has changed. regexp-substitute/global recognizes the
4120 same set of matches that list-matches does; see below.
4122 ** New function: list-matches REGEXP STRING [FLAGS]
4124 Return a list of match objects, one for every non-overlapping, maximal
4125 match of REGEXP in STRING. The matches appear in left-to-right order.
4126 list-matches only reports matches of the empty string if there are no
4127 other matches which begin on, end at, or include the empty match's
4130 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4132 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
4134 For each match of REGEXP in STRING, apply PROC to the match object,
4135 and the last value PROC returned, or INIT for the first call. Return
4136 the last value returned by PROC. We apply PROC to the matches as they
4137 appear from left to right.
4139 This function recognizes matches according to the same criteria as
4142 Thus, you could define list-matches like this:
4144 (define (list-matches regexp string . flags)
4145 (reverse! (apply fold-matches regexp string '() cons flags)))
4147 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4151 *** New function: hook? OBJ
4153 Return #t if OBJ is a hook, otherwise #f.
4155 *** New function: make-hook-with-name NAME [ARITY]
4157 Return a hook with name NAME and arity ARITY. The default value for
4158 ARITY is 0. The only effect of NAME is that it will appear when the
4159 hook object is printed to ease debugging.
4161 *** New function: hook-empty? HOOK
4163 Return #t if HOOK doesn't contain any procedures, otherwise #f.
4165 *** New function: hook->list HOOK
4167 Return a list of the procedures that are called when run-hook is
4170 ** `map' signals an error if its argument lists are not all the same length.
4172 This is the behavior required by R5RS, so this change is really a bug
4173 fix. But it seems to affect a lot of people's code, so we're
4174 mentioning it here anyway.
4176 ** Print-state handling has been made more transparent
4178 Under certain circumstances, ports are represented as a port with an
4179 associated print state. Earlier, this pair was represented as a pair
4180 (see "Some magic has been added to the printer" below). It is now
4181 indistinguishable (almost; see `get-print-state') from a port on the
4184 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
4186 Return a new port with the associated print state PRINT-STATE.
4188 *** New function: get-print-state OUTPUT-PORT
4190 Return the print state associated with this port if it exists,
4191 otherwise return #f.
4193 *** New function: directory-stream? OBJECT
4195 Returns true iff OBJECT is a directory stream --- the sort of object
4196 returned by `opendir'.
4198 ** New function: using-readline?
4200 Return #t if readline is in use in the current repl.
4202 ** structs will be removed in 1.4
4204 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
4205 and use GOOPS objects as the fundamental record type.
4207 * Changes to the scm_ interface
4209 ** structs will be removed in 1.4
4211 The entire current struct interface (struct.c, struct.h) will be
4212 replaced in Guile 1.4. We will merge GOOPS into libguile and use
4213 GOOPS objects as the fundamental record type.
4215 ** The internal representation of subr's has changed
4217 Instead of giving a hint to the subr name, the CAR field of the subr
4218 now contains an index to a subr entry in scm_subr_table.
4220 *** New variable: scm_subr_table
4222 An array of subr entries. A subr entry contains the name, properties
4223 and documentation associated with the subr. The properties and
4224 documentation slots are not yet used.
4226 ** A new scheme for "forwarding" calls to a builtin to a generic function
4228 It is now possible to extend the functionality of some Guile
4229 primitives by letting them defer a call to a GOOPS generic function on
4230 argument mismatch. This means that there is no loss of efficiency in
4235 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
4236 (define-method + ((x <string>) (y <string>))
4237 (string-append x y))
4239 + will still be as efficient as usual in numerical calculations, but
4240 can also be used for concatenating strings.
4242 Who will be the first one to extend Guile's numerical tower to
4243 rationals? :) [OK, there a few other things to fix before this can
4244 be made in a clean way.]
4246 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
4248 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
4250 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
4252 These do the same job as SCM_PROC and SCM_PROC1, but they also define
4253 a variable GENERIC which can be used by the dispatch macros below.
4255 [This is experimental code which may change soon.]
4257 *** New macros for forwarding control to a generic on arg type error
4259 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
4261 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
4263 These correspond to the scm_wta function call, and have the same
4264 behaviour until the user has called the GOOPS primitive
4265 `enable-primitive-generic!'. After that, these macros will apply the
4266 generic function GENERIC to the argument(s) instead of calling
4269 [This is experimental code which may change soon.]
4271 *** New macros for argument testing with generic dispatch
4273 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
4275 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
4277 These correspond to the SCM_ASSERT macro, but will defer control to
4278 GENERIC on error after `enable-primitive-generic!' has been called.
4280 [This is experimental code which may change soon.]
4282 ** New function: SCM scm_eval_body (SCM body, SCM env)
4284 Evaluates the body of a special form.
4286 ** The internal representation of struct's has changed
4288 Previously, four slots were allocated for the procedure(s) of entities
4289 and operators. The motivation for this representation had to do with
4290 the structure of the evaluator, the wish to support tail-recursive
4291 generic functions, and efficiency. Since the generic function
4292 dispatch mechanism has changed, there is no longer a need for such an
4293 expensive representation, and the representation has been simplified.
4295 This should not make any difference for most users.
4297 ** GOOPS support has been cleaned up.
4299 Some code has been moved from eval.c to objects.c and code in both of
4300 these compilation units has been cleaned up and better structured.
4302 *** New functions for applying generic functions
4304 New function: SCM scm_apply_generic (GENERIC, ARGS)
4305 New function: SCM scm_call_generic_0 (GENERIC)
4306 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
4307 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
4308 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
4310 ** Deprecated function: scm_make_named_hook
4312 It is now replaced by:
4314 ** New function: SCM scm_create_hook (const char *name, int arity)
4316 Creates a hook in the same way as make-hook above but also
4317 binds a variable named NAME to it.
4319 This is the typical way of creating a hook from C code.
4321 Currently, the variable is created in the "current" module.
4322 This might change when we get the new module system.
4324 [The behaviour is identical to scm_make_named_hook.]
4328 Changes since Guile 1.3:
4330 * Changes to mailing lists
4332 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
4334 See the README file to find current addresses for all the Guile
4337 * Changes to the distribution
4339 ** Readline support is no longer included with Guile by default.
4341 Based on the different license terms of Guile and Readline, we
4342 concluded that Guile should not *by default* cause the linking of
4343 Readline into an application program. Readline support is now offered
4344 as a separate module, which is linked into an application only when
4345 you explicitly specify it.
4347 Although Guile is GNU software, its distribution terms add a special
4348 exception to the usual GNU General Public License (GPL). Guile's
4349 license includes a clause that allows you to link Guile with non-free
4350 programs. We add this exception so as not to put Guile at a
4351 disadvantage vis-a-vis other extensibility packages that support other
4354 In contrast, the GNU Readline library is distributed under the GNU
4355 General Public License pure and simple. This means that you may not
4356 link Readline, even dynamically, into an application unless it is
4357 distributed under a free software license that is compatible the GPL.
4359 Because of this difference in distribution terms, an application that
4360 can use Guile may not be able to use Readline. Now users will be
4361 explicitly offered two independent decisions about the use of these
4364 You can activate the readline support by issuing
4366 (use-modules (readline-activator))
4369 from your ".guile" file, for example.
4371 * Changes to the stand-alone interpreter
4373 ** All builtins now print as primitives.
4374 Previously builtin procedures not belonging to the fundamental subr
4375 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
4376 Now, they print as #<primitive-procedure NAME>.
4378 ** Backtraces slightly more intelligible.
4379 gsubr-apply and macro transformer application frames no longer appear
4382 * Changes to Scheme functions and syntax
4384 ** Guile now correctly handles internal defines by rewriting them into
4385 their equivalent letrec. Previously, internal defines would
4386 incrementally add to the innermost environment, without checking
4387 whether the restrictions specified in RnRS were met. This lead to the
4388 correct behaviour when these restriction actually were met, but didn't
4389 catch all illegal uses. Such an illegal use could lead to crashes of
4390 the Guile interpreter or or other unwanted results. An example of
4391 incorrect internal defines that made Guile behave erratically:
4403 The problem with this example is that the definition of `c' uses the
4404 value of `b' directly. This confuses the meoization machine of Guile
4405 so that the second call of `b' (this time in a larger environment that
4406 also contains bindings for `c' and `d') refers to the binding of `c'
4407 instead of `a'. You could also make Guile crash with a variation on
4412 (define (b flag) (if flag a 1))
4413 (define c (1+ (b flag)))
4421 From now on, Guile will issue an `Unbound variable: b' error message
4426 A hook contains a list of functions which should be called on
4427 particular occasions in an existing program. Hooks are used for
4430 A window manager might have a hook before-window-map-hook. The window
4431 manager uses the function run-hooks to call all functions stored in
4432 before-window-map-hook each time a window is mapped. The user can
4433 store functions in the hook using add-hook!.
4435 In Guile, hooks are first class objects.
4437 *** New function: make-hook [N_ARGS]
4439 Return a hook for hook functions which can take N_ARGS arguments.
4440 The default value for N_ARGS is 0.
4442 (See also scm_make_named_hook below.)
4444 *** New function: add-hook! HOOK PROC [APPEND_P]
4446 Put PROC at the beginning of the list of functions stored in HOOK.
4447 If APPEND_P is supplied, and non-false, put PROC at the end instead.
4449 PROC must be able to take the number of arguments specified when the
4452 If PROC already exists in HOOK, then remove it first.
4454 *** New function: remove-hook! HOOK PROC
4456 Remove PROC from the list of functions in HOOK.
4458 *** New function: reset-hook! HOOK
4460 Clear the list of hook functions stored in HOOK.
4462 *** New function: run-hook HOOK ARG1 ...
4464 Run all hook functions stored in HOOK with arguments ARG1 ... .
4465 The number of arguments supplied must correspond to the number given
4466 when the hook was created.
4468 ** The function `dynamic-link' now takes optional keyword arguments.
4469 The only keyword argument that is currently defined is `:global
4470 BOOL'. With it, you can control whether the shared library will be
4471 linked in global mode or not. In global mode, the symbols from the
4472 linked library can be used to resolve references from other
4473 dynamically linked libraries. In non-global mode, the linked
4474 library is essentially invisible and can only be accessed via
4475 `dynamic-func', etc. The default is now to link in global mode.
4476 Previously, the default has been non-global mode.
4478 The `#:global' keyword is only effective on platforms that support
4479 the dlopen family of functions.
4481 ** New function `provided?'
4483 - Function: provided? FEATURE
4484 Return true iff FEATURE is supported by this installation of
4485 Guile. FEATURE must be a symbol naming a feature; the global
4486 variable `*features*' is a list of available features.
4488 ** Changes to the module (ice-9 expect):
4490 *** The expect-strings macro now matches `$' in a regular expression
4491 only at a line-break or end-of-file by default. Previously it would
4492 match the end of the string accumulated so far. The old behaviour
4493 can be obtained by setting the variable `expect-strings-exec-flags'
4496 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
4497 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
4498 in a regular expression will still match before a line-break or
4499 end-of-file. The default is `regexp/noteol'.
4501 *** The expect-strings macro now uses a variable
4502 `expect-strings-compile-flags' for the flags to be supplied to
4503 `make-regexp'. The default is `regexp/newline', which was previously
4506 *** The expect macro now supplies two arguments to a match procedure:
4507 the current accumulated string and a flag to indicate whether
4508 end-of-file has been reached. Previously only the string was supplied.
4509 If end-of-file is reached, the match procedure will be called an
4510 additional time with the same accumulated string as the previous call
4511 but with the flag set.
4513 ** New module (ice-9 format), implementing the Common Lisp `format' function.
4515 This code, and the documentation for it that appears here, was
4516 borrowed from SLIB, with minor adaptations for Guile.
4518 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
4519 An almost complete implementation of Common LISP format description
4520 according to the CL reference book `Common LISP' from Guy L.
4521 Steele, Digital Press. Backward compatible to most of the
4522 available Scheme format implementations.
4524 Returns `#t', `#f' or a string; has side effect of printing
4525 according to FORMAT-STRING. If DESTINATION is `#t', the output is
4526 to the current output port and `#t' is returned. If DESTINATION
4527 is `#f', a formatted string is returned as the result of the call.
4528 NEW: If DESTINATION is a string, DESTINATION is regarded as the
4529 format string; FORMAT-STRING is then the first argument and the
4530 output is returned as a string. If DESTINATION is a number, the
4531 output is to the current error port if available by the
4532 implementation. Otherwise DESTINATION must be an output port and
4535 FORMAT-STRING must be a string. In case of a formatting error
4536 format returns `#f' and prints a message on the current output or
4537 error port. Characters are output as if the string were output by
4538 the `display' function with the exception of those prefixed by a
4539 tilde (~). For a detailed description of the FORMAT-STRING syntax
4540 please consult a Common LISP format reference manual. For a test
4541 suite to verify this format implementation load `formatst.scm'.
4542 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
4544 Note: `format' is not reentrant, i.e. only one `format'-call may
4545 be executed at a time.
4548 *** Format Specification (Format version 3.0)
4550 Please consult a Common LISP format reference manual for a detailed
4551 description of the format string syntax. For a demonstration of the
4552 implemented directives see `formatst.scm'.
4554 This implementation supports directive parameters and modifiers (`:'
4555 and `@' characters). Multiple parameters must be separated by a comma
4556 (`,'). Parameters can be numerical parameters (positive or negative),
4557 character parameters (prefixed by a quote character (`''), variable
4558 parameters (`v'), number of rest arguments parameter (`#'), empty and
4559 default parameters. Directive characters are case independent. The
4560 general form of a directive is:
4562 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
4564 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
4566 *** Implemented CL Format Control Directives
4568 Documentation syntax: Uppercase characters represent the
4569 corresponding control directive characters. Lowercase characters
4570 represent control directive parameter descriptions.
4573 Any (print as `display' does).
4577 `~MINCOL,COLINC,MINPAD,PADCHARA'
4581 S-expression (print as `write' does).
4585 `~MINCOL,COLINC,MINPAD,PADCHARS'
4591 print number sign always.
4594 print comma separated.
4596 `~MINCOL,PADCHAR,COMMACHARD'
4602 print number sign always.
4605 print comma separated.
4607 `~MINCOL,PADCHAR,COMMACHARX'
4613 print number sign always.
4616 print comma separated.
4618 `~MINCOL,PADCHAR,COMMACHARO'
4624 print number sign always.
4627 print comma separated.
4629 `~MINCOL,PADCHAR,COMMACHARB'
4634 `~N,MINCOL,PADCHAR,COMMACHARR'
4638 print a number as a Roman numeral.
4641 print a number as an "old fashioned" Roman numeral.
4644 print a number as an ordinal English number.
4647 print a number as a cardinal English number.
4652 prints `y' and `ies'.
4655 as `~P but jumps 1 argument backward.'
4658 as `~@P but jumps 1 argument backward.'
4663 prints a character as the reader can understand it (i.e. `#\'
4667 prints a character as emacs does (eg. `^C' for ASCII 03).
4670 Fixed-format floating-point (prints a flonum like MMM.NNN).
4671 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4673 If the number is positive a plus sign is printed.
4676 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4677 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4679 If the number is positive a plus sign is printed.
4682 General floating-point (prints a flonum either fixed or
4684 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4686 If the number is positive a plus sign is printed.
4689 Dollars floating-point (prints a flonum in fixed with signs
4691 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4693 If the number is positive a plus sign is printed.
4696 A sign is always printed and appears before the padding.
4699 The sign appears before the padding.
4707 print newline if not at the beginning of the output line.
4709 prints `~&' and then N-1 newlines.
4714 print N page separators.
4724 newline is ignored, white space left.
4727 newline is left, white space ignored.
4732 relative tabulation.
4738 Indirection (expects indirect arguments as a list).
4740 extracts indirect arguments from format arguments.
4743 Case conversion (converts by `string-downcase').
4745 converts by `string-capitalize'.
4748 converts by `string-capitalize-first'.
4751 converts by `string-upcase'.
4754 Argument Jumping (jumps 1 argument forward).
4756 jumps N arguments forward.
4759 jumps 1 argument backward.
4762 jumps N arguments backward.
4765 jumps to the 0th argument.
4768 jumps to the Nth argument (beginning from 0)
4770 `~[STR0~;STR1~;...~;STRN~]'
4771 Conditional Expression (numerical clause conditional).
4773 take argument from N.
4776 true test conditional.
4779 if-else-then conditional.
4785 default clause follows.
4788 Iteration (args come from the next argument (a list)).
4790 at most N iterations.
4793 args from next arg (a list of lists).
4796 args from the rest of arguments.
4799 args from the rest args (lists).
4810 aborts if N <= M <= K
4812 *** Not Implemented CL Format Control Directives
4815 print `#f' as an empty list (see below).
4818 print `#f' as an empty list (see below).
4824 (sorry I don't understand its semantics completely)
4826 *** Extended, Replaced and Additional Control Directives
4828 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4829 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4830 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4831 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4832 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4833 COMMAWIDTH is the number of characters between two comma
4837 print a R4RS complex number as `~F~@Fi' with passed parameters for
4841 Pretty print formatting of an argument for scheme code lists.
4847 Flushes the output if format DESTINATION is a port.
4850 Print a `#\space' character
4852 print N `#\space' characters.
4855 Print a `#\tab' character
4857 print N `#\tab' characters.
4860 Takes N as an integer representation for a character. No arguments
4861 are consumed. N is converted to a character by `integer->char'. N
4862 must be a positive decimal number.
4865 Print out readproof. Prints out internal objects represented as
4866 `#<...>' as strings `"#<...>"' so that the format output can always
4867 be processed by `read'.
4870 Print out readproof. Prints out internal objects represented as
4871 `#<...>' as strings `"#<...>"' so that the format output can always
4872 be processed by `read'.
4875 Prints information and a copyright notice on the format
4878 prints format version.
4881 may also print number strings, i.e. passing a number as a string
4882 and format it accordingly.
4884 *** Configuration Variables
4886 The format module exports some configuration variables to suit the
4887 systems and users needs. There should be no modification necessary for
4888 the configuration that comes with Guile. Format detects automatically
4889 if the running scheme system implements floating point numbers and
4892 format:symbol-case-conv
4893 Symbols are converted by `symbol->string' so the case type of the
4894 printed symbols is implementation dependent.
4895 `format:symbol-case-conv' is a one arg closure which is either
4896 `#f' (no conversion), `string-upcase', `string-downcase' or
4897 `string-capitalize'. (default `#f')
4899 format:iobj-case-conv
4900 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4901 implementation internal objects. (default `#f')
4904 The character prefixing the exponent value in `~E' printing.
4907 *** Compatibility With Other Format Implementations
4913 Downward compatible except for padding support and `~A', `~S',
4914 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4915 `printf' padding support which is completely replaced by the CL
4916 `format' padding style.
4919 Downward compatible except for `~', which is not documented
4920 (ignores all characters inside the format string up to a newline
4921 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4922 numerical and variable parameters and `:/@' modifiers in the CL
4926 Downward compatible except for `~A' and `~S' which print in
4927 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4928 directive parameters or modifiers)).
4931 Downward compatible except for an optional destination parameter:
4932 S2C accepts a format call without a destination which returns a
4933 formatted string. This is equivalent to a #f destination in S2C.
4934 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4935 parameters or modifiers)).
4938 ** Changes to string-handling functions.
4940 These functions were added to support the (ice-9 format) module, above.
4942 *** New function: string-upcase STRING
4943 *** New function: string-downcase STRING
4945 These are non-destructive versions of the existing string-upcase! and
4946 string-downcase! functions.
4948 *** New function: string-capitalize! STRING
4949 *** New function: string-capitalize STRING
4951 These functions convert the first letter of each word in the string to
4954 (string-capitalize "howdy there")
4957 As with the other functions, string-capitalize! modifies the string in
4958 place, while string-capitalize returns a modified copy of its argument.
4960 *** New function: string-ci->symbol STRING
4962 Return a symbol whose name is STRING, but having the same case as if
4963 the symbol had be read by `read'.
4965 Guile can be configured to be sensitive or insensitive to case
4966 differences in Scheme identifiers. If Guile is case-insensitive, all
4967 symbols are converted to lower case on input. The `string-ci->symbol'
4968 function returns a symbol whose name in STRING, transformed as Guile
4969 would if STRING were input.
4971 *** New function: substring-move! STRING1 START END STRING2 START
4973 Copy the substring of STRING1 from START (inclusive) to END
4974 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4975 string, and the source and destination areas may overlap; in all
4976 cases, the function behaves as if all the characters were copied
4979 *** Extended functions: substring-move-left! substring-move-right!
4981 These functions now correctly copy arbitrarily overlapping substrings;
4982 they are both synonyms for substring-move!.
4985 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4987 getopt-long is a function for parsing command-line arguments in a
4988 manner consistent with other GNU programs.
4990 (getopt-long ARGS GRAMMAR)
4991 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4993 ARGS should be a list of strings. Its first element should be the
4994 name of the program; subsequent elements should be the arguments
4995 that were passed to the program on the command line. The
4996 `program-arguments' procedure returns a list of this form.
4998 GRAMMAR is a list of the form:
4999 ((OPTION (PROPERTY VALUE) ...) ...)
5001 Each OPTION should be a symbol. `getopt-long' will accept a
5002 command-line option named `--OPTION'.
5003 Each option can have the following (PROPERTY VALUE) pairs:
5005 (single-char CHAR) --- Accept `-CHAR' as a single-character
5006 equivalent to `--OPTION'. This is how to specify traditional
5008 (required? BOOL) --- If BOOL is true, the option is required.
5009 getopt-long will raise an error if it is not found in ARGS.
5010 (value BOOL) --- If BOOL is #t, the option accepts a value; if
5011 it is #f, it does not; and if it is the symbol
5012 `optional', the option may appear in ARGS with or
5014 (predicate FUNC) --- If the option accepts a value (i.e. you
5015 specified `(value #t)' for this option), then getopt
5016 will apply FUNC to the value, and throw an exception
5017 if it returns #f. FUNC should be a procedure which
5018 accepts a string and returns a boolean value; you may
5019 need to use quasiquotes to get it into GRAMMAR.
5021 The (PROPERTY VALUE) pairs may occur in any order, but each
5022 property may occur only once. By default, options do not have
5023 single-character equivalents, are not required, and do not take
5026 In ARGS, single-character options may be combined, in the usual
5027 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
5028 accepts values, then it must be the last option in the
5029 combination; the value is the next argument. So, for example, using
5030 the following grammar:
5031 ((apples (single-char #\a))
5032 (blimps (single-char #\b) (value #t))
5033 (catalexis (single-char #\c) (value #t)))
5034 the following argument lists would be acceptable:
5035 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
5036 for "blimps" and "catalexis")
5037 ("-ab" "bang" "-c" "couth") (same)
5038 ("-ac" "couth" "-b" "bang") (same)
5039 ("-abc" "couth" "bang") (an error, since `-b' is not the
5040 last option in its combination)
5042 If an option's value is optional, then `getopt-long' decides
5043 whether it has a value by looking at what follows it in ARGS. If
5044 the next element is a string, and it does not appear to be an
5045 option itself, then that string is the option's value.
5047 The value of a long option can appear as the next element in ARGS,
5048 or it can follow the option name, separated by an `=' character.
5049 Thus, using the same grammar as above, the following argument lists
5051 ("--apples" "Braeburn" "--blimps" "Goodyear")
5052 ("--apples=Braeburn" "--blimps" "Goodyear")
5053 ("--blimps" "Goodyear" "--apples=Braeburn")
5055 If the option "--" appears in ARGS, argument parsing stops there;
5056 subsequent arguments are returned as ordinary arguments, even if
5057 they resemble options. So, in the argument list:
5058 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
5059 `getopt-long' will recognize the `apples' option as having the
5060 value "Granny Smith", but it will not recognize the `blimp'
5061 option; it will return the strings "--blimp" and "Goodyear" as
5062 ordinary argument strings.
5064 The `getopt-long' function returns the parsed argument list as an
5065 assocation list, mapping option names --- the symbols from GRAMMAR
5066 --- onto their values, or #t if the option does not accept a value.
5067 Unused options do not appear in the alist.
5069 All arguments that are not the value of any option are returned
5070 as a list, associated with the empty list.
5072 `getopt-long' throws an exception if:
5073 - it finds an unrecognized option in ARGS
5074 - a required option is omitted
5075 - an option that requires an argument doesn't get one
5076 - an option that doesn't accept an argument does get one (this can
5077 only happen using the long option `--opt=value' syntax)
5078 - an option predicate fails
5083 `((lockfile-dir (required? #t)
5086 (predicate ,file-is-directory?))
5087 (verbose (required? #f)
5090 (x-includes (single-char #\x))
5091 (rnet-server (single-char #\y)
5092 (predicate ,string?))))
5094 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
5095 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
5097 => ((() "foo1" "-fred" "foo2" "foo3")
5098 (rnet-server . "lamprod")
5099 (x-includes . "/usr/include")
5100 (lockfile-dir . "/tmp")
5103 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
5105 It will be removed in a few releases.
5107 ** New syntax: lambda*
5108 ** New syntax: define*
5109 ** New syntax: define*-public
5110 ** New syntax: defmacro*
5111 ** New syntax: defmacro*-public
5112 Guile now supports optional arguments.
5114 `lambda*', `define*', `define*-public', `defmacro*' and
5115 `defmacro*-public' are identical to the non-* versions except that
5116 they use an extended type of parameter list that has the following BNF
5117 syntax (parentheses are literal, square brackets indicate grouping,
5118 and `*', `+' and `?' have the usual meaning):
5120 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
5121 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
5122 [[#&rest identifier]|[. identifier]]? ) | [identifier]
5124 ext-var-decl ::= identifier | ( identifier expression )
5126 The semantics are best illustrated with the following documentation
5127 and examples for `lambda*':
5130 lambda extended for optional and keyword arguments
5132 lambda* creates a procedure that takes optional arguments. These
5133 are specified by putting them inside brackets at the end of the
5134 paramater list, but before any dotted rest argument. For example,
5135 (lambda* (a b #&optional c d . e) '())
5136 creates a procedure with fixed arguments a and b, optional arguments c
5137 and d, and rest argument e. If the optional arguments are omitted
5138 in a call, the variables for them are unbound in the procedure. This
5139 can be checked with the bound? macro.
5141 lambda* can also take keyword arguments. For example, a procedure
5143 (lambda* (#&key xyzzy larch) '())
5144 can be called with any of the argument lists (#:xyzzy 11)
5145 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
5146 are given as keywords are bound to values.
5148 Optional and keyword arguments can also be given default values
5149 which they take on when they are not present in a call, by giving a
5150 two-item list in place of an optional argument, for example in:
5151 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
5152 foo is a fixed argument, bar is an optional argument with default
5153 value 42, and baz is a keyword argument with default value 73.
5154 Default value expressions are not evaluated unless they are needed
5155 and until the procedure is called.
5157 lambda* now supports two more special parameter list keywords.
5159 lambda*-defined procedures now throw an error by default if a
5160 keyword other than one of those specified is found in the actual
5161 passed arguments. However, specifying #&allow-other-keys
5162 immediately after the kyword argument declarations restores the
5163 previous behavior of ignoring unknown keywords. lambda* also now
5164 guarantees that if the same keyword is passed more than once, the
5165 last one passed is the one that takes effect. For example,
5166 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
5167 #:heads 37 #:tails 42 #:heads 99)
5168 would result in (99 47) being displayed.
5170 #&rest is also now provided as a synonym for the dotted syntax rest
5171 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
5172 all respects to lambda*. This is provided for more similarity to DSSSL,
5173 MIT-Scheme and Kawa among others, as well as for refugees from other
5176 Further documentation may be found in the optargs.scm file itself.
5178 The optional argument module also exports the macros `let-optional',
5179 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
5180 are not documented here because they may be removed in the future, but
5181 full documentation is still available in optargs.scm.
5183 ** New syntax: and-let*
5184 Guile now supports the `and-let*' form, described in the draft SRFI-2.
5186 Syntax: (land* (<clause> ...) <body> ...)
5187 Each <clause> should have one of the following forms:
5188 (<variable> <expression>)
5191 Each <variable> or <bound-variable> should be an identifier. Each
5192 <expression> should be a valid expression. The <body> should be a
5193 possibly empty sequence of expressions, like the <body> of a
5196 Semantics: A LAND* expression is evaluated by evaluating the
5197 <expression> or <bound-variable> of each of the <clause>s from
5198 left to right. The value of the first <expression> or
5199 <bound-variable> that evaluates to a false value is returned; the
5200 remaining <expression>s and <bound-variable>s are not evaluated.
5201 The <body> forms are evaluated iff all the <expression>s and
5202 <bound-variable>s evaluate to true values.
5204 The <expression>s and the <body> are evaluated in an environment
5205 binding each <variable> of the preceding (<variable> <expression>)
5206 clauses to the value of the <expression>. Later bindings
5207 shadow earlier bindings.
5209 Guile's and-let* macro was contributed by Michael Livshin.
5211 ** New sorting functions
5213 *** New function: sorted? SEQUENCE LESS?
5214 Returns `#t' when the sequence argument is in non-decreasing order
5215 according to LESS? (that is, there is no adjacent pair `... x y
5216 ...' for which `(less? y x)').
5218 Returns `#f' when the sequence contains at least one out-of-order
5219 pair. It is an error if the sequence is neither a list nor a
5222 *** New function: merge LIST1 LIST2 LESS?
5223 LIST1 and LIST2 are sorted lists.
5224 Returns the sorted list of all elements in LIST1 and LIST2.
5226 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
5227 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
5228 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
5229 (Here "<" should read "comes before".)
5231 *** New procedure: merge! LIST1 LIST2 LESS?
5232 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
5233 the result. If the code is compiled, and LESS? constructs no new
5234 pairs, no pairs at all will be allocated. The first pair of the
5235 result will be either the first pair of LIST1 or the first pair of
5238 *** New function: sort SEQUENCE LESS?
5239 Accepts either a list or a vector, and returns a new sequence
5240 which is sorted. The new sequence is the same type as the input.
5241 Always `(sorted? (sort sequence less?) less?)'. The original
5242 sequence is not altered in any way. The new sequence shares its
5243 elements with the old one; no elements are copied.
5245 *** New procedure: sort! SEQUENCE LESS
5246 Returns its sorted result in the original boxes. No new storage is
5247 allocated at all. Proper usage: (set! slist (sort! slist <))
5249 *** New function: stable-sort SEQUENCE LESS?
5250 Similar to `sort' but stable. That is, if "equal" elements are
5251 ordered a < b in the original sequence, they will have the same order
5254 *** New function: stable-sort! SEQUENCE LESS?
5255 Similar to `sort!' but stable.
5256 Uses temporary storage when sorting vectors.
5258 *** New functions: sort-list, sort-list!
5259 Added for compatibility with scsh.
5261 ** New built-in random number support
5263 *** New function: random N [STATE]
5264 Accepts a positive integer or real N and returns a number of the
5265 same type between zero (inclusive) and N (exclusive). The values
5266 returned have a uniform distribution.
5268 The optional argument STATE must be of the type produced by
5269 `copy-random-state' or `seed->random-state'. It defaults to the value
5270 of the variable `*random-state*'. This object is used to maintain the
5271 state of the pseudo-random-number generator and is altered as a side
5272 effect of the `random' operation.
5274 *** New variable: *random-state*
5275 Holds a data structure that encodes the internal state of the
5276 random-number generator that `random' uses by default. The nature
5277 of this data structure is implementation-dependent. It may be
5278 printed out and successfully read back in, but may or may not
5279 function correctly as a random-number state object in another
5282 *** New function: copy-random-state [STATE]
5283 Returns a new object of type suitable for use as the value of the
5284 variable `*random-state*' and as a second argument to `random'.
5285 If argument STATE is given, a copy of it is returned. Otherwise a
5286 copy of `*random-state*' is returned.
5288 *** New function: seed->random-state SEED
5289 Returns a new object of type suitable for use as the value of the
5290 variable `*random-state*' and as a second argument to `random'.
5291 SEED is a string or a number. A new state is generated and
5292 initialized using SEED.
5294 *** New function: random:uniform [STATE]
5295 Returns an uniformly distributed inexact real random number in the
5296 range between 0 and 1.
5298 *** New procedure: random:solid-sphere! VECT [STATE]
5299 Fills VECT with inexact real random numbers the sum of whose
5300 squares is less than 1.0. Thinking of VECT as coordinates in
5301 space of dimension N = `(vector-length VECT)', the coordinates are
5302 uniformly distributed within the unit N-shere. The sum of the
5303 squares of the numbers is returned. VECT can be either a vector
5304 or a uniform vector of doubles.
5306 *** New procedure: random:hollow-sphere! VECT [STATE]
5307 Fills VECT with inexact real random numbers the sum of whose squares
5308 is equal to 1.0. Thinking of VECT as coordinates in space of
5309 dimension n = `(vector-length VECT)', the coordinates are uniformly
5310 distributed over the surface of the unit n-shere. VECT can be either
5311 a vector or a uniform vector of doubles.
5313 *** New function: random:normal [STATE]
5314 Returns an inexact real in a normal distribution with mean 0 and
5315 standard deviation 1. For a normal distribution with mean M and
5316 standard deviation D use `(+ M (* D (random:normal)))'.
5318 *** New procedure: random:normal-vector! VECT [STATE]
5319 Fills VECT with inexact real random numbers which are independent and
5320 standard normally distributed (i.e., with mean 0 and variance 1).
5321 VECT can be either a vector or a uniform vector of doubles.
5323 *** New function: random:exp STATE
5324 Returns an inexact real in an exponential distribution with mean 1.
5325 For an exponential distribution with mean U use (* U (random:exp)).
5327 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
5329 These functions now operate on numbers in the range of a C unsigned
5332 These functions used to operate on numbers in the range of a C signed
5333 long; however, this seems inappropriate, because Guile integers don't
5336 ** New function: make-guardian
5337 This is an implementation of guardians as described in
5338 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
5339 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
5340 Programming Language Design and Implementation, June 1993
5341 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
5343 ** New functions: delq1!, delv1!, delete1!
5344 These procedures behave similar to delq! and friends but delete only
5345 one object if at all.
5347 ** New function: unread-string STRING PORT
5348 Unread STRING to PORT, that is, push it back onto the port so that
5349 next read operation will work on the pushed back characters.
5351 ** unread-char can now be called multiple times
5352 If unread-char is called multiple times, the unread characters will be
5353 read again in last-in first-out order.
5355 ** the procedures uniform-array-read! and uniform-array-write! now
5356 work on any kind of port, not just ports which are open on a file.
5358 ** Now 'l' in a port mode requests line buffering.
5360 ** The procedure truncate-file now works on string ports as well
5361 as file ports. If the size argument is omitted, the current
5362 file position is used.
5364 ** new procedure: seek PORT/FDES OFFSET WHENCE
5365 The arguments are the same as for the old fseek procedure, but it
5366 works on string ports as well as random-access file ports.
5368 ** the fseek procedure now works on string ports, since it has been
5369 redefined using seek.
5371 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
5372 size is not supplied.
5374 ** the newline procedure no longer flushes the port if it's not
5375 line-buffered: previously it did if it was the current output port.
5377 ** open-pipe and close-pipe are no longer primitive procedures, but
5378 an emulation can be obtained using `(use-modules (ice-9 popen))'.
5380 ** the freopen procedure has been removed.
5382 ** new procedure: drain-input PORT
5383 Drains PORT's read buffers (including any pushed-back characters)
5384 and returns the contents as a single string.
5386 ** New function: map-in-order PROC LIST1 LIST2 ...
5387 Version of `map' which guarantees that the procedure is applied to the
5388 lists in serial order.
5390 ** Renamed `serial-array-copy!' and `serial-array-map!' to
5391 `array-copy-in-order!' and `array-map-in-order!'. The old names are
5392 now obsolete and will go away in release 1.5.
5394 ** New syntax: collect BODY1 ...
5395 Version of `begin' which returns a list of the results of the body
5396 forms instead of the result of the last body form. In contrast to
5397 `begin', `collect' allows an empty body.
5399 ** New functions: read-history FILENAME, write-history FILENAME
5400 Read/write command line history from/to file. Returns #t on success
5401 and #f if an error occured.
5403 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
5405 These procedures return a list of definitions available in the specified
5406 argument, a relative module reference. In the case of no argument,
5407 `(current-module)' is now consulted for definitions to return, instead
5408 of simply returning #f, the former behavior.
5410 ** The #/ syntax for lists is no longer supported.
5412 Earlier versions of Scheme accepted this syntax, but printed a
5415 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
5417 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
5420 * Changes to the gh_ interface
5424 Now takes a second argument which is the result array. If this
5425 pointer is NULL, a new array is malloced (the old behaviour).
5427 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
5428 gh_scm2shorts, gh_scm2longs, gh_scm2floats
5432 * Changes to the scm_ interface
5434 ** Function: scm_make_named_hook (char* name, int n_args)
5436 Creates a hook in the same way as make-hook above but also
5437 binds a variable named NAME to it.
5439 This is the typical way of creating a hook from C code.
5441 Currently, the variable is created in the "current" module. This
5442 might change when we get the new module system.
5444 ** The smob interface
5446 The interface for creating smobs has changed. For documentation, see
5447 data-rep.info (made from guile-core/doc/data-rep.texi).
5449 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
5451 >>> This function will be removed in 1.3.4. <<<
5455 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
5456 This function adds a new smob type, named NAME, with instance size
5457 SIZE to the system. The return value is a tag that is used in
5458 creating instances of the type. If SIZE is 0, then no memory will
5459 be allocated when instances of the smob are created, and nothing
5460 will be freed by the default free function.
5462 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
5463 This function sets the smob marking procedure for the smob type
5464 specified by the tag TC. TC is the tag returned by
5465 `scm_make_smob_type'.
5467 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
5468 This function sets the smob freeing procedure for the smob type
5469 specified by the tag TC. TC is the tag returned by
5470 `scm_make_smob_type'.
5472 *** Function: void scm_set_smob_print (tc, print)
5474 - Function: void scm_set_smob_print (long tc,
5475 scm_sizet (*print) (SCM,
5479 This function sets the smob printing procedure for the smob type
5480 specified by the tag TC. TC is the tag returned by
5481 `scm_make_smob_type'.
5483 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
5484 This function sets the smob equality-testing predicate for the
5485 smob type specified by the tag TC. TC is the tag returned by
5486 `scm_make_smob_type'.
5488 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
5489 Make VALUE contain a smob instance of the type with type code TC and
5490 smob data DATA. VALUE must be previously declared as C type `SCM'.
5492 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
5493 This macro expands to a block of code that creates a smob instance
5494 of the type with type code TC and smob data DATA, and returns that
5495 `SCM' value. It should be the last piece of code in a block.
5497 ** The interfaces for using I/O ports and implementing port types
5498 (ptobs) have changed significantly. The new interface is based on
5499 shared access to buffers and a new set of ptob procedures.
5501 *** scm_newptob has been removed
5505 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
5507 - Function: SCM scm_make_port_type (char *type_name,
5508 int (*fill_buffer) (SCM port),
5509 void (*write_flush) (SCM port));
5511 Similarly to the new smob interface, there is a set of function
5512 setters by which the user can customize the behaviour of his port
5513 type. See ports.h (scm_set_port_XXX).
5515 ** scm_strport_to_string: New function: creates a new string from
5516 a string port's buffer.
5518 ** Plug in interface for random number generators
5519 The variable `scm_the_rng' in random.c contains a value and three
5520 function pointers which together define the current random number
5521 generator being used by the Scheme level interface and the random
5522 number library functions.
5524 The user is free to replace the default generator with the generator
5527 *** Variable: size_t scm_the_rng.rstate_size
5528 The size of the random state type used by the current RNG
5531 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
5532 Given the random STATE, return 32 random bits.
5534 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
5535 Seed random state STATE using string S of length N.
5537 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
5538 Given random state STATE, return a malloced copy.
5541 The default RNG is the MWC (Multiply With Carry) random number
5542 generator described by George Marsaglia at the Department of
5543 Statistics and Supercomputer Computations Research Institute, The
5544 Florida State University (http://stat.fsu.edu/~geo).
5546 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
5547 passes all tests in the DIEHARD test suite
5548 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
5549 costs one multiply and one add on platforms which either supports long
5550 longs (gcc does this on most systems) or have 64 bit longs. The cost
5551 is four multiply on other systems but this can be optimized by writing
5552 scm_i_uniform32 in assembler.
5554 These functions are provided through the scm_the_rng interface for use
5555 by libguile and the application.
5557 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
5558 Given the random STATE, return 32 random bits.
5559 Don't use this function directly. Instead go through the plugin
5560 interface (see "Plug in interface" above).
5562 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
5563 Initialize STATE using SEED of length N.
5565 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
5566 Return a malloc:ed copy of STATE. This function can easily be re-used
5567 in the interfaces to other RNGs.
5569 ** Random number library functions
5570 These functions use the current RNG through the scm_the_rng interface.
5571 It might be a good idea to use these functions from your C code so
5572 that only one random generator is used by all code in your program.
5574 The default random state is stored in:
5576 *** Variable: SCM scm_var_random_state
5577 Contains the vcell of the Scheme variable "*random-state*" which is
5578 used as default state by all random number functions in the Scheme
5583 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
5585 *** Function: scm_rstate *scm_c_default_rstate (void)
5586 This is a convenience function which returns the value of
5587 scm_var_random_state. An error message is generated if this value
5588 isn't a random state.
5590 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
5591 Make a new random state from the string SEED of length LENGTH.
5593 It is generally not a good idea to use multiple random states in a
5594 program. While subsequent random numbers generated from one random
5595 state are guaranteed to be reasonably independent, there is no such
5596 guarantee for numbers generated from different random states.
5598 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
5599 Return 32 random bits.
5601 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
5602 Return a sample from the uniform(0,1) distribution.
5604 *** Function: double scm_c_normal01 (scm_rstate *STATE)
5605 Return a sample from the normal(0,1) distribution.
5607 *** Function: double scm_c_exp1 (scm_rstate *STATE)
5608 Return a sample from the exp(1) distribution.
5610 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
5611 Return a sample from the discrete uniform(0,M) distribution.
5613 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
5614 Return a sample from the discrete uniform(0,M) distribution.
5615 M must be a bignum object. The returned value may be an INUM.
5619 Changes in Guile 1.3 (released Monday, October 19, 1998):
5621 * Changes to the distribution
5623 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
5624 To avoid conflicts, programs should name environment variables after
5625 themselves, except when there's a common practice establishing some
5628 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5629 giving the former precedence, and printing a warning message if the
5630 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5632 ** The header files related to multi-byte characters have been removed.
5633 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5634 which referred to these explicitly will probably need to be rewritten,
5635 since the support for the variant string types has been removed; see
5638 ** The header files append.h and sequences.h have been removed. These
5639 files implemented non-R4RS operations which would encourage
5640 non-portable programming style and less easy-to-read code.
5642 * Changes to the stand-alone interpreter
5644 ** New procedures have been added to implement a "batch mode":
5646 *** Function: batch-mode?
5648 Returns a boolean indicating whether the interpreter is in batch
5651 *** Function: set-batch-mode?! ARG
5653 If ARG is true, switches the interpreter to batch mode. The `#f'
5654 case has not been implemented.
5656 ** Guile now provides full command-line editing, when run interactively.
5657 To use this feature, you must have the readline library installed.
5658 The Guile build process will notice it, and automatically include
5661 The readline library is available via anonymous FTP from any GNU
5662 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5664 ** the-last-stack is now a fluid.
5666 * Changes to the procedure for linking libguile with your programs
5668 ** You can now use the `guile-config' utility to build programs that use Guile.
5670 Guile now includes a command-line utility called `guile-config', which
5671 can provide information about how to compile and link programs that
5674 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5675 You should include this command's output on the command line you use
5676 to compile C or C++ code that #includes the Guile header files. It's
5677 usually just a `-I' flag to help the compiler find the Guile headers.
5680 *** `guile-config link' prints any linker flags necessary to link with Guile.
5682 This command writes to its standard output a list of flags which you
5683 must pass to the linker to link your code against the Guile library.
5684 The flags include '-lguile' itself, any other libraries the Guile
5685 library depends upon, and any `-L' flags needed to help the linker
5686 find those libraries.
5688 For example, here is a Makefile rule that builds a program named 'foo'
5689 from the object files ${FOO_OBJECTS}, and links them against Guile:
5692 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5694 Previous Guile releases recommended that you use autoconf to detect
5695 which of a predefined set of libraries were present on your system.
5696 It is more robust to use `guile-config', since it records exactly which
5697 libraries the installed Guile library requires.
5699 This was originally called `build-guile', but was renamed to
5700 `guile-config' before Guile 1.3 was released, to be consistent with
5701 the analogous script for the GTK+ GUI toolkit, which is called
5705 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5707 If you are using the GNU autoconf package to configure your program,
5708 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5709 (described above) and gather the necessary values for use in your
5712 The GUILE_FLAGS macro expands to configure script code which runs the
5713 `guile-config' script, to find out where Guile's header files and
5714 libraries are installed. It sets two variables, marked for
5715 substitution, as by AC_SUBST.
5717 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5718 code that uses Guile header files. This is almost always just a
5721 GUILE_LDFLAGS --- flags to pass to the linker to link a
5722 program against Guile. This includes `-lguile' for the Guile
5723 library itself, any libraries that Guile itself requires (like
5724 -lqthreads), and so on. It may also include a -L flag to tell the
5725 compiler where to find the libraries.
5727 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5728 directory of the Guile distribution. You can copy it into your
5729 package's aclocal.m4 file, and then use it in your configure.in file.
5731 If you are using the `aclocal' program, distributed with GNU automake,
5732 to maintain your aclocal.m4 file, the Guile installation process
5733 installs guile.m4 where aclocal will find it. All you need to do is
5734 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5735 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5739 * Changes to Scheme functions and syntax
5741 ** Multi-byte strings have been removed, as have multi-byte and wide
5742 ports. We felt that these were the wrong approach to
5743 internationalization support.
5745 ** New function: readline [PROMPT]
5746 Read a line from the terminal, and allow the user to edit it,
5747 prompting with PROMPT. READLINE provides a large set of Emacs-like
5748 editing commands, lets the user recall previously typed lines, and
5749 works on almost every kind of terminal, including dumb terminals.
5751 READLINE assumes that the cursor is at the beginning of the line when
5752 it is invoked. Thus, you can't print a prompt yourself, and then call
5753 READLINE; you need to package up your prompt as a string, pass it to
5754 the function, and let READLINE print the prompt itself. This is
5755 because READLINE needs to know the prompt's screen width.
5757 For Guile to provide this function, you must have the readline
5758 library, version 2.1 or later, installed on your system. Readline is
5759 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5760 any GNU mirror site.
5762 See also ADD-HISTORY function.
5764 ** New function: add-history STRING
5765 Add STRING as the most recent line in the history used by the READLINE
5766 command. READLINE does not add lines to the history itself; you must
5767 call ADD-HISTORY to make previous input available to the user.
5769 ** The behavior of the read-line function has changed.
5771 This function now uses standard C library functions to read the line,
5772 for speed. This means that it doesn not respect the value of
5773 scm-line-incrementors; it assumes that lines are delimited with
5776 (Note that this is read-line, the function that reads a line of text
5777 from a port, not readline, the function that reads a line from a
5778 terminal, providing full editing capabilities.)
5780 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5782 This module provides some simple argument parsing. It exports one
5785 Function: getopt-gnu-style ARG-LS
5786 Parse a list of program arguments into an alist of option
5789 Each item in the list of program arguments is examined to see if
5790 it meets the syntax of a GNU long-named option. An argument like
5791 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5792 returned alist, where MUMBLE is a keyword object with the same
5793 name as the argument. An argument like `--MUMBLE=FROB' produces
5794 an element of the form (MUMBLE . FROB), where FROB is a string.
5796 As a special case, the returned alist also contains a pair whose
5797 car is the symbol `rest'. The cdr of this pair is a list
5798 containing all the items in the argument list that are not options
5799 of the form mentioned above.
5801 The argument `--' is treated specially: all items in the argument
5802 list appearing after such an argument are not examined, and are
5803 returned in the special `rest' list.
5805 This function does not parse normal single-character switches.
5806 You will need to parse them out of the `rest' list yourself.
5808 ** The read syntax for byte vectors and short vectors has changed.
5810 Instead of #bytes(...), write #y(...).
5812 Instead of #short(...), write #h(...).
5814 This may seem nutty, but, like the other uniform vectors, byte vectors
5815 and short vectors want to have the same print and read syntax (and,
5816 more basic, want to have read syntax!). Changing the read syntax to
5817 use multiple characters after the hash sign breaks with the
5818 conventions used in R5RS and the conventions used for the other
5819 uniform vectors. It also introduces complexity in the current reader,
5820 both on the C and Scheme levels. (The Right solution is probably to
5821 change the syntax and prototypes for uniform vectors entirely.)
5824 ** The new module (ice-9 session) provides useful interactive functions.
5826 *** New procedure: (apropos REGEXP OPTION ...)
5828 Display a list of top-level variables whose names match REGEXP, and
5829 the modules they are imported from. Each OPTION should be one of the
5832 value --- Show the value of each matching variable.
5833 shadow --- Show bindings shadowed by subsequently imported modules.
5834 full --- Same as both `shadow' and `value'.
5838 guile> (apropos "trace" 'full)
5839 debug: trace #<procedure trace args>
5840 debug: untrace #<procedure untrace args>
5841 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5842 the-scm-module: before-backtrace-hook ()
5843 the-scm-module: backtrace #<primitive-procedure backtrace>
5844 the-scm-module: after-backtrace-hook ()
5845 the-scm-module: has-shown-backtrace-hint? #f
5848 ** There are new functions and syntax for working with macros.
5850 Guile implements macros as a special object type. Any variable whose
5851 top-level binding is a macro object acts as a macro. The macro object
5852 specifies how the expression should be transformed before evaluation.
5854 *** Macro objects now print in a reasonable way, resembling procedures.
5856 *** New function: (macro? OBJ)
5857 True iff OBJ is a macro object.
5859 *** New function: (primitive-macro? OBJ)
5860 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5861 macro transformers, implemented in eval.c rather than Scheme code.
5863 Why do we have this function?
5864 - For symmetry with procedure? and primitive-procedure?,
5865 - to allow custom print procedures to tell whether a macro is
5866 primitive, and display it differently, and
5867 - to allow compilers and user-written evaluators to distinguish
5868 builtin special forms from user-defined ones, which could be
5871 *** New function: (macro-type OBJ)
5872 Return a value indicating what kind of macro OBJ is. Possible return
5875 The symbol `syntax' --- a macro created by procedure->syntax.
5876 The symbol `macro' --- a macro created by procedure->macro.
5877 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5878 The boolean #f --- if OBJ is not a macro object.
5880 *** New function: (macro-name MACRO)
5881 Return the name of the macro object MACRO's procedure, as returned by
5884 *** New function: (macro-transformer MACRO)
5885 Return the transformer procedure for MACRO.
5887 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5889 Specify a new macro expander to use in the current module. Each
5890 MODULE is a module name, with the same meaning as in the `use-modules'
5891 form; each named module's exported bindings are added to the current
5892 top-level environment. TRANSFORMER is an expression evaluated in the
5893 resulting environment which must yield a procedure to use as the
5894 module's eval transformer: every expression evaluated in this module
5895 is passed to this function, and the result passed to the Guile
5898 *** macro-eval! is removed. Use local-eval instead.
5900 ** Some magic has been added to the printer to better handle user
5901 written printing routines (like record printers, closure printers).
5903 The problem is that these user written routines must have access to
5904 the current `print-state' to be able to handle fancy things like
5905 detection of circular references. These print-states have to be
5906 passed to the builtin printing routines (display, write, etc) to
5907 properly continue the print chain.
5909 We didn't want to change all existing print code so that it
5910 explicitly passes thru a print state in addition to a port. Instead,
5911 we extented the possible values that the builtin printing routines
5912 accept as a `port'. In addition to a normal port, they now also take
5913 a pair of a normal port and a print-state. Printing will go to the
5914 port and the print-state will be used to control the detection of
5915 circular references, etc. If the builtin function does not care for a
5916 print-state, it is simply ignored.
5918 User written callbacks are now called with such a pair as their
5919 `port', but because every function now accepts this pair as a PORT
5920 argument, you don't have to worry about that. In fact, it is probably
5921 safest to not check for these pairs.
5923 However, it is sometimes necessary to continue a print chain on a
5924 different port, for example to get a intermediate string
5925 representation of the printed value, mangle that string somehow, and
5926 then to finally print the mangled string. Use the new function
5928 inherit-print-state OLD-PORT NEW-PORT
5930 for this. It constructs a new `port' that prints to NEW-PORT but
5931 inherits the print-state of OLD-PORT.
5933 ** struct-vtable-offset renamed to vtable-offset-user
5935 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5937 ** There is now a third optional argument to make-vtable-vtable
5938 (and fourth to make-struct) when constructing new types (vtables).
5939 This argument initializes field vtable-index-printer of the vtable.
5941 ** The detection of circular references has been extended to structs.
5942 That is, a structure that -- in the process of being printed -- prints
5943 itself does not lead to infinite recursion.
5945 ** There is now some basic support for fluids. Please read
5946 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5947 the following functions and macros:
5949 Function: make-fluid
5951 Create a new fluid object. Fluids are not special variables or
5952 some other extension to the semantics of Scheme, but rather
5953 ordinary Scheme objects. You can store them into variables (that
5954 are still lexically scoped, of course) or into any other place you
5955 like. Every fluid has a initial value of `#f'.
5957 Function: fluid? OBJ
5959 Test whether OBJ is a fluid.
5961 Function: fluid-ref FLUID
5962 Function: fluid-set! FLUID VAL
5964 Access/modify the fluid FLUID. Modifications are only visible
5965 within the current dynamic root (that includes threads).
5967 Function: with-fluids* FLUIDS VALUES THUNK
5969 FLUIDS is a list of fluids and VALUES a corresponding list of
5970 values for these fluids. Before THUNK gets called the values are
5971 installed in the fluids and the old values of the fluids are
5972 saved in the VALUES list. When the flow of control leaves THUNK
5973 or reenters it, the values get swapped again. You might think of
5974 this as a `safe-fluid-excursion'. Note that the VALUES list is
5975 modified by `with-fluids*'.
5977 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5979 The same as `with-fluids*' but with a different syntax. It looks
5980 just like `let', but both FLUID and VALUE are evaluated. Remember,
5981 fluids are not special variables but ordinary objects. FLUID
5982 should evaluate to a fluid.
5984 ** Changes to system call interfaces:
5986 *** close-port, close-input-port and close-output-port now return a
5987 boolean instead of an `unspecified' object. #t means that the port
5988 was successfully closed, while #f means it was already closed. It is
5989 also now possible for these procedures to raise an exception if an
5990 error occurs (some errors from write can be delayed until close.)
5992 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5995 *** the third argument to fcntl is now optional.
5997 *** the first argument to chown can now be a file descriptor or a port.
5999 *** the argument to stat can now be a port.
6001 *** The following new procedures have been added (most use scsh
6004 *** procedure: close PORT/FD
6005 Similar to close-port (*note close-port: Closing Ports.), but also
6006 works on file descriptors. A side effect of closing a file
6007 descriptor is that any ports using that file descriptor are moved
6008 to a different file descriptor and have their revealed counts set
6011 *** procedure: port->fdes PORT
6012 Returns the integer file descriptor underlying PORT. As a side
6013 effect the revealed count of PORT is incremented.
6015 *** procedure: fdes->ports FDES
6016 Returns a list of existing ports which have FDES as an underlying
6017 file descriptor, without changing their revealed counts.
6019 *** procedure: fdes->inport FDES
6020 Returns an existing input port which has FDES as its underlying
6021 file descriptor, if one exists, and increments its revealed count.
6022 Otherwise, returns a new input port with a revealed count of 1.
6024 *** procedure: fdes->outport FDES
6025 Returns an existing output port which has FDES as its underlying
6026 file descriptor, if one exists, and increments its revealed count.
6027 Otherwise, returns a new output port with a revealed count of 1.
6029 The next group of procedures perform a `dup2' system call, if NEWFD
6030 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
6031 duplicated can be supplied as an integer or contained in a port. The
6032 type of value returned varies depending on which procedure is used.
6034 All procedures also have the side effect when performing `dup2' that
6035 any ports using NEWFD are moved to a different file descriptor and have
6036 their revealed counts set to zero.
6038 *** procedure: dup->fdes PORT/FD [NEWFD]
6039 Returns an integer file descriptor.
6041 *** procedure: dup->inport PORT/FD [NEWFD]
6042 Returns a new input port using the new file descriptor.
6044 *** procedure: dup->outport PORT/FD [NEWFD]
6045 Returns a new output port using the new file descriptor.
6047 *** procedure: dup PORT/FD [NEWFD]
6048 Returns a new port if PORT/FD is a port, with the same mode as the
6049 supplied port, otherwise returns an integer file descriptor.
6051 *** procedure: dup->port PORT/FD MODE [NEWFD]
6052 Returns a new port using the new file descriptor. MODE supplies a
6053 mode string for the port (*note open-file: File Ports.).
6055 *** procedure: setenv NAME VALUE
6056 Modifies the environment of the current process, which is also the
6057 default environment inherited by child processes.
6059 If VALUE is `#f', then NAME is removed from the environment.
6060 Otherwise, the string NAME=VALUE is added to the environment,
6061 replacing any existing string with name matching NAME.
6063 The return value is unspecified.
6065 *** procedure: truncate-file OBJ SIZE
6066 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
6067 can be a string containing a file name or an integer file
6068 descriptor or port open for output on the file. The underlying
6069 system calls are `truncate' and `ftruncate'.
6071 The return value is unspecified.
6073 *** procedure: setvbuf PORT MODE [SIZE]
6074 Set the buffering mode for PORT. MODE can be:
6082 block buffered, using a newly allocated buffer of SIZE bytes.
6083 However if SIZE is zero or unspecified, the port will be made
6086 This procedure should not be used after I/O has been performed with
6089 Ports are usually block buffered by default, with a default buffer
6090 size. Procedures e.g., *Note open-file: File Ports, which accept a
6091 mode string allow `0' to be added to request an unbuffered port.
6093 *** procedure: fsync PORT/FD
6094 Copies any unwritten data for the specified output file descriptor
6095 to disk. If PORT/FD is a port, its buffer is flushed before the
6096 underlying file descriptor is fsync'd. The return value is
6099 *** procedure: open-fdes PATH FLAGS [MODES]
6100 Similar to `open' but returns a file descriptor instead of a port.
6102 *** procedure: execle PATH ENV [ARG] ...
6103 Similar to `execl', but the environment of the new process is
6104 specified by ENV, which must be a list of strings as returned by
6105 the `environ' procedure.
6107 This procedure is currently implemented using the `execve' system
6108 call, but we call it `execle' because of its Scheme calling
6111 *** procedure: strerror ERRNO
6112 Returns the Unix error message corresponding to ERRNO, an integer.
6114 *** procedure: primitive-exit [STATUS]
6115 Terminate the current process without unwinding the Scheme stack.
6116 This is would typically be useful after a fork. The exit status
6117 is STATUS if supplied, otherwise zero.
6119 *** procedure: times
6120 Returns an object with information about real and processor time.
6121 The following procedures accept such an object as an argument and
6122 return a selected component:
6125 The current real time, expressed as time units relative to an
6129 The CPU time units used by the calling process.
6132 The CPU time units used by the system on behalf of the
6136 The CPU time units used by terminated child processes of the
6137 calling process, whose status has been collected (e.g., using
6141 Similarly, the CPU times units used by the system on behalf of
6142 terminated child processes.
6144 ** Removed: list-length
6145 ** Removed: list-append, list-append!
6146 ** Removed: list-reverse, list-reverse!
6148 ** array-map renamed to array-map!
6150 ** serial-array-map renamed to serial-array-map!
6152 ** catch doesn't take #f as first argument any longer
6154 Previously, it was possible to pass #f instead of a key to `catch'.
6155 That would cause `catch' to pass a jump buffer object to the procedure
6156 passed as second argument. The procedure could then use this jump
6157 buffer objekt as an argument to throw.
6159 This mechanism has been removed since its utility doesn't motivate the
6160 extra complexity it introduces.
6162 ** The `#/' notation for lists now provokes a warning message from Guile.
6163 This syntax will be removed from Guile in the near future.
6165 To disable the warning message, set the GUILE_HUSH environment
6166 variable to any non-empty value.
6168 ** The newline character now prints as `#\newline', following the
6169 normal Scheme notation, not `#\nl'.
6171 * Changes to the gh_ interface
6173 ** The gh_enter function now takes care of loading the Guile startup files.
6174 gh_enter works by calling scm_boot_guile; see the remarks below.
6176 ** Function: void gh_write (SCM x)
6178 Write the printed representation of the scheme object x to the current
6179 output port. Corresponds to the scheme level `write'.
6181 ** gh_list_length renamed to gh_length.
6183 ** vector handling routines
6185 Several major changes. In particular, gh_vector() now resembles
6186 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
6187 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
6188 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
6189 vector-related gh_ functions have been implemented.
6191 ** pair and list routines
6193 Implemented several of the R4RS pair and list functions that were
6196 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
6198 New function. Converts double arrays back and forth between Scheme
6201 * Changes to the scm_ interface
6203 ** The function scm_boot_guile now takes care of loading the startup files.
6205 Guile's primary initialization function, scm_boot_guile, now takes
6206 care of loading `boot-9.scm', in the `ice-9' module, to initialize
6207 Guile, define the module system, and put together some standard
6208 bindings. It also loads `init.scm', which is intended to hold
6209 site-specific initialization code.
6211 Since Guile cannot operate properly until boot-9.scm is loaded, there
6212 is no reason to separate loading boot-9.scm from Guile's other
6213 initialization processes.
6215 This job used to be done by scm_compile_shell_switches, which didn't
6216 make much sense; in particular, it meant that people using Guile for
6217 non-shell-like applications had to jump through hoops to get Guile
6218 initialized properly.
6220 ** The function scm_compile_shell_switches no longer loads the startup files.
6221 Now, Guile always loads the startup files, whenever it is initialized;
6222 see the notes above for scm_boot_guile and scm_load_startup_files.
6224 ** Function: scm_load_startup_files
6225 This new function takes care of loading Guile's initialization file
6226 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
6227 this is always called by the Guile initialization process, it's
6228 probably not too useful to call this yourself, but it's there anyway.
6230 ** The semantics of smob marking have changed slightly.
6232 The smob marking function (the `mark' member of the scm_smobfuns
6233 structure) is no longer responsible for setting the mark bit on the
6234 smob. The generic smob handling code in the garbage collector will
6235 set this bit. The mark function need only ensure that any other
6236 objects the smob refers to get marked.
6238 Note that this change means that the smob's GC8MARK bit is typically
6239 already set upon entry to the mark function. Thus, marking functions
6240 which look like this:
6243 if (SCM_GC8MARKP (ptr))
6245 SCM_SETGC8MARK (ptr);
6246 ... mark objects to which the smob refers ...
6249 are now incorrect, since they will return early, and fail to mark any
6250 other objects the smob refers to. Some code in the Guile library used
6253 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
6255 If you have implemented your own I/O port type, by writing the
6256 functions required by the scm_ptobfuns and then calling scm_newptob,
6257 you will need to change your functions slightly.
6259 The functions in a scm_ptobfuns structure now expect the port itself
6260 as their argument; they used to expect the `stream' member of the
6261 port's scm_port_table structure. This allows functions in an
6262 scm_ptobfuns structure to easily access the port's cell (and any flags
6263 it its CAR), and the port's scm_port_table structure.
6265 Guile now passes the I/O port itself as the `port' argument in the
6266 following scm_ptobfuns functions:
6268 int (*free) (SCM port);
6269 int (*fputc) (int, SCM port);
6270 int (*fputs) (char *, SCM port);
6271 scm_sizet (*fwrite) SCM_P ((char *ptr,
6275 int (*fflush) (SCM port);
6276 int (*fgetc) (SCM port);
6277 int (*fclose) (SCM port);
6279 The interfaces to the `mark', `print', `equalp', and `fgets' methods
6282 If you have existing code which defines its own port types, it is easy
6283 to convert your code to the new interface; simply apply SCM_STREAM to
6284 the port argument to yield the value you code used to expect.
6286 Note that since both the port and the stream have the same type in the
6287 C code --- they are both SCM values --- the C compiler will not remind
6288 you if you forget to update your scm_ptobfuns functions.
6291 ** Function: int scm_internal_select (int fds,
6295 struct timeval *timeout);
6297 This is a replacement for the `select' function provided by the OS.
6298 It enables I/O blocking and sleeping to happen for one cooperative
6299 thread without blocking other threads. It also avoids busy-loops in
6300 these situations. It is intended that all I/O blocking and sleeping
6301 will finally go through this function. Currently, this function is
6302 only available on systems providing `gettimeofday' and `select'.
6304 ** Function: SCM scm_internal_stack_catch (SCM tag,
6305 scm_catch_body_t body,
6307 scm_catch_handler_t handler,
6310 A new sibling to the other two C level `catch' functions
6311 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
6312 the stack to be saved automatically into the variable `the-last-stack'
6313 (scm_the_last_stack_var) on error. This is necessary if you want to
6314 use advanced error reporting, such as calling scm_display_error and
6315 scm_display_backtrace. (They both take a stack object as argument.)
6317 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
6319 scm_catch_handler_t handler,
6322 Spawns a new thread. It does a job similar to
6323 scm_call_with_new_thread but takes arguments more suitable when
6324 spawning threads from application C code.
6326 ** The hook scm_error_callback has been removed. It was originally
6327 intended as a way for the user to install his own error handler. But
6328 that method works badly since it intervenes between throw and catch,
6329 thereby changing the semantics of expressions like (catch #t ...).
6330 The correct way to do it is to use one of the C level catch functions
6331 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
6333 ** Removed functions:
6335 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
6336 scm_list_reverse, scm_list_reverse_x
6338 ** New macros: SCM_LISTn where n is one of the integers 0-9.
6340 These can be used for pretty list creation from C. The idea is taken
6341 from Erick Gallesio's STk.
6343 ** scm_array_map renamed to scm_array_map_x
6345 ** mbstrings are now removed
6347 This means that the type codes scm_tc7_mb_string and
6348 scm_tc7_mb_substring has been removed.
6350 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
6352 Since we no longer support multi-byte strings, these I/O functions
6353 have been simplified, and renamed. Here are their old names, and
6354 their new names and arguments:
6356 scm_gen_putc -> void scm_putc (int c, SCM port);
6357 scm_gen_puts -> void scm_puts (char *s, SCM port);
6358 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
6359 scm_gen_getc -> void scm_getc (SCM port);
6362 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
6364 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
6366 SCM_TYP7S now masks away the bit which distinguishes substrings from
6369 ** scm_catch_body_t: Backward incompatible change!
6371 Body functions to scm_internal_catch and friends do not any longer
6372 take a second argument. This is because it is no longer possible to
6373 pass a #f arg to catch.
6375 ** Calls to scm_protect_object and scm_unprotect now nest properly.
6377 The function scm_protect_object protects its argument from being freed
6378 by the garbage collector. scm_unprotect_object removes that
6381 These functions now nest properly. That is, for every object O, there
6382 is a counter which scm_protect_object(O) increments and
6383 scm_unprotect_object(O) decrements, if the counter is greater than
6384 zero. Every object's counter is zero when it is first created. If an
6385 object's counter is greater than zero, the garbage collector will not
6386 reclaim its storage.
6388 This allows you to use scm_protect_object in your code without
6389 worrying that some other function you call will call
6390 scm_unprotect_object, and allow it to be freed. Assuming that the
6391 functions you call are well-behaved, and unprotect only those objects
6392 they protect, you can follow the same rule and have confidence that
6393 objects will be freed only at appropriate times.
6396 Changes in Guile 1.2 (released Tuesday, June 24 1997):
6398 * Changes to the distribution
6400 ** Nightly snapshots are now available from ftp.red-bean.com.
6401 The old server, ftp.cyclic.com, has been relinquished to its rightful
6404 Nightly snapshots of the Guile development sources are now available via
6405 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
6407 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
6408 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
6410 ** To run Guile without installing it, the procedure has changed a bit.
6412 If you used a separate build directory to compile Guile, you'll need
6413 to include the build directory in SCHEME_LOAD_PATH, as well as the
6414 source directory. See the `INSTALL' file for examples.
6416 * Changes to the procedure for linking libguile with your programs
6418 ** The standard Guile load path for Scheme code now includes
6419 $(datadir)/guile (usually /usr/local/share/guile). This means that
6420 you can install your own Scheme files there, and Guile will find them.
6421 (Previous versions of Guile only checked a directory whose name
6422 contained the Guile version number, so you had to re-install or move
6423 your Scheme sources each time you installed a fresh version of Guile.)
6425 The load path also includes $(datadir)/guile/site; we recommend
6426 putting individual Scheme files there. If you want to install a
6427 package with multiple source files, create a directory for them under
6430 ** Guile 1.2 will now use the Rx regular expression library, if it is
6431 installed on your system. When you are linking libguile into your own
6432 programs, this means you will have to link against -lguile, -lqt (if
6433 you configured Guile with thread support), and -lrx.
6435 If you are using autoconf to generate configuration scripts for your
6436 application, the following lines should suffice to add the appropriate
6437 libraries to your link command:
6439 ### Find Rx, quickthreads and libguile.
6440 AC_CHECK_LIB(rx, main)
6441 AC_CHECK_LIB(qt, main)
6442 AC_CHECK_LIB(guile, scm_shell)
6444 The Guile 1.2 distribution does not contain sources for the Rx
6445 library, as Guile 1.0 did. If you want to use Rx, you'll need to
6446 retrieve it from a GNU FTP site and install it separately.
6448 * Changes to Scheme functions and syntax
6450 ** The dynamic linking features of Guile are now enabled by default.
6451 You can disable them by giving the `--disable-dynamic-linking' option
6454 (dynamic-link FILENAME)
6456 Find the object file denoted by FILENAME (a string) and link it
6457 into the running Guile application. When everything works out,
6458 return a Scheme object suitable for representing the linked object
6459 file. Otherwise an error is thrown. How object files are
6460 searched is system dependent.
6462 (dynamic-object? VAL)
6464 Determine whether VAL represents a dynamically linked object file.
6466 (dynamic-unlink DYNOBJ)
6468 Unlink the indicated object file from the application. DYNOBJ
6469 should be one of the values returned by `dynamic-link'.
6471 (dynamic-func FUNCTION DYNOBJ)
6473 Search the C function indicated by FUNCTION (a string or symbol)
6474 in DYNOBJ and return some Scheme object that can later be used
6475 with `dynamic-call' to actually call this function. Right now,
6476 these Scheme objects are formed by casting the address of the
6477 function to `long' and converting this number to its Scheme
6480 (dynamic-call FUNCTION DYNOBJ)
6482 Call the C function indicated by FUNCTION and DYNOBJ. The
6483 function is passed no arguments and its return value is ignored.
6484 When FUNCTION is something returned by `dynamic-func', call that
6485 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
6486 etc.), look it up in DYNOBJ; this is equivalent to
6488 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
6490 Interrupts are deferred while the C function is executing (with
6491 SCM_DEFER_INTS/SCM_ALLOW_INTS).
6493 (dynamic-args-call FUNCTION DYNOBJ ARGS)
6495 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
6496 some arguments and return its return value. The C function is
6497 expected to take two arguments and return an `int', just like
6500 int c_func (int argc, char **argv);
6502 ARGS must be a list of strings and is converted into an array of
6503 `char *'. The array is passed in ARGV and its size in ARGC. The
6504 return value is converted to a Scheme number and returned from the
6505 call to `dynamic-args-call'.
6507 When dynamic linking is disabled or not supported on your system,
6508 the above functions throw errors, but they are still available.
6510 Here is a small example that works on GNU/Linux:
6512 (define libc-obj (dynamic-link "libc.so"))
6513 (dynamic-args-call 'rand libc-obj '())
6515 See the file `libguile/DYNAMIC-LINKING' for additional comments.
6517 ** The #/ syntax for module names is depreciated, and will be removed
6518 in a future version of Guile. Instead of
6526 The latter syntax is more consistent with existing Lisp practice.
6528 ** Guile now does fancier printing of structures. Structures are the
6529 underlying implementation for records, which in turn are used to
6530 implement modules, so all of these object now print differently and in
6531 a more informative way.
6533 The Scheme printer will examine the builtin variable *struct-printer*
6534 whenever it needs to print a structure object. When this variable is
6535 not `#f' it is deemed to be a procedure and will be applied to the
6536 structure object and the output port. When *struct-printer* is `#f'
6537 or the procedure return `#f' the structure object will be printed in
6538 the boring #<struct 80458270> form.
6540 This hook is used by some routines in ice-9/boot-9.scm to implement
6541 type specific printing routines. Please read the comments there about
6544 One of the more specific uses of structs are records. The printing
6545 procedure that could be passed to MAKE-RECORD-TYPE is now actually
6546 called. It should behave like a *struct-printer* procedure (described
6549 ** Guile now supports a new R4RS-compliant syntax for keywords. A
6550 token of the form #:NAME, where NAME has the same syntax as a Scheme
6551 symbol, is the external representation of the keyword named NAME.
6552 Keyword objects print using this syntax as well, so values containing
6553 keyword objects can be read back into Guile. When used in an
6554 expression, keywords are self-quoting objects.
6556 Guile suports this read syntax, and uses this print syntax, regardless
6557 of the current setting of the `keyword' read option. The `keyword'
6558 read option only controls whether Guile recognizes the `:NAME' syntax,
6559 which is incompatible with R4RS. (R4RS says such token represent
6562 ** Guile has regular expression support again. Guile 1.0 included
6563 functions for matching regular expressions, based on the Rx library.
6564 In Guile 1.1, the Guile/Rx interface was removed to simplify the
6565 distribution, and thus Guile had no regular expression support. Guile
6566 1.2 again supports the most commonly used functions, and supports all
6567 of SCSH's regular expression functions.
6569 If your system does not include a POSIX regular expression library,
6570 and you have not linked Guile with a third-party regexp library such as
6571 Rx, these functions will not be available. You can tell whether your
6572 Guile installation includes regular expression support by checking
6573 whether the `*features*' list includes the `regex' symbol.
6575 *** regexp functions
6577 By default, Guile supports POSIX extended regular expressions. That
6578 means that the characters `(', `)', `+' and `?' are special, and must
6579 be escaped if you wish to match the literal characters.
6581 This regular expression interface was modeled after that implemented
6582 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
6583 with SCSH regular expressions.
6585 **** Function: string-match PATTERN STR [START]
6586 Compile the string PATTERN into a regular expression and compare
6587 it with STR. The optional numeric argument START specifies the
6588 position of STR at which to begin matching.
6590 `string-match' returns a "match structure" which describes what,
6591 if anything, was matched by the regular expression. *Note Match
6592 Structures::. If STR does not match PATTERN at all,
6593 `string-match' returns `#f'.
6595 Each time `string-match' is called, it must compile its PATTERN
6596 argument into a regular expression structure. This operation is
6597 expensive, which makes `string-match' inefficient if the same regular
6598 expression is used several times (for example, in a loop). For better
6599 performance, you can compile a regular expression in advance and then
6600 match strings against the compiled regexp.
6602 **** Function: make-regexp STR [FLAGS]
6603 Compile the regular expression described by STR, and return the
6604 compiled regexp structure. If STR does not describe a legal
6605 regular expression, `make-regexp' throws a
6606 `regular-expression-syntax' error.
6608 FLAGS may be the bitwise-or of one or more of the following:
6610 **** Constant: regexp/extended
6611 Use POSIX Extended Regular Expression syntax when interpreting
6612 STR. If not set, POSIX Basic Regular Expression syntax is used.
6613 If the FLAGS argument is omitted, we assume regexp/extended.
6615 **** Constant: regexp/icase
6616 Do not differentiate case. Subsequent searches using the
6617 returned regular expression will be case insensitive.
6619 **** Constant: regexp/newline
6620 Match-any-character operators don't match a newline.
6622 A non-matching list ([^...]) not containing a newline matches a
6625 Match-beginning-of-line operator (^) matches the empty string
6626 immediately after a newline, regardless of whether the FLAGS
6627 passed to regexp-exec contain regexp/notbol.
6629 Match-end-of-line operator ($) matches the empty string
6630 immediately before a newline, regardless of whether the FLAGS
6631 passed to regexp-exec contain regexp/noteol.
6633 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6634 Match the compiled regular expression REGEXP against `str'. If
6635 the optional integer START argument is provided, begin matching
6636 from that position in the string. Return a match structure
6637 describing the results of the match, or `#f' if no match could be
6640 FLAGS may be the bitwise-or of one or more of the following:
6642 **** Constant: regexp/notbol
6643 The match-beginning-of-line operator always fails to match (but
6644 see the compilation flag regexp/newline above) This flag may be
6645 used when different portions of a string are passed to
6646 regexp-exec and the beginning of the string should not be
6647 interpreted as the beginning of the line.
6649 **** Constant: regexp/noteol
6650 The match-end-of-line operator always fails to match (but see the
6651 compilation flag regexp/newline above)
6653 **** Function: regexp? OBJ
6654 Return `#t' if OBJ is a compiled regular expression, or `#f'
6657 Regular expressions are commonly used to find patterns in one string
6658 and replace them with the contents of another string.
6660 **** Function: regexp-substitute PORT MATCH [ITEM...]
6661 Write to the output port PORT selected contents of the match
6662 structure MATCH. Each ITEM specifies what should be written, and
6663 may be one of the following arguments:
6665 * A string. String arguments are written out verbatim.
6667 * An integer. The submatch with that number is written.
6669 * The symbol `pre'. The portion of the matched string preceding
6670 the regexp match is written.
6672 * The symbol `post'. The portion of the matched string
6673 following the regexp match is written.
6675 PORT may be `#f', in which case nothing is written; instead,
6676 `regexp-substitute' constructs a string from the specified ITEMs
6679 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6680 Similar to `regexp-substitute', but can be used to perform global
6681 substitutions on STR. Instead of taking a match structure as an
6682 argument, `regexp-substitute/global' takes two string arguments: a
6683 REGEXP string describing a regular expression, and a TARGET string
6684 which should be matched against this regular expression.
6686 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6689 * A function may be supplied. When this function is called, it
6690 will be passed one argument: a match structure for a given
6691 regular expression match. It should return a string to be
6692 written out to PORT.
6694 * The `post' symbol causes `regexp-substitute/global' to recurse
6695 on the unmatched portion of STR. This *must* be supplied in
6696 order to perform global search-and-replace on STR; if it is
6697 not present among the ITEMs, then `regexp-substitute/global'
6698 will return after processing a single match.
6700 *** Match Structures
6702 A "match structure" is the object returned by `string-match' and
6703 `regexp-exec'. It describes which portion of a string, if any, matched
6704 the given regular expression. Match structures include: a reference to
6705 the string that was checked for matches; the starting and ending
6706 positions of the regexp match; and, if the regexp included any
6707 parenthesized subexpressions, the starting and ending positions of each
6710 In each of the regexp match functions described below, the `match'
6711 argument must be a match structure returned by a previous call to
6712 `string-match' or `regexp-exec'. Most of these functions return some
6713 information about the original target string that was matched against a
6714 regular expression; we will call that string TARGET for easy reference.
6716 **** Function: regexp-match? OBJ
6717 Return `#t' if OBJ is a match structure returned by a previous
6718 call to `regexp-exec', or `#f' otherwise.
6720 **** Function: match:substring MATCH [N]
6721 Return the portion of TARGET matched by subexpression number N.
6722 Submatch 0 (the default) represents the entire regexp match. If
6723 the regular expression as a whole matched, but the subexpression
6724 number N did not match, return `#f'.
6726 **** Function: match:start MATCH [N]
6727 Return the starting position of submatch number N.
6729 **** Function: match:end MATCH [N]
6730 Return the ending position of submatch number N.
6732 **** Function: match:prefix MATCH
6733 Return the unmatched portion of TARGET preceding the regexp match.
6735 **** Function: match:suffix MATCH
6736 Return the unmatched portion of TARGET following the regexp match.
6738 **** Function: match:count MATCH
6739 Return the number of parenthesized subexpressions from MATCH.
6740 Note that the entire regular expression match itself counts as a
6741 subexpression, and failed submatches are included in the count.
6743 **** Function: match:string MATCH
6744 Return the original TARGET string.
6746 *** Backslash Escapes
6748 Sometimes you will want a regexp to match characters like `*' or `$'
6749 exactly. For example, to check whether a particular string represents
6750 a menu entry from an Info node, it would be useful to match it against
6751 a regexp like `^* [^:]*::'. However, this won't work; because the
6752 asterisk is a metacharacter, it won't match the `*' at the beginning of
6753 the string. In this case, we want to make the first asterisk un-magic.
6755 You can do this by preceding the metacharacter with a backslash
6756 character `\'. (This is also called "quoting" the metacharacter, and
6757 is known as a "backslash escape".) When Guile sees a backslash in a
6758 regular expression, it considers the following glyph to be an ordinary
6759 character, no matter what special meaning it would ordinarily have.
6760 Therefore, we can make the above example work by changing the regexp to
6761 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6762 to match only a single asterisk in the target string.
6764 Since the backslash is itself a metacharacter, you may force a
6765 regexp to match a backslash in the target string by preceding the
6766 backslash with itself. For example, to find variable references in a
6767 TeX program, you might want to find occurrences of the string `\let\'
6768 followed by any number of alphabetic characters. The regular expression
6769 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6770 each match a single backslash in the target string.
6772 **** Function: regexp-quote STR
6773 Quote each special character found in STR with a backslash, and
6774 return the resulting string.
6776 *Very important:* Using backslash escapes in Guile source code (as
6777 in Emacs Lisp or C) can be tricky, because the backslash character has
6778 special meaning for the Guile reader. For example, if Guile encounters
6779 the character sequence `\n' in the middle of a string while processing
6780 Scheme code, it replaces those characters with a newline character.
6781 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6782 Several of these "escape sequences" are processed by the Guile reader
6783 before your code is executed. Unrecognized escape sequences are
6784 ignored: if the characters `\*' appear in a string, they will be
6785 translated to the single character `*'.
6787 This translation is obviously undesirable for regular expressions,
6788 since we want to be able to include backslashes in a string in order to
6789 escape regexp metacharacters. Therefore, to make sure that a backslash
6790 is preserved in a string in your Guile program, you must use *two*
6791 consecutive backslashes:
6793 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6795 The string in this example is preprocessed by the Guile reader before
6796 any code is executed. The resulting argument to `make-regexp' is the
6797 string `^\* [^:]*', which is what we really want.
6799 This also means that in order to write a regular expression that
6800 matches a single backslash character, the regular expression string in
6801 the source code must include *four* backslashes. Each consecutive pair
6802 of backslashes gets translated by the Guile reader to a single
6803 backslash, and the resulting double-backslash is interpreted by the
6804 regexp engine as matching a single backslash character. Hence:
6806 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6808 The reason for the unwieldiness of this syntax is historical. Both
6809 regular expression pattern matchers and Unix string processing systems
6810 have traditionally used backslashes with the special meanings described
6811 above. The POSIX regular expression specification and ANSI C standard
6812 both require these semantics. Attempting to abandon either convention
6813 would cause other kinds of compatibility problems, possibly more severe
6814 ones. Therefore, without extending the Scheme reader to support
6815 strings with different quoting conventions (an ungainly and confusing
6816 extension when implemented in other languages), we must adhere to this
6817 cumbersome escape syntax.
6819 * Changes to the gh_ interface
6821 * Changes to the scm_ interface
6823 * Changes to system call interfaces:
6825 ** The value returned by `raise' is now unspecified. It throws an exception
6828 *** A new procedure `sigaction' can be used to install signal handlers
6830 (sigaction signum [action] [flags])
6832 signum is the signal number, which can be specified using the value
6835 If action is omitted, sigaction returns a pair: the CAR is the current
6836 signal hander, which will be either an integer with the value SIG_DFL
6837 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6838 handles the signal, or #f if a non-Scheme procedure handles the
6839 signal. The CDR contains the current sigaction flags for the handler.
6841 If action is provided, it is installed as the new handler for signum.
6842 action can be a Scheme procedure taking one argument, or the value of
6843 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6844 whatever signal handler was installed before sigaction was first used.
6845 Flags can optionally be specified for the new handler (SA_RESTART is
6846 always used if the system provides it, so need not be specified.) The
6847 return value is a pair with information about the old handler as
6850 This interface does not provide access to the "signal blocking"
6851 facility. Maybe this is not needed, since the thread support may
6852 provide solutions to the problem of consistent access to data
6855 *** A new procedure `flush-all-ports' is equivalent to running
6856 `force-output' on every port open for output.
6858 ** Guile now provides information on how it was built, via the new
6859 global variable, %guile-build-info. This variable records the values
6860 of the standard GNU makefile directory variables as an assocation
6861 list, mapping variable names (symbols) onto directory paths (strings).
6862 For example, to find out where the Guile link libraries were
6863 installed, you can say:
6865 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6868 * Changes to the scm_ interface
6870 ** The new function scm_handle_by_message_noexit is just like the
6871 existing scm_handle_by_message function, except that it doesn't call
6872 exit to terminate the process. Instead, it prints a message and just
6873 returns #f. This might be a more appropriate catch-all handler for
6874 new dynamic roots and threads.
6877 Changes in Guile 1.1 (released Friday, May 16 1997):
6879 * Changes to the distribution.
6881 The Guile 1.0 distribution has been split up into several smaller
6883 guile-core --- the Guile interpreter itself.
6884 guile-tcltk --- the interface between the Guile interpreter and
6885 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6886 is a toolkit for building graphical user interfaces.
6887 guile-rgx-ctax --- the interface between Guile and the Rx regular
6888 expression matcher, and the translator for the Ctax
6889 programming language. These are packaged together because the
6890 Ctax translator uses Rx to parse Ctax source code.
6892 This NEWS file describes the changes made to guile-core since the 1.0
6895 We no longer distribute the documentation, since it was either out of
6896 date, or incomplete. As soon as we have current documentation, we
6901 * Changes to the stand-alone interpreter
6903 ** guile now accepts command-line arguments compatible with SCSH, Olin
6904 Shivers' Scheme Shell.
6906 In general, arguments are evaluated from left to right, but there are
6907 exceptions. The following switches stop argument processing, and
6908 stash all remaining command-line arguments as the value returned by
6909 the (command-line) function.
6910 -s SCRIPT load Scheme source code from FILE, and exit
6911 -c EXPR evalute Scheme expression EXPR, and exit
6912 -- stop scanning arguments; run interactively
6914 The switches below are processed as they are encountered.
6915 -l FILE load Scheme source code from FILE
6916 -e FUNCTION after reading script, apply FUNCTION to
6917 command line arguments
6918 -ds do -s script at this point
6919 --emacs enable Emacs protocol (experimental)
6920 -h, --help display this help and exit
6921 -v, --version display version information and exit
6922 \ read arguments from following script lines
6924 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6925 which re-implements the traditional "echo" command:
6927 #!/usr/local/bin/guile -s
6930 (map (lambda (arg) (display arg) (display " "))
6934 (main (command-line))
6936 Suppose we invoke this script as follows:
6938 ekko a speckled gecko
6940 Through the magic of Unix script processing (triggered by the `#!'
6941 token at the top of the file), /usr/local/bin/guile receives the
6942 following list of command-line arguments:
6944 ("-s" "./ekko" "a" "speckled" "gecko")
6946 Unix inserts the name of the script after the argument specified on
6947 the first line of the file (in this case, "-s"), and then follows that
6948 with the arguments given to the script. Guile loads the script, which
6949 defines the `main' function, and then applies it to the list of
6950 remaining command-line arguments, ("a" "speckled" "gecko").
6952 In Unix, the first line of a script file must take the following form:
6954 #!INTERPRETER ARGUMENT
6956 where INTERPRETER is the absolute filename of the interpreter
6957 executable, and ARGUMENT is a single command-line argument to pass to
6960 You may only pass one argument to the interpreter, and its length is
6961 limited. These restrictions can be annoying to work around, so Guile
6962 provides a general mechanism (borrowed from, and compatible with,
6963 SCSH) for circumventing them.
6965 If the ARGUMENT in a Guile script is a single backslash character,
6966 `\', Guile will open the script file, parse arguments from its second
6967 and subsequent lines, and replace the `\' with them. So, for example,
6968 here is another implementation of the `ekko' script:
6970 #!/usr/local/bin/guile \
6974 (for-each (lambda (arg) (display arg) (display " "))
6978 If the user invokes this script as follows:
6980 ekko a speckled gecko
6982 Unix expands this into
6984 /usr/local/bin/guile \ ekko a speckled gecko
6986 When Guile sees the `\' argument, it replaces it with the arguments
6987 read from the second line of the script, producing:
6989 /usr/local/bin/guile -e main -s ekko a speckled gecko
6991 This tells Guile to load the `ekko' script, and apply the function
6992 `main' to the argument list ("a" "speckled" "gecko").
6994 Here is how Guile parses the command-line arguments:
6995 - Each space character terminates an argument. This means that two
6996 spaces in a row introduce an empty-string argument.
6997 - The tab character is not permitted (unless you quote it with the
6998 backslash character, as described below), to avoid confusion.
6999 - The newline character terminates the sequence of arguments, and will
7000 also terminate a final non-empty argument. (However, a newline
7001 following a space will not introduce a final empty-string argument;
7002 it only terminates the argument list.)
7003 - The backslash character is the escape character. It escapes
7004 backslash, space, tab, and newline. The ANSI C escape sequences
7005 like \n and \t are also supported. These produce argument
7006 constituents; the two-character combination \n doesn't act like a
7007 terminating newline. The escape sequence \NNN for exactly three
7008 octal digits reads as the character whose ASCII code is NNN. As
7009 above, characters produced this way are argument constituents.
7010 Backslash followed by other characters is not allowed.
7012 * Changes to the procedure for linking libguile with your programs
7014 ** Guile now builds and installs a shared guile library, if your
7015 system support shared libraries. (It still builds a static library on
7016 all systems.) Guile automatically detects whether your system
7017 supports shared libraries. To prevent Guile from buildisg shared
7018 libraries, pass the `--disable-shared' flag to the configure script.
7020 Guile takes longer to compile when it builds shared libraries, because
7021 it must compile every file twice --- once to produce position-
7022 independent object code, and once to produce normal object code.
7024 ** The libthreads library has been merged into libguile.
7026 To link a program against Guile, you now need only link against
7027 -lguile and -lqt; -lthreads is no longer needed. If you are using
7028 autoconf to generate configuration scripts for your application, the
7029 following lines should suffice to add the appropriate libraries to
7032 ### Find quickthreads and libguile.
7033 AC_CHECK_LIB(qt, main)
7034 AC_CHECK_LIB(guile, scm_shell)
7036 * Changes to Scheme functions
7038 ** Guile Scheme's special syntax for keyword objects is now optional,
7039 and disabled by default.
7041 The syntax variation from R4RS made it difficult to port some
7042 interesting packages to Guile. The routines which accepted keyword
7043 arguments (mostly in the module system) have been modified to also
7044 accept symbols whose names begin with `:'.
7046 To change the keyword syntax, you must first import the (ice-9 debug)
7048 (use-modules (ice-9 debug))
7050 Then you can enable the keyword syntax as follows:
7051 (read-set! keywords 'prefix)
7053 To disable keyword syntax, do this:
7054 (read-set! keywords #f)
7056 ** Many more primitive functions accept shared substrings as
7057 arguments. In the past, these functions required normal, mutable
7058 strings as arguments, although they never made use of this
7061 ** The uniform array functions now operate on byte vectors. These
7062 functions are `array-fill!', `serial-array-copy!', `array-copy!',
7063 `serial-array-map', `array-map', `array-for-each', and
7066 ** The new functions `trace' and `untrace' implement simple debugging
7067 support for Scheme functions.
7069 The `trace' function accepts any number of procedures as arguments,
7070 and tells the Guile interpreter to display each procedure's name and
7071 arguments each time the procedure is invoked. When invoked with no
7072 arguments, `trace' returns the list of procedures currently being
7075 The `untrace' function accepts any number of procedures as arguments,
7076 and tells the Guile interpreter not to trace them any more. When
7077 invoked with no arguments, `untrace' untraces all curretly traced
7080 The tracing in Guile has an advantage over most other systems: we
7081 don't create new procedure objects, but mark the procedure objects
7082 themselves. This means that anonymous and internal procedures can be
7085 ** The function `assert-repl-prompt' has been renamed to
7086 `set-repl-prompt!'. It takes one argument, PROMPT.
7087 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
7088 - If PROMPT is a string, we use it as a prompt.
7089 - If PROMPT is a procedure accepting no arguments, we call it, and
7090 display the result as a prompt.
7091 - Otherwise, we display "> ".
7093 ** The new function `eval-string' reads Scheme expressions from a
7094 string and evaluates them, returning the value of the last expression
7095 in the string. If the string contains no expressions, it returns an
7098 ** The new function `thunk?' returns true iff its argument is a
7099 procedure of zero arguments.
7101 ** `defined?' is now a builtin function, instead of syntax. This
7102 means that its argument should be quoted. It returns #t iff its
7103 argument is bound in the current module.
7105 ** The new syntax `use-modules' allows you to add new modules to your
7106 environment without re-typing a complete `define-module' form. It
7107 accepts any number of module names as arguments, and imports their
7108 public bindings into the current module.
7110 ** The new function (module-defined? NAME MODULE) returns true iff
7111 NAME, a symbol, is defined in MODULE, a module object.
7113 ** The new function `builtin-bindings' creates and returns a hash
7114 table containing copies of all the root module's bindings.
7116 ** The new function `builtin-weak-bindings' does the same as
7117 `builtin-bindings', but creates a doubly-weak hash table.
7119 ** The `equal?' function now considers variable objects to be
7120 equivalent if they have the same name and the same value.
7122 ** The new function `command-line' returns the command-line arguments
7123 given to Guile, as a list of strings.
7125 When using guile as a script interpreter, `command-line' returns the
7126 script's arguments; those processed by the interpreter (like `-s' or
7127 `-c') are omitted. (In other words, you get the normal, expected
7128 behavior.) Any application that uses scm_shell to process its
7129 command-line arguments gets this behavior as well.
7131 ** The new function `load-user-init' looks for a file called `.guile'
7132 in the user's home directory, and loads it if it exists. This is
7133 mostly for use by the code generated by scm_compile_shell_switches,
7134 but we thought it might also be useful in other circumstances.
7136 ** The new function `log10' returns the base-10 logarithm of its
7139 ** Changes to I/O functions
7141 *** The functions `read', `primitive-load', `read-and-eval!', and
7142 `primitive-load-path' no longer take optional arguments controlling
7143 case insensitivity and a `#' parser.
7145 Case sensitivity is now controlled by a read option called
7146 `case-insensitive'. The user can add new `#' syntaxes with the
7147 `read-hash-extend' function (see below).
7149 *** The new function `read-hash-extend' allows the user to change the
7150 syntax of Guile Scheme in a somewhat controlled way.
7152 (read-hash-extend CHAR PROC)
7153 When parsing S-expressions, if we read a `#' character followed by
7154 the character CHAR, use PROC to parse an object from the stream.
7155 If PROC is #f, remove any parsing procedure registered for CHAR.
7157 The reader applies PROC to two arguments: CHAR and an input port.
7159 *** The new functions read-delimited and read-delimited! provide a
7160 general mechanism for doing delimited input on streams.
7162 (read-delimited DELIMS [PORT HANDLE-DELIM])
7163 Read until we encounter one of the characters in DELIMS (a string),
7164 or end-of-file. PORT is the input port to read from; it defaults to
7165 the current input port. The HANDLE-DELIM parameter determines how
7166 the terminating character is handled; it should be one of the
7169 'trim omit delimiter from result
7170 'peek leave delimiter character in input stream
7171 'concat append delimiter character to returned value
7172 'split return a pair: (RESULT . TERMINATOR)
7174 HANDLE-DELIM defaults to 'peek.
7176 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
7177 A side-effecting variant of `read-delimited'.
7179 The data is written into the string BUF at the indices in the
7180 half-open interval [START, END); the default interval is the whole
7181 string: START = 0 and END = (string-length BUF). The values of
7182 START and END must specify a well-defined interval in BUF, i.e.
7183 0 <= START <= END <= (string-length BUF).
7185 It returns NBYTES, the number of bytes read. If the buffer filled
7186 up without a delimiter character being found, it returns #f. If the
7187 port is at EOF when the read starts, it returns the EOF object.
7189 If an integer is returned (i.e., the read is successfully terminated
7190 by reading a delimiter character), then the HANDLE-DELIM parameter
7191 determines how to handle the terminating character. It is described
7192 above, and defaults to 'peek.
7194 (The descriptions of these functions were borrowed from the SCSH
7195 manual, by Olin Shivers and Brian Carlstrom.)
7197 *** The `%read-delimited!' function is the primitive used to implement
7198 `read-delimited' and `read-delimited!'.
7200 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
7202 This returns a pair of values: (TERMINATOR . NUM-READ).
7203 - TERMINATOR describes why the read was terminated. If it is a
7204 character or the eof object, then that is the value that terminated
7205 the read. If it is #f, the function filled the buffer without finding
7206 a delimiting character.
7207 - NUM-READ is the number of characters read into BUF.
7209 If the read is successfully terminated by reading a delimiter
7210 character, then the gobble? parameter determines what to do with the
7211 terminating character. If true, the character is removed from the
7212 input stream; if false, the character is left in the input stream
7213 where a subsequent read operation will retrieve it. In either case,
7214 the character is also the first value returned by the procedure call.
7216 (The descriptions of this function was borrowed from the SCSH manual,
7217 by Olin Shivers and Brian Carlstrom.)
7219 *** The `read-line' and `read-line!' functions have changed; they now
7220 trim the terminator by default; previously they appended it to the
7221 returned string. For the old behavior, use (read-line PORT 'concat).
7223 *** The functions `uniform-array-read!' and `uniform-array-write!' now
7224 take new optional START and END arguments, specifying the region of
7225 the array to read and write.
7227 *** The `ungetc-char-ready?' function has been removed. We feel it's
7228 inappropriate for an interface to expose implementation details this
7231 ** Changes to the Unix library and system call interface
7233 *** The new fcntl function provides access to the Unix `fcntl' system
7236 (fcntl PORT COMMAND VALUE)
7237 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
7238 Values for COMMAND are:
7240 F_DUPFD duplicate a file descriptor
7241 F_GETFD read the descriptor's close-on-exec flag
7242 F_SETFD set the descriptor's close-on-exec flag to VALUE
7243 F_GETFL read the descriptor's flags, as set on open
7244 F_SETFL set the descriptor's flags, as set on open to VALUE
7245 F_GETOWN return the process ID of a socket's owner, for SIGIO
7246 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
7247 FD_CLOEXEC not sure what this is
7249 For details, see the documentation for the fcntl system call.
7251 *** The arguments to `select' have changed, for compatibility with
7252 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
7253 expected behavior. The MILLISECONDS parameter has been changed to
7254 MICROSECONDS, to more closely resemble the underlying system call.
7255 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
7256 corresponding return set will be the same.
7258 *** The arguments to the `mknod' system call have changed. They are
7261 (mknod PATH TYPE PERMS DEV)
7262 Create a new file (`node') in the file system. PATH is the name of
7263 the file to create. TYPE is the kind of file to create; it should
7264 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
7265 permission bits to give the newly created file. If TYPE is
7266 'block-special or 'char-special, DEV specifies which device the
7267 special file refers to; its interpretation depends on the kind of
7268 special file being created.
7270 *** The `fork' function has been renamed to `primitive-fork', to avoid
7271 clashing with various SCSH forks.
7273 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
7274 and `recvfrom!'. They no longer accept a size for a second argument;
7275 you must pass a string to hold the received value. They no longer
7276 return the buffer. Instead, `recv' returns the length of the message
7277 received, and `recvfrom' returns a pair containing the packet's length
7278 and originating address.
7280 *** The file descriptor datatype has been removed, as have the
7281 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
7282 We plan to replace these functions with a SCSH-compatible interface.
7284 *** The `create' function has been removed; it's just a special case
7287 *** There are new functions to break down process termination status
7288 values. In the descriptions below, STATUS is a value returned by
7291 (status:exit-val STATUS)
7292 If the child process exited normally, this function returns the exit
7293 code for the child process (i.e., the value passed to exit, or
7294 returned from main). If the child process did not exit normally,
7295 this function returns #f.
7297 (status:stop-sig STATUS)
7298 If the child process was suspended by a signal, this function
7299 returns the signal that suspended the child. Otherwise, it returns
7302 (status:term-sig STATUS)
7303 If the child process terminated abnormally, this function returns
7304 the signal that terminated the child. Otherwise, this function
7307 POSIX promises that exactly one of these functions will return true on
7308 a valid STATUS value.
7310 These functions are compatible with SCSH.
7312 *** There are new accessors and setters for the broken-out time vectors
7313 returned by `localtime', `gmtime', and that ilk. They are:
7315 Component Accessor Setter
7316 ========================= ============ ============
7317 seconds tm:sec set-tm:sec
7318 minutes tm:min set-tm:min
7319 hours tm:hour set-tm:hour
7320 day of the month tm:mday set-tm:mday
7321 month tm:mon set-tm:mon
7322 year tm:year set-tm:year
7323 day of the week tm:wday set-tm:wday
7324 day in the year tm:yday set-tm:yday
7325 daylight saving time tm:isdst set-tm:isdst
7326 GMT offset, seconds tm:gmtoff set-tm:gmtoff
7327 name of time zone tm:zone set-tm:zone
7329 *** There are new accessors for the vectors returned by `uname',
7330 describing the host system:
7333 ============================================== ================
7334 name of the operating system implementation utsname:sysname
7335 network name of this machine utsname:nodename
7336 release level of the operating system utsname:release
7337 version level of the operating system utsname:version
7338 machine hardware platform utsname:machine
7340 *** There are new accessors for the vectors returned by `getpw',
7341 `getpwnam', `getpwuid', and `getpwent', describing entries from the
7342 system's user database:
7345 ====================== =================
7346 user name passwd:name
7347 user password passwd:passwd
7350 real name passwd:gecos
7351 home directory passwd:dir
7352 shell program passwd:shell
7354 *** There are new accessors for the vectors returned by `getgr',
7355 `getgrnam', `getgrgid', and `getgrent', describing entries from the
7356 system's group database:
7359 ======================= ============
7360 group name group:name
7361 group password group:passwd
7363 group members group:mem
7365 *** There are new accessors for the vectors returned by `gethost',
7366 `gethostbyaddr', `gethostbyname', and `gethostent', describing
7370 ========================= ===============
7371 official name of host hostent:name
7372 alias list hostent:aliases
7373 host address type hostent:addrtype
7374 length of address hostent:length
7375 list of addresses hostent:addr-list
7377 *** There are new accessors for the vectors returned by `getnet',
7378 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
7382 ========================= ===============
7383 official name of net netent:name
7384 alias list netent:aliases
7385 net number type netent:addrtype
7386 net number netent:net
7388 *** There are new accessors for the vectors returned by `getproto',
7389 `getprotobyname', `getprotobynumber', and `getprotoent', describing
7393 ========================= ===============
7394 official protocol name protoent:name
7395 alias list protoent:aliases
7396 protocol number protoent:proto
7398 *** There are new accessors for the vectors returned by `getserv',
7399 `getservbyname', `getservbyport', and `getservent', describing
7403 ========================= ===============
7404 official service name servent:name
7405 alias list servent:aliases
7406 port number servent:port
7407 protocol to use servent:proto
7409 *** There are new accessors for the sockaddr structures returned by
7410 `accept', `getsockname', `getpeername', `recvfrom!':
7413 ======================================== ===============
7414 address format (`family') sockaddr:fam
7415 path, for file domain addresses sockaddr:path
7416 address, for internet domain addresses sockaddr:addr
7417 TCP or UDP port, for internet sockaddr:port
7419 *** The `getpwent', `getgrent', `gethostent', `getnetent',
7420 `getprotoent', and `getservent' functions now return #f at the end of
7421 the user database. (They used to throw an exception.)
7423 Note that calling MUMBLEent function is equivalent to calling the
7424 corresponding MUMBLE function with no arguments.
7426 *** The `setpwent', `setgrent', `sethostent', `setnetent',
7427 `setprotoent', and `setservent' routines now take no arguments.
7429 *** The `gethost', `getproto', `getnet', and `getserv' functions now
7430 provide more useful information when they throw an exception.
7432 *** The `lnaof' function has been renamed to `inet-lnaof'.
7434 *** Guile now claims to have the `current-time' feature.
7436 *** The `mktime' function now takes an optional second argument ZONE,
7437 giving the time zone to use for the conversion. ZONE should be a
7438 string, in the same format as expected for the "TZ" environment variable.
7440 *** The `strptime' function now returns a pair (TIME . COUNT), where
7441 TIME is the parsed time as a vector, and COUNT is the number of
7442 characters from the string left unparsed. This function used to
7443 return the remaining characters as a string.
7445 *** The `gettimeofday' function has replaced the old `time+ticks' function.
7446 The return value is now (SECONDS . MICROSECONDS); the fractional
7447 component is no longer expressed in "ticks".
7449 *** The `ticks/sec' constant has been removed, in light of the above change.
7451 * Changes to the gh_ interface
7453 ** gh_eval_str() now returns an SCM object which is the result of the
7456 ** gh_scm2str() now copies the Scheme data to a caller-provided C
7459 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
7460 and returns the array
7462 ** gh_scm2str0() is gone: there is no need to distinguish
7463 null-terminated from non-null-terminated, since gh_scm2newstr() allows
7464 the user to interpret the data both ways.
7466 * Changes to the scm_ interface
7468 ** The new function scm_symbol_value0 provides an easy way to get a
7469 symbol's value from C code:
7471 SCM scm_symbol_value0 (char *NAME)
7472 Return the value of the symbol named by the null-terminated string
7473 NAME in the current module. If the symbol named NAME is unbound in
7474 the current module, return SCM_UNDEFINED.
7476 ** The new function scm_sysintern0 creates new top-level variables,
7477 without assigning them a value.
7479 SCM scm_sysintern0 (char *NAME)
7480 Create a new Scheme top-level variable named NAME. NAME is a
7481 null-terminated string. Return the variable's value cell.
7483 ** The function scm_internal_catch is the guts of catch. It handles
7484 all the mechanics of setting up a catch target, invoking the catch
7485 body, and perhaps invoking the handler if the body does a throw.
7487 The function is designed to be usable from C code, but is general
7488 enough to implement all the semantics Guile Scheme expects from throw.
7490 TAG is the catch tag. Typically, this is a symbol, but this function
7491 doesn't actually care about that.
7493 BODY is a pointer to a C function which runs the body of the catch;
7494 this is the code you can throw from. We call it like this:
7495 BODY (BODY_DATA, JMPBUF)
7497 BODY_DATA is just the BODY_DATA argument we received; we pass it
7498 through to BODY as its first argument. The caller can make
7499 BODY_DATA point to anything useful that BODY might need.
7500 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
7501 which we have just created and initialized.
7503 HANDLER is a pointer to a C function to deal with a throw to TAG,
7504 should one occur. We call it like this:
7505 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
7507 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
7508 same idea as BODY_DATA above.
7509 THROWN_TAG is the tag that the user threw to; usually this is
7510 TAG, but it could be something else if TAG was #t (i.e., a
7511 catch-all), or the user threw to a jmpbuf.
7512 THROW_ARGS is the list of arguments the user passed to the THROW
7515 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
7516 is just a pointer we pass through to HANDLER. We don't actually
7517 use either of those pointers otherwise ourselves. The idea is
7518 that, if our caller wants to communicate something to BODY or
7519 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
7520 HANDLER can then use. Think of it as a way to make BODY and
7521 HANDLER closures, not just functions; MUMBLE_DATA points to the
7524 Of course, it's up to the caller to make sure that any data a
7525 MUMBLE_DATA needs is protected from GC. A common way to do this is
7526 to make MUMBLE_DATA a pointer to data stored in an automatic
7527 structure variable; since the collector must scan the stack for
7528 references anyway, this assures that any references in MUMBLE_DATA
7531 ** The new function scm_internal_lazy_catch is exactly like
7532 scm_internal_catch, except:
7534 - It does not unwind the stack (this is the major difference).
7535 - If handler returns, its value is returned from the throw.
7536 - BODY always receives #f as its JMPBUF argument (since there's no
7537 jmpbuf associated with a lazy catch, because we don't unwind the
7540 ** scm_body_thunk is a new body function you can pass to
7541 scm_internal_catch if you want the body to be like Scheme's `catch'
7542 --- a thunk, or a function of one argument if the tag is #f.
7544 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
7545 contains the Scheme procedure to invoke as the body, and the tag
7546 we're catching. If the tag is #f, then we pass JMPBUF (created by
7547 scm_internal_catch) to the body procedure; otherwise, the body gets
7550 ** scm_handle_by_proc is a new handler function you can pass to
7551 scm_internal_catch if you want the handler to act like Scheme's catch
7552 --- call a procedure with the tag and the throw arguments.
7554 If the user does a throw to this catch, this function runs a handler
7555 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
7556 variable holding the Scheme procedure object to invoke. It ought to
7557 be a pointer to an automatic variable (i.e., one living on the stack),
7558 or the procedure object should be otherwise protected from GC.
7560 ** scm_handle_by_message is a new handler function to use with
7561 `scm_internal_catch' if you want Guile to print a message and die.
7562 It's useful for dealing with throws to uncaught keys at the top level.
7564 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
7565 message header to print; if zero, we use "guile" instead. That
7566 text is followed by a colon, then the message described by ARGS.
7568 ** The return type of scm_boot_guile is now void; the function does
7569 not return a value, and indeed, never returns at all.
7571 ** The new function scm_shell makes it easy for user applications to
7572 process command-line arguments in a way that is compatible with the
7573 stand-alone guile interpreter (which is in turn compatible with SCSH,
7576 To use the scm_shell function, first initialize any guile modules
7577 linked into your application, and then call scm_shell with the values
7578 of ARGC and ARGV your `main' function received. scm_shell will add
7579 any SCSH-style meta-arguments from the top of the script file to the
7580 argument vector, and then process the command-line arguments. This
7581 generally means loading a script file or starting up an interactive
7582 command interpreter. For details, see "Changes to the stand-alone
7585 ** The new functions scm_get_meta_args and scm_count_argv help you
7586 implement the SCSH-style meta-argument, `\'.
7588 char **scm_get_meta_args (int ARGC, char **ARGV)
7589 If the second element of ARGV is a string consisting of a single
7590 backslash character (i.e. "\\" in Scheme notation), open the file
7591 named by the following argument, parse arguments from it, and return
7592 the spliced command line. The returned array is terminated by a
7595 For details of argument parsing, see above, under "guile now accepts
7596 command-line arguments compatible with SCSH..."
7598 int scm_count_argv (char **ARGV)
7599 Count the arguments in ARGV, assuming it is terminated by a null
7602 For an example of how these functions might be used, see the source
7603 code for the function scm_shell in libguile/script.c.
7605 You will usually want to use scm_shell instead of calling this
7608 ** The new function scm_compile_shell_switches turns an array of
7609 command-line arguments into Scheme code to carry out the actions they
7610 describe. Given ARGC and ARGV, it returns a Scheme expression to
7611 evaluate, and calls scm_set_program_arguments to make any remaining
7612 command-line arguments available to the Scheme code. For example,
7613 given the following arguments:
7615 -e main -s ekko a speckled gecko
7617 scm_set_program_arguments will return the following expression:
7619 (begin (load "ekko") (main (command-line)) (quit))
7621 You will usually want to use scm_shell instead of calling this
7624 ** The function scm_shell_usage prints a usage message appropriate for
7625 an interpreter that uses scm_compile_shell_switches to handle its
7626 command-line arguments.
7628 void scm_shell_usage (int FATAL, char *MESSAGE)
7629 Print a usage message to the standard error output. If MESSAGE is
7630 non-zero, write it before the usage message, followed by a newline.
7631 If FATAL is non-zero, exit the process, using FATAL as the
7632 termination status. (If you want to be compatible with Guile,
7633 always use 1 as the exit status when terminating due to command-line
7636 You will usually want to use scm_shell instead of calling this
7639 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7640 expressions. It used to return SCM_EOL. Earth-shattering.
7642 ** The macros for declaring scheme objects in C code have been
7643 rearranged slightly. They are now:
7645 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7646 Declare a static SCM variable named C_NAME, and initialize it to
7647 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7648 be a C identifier, and SCHEME_NAME should be a C string.
7650 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7651 Just like SCM_SYMBOL, but make C_NAME globally visible.
7653 SCM_VCELL (C_NAME, SCHEME_NAME)
7654 Create a global variable at the Scheme level named SCHEME_NAME.
7655 Declare a static SCM variable named C_NAME, and initialize it to
7656 point to the Scheme variable's value cell.
7658 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7659 Just like SCM_VCELL, but make C_NAME globally visible.
7661 The `guile-snarf' script writes initialization code for these macros
7662 to its standard output, given C source code as input.
7664 The SCM_GLOBAL macro is gone.
7666 ** The scm_read_line and scm_read_line_x functions have been replaced
7667 by Scheme code based on the %read-delimited! procedure (known to C
7668 code as scm_read_delimited_x). See its description above for more
7671 ** The function scm_sys_open has been renamed to scm_open. It now
7672 returns a port instead of an FD object.
7674 * The dynamic linking support has changed. For more information, see
7675 libguile/DYNAMIC-LINKING.
7680 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7683 * Changes to the 'guile' program:
7685 ** Guile now loads some new files when it starts up. Guile first
7686 searches the load path for init.scm, and loads it if found. Then, if
7687 Guile is not being used to execute a script, and the user's home
7688 directory contains a file named `.guile', Guile loads that.
7690 ** You can now use Guile as a shell script interpreter.
7692 To paraphrase the SCSH manual:
7694 When Unix tries to execute an executable file whose first two
7695 characters are the `#!', it treats the file not as machine code to
7696 be directly executed by the native processor, but as source code
7697 to be executed by some interpreter. The interpreter to use is
7698 specified immediately after the #! sequence on the first line of
7699 the source file. The kernel reads in the name of the interpreter,
7700 and executes that instead. It passes the interpreter the source
7701 filename as its first argument, with the original arguments
7702 following. Consult the Unix man page for the `exec' system call
7703 for more information.
7705 Now you can use Guile as an interpreter, using a mechanism which is a
7706 compatible subset of that provided by SCSH.
7708 Guile now recognizes a '-s' command line switch, whose argument is the
7709 name of a file of Scheme code to load. It also treats the two
7710 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7711 to make a file of Scheme code directly executable by Unix, insert the
7712 following two lines at the top of the file:
7714 #!/usr/local/bin/guile -s
7717 Guile treats the argument of the `-s' command-line switch as the name
7718 of a file of Scheme code to load, and treats the sequence `#!' as the
7719 start of a block comment, terminated by `!#'.
7721 For example, here's a version of 'echo' written in Scheme:
7723 #!/usr/local/bin/guile -s
7725 (let loop ((args (cdr (program-arguments))))
7728 (display (car args))
7729 (if (pair? (cdr args))
7731 (loop (cdr args)))))
7734 Why does `#!' start a block comment terminated by `!#', instead of the
7735 end of the line? That is the notation SCSH uses, and although we
7736 don't yet support the other SCSH features that motivate that choice,
7737 we would like to be backward-compatible with any existing Guile
7738 scripts once we do. Furthermore, if the path to Guile on your system
7739 is too long for your kernel, you can start the script with this
7743 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7746 Note that some very old Unix systems don't support the `#!' syntax.
7749 ** You can now run Guile without installing it.
7751 Previous versions of the interactive Guile interpreter (`guile')
7752 couldn't start up unless Guile's Scheme library had been installed;
7753 they used the value of the environment variable `SCHEME_LOAD_PATH'
7754 later on in the startup process, but not to find the startup code
7755 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7758 To run Guile without installing it, build it in the normal way, and
7759 then set the environment variable `SCHEME_LOAD_PATH' to a
7760 colon-separated list of directories, including the top-level directory
7761 of the Guile sources. For example, if you unpacked Guile so that the
7762 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7765 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7768 ** Guile's read-eval-print loop no longer prints #<unspecified>
7769 results. If the user wants to see this, she can evaluate the
7770 expression (assert-repl-print-unspecified #t), perhaps in her startup
7773 ** Guile no longer shows backtraces by default when an error occurs;
7774 however, it does display a message saying how to get one, and how to
7775 request that they be displayed by default. After an error, evaluate
7777 to see a backtrace, and
7778 (debug-enable 'backtrace)
7779 to see them by default.
7783 * Changes to Guile Scheme:
7785 ** Guile now distinguishes between #f and the empty list.
7787 This is for compatibility with the IEEE standard, the (possibly)
7788 upcoming Revised^5 Report on Scheme, and many extant Scheme
7791 Guile used to have #f and '() denote the same object, to make Scheme's
7792 type system more compatible with Emacs Lisp's. However, the change
7793 caused too much trouble for Scheme programmers, and we found another
7794 way to reconcile Emacs Lisp with Scheme that didn't require this.
7797 ** Guile's delq, delv, delete functions, and their destructive
7798 counterparts, delq!, delv!, and delete!, now remove all matching
7799 elements from the list, not just the first. This matches the behavior
7800 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7801 functions which inspired them.
7803 I recognize that this change may break code in subtle ways, but it
7804 seems best to make the change before the FSF's first Guile release,
7808 ** The compiled-library-path function has been deleted from libguile.
7810 ** The facilities for loading Scheme source files have changed.
7812 *** The variable %load-path now tells Guile which directories to search
7813 for Scheme code. Its value is a list of strings, each of which names
7816 *** The variable %load-extensions now tells Guile which extensions to
7817 try appending to a filename when searching the load path. Its value
7818 is a list of strings. Its default value is ("" ".scm").
7820 *** (%search-load-path FILENAME) searches the directories listed in the
7821 value of the %load-path variable for a Scheme file named FILENAME,
7822 with all the extensions listed in %load-extensions. If it finds a
7823 match, then it returns its full filename. If FILENAME is absolute, it
7824 returns it unchanged. Otherwise, it returns #f.
7826 %search-load-path will not return matches that refer to directories.
7828 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7829 uses %seach-load-path to find a file named FILENAME, and loads it if
7830 it finds it. If it can't read FILENAME for any reason, it throws an
7833 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7836 *** load uses the same searching semantics as primitive-load.
7838 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7839 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7840 path, and load-module-with-path have been deleted. The functions
7841 above should serve their purposes.
7843 *** If the value of the variable %load-hook is a procedure,
7844 `primitive-load' applies its value to the name of the file being
7845 loaded (without the load path directory name prepended). If its value
7846 is #f, it is ignored. Otherwise, an error occurs.
7848 This is mostly useful for printing load notification messages.
7851 ** The function `eval!' is no longer accessible from the scheme level.
7852 We can't allow operations which introduce glocs into the scheme level,
7853 because Guile's type system can't handle these as data. Use `eval' or
7854 `read-and-eval!' (see below) as replacement.
7856 ** The new function read-and-eval! reads an expression from PORT,
7857 evaluates it, and returns the result. This is more efficient than
7858 simply calling `read' and `eval', since it is not necessary to make a
7859 copy of the expression for the evaluator to munge.
7861 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7862 for the `read' function.
7865 ** The function `int?' has been removed; its definition was identical
7866 to that of `integer?'.
7868 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7869 use the R4RS names for these functions.
7871 ** The function object-properties no longer returns the hash handle;
7872 it simply returns the object's property list.
7874 ** Many functions have been changed to throw errors, instead of
7875 returning #f on failure. The point of providing exception handling in
7876 the language is to simplify the logic of user code, but this is less
7877 useful if Guile's primitives don't throw exceptions.
7879 ** The function `fileno' has been renamed from `%fileno'.
7881 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7884 * Changes to Guile's C interface:
7886 ** The library's initialization procedure has been simplified.
7887 scm_boot_guile now has the prototype:
7889 void scm_boot_guile (int ARGC,
7891 void (*main_func) (),
7894 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7895 MAIN_FUNC should do all the work of the program (initializing other
7896 packages, reading user input, etc.) before returning. When MAIN_FUNC
7897 returns, call exit (0); this function never returns. If you want some
7898 other exit value, MAIN_FUNC may call exit itself.
7900 scm_boot_guile arranges for program-arguments to return the strings
7901 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7902 scm_set_program_arguments with the final list, so Scheme code will
7903 know which arguments have been processed.
7905 scm_boot_guile establishes a catch-all catch handler which prints an
7906 error message and exits the process. This means that Guile exits in a
7907 coherent way when system errors occur and the user isn't prepared to
7908 handle it. If the user doesn't like this behavior, they can establish
7909 their own universal catcher in MAIN_FUNC to shadow this one.
7911 Why must the caller do all the real work from MAIN_FUNC? The garbage
7912 collector assumes that all local variables of type SCM will be above
7913 scm_boot_guile's stack frame on the stack. If you try to manipulate
7914 SCM values after this function returns, it's the luck of the draw
7915 whether the GC will be able to find the objects you allocate. So,
7916 scm_boot_guile function exits, rather than returning, to discourage
7917 people from making that mistake.
7919 The IN, OUT, and ERR arguments were removed; there are other
7920 convenient ways to override these when desired.
7922 The RESULT argument was deleted; this function should never return.
7924 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7928 ** Guile's header files should no longer conflict with your system's
7931 In order to compile code which #included <libguile.h>, previous
7932 versions of Guile required you to add a directory containing all the
7933 Guile header files to your #include path. This was a problem, since
7934 Guile's header files have names which conflict with many systems'
7937 Now only <libguile.h> need appear in your #include path; you must
7938 refer to all Guile's other header files as <libguile/mumble.h>.
7939 Guile's installation procedure puts libguile.h in $(includedir), and
7940 the rest in $(includedir)/libguile.
7943 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7944 have been added to the Guile library.
7946 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7947 OBJ will not be freed, even if all other references are dropped,
7948 until someone does scm_unprotect_object (OBJ). Both functions
7951 Note that calls to scm_protect_object do not nest. You can call
7952 scm_protect_object any number of times on a given object, and the
7953 next call to scm_unprotect_object will unprotect it completely.
7955 Basically, scm_protect_object and scm_unprotect_object just
7956 maintain a list of references to things. Since the GC knows about
7957 this list, all objects it mentions stay alive. scm_protect_object
7958 adds its argument to the list; scm_unprotect_object remove its
7959 argument from the list.
7962 ** scm_eval_0str now returns the value of the last expression
7965 ** The new function scm_read_0str reads an s-expression from a
7966 null-terminated string, and returns it.
7968 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7969 to a Scheme port object.
7971 ** The new function `scm_set_program_arguments' allows C code to set
7972 the value returned by the Scheme `program-arguments' function.
7977 * Guile no longer includes sophisticated Tcl/Tk support.
7979 The old Tcl/Tk support was unsatisfying to us, because it required the
7980 user to link against the Tcl library, as well as Tk and Guile. The
7981 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7982 referring to widgets by names, rather than exporting widgets to Scheme
7983 code as a special datatype.
7985 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7986 maintainers described some very interesting changes in progress to the
7987 Tcl/Tk internals, which would facilitate clean interfaces between lone
7988 Tk and other interpreters --- even for garbage-collected languages
7989 like Scheme. They expected the new Tk to be publicly available in the
7992 Since it seems that Guile might soon have a new, cleaner interface to
7993 lone Tk, and that the old Guile/Tk glue code would probably need to be
7994 completely rewritten, we (Jim Blandy and Richard Stallman) have
7995 decided not to support the old code. We'll spend the time instead on
7996 a good interface to the newer Tk, as soon as it is available.
7998 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
8001 Copyright information:
8003 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
8005 Permission is granted to anyone to make or distribute verbatim copies
8006 of this document as received, in any medium, provided that the
8007 copyright notice and this permission notice are preserved,
8008 thus giving the recipient permission to redistribute in turn.
8010 Permission is granted to distribute modified versions
8011 of this document, or of portions of it,
8012 under the above conditions, provided also that they
8013 carry prominent notices stating who last changed them.
8018 paragraph-separate: "[
\f]*$"