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 1.9.3 (since the 1.9.2 prerelease):
13 ** Removed deprecated uniform array procedures: scm_make_uve,
14 scm_array_prototype, scm_list_to_uniform_array,
15 scm_dimensions_to_uniform_array, scm_make_ra, scm_shap2ra, scm_cvref,
16 scm_ra_set_contp, scm_aind, scm_raprin1
18 These functions have been deprecated since early 2005.
20 ** scm_array_p has one argument, not two
22 Use of the second argument produced a deprecation warning, so it is
23 unlikely that any code out there actually used this functionality.
25 ** Removed deprecated uniform array procedures:
26 dimensions->uniform-array, list->uniform-array, array-prototype
28 Instead, use make-typed-array, list->typed-array, or array-type,
31 ** And of course, the usual collection of bugfixes
33 Interested users should see the ChangeLog for more information.
35 * Changes to Scheme functions and syntax
37 ** New readline history functions
39 The (ice-9 readline) module now provides add-history, read-history,
40 write-history and clear-history, which wrap the corresponding GNU
41 History library functions.
44 Changes in 1.9.x (since the 1.8.x series):
46 * New modules (see the manual for details)
48 ** `(srfi srfi-18)', more sophisticated multithreading support
49 ** `(ice-9 i18n)', internationalization support
50 ** `(rnrs bytevector)', the R6RS bytevector API
51 ** `(rnrs io ports)', a subset of the R6RS I/O port API
52 ** `(system xref)', a cross-referencing facility (FIXME undocumented)
54 * Changes to the stand-alone interpreter
56 ** Guile now can compile Scheme to bytecode for a custom virtual machine.
58 Compiled code loads much faster than Scheme source code, and runs around
59 3 or 4 times as fast, generating much less garbage in the process.
61 ** The stack limit is now initialized from the environment.
63 If getrlimit(2) is available and a stack limit is set, Guile will set
64 its stack limit to 80% of the rlimit. Otherwise the limit is 160000
65 words, a four-fold increase from the earlier default limit.
67 ** New environment variables: GUILE_LOAD_COMPILED_PATH,
68 GUILE_SYSTEM_LOAD_COMPILED_PATH
70 GUILE_LOAD_COMPILED_PATH is for compiled files what GUILE_LOAD_PATH is
71 for source files. It is a different path, however, because compiled
72 files are architecture-specific. GUILE_SYSTEM_LOAD_COMPILED_PATH is like
75 ** New read-eval-print loop (REPL) implementation
77 Running Guile with no arguments drops the user into the new REPL. While
78 it is self-documenting to an extent, the new REPL has not yet been
79 documented in the manual. This will be fixed before 2.0.
81 ** New `guile-tools' commands: `compile', `disassemble'
83 Pass the `--help' command-line option to these commands for more
86 * Changes to Scheme functions and syntax
88 ** Procedure removed: `the-environment'
90 This procedure was part of the interpreter's execution model, and does
91 not apply to the compiler.
93 ** Files loaded with `primitive-load-path' will now be compiled
96 If a compiled .go file corresponding to a .scm file is not found or is
97 not fresh, the .scm file will be compiled on the fly, and the resulting
98 .go file stored away. An advisory note will be printed on the console.
100 Note that this mechanism depends on preservation of the .scm and .go
101 modification times; if the .scm or .go files are moved after
102 installation, care should be taken to preserve their original
105 Autocompiled files will be stored in the $XDG_CACHE_HOME/guile/ccache
106 directory, where $XDG_CACHE_HOME defaults to ~/.cache. This directory
107 will be created if needed.
109 To inhibit autocompilation, set the GUILE_AUTO_COMPILE environment
110 variable to 0, or pass --no-autocompile on the Guile command line.
112 Note that there is currently a bug here: automatic compilation will
113 sometimes be attempted when it shouldn't.
115 For example, the old (lang elisp) modules are meant to be interpreted,
116 not compiled. This bug will be fixed before 2.0. FIXME 2.0: Should say
117 something here about module-transformer called for compile.
119 ** New POSIX procedures: `getrlimit' and `setrlimit'
121 Note however that the interface of these functions is likely to change
122 in the next prerelease.
124 ** New procedure in `(oops goops)': `method-formals'
126 ** BUG: (procedure-property func 'arity) does not work on compiled
129 This will be fixed one way or another before 2.0.
131 ** New procedures in (ice-9 session): `add-value-help-handler!',
132 `remove-value-help-handler!', `add-name-help-handler!'
133 `remove-name-help-handler!', `procedure-arguments',
135 The value and name help handlers provide some minimal extensibility to
136 the help interface. Guile-lib's `(texinfo reflection)' uses them, for
137 example, to make stexinfo help documentation available. See those
138 procedures' docstrings for more information.
140 `procedure-arguments' describes the arguments that a procedure can take,
141 combining arity and formals. For example:
143 (procedure-arguments resolve-interface)
144 => ((required . (name)) (rest . args))
146 Additionally, `module-commentary' is now publically exported from
149 ** Deprecated: `procedure->memoizing-macro', `procedure->syntax'
151 These procedures will not work with syncase expansion, and indeed are
152 not used in the normal course of Guile. They are still used by the old
153 Emacs Lisp support, however.
155 ** New language: ECMAScript
157 Guile now ships with one other high-level language supported,
158 ECMAScript. The goal is to support all of version 3.1 of the standard,
159 but not all of the libraries are there yet. This support is not yet
160 documented; ask on the mailing list if you are interested.
162 ** New language: Brainfuck
164 Brainfuck is a toy language that closely models Turing machines. Guile's
165 brainfuck compiler is meant to be an example of implementing other
166 languages. See the manual for details, or
167 http://en.wikipedia.org/wiki/Brainfuck for more information about the
168 Brainfuck language itself.
170 ** Defmacros may now have docstrings.
172 Indeed, any macro may have a docstring. `object-documentation' from
173 `(ice-9 documentation)' may be used to retrieve the docstring, once you
174 have a macro value -- but see the above note about first-class macros.
175 Docstrings are associated with the syntax transformer procedures.
177 ** The psyntax expander now knows how to interpret the @ and @@ special
180 ** The psyntax expander is now hygienic with respect to modules.
182 Free variables in a macro are scoped in the module that the macro was
183 defined in, not in the module the macro is used in. For example, code
186 (define-module (foo) #:export (bar))
187 (define (helper x) ...)
189 (syntax-rules () ((_ x) (helper x))))
191 (define-module (baz) #:use-module (foo))
194 It used to be you had to export `helper' from `(foo)' as well.
195 Thankfully, this has been fixed.
197 ** New function, `procedure-module'
199 While useful on its own, `procedure-module' is used by psyntax on syntax
200 transformers to determine the module in which to scope introduced
203 ** `eval-case' has been deprecated, and replaced by `eval-when'.
205 The semantics of `eval-when' are easier to understand. It is still
206 missing documentation, however.
208 ** Guile is now more strict about prohibiting definitions in expression
211 Although previous versions of Guile accepted it, the following
212 expression is not valid, in R5RS or R6RS:
214 (if test (define foo 'bar) (define foo 'baz))
216 In this specific case, it would be better to do:
218 (define foo (if test 'bar 'baz))
220 It is certainly possible to circumvent this resriction with e.g.
221 `(module-define! (current-module) 'foo 'baz)'. We would appreciate
222 feedback about this change (a consequence of using psyntax as the
223 default expander), and may choose to revisit this situation before 2.0
224 in response to user feedback.
226 ** Defmacros must now produce valid Scheme expressions.
228 It used to be that defmacros could unquote in Scheme values, as a way of
229 supporting partial evaluation, and avoiding some hygiene issues. For
232 (define (helper x) ...)
233 (define-macro (foo bar)
236 Assuming this macro is in the `(baz)' module, the direct translation of
239 (define (helper x) ...)
240 (define-macro (foo bar)
241 `((@@ (baz) helper) ,bar))
243 Of course, one could just use a hygienic macro instead:
247 ((_ bar) (helper bar))))
249 ** Guile's psyntax now supports docstrings and internal definitions.
251 The following Scheme is not strictly legal:
258 However its intent is fairly clear. Guile interprets "bar" to be the
259 docstring of `foo', and the definition of `baz' is still in definition
262 ** Macros need to be defined before their first use.
264 It used to be that with lazy memoization, this might work:
268 (define-macro (ref x) x)
271 But now, the body of `foo' is interpreted to mean a call to the toplevel
272 `ref' function, instead of a macro expansion. The solution is to define
273 macros before code that uses them.
275 ** Functions needed by macros at expand-time need to be present at
278 For example, this code will work at the REPL:
280 (define (double-helper x) (* x x))
281 (define-macro (double-literal x) (double-helper x))
282 (double-literal 2) => 4
284 But it will not work when a file is compiled, because the definition of
285 `double-helper' is not present at expand-time. The solution is to wrap
286 the definition of `double-helper' in `eval-when':
288 (eval-when (load compile eval)
289 (define (double-helper x) (* x x)))
290 (define-macro (double-literal x) (double-helper x))
291 (double-literal 2) => 4
293 See the (currently missing) documentation for eval-when for more
296 ** New variable, %pre-modules-transformer
298 Need to document this one some more.
300 ** Temporarily removed functions: `macroexpand', `macroexpand-1'
302 `macroexpand' will be added back before 2.0. It is unclear how to
303 implement `macroexpand-1' with syntax-case, though PLT Scheme does prove
306 ** New reader macros: #' #` #, #,@
308 These macros translate, respectively, to `syntax', `quasisyntax',
309 `unsyntax', and `unsyntax-splicing'. See the R6RS for more information.
310 These reader macros may be overridden by `read-hash-extend'.
312 ** Incompatible change to #'
314 Guile did have a #' hash-extension, by default, which just returned the
315 subsequent datum: #'foo => foo. In the unlikely event that anyone
316 actually used this, this behavior may be reinstated via the
317 `read-hash-extend' mechanism.
319 ** Scheme expresssions may be commented out with #;
321 #; comments out an entire expression. See SRFI-62 or the R6RS for more
324 ** `make-stack' with a tail-called procedural narrowing argument no longer
325 works (with compiled procedures)
327 It used to be the case that a captured stack could be narrowed to select
328 calls only up to or from a certain procedure, even if that procedure
329 already tail-called another procedure. This was because the debug
330 information from the original procedure was kept on the stack.
332 Now with the new compiler, the stack only contains active frames from
333 the current continuation. A narrow to a procedure that is not in the
334 stack will result in an empty stack. To fix this, narrow to a procedure
335 that is active in the current continuation, or narrow to a specific
336 number of stack frames.
338 ** backtraces through compiled procedures only show procedures that are
339 active in the current continuation
341 Similarly to the previous issue, backtraces in compiled code may be
342 different from backtraces in interpreted code. There are no semantic
343 differences, however. Please mail bug-guile@gnu.org if you see any
344 deficiencies with Guile's backtraces.
346 ** syntax-rules and syntax-case macros now propagate source information
347 through to the expanded code
349 This should result in better backtraces.
351 ** The currying behavior of `define' has been removed.
353 Before, `(define ((f a) b) (* a b))' would translate to
355 (define f (lambda (a) (lambda (b) (* a b))))
357 Now a syntax error is signaled, as this syntax is not supported by
358 default. If there is sufficient demand, this syntax can be supported
361 ** All modules have names now
363 Before, you could have anonymous modules: modules without names. Now,
364 because of hygiene and macros, all modules have names. If a module was
365 created without a name, the first time `module-name' is called on it, a
366 fresh name will be lazily generated for it.
368 ** Many syntax errors have different texts now
370 Syntax errors still throw to the `syntax-error' key, but the arguments
371 are often different now. Perhaps in the future, Guile will switch to
372 using standard SRFI-35 conditions.
374 ** Returning multiple values to compiled code will silently truncate the
375 values to the expected number
377 For example, the interpreter would raise an error evaluating the form,
378 `(+ (values 1 2) (values 3 4))', because it would see the operands as
379 being two compound "values" objects, to which `+' does not apply.
381 The compiler, on the other hand, receives multiple values on the stack,
382 not as a compound object. Given that it must check the number of values
383 anyway, if too many values are provided for a continuation, it chooses
384 to truncate those values, effectively evaluating `(+ 1 3)' instead.
386 The idea is that the semantics that the compiler implements is more
387 intuitive, and the use of the interpreter will fade out with time.
388 This behavior is allowed both by the R5RS and the R6RS.
390 ** Multiple values in compiled code are not represented by compound
393 This change may manifest itself in the following situation:
395 (let ((val (foo))) (do-something) val)
397 In the interpreter, if `foo' returns multiple values, multiple values
398 are produced from the `let' expression. In the compiler, those values
399 are truncated to the first value, and that first value is returned. In
400 the compiler, if `foo' returns no values, an error will be raised, while
401 the interpreter would proceed.
403 Both of these behaviors are allowed by R5RS and R6RS. The compiler's
404 behavior is more correct, however. If you wish to preserve a potentially
405 multiply-valued return, you will need to set up a multiple-value
406 continuation, using `call-with-values'.
408 ** Defmacros are now implemented in terms of syntax-case.
410 The practical ramification of this is that the `defmacro?' predicate has
411 been removed, along with `defmacro-transformer', `macro-table',
412 `xformer-table', `assert-defmacro?!', `set-defmacro-transformer!' and
413 `defmacro:transformer'. This is because defmacros are simply macros. If
414 any of these procedures provided useful facilities to you, we encourage
415 you to contact the Guile developers.
417 ** psyntax is now the default expander
419 Scheme code is now expanded by default by the psyntax hygienic macro
420 expander. Expansion is performed completely before compilation or
423 Notably, syntax errors will be signalled before interpretation begins.
424 In the past, many syntax errors were only detected at runtime if the
425 code in question was memoized.
427 As part of its expansion, psyntax renames all lexically-bound
428 identifiers. Original identifier names are preserved and given to the
429 compiler, but the interpreter will see the renamed variables, e.g.,
430 `x432' instead of `x'.
432 Note that the psyntax that Guile uses is a fork, as Guile already had
433 modules before incompatible modules were added to psyntax -- about 10
434 years ago! Thus there are surely a number of bugs that have been fixed
435 in psyntax since then. If you find one, please notify bug-guile@gnu.org.
437 ** syntax-rules and syntax-case are available by default.
439 There is no longer any need to import the `(ice-9 syncase)' module
440 (which is now deprecated). The expander may be invoked directly via
441 `sc-expand', though it is normally searched for via the current module
444 Also, the helper routines for syntax-case are available in the default
445 environment as well: `syntax->datum', `datum->syntax',
446 `bound-identifier=?', `free-identifier=?', `generate-temporaries',
447 `identifier?', and `syntax-violation'. See the R6RS for documentation.
449 ** Lexical bindings introduced by hygienic macros may not be referenced
450 by nonhygienic macros.
452 If a lexical binding is introduced by a hygienic macro, it may not be
453 referenced by a nonhygienic macro. For example, this works:
456 (define-macro (bind-x val body)
457 `(let ((x ,val)) ,body))
458 (define-macro (ref x)
465 (define-syntax bind-x
467 ((_ val body) (let ((x val)) body))))
468 (define-macro (ref x)
472 It is not normal to run into this situation with existing code. However,
473 as code is ported over from defmacros to syntax-case, it is possible to
474 run into situations like this. In the future, Guile will probably port
475 its `while' macro to syntax-case, which makes this issue one to know
478 ** Macros may no longer be referenced as first-class values.
480 In the past, you could evaluate e.g. `if', and get its macro value. Now,
481 expanding this form raises a syntax error.
483 Macros still /exist/ as first-class values, but they must be
484 /referenced/ via the module system, e.g. `(module-ref (current-module)
487 This decision may be revisited before the 2.0 release. Feedback welcome
488 to guile-devel@gnu.org (subscription required) or bug-guile@gnu.org (no
489 subscription required).
491 ** Unicode characters
493 Unicode characters may be entered in octal format via e.g. `#\454', or
494 created via (integer->char 300). A hex external representation will
495 probably be introduced at some point.
499 Internally, strings are now represented either in the `latin-1'
500 encoding, one byte per character, or in UTF-32, with four bytes per
501 character. Strings manage their own allocation, switching if needed.
503 Currently no locale conversion is performed. Extended characters may be
504 written in a string using the hexadecimal escapes `\xXX', `\uXXXX', or
505 `\UXXXXXX', for 8-bit, 16-bit, or 24-bit codepoints, respectively.
507 ** Global variables `scm_charnames' and `scm_charnums' are removed
509 These variables contained the names of control characters and were
510 used when writing characters. While these were global, they were
511 never intended to be public API. They have been replaced with private
514 ** EBCDIC support is removed
516 There was an EBCDIC compile flag that altered some of the character
517 processing. It appeared that full EBCDIC support was never completed
518 and was unmaintained.
520 ** New macro type: syncase-macro
522 XXX Need to decide whether to document this for 2.0, probably should:
523 make-syncase-macro, make-extended-syncase-macro, macro-type,
524 syncase-macro-type, syncase-macro-binding
526 ** A new `memoize-symbol' evaluator trap has been added.
528 This trap can be used for efficiently implementing a Scheme code
531 ** Duplicate bindings among used modules are resolved lazily.
533 This slightly improves program startup times.
535 ** New thread cancellation and thread cleanup API
537 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
539 ** Fix bad interaction between `false-if-exception' and stack-call.
541 Exceptions thrown by `false-if-exception' were erronously causing the
542 stack to be saved, causing later errors to show the incorrectly-saved
543 backtrace. This has been fixed.
545 ** New global variables: %load-compiled-path, %load-compiled-extensions
547 These are analogous to %load-path and %load-extensions.
549 ** New procedure, `make-promise'
551 `(make-promise (lambda () foo))' is equivalent to `(delay foo)'.
553 ** `defined?' may accept a module as its second argument
555 Previously it only accepted internal structures from the evaluator.
557 ** New entry into %guile-build-info: `ccachedir'
559 ** Fix bug in `module-bound?'.
561 `module-bound?' was returning true if a module did have a local
562 variable, but one that was unbound, but another imported module bound
563 the variable. This was an error, and was fixed.
565 ** `(ice-9 syncase)' has been deprecated.
567 As syntax-case is available by default, importing `(ice-9 syncase)' has
568 no effect, and will trigger a deprecation warning.
570 ** Removed deprecated uniform array procedures:
571 dimensions->uniform-array, list->uniform-array, array-prototype
573 Instead, use make-typed-array, list->typed-array, or array-type,
576 * Changes to the C interface
578 ** The GH interface (deprecated in version 1.6, 2001) was removed.
580 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
582 This makes these internal functions technically not callable from
585 ** Functions for handling `scm_option' now no longer require an argument
586 indicating length of the `scm_t_option' array.
588 ** scm_primitive_load_path has additional argument, exception_on_error
590 ** New C function: scm_module_public_interface
592 This procedure corresponds to Scheme's `module-public-interface'.
594 ** `scm_stat' has an additional argument, `exception_on_error'
595 ** `scm_primitive_load_path' has an additional argument `exception_on_not_found'
597 ** `scm_set_port_seek' and `scm_set_port_truncate' use the `scm_t_off' type
599 Previously they would use the `off_t' type, which is fragile since its
600 definition depends on the application's value for `_FILE_OFFSET_BITS'.
602 ** The `long_long' C type, deprecated in 1.8, has been removed
604 ** Removed deprecated uniform array procedures: scm_make_uve,
605 scm_array_prototype, scm_list_to_uniform_array,
606 scm_dimensions_to_uniform_array, scm_make_ra, scm_shap2ra, scm_cvref,
607 scm_ra_set_contp, scm_aind, scm_raprin1
609 These functions have been deprecated since early 2005.
611 ** scm_array_p has one argument, not two
613 Use of the second argument produced a deprecation warning, so it is
614 unlikely that any code out there actually used this functionality.
616 * Changes to the distribution
618 ** Guile's license is now LGPLv3+
620 In other words the GNU Lesser General Public License, version 3 or
621 later (at the discretion of each person that chooses to redistribute
624 ** `guile-config' will be deprecated in favor of `pkg-config'
626 `guile-config' has been rewritten to get its information from
627 `pkg-config', so this should be a transparent change. Note however that
628 guile.m4 has yet to be modified to call pkg-config instead of
631 ** Guile now provides `guile-2.0.pc' instead of `guile-1.8.pc'
633 Programs that use `pkg-config' to find Guile or one of its Autoconf
634 macros should now require `guile-2.0' instead of `guile-1.8'.
636 ** New installation directory: $(pkglibdir)/1.9/ccache
638 If $(libdir) is /usr/lib, for example, Guile will install its .go files
639 to /usr/lib/guile/1.9/ccache. These files are architecture-specific.
641 ** New dependency: GNU libunistring.
643 See http://www.gnu.org/software/libunistring/, for more information. Our
644 unicode support uses routines from libunistring.
648 Changes in 1.8.8 (since 1.8.7)
652 ** Fix possible buffer overruns when parsing numbers
653 ** Avoid clash with system setjmp/longjmp on IA64
656 Changes in 1.8.7 (since 1.8.6)
658 * New modules (see the manual for details)
660 ** `(srfi srfi-98)', an interface to access environment variables
664 ** Fix compilation with `--disable-deprecated'
665 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
666 ** Fix MinGW build problem caused by HAVE_STRUCT_TIMESPEC confusion
667 ** Fix build problem when scm_t_timespec is different from struct timespec
668 ** Fix build when compiled with -Wundef -Werror
669 ** More build fixes for `alphaev56-dec-osf5.1b' (Tru64)
670 ** Build fixes for `powerpc-ibm-aix5.3.0.0' (AIX 5.3)
671 ** With GCC, always compile with `-mieee' on `alpha*' and `sh*'
672 ** Better diagnose broken `(strftime "%z" ...)' in `time.test' (bug #24130)
673 ** Fix parsing of SRFI-88/postfix keywords longer than 128 characters
674 ** Fix reading of complex numbers where both parts are inexact decimals
676 ** Allow @ macro to work with (ice-9 syncase)
678 Previously, use of the @ macro in a module whose code is being
679 transformed by (ice-9 syncase) would cause an "Invalid syntax" error.
680 Now it works as you would expect (giving the value of the specified
683 ** Have `scm_take_locale_symbol ()' return an interned symbol (bug #25865)
686 Changes in 1.8.6 (since 1.8.5)
688 * New features (see the manual for details)
690 ** New convenience function `scm_c_symbol_length ()'
692 ** Single stepping through code from Emacs
694 When you use GDS to evaluate Scheme code from Emacs, you can now use
695 `C-u' to indicate that you want to single step through that code. See
696 `Evaluating Scheme Code' in the manual for more details.
698 ** New "guile(1)" man page!
700 * Changes to the distribution
702 ** Automake's `AM_MAINTAINER_MODE' is no longer used
704 Thus, the `--enable-maintainer-mode' configure option is no longer
705 available: Guile is now always configured in "maintainer mode".
707 ** `ChangeLog' files are no longer updated
709 Instead, changes are detailed in the version control system's logs. See
710 the top-level `ChangeLog' files for details.
715 ** `symbol->string' now returns a read-only string, as per R5RS
716 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
717 ** `guile-config link' now prints `-L$libdir' before `-lguile'
718 ** Fix memory corruption involving GOOPS' `class-redefinition'
719 ** Fix possible deadlock in `mutex-lock'
720 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
721 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
722 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
723 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
724 ** Fix misleading output from `(help rationalize)'
725 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
726 ** Fix `gcd' when called with a single, negative argument.
727 ** Fix `Stack overflow' errors seen when building on some platforms
728 ** Fix bug when `scm_with_guile ()' was called several times from the
730 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
731 dynamic environment of the call to `raise'
732 ** Fix potential deadlock in `make-struct'
733 ** Fix compilation problem with libltdl from Libtool 2.2.x
734 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
737 Changes in 1.8.5 (since 1.8.4)
739 * Infrastructure changes
741 ** Guile repository switched from CVS to Git
743 The new repository can be accessed using
744 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
745 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
747 ** Add support for `pkg-config'
749 See "Autoconf Support" in the manual for details.
751 * New modules (see the manual for details)
755 * New features (see the manual for details)
757 ** New `postfix' read option, for SRFI-88 keyword syntax
758 ** Some I/O primitives have been inlined, which improves I/O performance
759 ** New object-based traps infrastructure
761 This is a GOOPS-based infrastructure that builds on Guile's low-level
762 evaluator trap calls and facilitates the development of debugging
763 features like single-stepping, breakpoints, tracing and profiling.
764 See the `Traps' node of the manual for details.
766 ** New support for working on Guile code from within Emacs
768 Guile now incorporates the `GDS' library (previously distributed
769 separately) for working on Guile code from within Emacs. See the
770 `Using Guile In Emacs' node of the manual for details.
774 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
775 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
777 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
778 would trigger an unbound variable error for `match:andmap'.
780 ** `(oop goops describe)' now properly provides the `describe' feature
781 ** Fixed `args-fold' from `(srfi srfi-37)'
783 Previously, parsing short option names of argument-less options would
784 lead to a stack overflow.
786 ** `(srfi srfi-35)' is now visible through `cond-expand'
787 ** Fixed type-checking for the second argument of `eval'
788 ** Fixed type-checking for SRFI-1 `partition'
789 ** Fixed `struct-ref' and `struct-set!' on "light structs"
790 ** Honor struct field access rights in GOOPS
791 ** Changed the storage strategy of source properties, which fixes a deadlock
792 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
793 ** Fixed build issue for GNU/Linux on IA64
794 ** Fixed build issues on NetBSD 1.6
795 ** Fixed build issue on Solaris 2.10 x86_64
796 ** Fixed build issue with DEC/Compaq/HP's compiler
797 ** Fixed `scm_from_complex_double' build issue on FreeBSD
798 ** Fixed `alloca' build issue on FreeBSD 6
799 ** Removed use of non-portable makefile constructs
800 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
801 ** Make sure all tests honor `$TMPDIR'
804 Changes in 1.8.4 (since 1.8.3)
808 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
809 ** Fixed a segmentation fault which occurred when displaying the
810 backtrace of a stack with a promise object (made by `delay') in it.
811 ** Make `accept' leave guile mode while blocking
812 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
813 ** Fixed a build problem on AIX (use of func_data identifier)
814 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
815 called with an associator proc that returns neither a pair nor #f.
816 ** Secondary threads now always return a valid module for (current-module).
817 ** Avoid MacOS build problems caused by incorrect combination of "64"
818 system and library calls.
819 ** `guile-snarf' now honors `$TMPDIR'
820 ** `guile-config compile' now reports CPPFLAGS used at compile-time
821 ** Fixed build with Sun Studio (Solaris 9)
822 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
823 uniform vectors on AIX.
824 ** Fixed a deadlock that occurs upon GC with multiple threads.
825 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
826 ** Fixed autotool-derived build problems on AIX 6.1.
827 ** Fixed NetBSD/alpha support
828 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
830 * New modules (see the manual for details)
834 * Documentation fixes and improvements
836 ** Removed premature breakpoint documentation
838 The features described are not available in the series of 1.8.x
839 releases, so the documentation was misleading and has been removed.
841 ** More about Guile's default *random-state* variable
843 ** GOOPS: more about how to use `next-method'
845 * Changes to the distribution
847 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
849 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
850 General Public License, and the few incorrect files have now been
851 fixed to agree with the rest of the Guile distribution.
853 ** Removed unnecessary extra copies of COPYING*
855 The distribution now contains a single COPYING.LESSER at its top level.
858 Changes in 1.8.3 (since 1.8.2)
860 * New modules (see the manual for details)
867 ** The `(ice-9 slib)' module now works as expected
868 ** Expressions like "(set! 'x #t)" no longer yield a crash
869 ** Warnings about duplicate bindings now go to stderr
870 ** A memory leak in `make-socket-address' was fixed
871 ** Alignment issues (e.g., on SPARC) in network routines were fixed
872 ** A threading issue that showed up at least on NetBSD was fixed
873 ** Build problems on Solaris and IRIX fixed
875 * Implementation improvements
877 ** The reader is now faster, which reduces startup time
878 ** Procedures returned by `record-accessor' and `record-modifier' are faster
881 Changes in 1.8.2 (since 1.8.1):
883 * New procedures (see the manual for details)
885 ** set-program-arguments
888 * Incompatible changes
890 ** The body of a top-level `define' no longer sees the binding being created
892 In a top-level `define', the binding being created is no longer visible
893 from the `define' body. This breaks code like
894 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
895 unbound in the body. However, such code was not R5RS-compliant anyway,
900 ** Fractions were not `equal?' if stored in unreduced form.
901 (A subtle problem, since printing a value reduced it, making it work.)
902 ** srfi-60 `copy-bit' failed on 64-bit systems
903 ** "guile --use-srfi" option at the REPL can replace core functions
904 (Programs run with that option were ok, but in the interactive REPL
905 the core bindings got priority, preventing SRFI replacements or
907 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
908 ** `kill' on mingw throws an error for a PID other than oneself
909 ** Procedure names are attached to procedure-with-setters
910 ** Array read syntax works with negative lower bound
911 ** `array-in-bounds?' fix if an array has different lower bounds on each index
912 ** `*' returns exact 0 for "(* inexact 0)"
913 This follows what it always did for "(* 0 inexact)".
914 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
915 ** SRFI-19: `date->julian-day' did not account for timezone offset
916 ** `ttyname' no longer crashes when passed a non-tty argument
917 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
918 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
919 ** GOOPS: Fixed a bug in `method-more-specific?'
920 ** Build problems on Solaris fixed
921 ** Build problems on HP-UX IA64 fixed
922 ** Build problems on MinGW fixed
925 Changes in 1.8.1 (since 1.8.0):
927 * LFS functions are now used to access 64-bit files on 32-bit systems.
929 * New procedures (see the manual for details)
931 ** primitive-_exit - [Scheme] the-root-module
932 ** scm_primitive__exit - [C]
933 ** make-completion-function - [Scheme] (ice-9 readline)
934 ** scm_c_locale_stringn_to_number - [C]
935 ** scm_srfi1_append_reverse [C]
936 ** scm_srfi1_append_reverse_x [C]
944 ** Build problems have been fixed on MacOS, SunOS, and QNX.
946 ** `strftime' fix sign of %z timezone offset.
948 ** A one-dimensional array can now be 'equal?' to a vector.
950 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
952 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
954 ** `record-accessor' and `record-modifier' now have strict type checks.
956 Record accessor and modifier procedures now throw an error if the
957 record type of the record they're given is not the type expected.
958 (Previously accessors returned #f and modifiers silently did nothing).
960 ** It is now OK to use both autoload and use-modules on a given module.
962 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
964 Previously there was no checking on primatives like make-vector that
965 accept "one or two" arguments. Now there is.
967 ** The srfi-1 assoc function now calls its equality predicate properly.
969 Previously srfi-1 assoc would call the equality predicate with the key
970 last. According to the SRFI, the key should be first.
972 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
974 ** The array-set! procedure no longer segfaults when given a bit vector.
976 ** Bugs in make-shared-array have been fixed.
978 ** string<? and friends now follow char<? etc order on 8-bit chars.
980 ** The format procedure now handles inf and nan values for ~f correctly.
982 ** exact->inexact should no longer overflow when given certain large fractions.
984 ** srfi-9 accessor and modifier procedures now have strict record type checks.
986 This matches the srfi-9 specification.
988 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
990 Previously the (ice-9 ftw) procedures would ignore any file that had
991 the same inode number as a file they had already seen, even if that
992 file was on a different device.
995 Changes in 1.8.0 (changes since the 1.6.x series):
997 * Changes to the distribution
999 ** Guile is now licensed with the GNU Lesser General Public License.
1001 ** The manual is now licensed with the GNU Free Documentation License.
1003 ** Guile now requires GNU MP (http://swox.com/gmp).
1005 Guile now uses the GNU MP library for arbitrary precision arithmetic.
1007 ** Guile now has separate private and public configuration headers.
1009 That is, things like HAVE_STRING_H no longer leak from Guile's
1012 ** Guile now provides and uses an "effective" version number.
1014 Guile now provides scm_effective_version and effective-version
1015 functions which return the "effective" version number. This is just
1016 the normal full version string without the final micro-version number,
1017 so the current effective-version is "1.8". The effective version
1018 should remain unchanged during a stable series, and should be used for
1019 items like the versioned share directory name
1020 i.e. /usr/share/guile/1.8.
1022 Providing an unchanging version number during a stable release for
1023 things like the versioned share directory can be particularly
1024 important for Guile "add-on" packages, since it provides a directory
1025 that they can install to that won't be changed out from under them
1026 with each micro release during a stable series.
1028 ** Thread implementation has changed.
1030 When you configure "--with-threads=null", you will get the usual
1031 threading API (call-with-new-thread, make-mutex, etc), but you can't
1032 actually create new threads. Also, "--with-threads=no" is now
1033 equivalent to "--with-threads=null". This means that the thread API
1034 is always present, although you might not be able to create new
1037 When you configure "--with-threads=pthreads" or "--with-threads=yes",
1038 you will get threads that are implemented with the portable POSIX
1039 threads. These threads can run concurrently (unlike the previous
1040 "coop" thread implementation), but need to cooperate for things like
1043 The default is "pthreads", unless your platform doesn't have pthreads,
1044 in which case "null" threads are used.
1046 See the manual for details, nodes "Initialization", "Multi-Threading",
1047 "Blocking", and others.
1049 ** There is the new notion of 'discouraged' features.
1051 This is a milder form of deprecation.
1053 Things that are discouraged should not be used in new code, but it is
1054 OK to leave them in old code for now. When a discouraged feature is
1055 used, no warning message is printed like there is for 'deprecated'
1056 features. Also, things that are merely discouraged are nevertheless
1057 implemented efficiently, while deprecated features can be very slow.
1059 You can omit discouraged features from libguile by configuring it with
1060 the '--disable-discouraged' option.
1062 ** Deprecation warnings can be controlled at run-time.
1064 (debug-enable 'warn-deprecated) switches them on and (debug-disable
1065 'warn-deprecated) switches them off.
1067 ** Support for SRFI 61, extended cond syntax for multiple values has
1070 This SRFI is always available.
1072 ** Support for require-extension, SRFI-55, has been added.
1074 The SRFI-55 special form `require-extension' has been added. It is
1075 available at startup, and provides a portable way to load Scheme
1076 extensions. SRFI-55 only requires support for one type of extension,
1077 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
1080 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
1082 The (srfi srfi-26) module is an implementation of SRFI-26 which
1083 provides the `cut' and `cute' syntax. These may be used to specialize
1084 parameters without currying.
1086 ** New module (srfi srfi-31)
1088 This is an implementation of SRFI-31 which provides a special form
1089 `rec' for recursive evaluation.
1091 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
1092 been merged with the core, making their functionality always
1095 The modules are still available, tho, and you could use them together
1096 with a renaming import, for example.
1098 ** Guile no longer includes its own version of libltdl.
1100 The official version is good enough now.
1102 ** The --enable-htmldoc option has been removed from 'configure'.
1104 Support for translating the documentation into HTML is now always
1105 provided. Use 'make html'.
1107 ** New module (ice-9 serialize):
1109 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
1110 don't trust the thread safety of most of your program, but where you
1111 have some section(s) of code which you consider can run in parallel to
1112 other sections. See ice-9/serialize.scm for more information.
1114 ** The configure option '--disable-arrays' has been removed.
1116 Support for arrays and uniform numeric arrays is now always included
1119 * Changes to the stand-alone interpreter
1121 ** New command line option `-L'.
1123 This option adds a directory to the front of the load path.
1125 ** New command line option `--no-debug'.
1127 Specifying `--no-debug' on the command line will keep the debugging
1128 evaluator turned off, even for interactive sessions.
1130 ** User-init file ~/.guile is now loaded with the debugging evaluator.
1132 Previously, the normal evaluator would have been used. Using the
1133 debugging evaluator gives better error messages.
1135 ** The '-e' option now 'read's its argument.
1137 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
1138 be used with '-e'. For example, you can now write a script like
1141 exec guile -e '(@ (demo) main)' -s "$0" "$@"
1144 (define-module (demo)
1148 (format #t "Demo: ~a~%" args))
1151 * Changes to Scheme functions and syntax
1153 ** Guardians have changed back to their original semantics
1155 Guardians now behave like described in the paper by Dybvig et al. In
1156 particular, they no longer make guarantees about the order in which
1157 they return objects, and they can no longer be greedy.
1159 They no longer drop cyclic data structures.
1161 The C function scm_make_guardian has been changed incompatibly and no
1162 longer takes the 'greedy_p' argument.
1164 ** New function hashx-remove!
1166 This function completes the set of 'hashx' functions.
1168 ** The concept of dynamic roots has been factored into continuation
1169 barriers and dynamic states.
1171 Each thread has a current dynamic state that carries the values of the
1172 fluids. You can create and copy dynamic states and use them as the
1173 second argument for 'eval'. See "Fluids and Dynamic States" in the
1176 To restrict the influence that captured continuations can have on the
1177 control flow, you can errect continuation barriers. See "Continuation
1178 Barriers" in the manual.
1180 The function call-with-dynamic-root now essentially temporarily
1181 installs a new dynamic state and errects a continuation barrier.
1183 ** The default load path no longer includes "." at the end.
1185 Automatically loading modules from the current directory should not
1186 happen by default. If you want to allow it in a more controlled
1187 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
1188 variable %load-path.
1190 ** The uniform vector and array support has been overhauled.
1192 It now complies with SRFI-4 and the weird prototype based uniform
1193 array creation has been deprecated. See the manual for more details.
1195 Some non-compatible changes have been made:
1196 - characters can no longer be stored into byte arrays.
1197 - strings and bit vectors are no longer considered to be uniform numeric
1199 - array-rank throws an error for non-arrays instead of returning zero.
1200 - array-ref does no longer accept non-arrays when no indices are given.
1202 There is the new notion of 'generalized vectors' and corresponding
1203 procedures like 'generalized-vector-ref'. Generalized vectors include
1204 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
1206 Arrays use generalized vectors as their storage, so that you still
1207 have arrays of characters, bits, etc. However, uniform-array-read!
1208 and uniform-array-write can no longer read/write strings and
1211 ** There is now support for copy-on-write substrings, mutation-sharing
1212 substrings and read-only strings.
1214 Three new procedures are related to this: substring/shared,
1215 substring/copy, and substring/read-only. See the manual for more
1218 ** Backtraces will now highlight the value that caused the error.
1220 By default, these values are enclosed in "{...}", such as in this
1229 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
1230 <unnamed port>:1:1: Wrong type (expecting pair): a
1231 ABORT: (wrong-type-arg)
1233 The prefix and suffix used for highlighting can be set via the two new
1234 printer options 'highlight-prefix' and 'highlight-suffix'. For
1235 example, putting this into ~/.guile will output the bad value in bold
1236 on an ANSI terminal:
1238 (print-set! highlight-prefix "\x1b[1m")
1239 (print-set! highlight-suffix "\x1b[22m")
1242 ** 'gettext' support for internationalization has been added.
1244 See the manual for details.
1246 ** New syntax '@' and '@@':
1248 You can now directly refer to variables exported from a module by
1251 (@ MODULE-NAME VARIABLE-NAME)
1253 For example (@ (ice-9 pretty-print) pretty-print) will directly access
1254 the pretty-print variable exported from the (ice-9 pretty-print)
1255 module. You don't need to 'use' that module first. You can also use
1256 '@' as a target of 'set!', as in (set! (@ mod var) val).
1258 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
1259 but it can also access variables that have not been exported. It is
1260 intended only for kluges and temporary fixes and for debugging, not
1263 ** Keyword syntax has been made more disciplined.
1265 Previously, the name of a keyword was read as a 'token' but printed as
1266 a symbol. Now, it is read as a general Scheme datum which must be a
1277 ERROR: In expression (a b c):
1281 ERROR: Unbound variable: foo
1286 ERROR: Wrong type (expecting symbol): 12
1290 ERROR: Wrong type (expecting symbol): (a b c)
1294 ** The printing of symbols that might look like keywords can be
1297 The new printer option 'quote-keywordish-symbols' controls how symbols
1298 are printed that have a colon as their first or last character. The
1299 default now is to only quote a symbol with #{...}# when the read
1300 option 'keywords' is not '#f'. Thus:
1302 guile> (define foo (string->symbol ":foo"))
1303 guile> (read-set! keywords #f)
1306 guile> (read-set! keywords 'prefix)
1309 guile> (print-set! quote-keywordish-symbols #f)
1313 ** 'while' now provides 'break' and 'continue'
1315 break and continue were previously bound in a while loop, but not
1316 documented, and continue didn't quite work properly. The undocumented
1317 parameter to break which gave a return value for the while has been
1320 ** 'call-with-current-continuation' is now also available under the name
1323 ** The module system now checks for duplicate bindings.
1325 The module system now can check for name conflicts among imported
1328 The behavior can be controlled by specifying one or more 'duplicates'
1329 handlers. For example, to make Guile return an error for every name
1332 (define-module (foo)
1337 The new default behavior of the module system when a name collision
1338 has been detected is to
1340 1. Give priority to bindings marked as a replacement.
1341 2. Issue a warning (different warning if overriding core binding).
1342 3. Give priority to the last encountered binding (this corresponds to
1345 If you want the old behavior back without replacements or warnings you
1348 (default-duplicate-binding-handler 'last)
1350 to your .guile init file.
1352 ** New define-module option: :replace
1354 :replace works as :export, but, in addition, marks the binding as a
1357 A typical example is `format' in (ice-9 format) which is a replacement
1358 for the core binding `format'.
1360 ** Adding prefixes to imported bindings in the module system
1362 There is now a new :use-module option :prefix. It can be used to add
1363 a prefix to all imported bindings.
1365 (define-module (foo)
1366 :use-module ((bar) :prefix bar:))
1368 will import all bindings exported from bar, but rename them by adding
1371 ** Conflicting generic functions can be automatically merged.
1373 When two imported bindings conflict and they are both generic
1374 functions, the two functions can now be merged automatically. This is
1375 activated with the 'duplicates' handler 'merge-generics'.
1377 ** New function: effective-version
1379 Returns the "effective" version number. This is just the normal full
1380 version string without the final micro-version number. See "Changes
1381 to the distribution" above.
1383 ** New threading functions: parallel, letpar, par-map, and friends
1385 These are convenient ways to run calculations in parallel in new
1386 threads. See "Parallel forms" in the manual for details.
1388 ** New function 'try-mutex'.
1390 This function will attempt to lock a mutex but will return immediately
1391 instead of blocking and indicate failure.
1393 ** Waiting on a condition variable can have a timeout.
1395 The function 'wait-condition-variable' now takes a third, optional
1396 argument that specifies the point in time where the waiting should be
1399 ** New function 'broadcast-condition-variable'.
1401 ** New functions 'all-threads' and 'current-thread'.
1403 ** Signals and system asyncs work better with threads.
1405 The function 'sigaction' now takes a fourth, optional, argument that
1406 specifies the thread that the handler should run in. When the
1407 argument is omitted, the handler will run in the thread that called
1410 Likewise, 'system-async-mark' takes a second, optional, argument that
1411 specifies the thread that the async should run in. When it is
1412 omitted, the async will run in the thread that called
1413 'system-async-mark'.
1415 C code can use the new functions scm_sigaction_for_thread and
1416 scm_system_async_mark_for_thread to pass the new thread argument.
1418 When a thread blocks on a mutex, a condition variable or is waiting
1419 for IO to be possible, it will still execute system asyncs. This can
1420 be used to interrupt such a thread by making it execute a 'throw', for
1423 ** The function 'system-async' is deprecated.
1425 You can now pass any zero-argument procedure to 'system-async-mark'.
1426 The function 'system-async' will just return its argument unchanged
1429 ** New functions 'call-with-blocked-asyncs' and
1430 'call-with-unblocked-asyncs'
1432 The expression (call-with-blocked-asyncs PROC) will call PROC and will
1433 block execution of system asyncs for the current thread by one level
1434 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
1435 procedure and will unblock the execution of system asyncs by one
1436 level for the current thread.
1438 Only system asyncs are affected by these functions.
1440 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
1442 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
1443 instead. Those functions are easier to use correctly and can be
1446 ** New function 'unsetenv'.
1448 ** New macro 'define-syntax-public'.
1450 It works like 'define-syntax' and also exports the defined macro (but
1453 ** There is support for Infinity and NaNs.
1455 Following PLT Scheme, Guile can now work with infinite numbers, and
1458 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
1459 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
1460 "+nan.0"). These numbers are inexact and have no exact counterpart.
1462 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
1463 sign of the dividend. The infinities are integers, and they answer #t
1464 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
1465 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
1476 ERROR: Numerical overflow
1478 Two new predicates 'inf?' and 'nan?' can be used to test for the
1481 ** Inexact zero can have a sign.
1483 Guile can now distinguish between plus and minus inexact zero, if your
1484 platform supports this, too. The two zeros are equal according to
1485 '=', but not according to 'eqv?'. For example
1496 ** Guile now has exact rationals.
1498 Guile can now represent fractions such as 1/3 exactly. Computing with
1499 them is also done exactly, of course:
1504 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
1505 for exact arguments.
1507 For example: (floor 2) now returns an exact 2 where in the past it
1508 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
1510 ** inexact->exact no longer returns only integers.
1512 Without exact rationals, the closest exact number was always an
1513 integer, but now inexact->exact returns the fraction that is exactly
1514 equal to a floating point number. For example:
1516 (inexact->exact 1.234)
1517 => 694680242521899/562949953421312
1519 When you want the old behavior, use 'round' explicitly:
1521 (inexact->exact (round 1.234))
1524 ** New function 'rationalize'.
1526 This function finds a simple fraction that is close to a given real
1527 number. For example (and compare with inexact->exact above):
1529 (rationalize (inexact->exact 1.234) 1/2000)
1532 Note that, as required by R5RS, rationalize returns only then an exact
1533 result when both its arguments are exact.
1535 ** 'odd?' and 'even?' work also for inexact integers.
1537 Previously, (odd? 1.0) would signal an error since only exact integers
1538 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
1539 returns #f and (odd? 1.5) signals an error.
1541 ** Guile now has uninterned symbols.
1543 The new function 'make-symbol' will return an uninterned symbol. This
1544 is a symbol that is unique and is guaranteed to remain unique.
1545 However, uninterned symbols can not yet be read back in.
1547 Use the new function 'symbol-interned?' to check whether a symbol is
1550 ** pretty-print has more options.
1552 The function pretty-print from the (ice-9 pretty-print) module can now
1553 also be invoked with keyword arguments that control things like
1554 maximum output width. See the manual for details.
1556 ** Variables have no longer a special behavior for `equal?'.
1558 Previously, comparing two variables with `equal?' would recursivly
1559 compare their values. This is no longer done. Variables are now only
1560 `equal?' if they are `eq?'.
1562 ** `(begin)' is now valid.
1564 You can now use an empty `begin' form. It will yield #<unspecified>
1565 when evaluated and simply be ignored in a definition context.
1567 ** Deprecated: procedure->macro
1569 Change your code to use 'define-macro' or r5rs macros. Also, be aware
1570 that macro expansion will not be done during evaluation, but prior to
1573 ** Soft ports now allow a `char-ready?' procedure
1575 The vector argument to `make-soft-port' can now have a length of
1576 either 5 or 6. (Previously the length had to be 5.) The optional 6th
1577 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
1578 that returns the number of characters that can be read immediately
1579 without the soft port blocking.
1581 ** Deprecated: undefine
1583 There is no replacement for undefine.
1585 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
1586 have been discouraged.
1588 They are relics from a time where a keyword like #:foo was used
1589 directly as a Tcl option "-foo" and thus keywords were internally
1590 stored as a symbol with a starting dash. We now store a symbol
1593 Use symbol->keyword and keyword->symbol instead.
1595 ** The `cheap' debug option is now obsolete
1597 Evaluator trap calls are now unconditionally "cheap" - in other words,
1598 they pass a debug object to the trap handler rather than a full
1599 continuation. The trap handler code can capture a full continuation
1600 by using `call-with-current-continuation' in the usual way, if it so
1603 The `cheap' option is retained for now so as not to break existing
1604 code which gets or sets it, but setting it now has no effect. It will
1605 be removed in the next major Guile release.
1607 ** Evaluator trap calls now support `tweaking'
1609 `Tweaking' means that the trap handler code can modify the Scheme
1610 expression that is about to be evaluated (in the case of an
1611 enter-frame trap) or the value that is being returned (in the case of
1612 an exit-frame trap). The trap handler code indicates that it wants to
1613 do this by returning a pair whose car is the symbol 'instead and whose
1614 cdr is the modified expression or return value.
1616 * Changes to the C interface
1618 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
1619 take a 'delete' function argument.
1621 This argument makes no sense since the delete function is used to
1622 remove a pair from an alist, and this must not be configurable.
1624 This is an incompatible change.
1626 ** The GH interface is now subject to the deprecation mechanism
1628 The GH interface has been deprecated for quite some time but now it is
1629 actually removed from Guile when it is configured with
1630 --disable-deprecated.
1632 See the manual "Transitioning away from GH" for more information.
1634 ** A new family of functions for converting between C values and
1635 Scheme values has been added.
1637 These functions follow a common naming scheme and are designed to be
1638 easier to use, thread-safe and more future-proof than the older
1641 - int scm_is_* (...)
1643 These are predicates that return a C boolean: 1 or 0. Instead of
1644 SCM_NFALSEP, you can now use scm_is_true, for example.
1646 - <type> scm_to_<type> (SCM val, ...)
1648 These are functions that convert a Scheme value into an appropriate
1649 C value. For example, you can use scm_to_int to safely convert from
1652 - SCM scm_from_<type> (<type> val, ...)
1654 These functions convert from a C type to a SCM value; for example,
1655 scm_from_int for ints.
1657 There is a huge number of these functions, for numbers, strings,
1658 symbols, vectors, etc. They are documented in the reference manual in
1659 the API section together with the types that they apply to.
1661 ** New functions for dealing with complex numbers in C have been added.
1663 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1664 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1665 They work like scm_make_rectangular etc but take or return doubles
1668 ** The function scm_make_complex has been discouraged.
1670 Use scm_c_make_rectangular instead.
1672 ** The INUM macros have been deprecated.
1674 A lot of code uses these macros to do general integer conversions,
1675 although the macros only work correctly with fixnums. Use the
1676 following alternatives.
1678 SCM_INUMP -> scm_is_integer or similar
1679 SCM_NINUMP -> !scm_is_integer or similar
1680 SCM_MAKINUM -> scm_from_int or similar
1681 SCM_INUM -> scm_to_int or similar
1683 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1684 do the validating for you.
1686 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1687 have been discouraged.
1689 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1690 new code. The functions have been discouraged since they don't fit
1693 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1695 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1696 evaluates its argument twice. Use scm_is_true, etc. instead for new
1699 ** The macro SCM_EQ_P has been discouraged.
1701 Use scm_is_eq for new code, which fits better into the naming
1704 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1707 Use the function scm_is_pair or scm_is_null instead.
1709 ** The functions scm_round and scm_truncate have been deprecated and
1710 are now available as scm_c_round and scm_c_truncate, respectively.
1712 These functions occupy the names that scm_round_number and
1713 scm_truncate_number should have.
1715 ** The functions scm_c_string2str, scm_c_substring2str, and
1716 scm_c_symbol2str have been deprecated.
1718 Use scm_to_locale_stringbuf or similar instead, maybe together with
1721 ** New functions scm_c_make_string, scm_c_string_length,
1722 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1723 scm_c_substring_shared, scm_c_substring_copy.
1725 These are like scm_make_string, scm_length, etc. but are slightly
1726 easier to use from C.
1728 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1729 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1731 They export too many assumptions about the implementation of strings
1732 and symbols that are no longer true in the presence of
1733 mutation-sharing substrings and when Guile switches to some form of
1736 When working with strings, it is often best to use the normal string
1737 functions provided by Guile, such as scm_c_string_ref,
1738 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1739 manual since many more such functions are now provided than
1742 When you want to convert a SCM string to a C string, use the
1743 scm_to_locale_string function or similar instead. For symbols, use
1744 scm_symbol_to_string and then work with that string. Because of the
1745 new string representation, scm_symbol_to_string does not need to copy
1746 and is thus quite efficient.
1748 ** Some string, symbol and keyword functions have been discouraged.
1750 They don't fit into the uniform naming scheme and are not explicit
1751 about the character encoding.
1753 Replace according to the following table:
1755 scm_allocate_string -> scm_c_make_string
1756 scm_take_str -> scm_take_locale_stringn
1757 scm_take0str -> scm_take_locale_string
1758 scm_mem2string -> scm_from_locale_stringn
1759 scm_str2string -> scm_from_locale_string
1760 scm_makfrom0str -> scm_from_locale_string
1761 scm_mem2symbol -> scm_from_locale_symboln
1762 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1763 scm_str2symbol -> scm_from_locale_symbol
1765 SCM_SYMBOL_HASH -> scm_hashq
1766 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1768 scm_c_make_keyword -> scm_from_locale_keyword
1770 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1771 now also available to C code.
1773 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1775 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1776 the latter returns the true name of the keyword, not the 'dash name',
1777 as SCM_KEYWORDSYM used to do.
1779 ** A new way to access arrays in a thread-safe and efficient way has
1782 See the manual, node "Accessing Arrays From C".
1784 ** The old uniform vector and bitvector implementations have been
1785 unceremoniously removed.
1787 This implementation exposed the details of the tagging system of
1788 Guile. Use the new C API explained in the manual in node "Uniform
1789 Numeric Vectors" and "Bit Vectors", respectively.
1791 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1792 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1793 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1794 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1795 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1796 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1799 ** The macros dealing with vectors have been deprecated.
1801 Use the new functions scm_is_vector, scm_vector_elements,
1802 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1803 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1804 manual for more details.
1806 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1807 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1809 The following macros have been removed: SCM_VECTOR_BASE,
1810 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1811 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1813 ** Some C functions and macros related to arrays have been deprecated.
1815 Migrate according to the following table:
1817 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1818 scm_make_ra -> scm_make_array
1819 scm_shap2ra -> scm_make_array
1820 scm_cvref -> scm_c_generalized_vector_ref
1821 scm_ra_set_contp -> do not use
1822 scm_aind -> scm_array_handle_pos
1823 scm_raprin1 -> scm_display or scm_write
1825 SCM_ARRAYP -> scm_is_array
1826 SCM_ARRAY_NDIM -> scm_c_array_rank
1827 SCM_ARRAY_DIMS -> scm_array_handle_dims
1828 SCM_ARRAY_CONTP -> do not use
1829 SCM_ARRAY_MEM -> do not use
1830 SCM_ARRAY_V -> scm_array_handle_elements or similar
1831 SCM_ARRAY_BASE -> do not use
1833 ** SCM_CELL_WORD_LOC has been deprecated.
1835 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1836 to a SCM, as opposed to a pointer to a scm_t_bits.
1838 This was done to allow the correct use of pointers into the Scheme
1839 heap. Previously, the heap words were of type scm_t_bits and local
1840 variables and function arguments were of type SCM, making it
1841 non-standards-conformant to have a pointer that can point to both.
1843 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1845 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1846 second and third words of double smobs. Likewise for
1847 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1849 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1850 used to get and set the 16 exra bits in the zeroth word of a smob.
1852 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1853 accesing the first immediate word of a smob as a SCM value, and there
1854 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1855 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1857 ** New way to deal with non-local exits and re-entries.
1859 There is a new set of functions that essentially do what
1860 scm_internal_dynamic_wind does, but in a way that is more convenient
1861 for C code in some situations. Here is a quick example of how to
1862 prevent a potential memory leak:
1869 scm_dynwind_begin (0);
1871 mem = scm_malloc (100);
1872 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1874 /* MEM would leak if BAR throws an error.
1875 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1882 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1883 SCM_DYNWIND_END as well.
1887 For full documentation, see the node "Dynamic Wind" in the manual.
1889 ** New function scm_dynwind_free
1891 This function calls 'free' on a given pointer when a dynwind context
1892 is left. Thus the call to scm_dynwind_unwind_handler above could be
1893 replaced with simply scm_dynwind_free (mem).
1895 ** New functions scm_c_call_with_blocked_asyncs and
1896 scm_c_call_with_unblocked_asyncs
1898 Like scm_call_with_blocked_asyncs etc. but for C functions.
1900 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1902 In addition to scm_c_call_with_blocked_asyncs you can now also use
1903 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1904 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1906 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1907 SCM_REALLOW_INTS have been deprecated.
1909 They do no longer fulfill their original role of blocking signal
1910 delivery. Depending on what you want to achieve, replace a pair of
1911 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1912 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1915 ** The value 'scm_mask_ints' is no longer writable.
1917 Previously, you could set scm_mask_ints directly. This is no longer
1918 possible. Use scm_c_call_with_blocked_asyncs and
1919 scm_c_call_with_unblocked_asyncs instead.
1921 ** New way to temporarily set the current input, output or error ports
1923 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1924 context' (see above). <foo> is one of "input", "output" or "error".
1926 ** New way to temporarily set fluids
1928 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1929 above) to temporarily set the value of a fluid.
1931 ** New types scm_t_intmax and scm_t_uintmax.
1933 On platforms that have them, these types are identical to intmax_t and
1934 uintmax_t, respectively. On other platforms, they are identical to
1935 the largest integer types that Guile knows about.
1937 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1939 You should not have used them.
1941 ** Many public #defines with generic names have been made private.
1943 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1944 private or renamed with a more suitable public name.
1946 ** The macro SCM_TYP16S has been deprecated.
1948 This macro is not intended for public use.
1950 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1952 Use scm_is_true (scm_inexact_p (...)) instead.
1954 ** The macro SCM_SLOPPY_REALP has been deprecated.
1956 Use scm_is_real instead.
1958 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1960 Use scm_is_complex instead.
1962 ** Some preprocessor defines have been deprecated.
1964 These defines indicated whether a certain feature was present in Guile
1965 or not. Going forward, assume that the features are always present.
1967 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1968 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1970 The following macros have been removed completely: MEMOIZE_LOCALS,
1971 SCM_RECKLESS, SCM_CAUTIOUS.
1973 ** The preprocessor define STACK_DIRECTION has been deprecated.
1975 There should be no need to know about the stack direction for ordinary
1978 ** New function: scm_effective_version
1980 Returns the "effective" version number. This is just the normal full
1981 version string without the final micro-version number. See "Changes
1982 to the distribution" above.
1984 ** The function scm_call_with_new_thread has a new prototype.
1986 Instead of taking a list with the thunk and handler, these two
1987 arguments are now passed directly:
1989 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1991 This is an incompatible change.
1993 ** New snarfer macro SCM_DEFINE_PUBLIC.
1995 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1996 function in the init section.
1998 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
2000 ** Garbage collector rewrite.
2002 The garbage collector is cleaned up a lot, and now uses lazy
2003 sweeping. This is reflected in the output of (gc-stats); since cells
2004 are being freed when they are allocated, the cells-allocated field
2005 stays roughly constant.
2007 For malloc related triggers, the behavior is changed. It uses the same
2008 heuristic as the cell-triggered collections. It may be tuned with the
2009 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
2010 for minimum yield of malloc related triggers. The default is 40.
2011 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
2014 Debugging operations for the freelist have been deprecated, along with
2015 the C variables that control garbage collection. The environment
2016 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
2017 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
2019 For understanding the memory usage of a GUILE program, the routine
2020 gc-live-object-stats returns an alist containing the number of live
2021 objects for every type.
2024 ** The function scm_definedp has been renamed to scm_defined_p
2026 The name scm_definedp is deprecated.
2028 ** The struct scm_cell type has been renamed to scm_t_cell
2030 This is in accordance to Guile's naming scheme for types. Note that
2031 the name scm_cell is now used for a function that allocates and
2032 initializes a new cell (see below).
2034 ** New functions for memory management
2036 A new set of functions for memory management has been added since the
2037 old way (scm_must_malloc, scm_must_free, etc) was error prone and
2038 indeed, Guile itself contained some long standing bugs that could
2039 cause aborts in long running programs.
2041 The new functions are more symmetrical and do not need cooperation
2042 from smob free routines, among other improvements.
2044 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
2045 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
2046 scm_gc_free, scm_gc_register_collectable_memory, and
2047 scm_gc_unregister_collectable_memory. Refer to the manual for more
2048 details and for upgrading instructions.
2050 The old functions for memory management have been deprecated. They
2051 are: scm_must_malloc, scm_must_realloc, scm_must_free,
2052 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
2054 ** Declarations of exported features are marked with SCM_API.
2056 Every declaration of a feature that belongs to the exported Guile API
2057 has been marked by adding the macro "SCM_API" to the start of the
2058 declaration. This macro can expand into different things, the most
2059 common of which is just "extern" for Unix platforms. On Win32, it can
2060 be used to control which symbols are exported from a DLL.
2062 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
2063 will expand into "__declspec (dllimport) extern", which is needed for
2064 linking to the Guile DLL in Windows.
2066 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
2067 SCM_SRFI4_IMPORT, for the corresponding libraries.
2069 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
2071 Use the new functions scm_cell and scm_double_cell instead. The old
2072 macros had problems because with them allocation and initialization
2073 was separated and the GC could sometimes observe half initialized
2074 cells. Only careful coding by the user of SCM_NEWCELL and
2075 SCM_NEWCELL2 could make this safe and efficient.
2077 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
2079 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
2082 ** SRCBRKP has been deprecated.
2084 Use scm_c_source_property_breakpoint_p instead.
2086 ** Deprecated: scm_makmacro
2088 Change your code to use either scm_makmmacro or to define macros in
2089 Scheme, using 'define-macro'.
2091 ** New function scm_c_port_for_each.
2093 This function is like scm_port_for_each but takes a pointer to a C
2094 function as the callback instead of a SCM value.
2096 ** The names scm_internal_select, scm_thread_sleep, and
2097 scm_thread_usleep have been discouraged.
2099 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
2101 ** The GC can no longer be blocked.
2103 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
2104 The GC can now run (partially) concurrently with other code and thus
2105 blocking it is not well defined.
2107 ** Many definitions have been removed that were previously deprecated.
2109 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
2110 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
2111 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
2112 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
2113 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
2114 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
2115 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
2116 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
2117 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
2118 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
2119 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
2120 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
2121 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
2122 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
2123 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
2124 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
2125 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
2126 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
2127 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
2128 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
2129 scm_vector_set_length_x, scm_contregs, scm_debug_info,
2130 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
2131 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
2132 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
2133 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
2134 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
2135 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
2136 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
2137 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
2138 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
2139 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
2141 * Changes to bundled modules
2145 Using the (ice-9 debug) module no longer automatically switches Guile
2146 to use the debugging evaluator. If you want to switch to the
2147 debugging evaluator (which is needed for backtrace information if you
2148 hit an error), please add an explicit "(debug-enable 'debug)" to your
2149 code just after the code to use (ice-9 debug).
2152 Changes since Guile 1.4:
2154 * Changes to the distribution
2156 ** A top-level TODO file is included.
2158 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
2160 Guile now always uses three numbers to represent the version,
2161 i.e. "1.6.5". The first number, 1, is the major version number, the
2162 second number, 6, is the minor version number, and the third number,
2163 5, is the micro version number. Changes in major version number
2164 indicate major changes in Guile.
2166 Minor version numbers that are even denote stable releases, and odd
2167 minor version numbers denote development versions (which may be
2168 unstable). The micro version number indicates a minor sub-revision of
2169 a given MAJOR.MINOR release.
2171 In keeping with the new scheme, (minor-version) and scm_minor_version
2172 no longer return everything but the major version number. They now
2173 just return the minor version number. Two new functions
2174 (micro-version) and scm_micro_version have been added to report the
2175 micro version number.
2177 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
2179 ** New preprocessor definitions are available for checking versions.
2181 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
2182 SCM_MICRO_VERSION to the appropriate integer values.
2184 ** Guile now actively warns about deprecated features.
2186 The new configure option `--enable-deprecated=LEVEL' and the
2187 environment variable GUILE_WARN_DEPRECATED control this mechanism.
2188 See INSTALL and README for more information.
2190 ** Guile is much more likely to work on 64-bit architectures.
2192 Guile now compiles and passes "make check" with only two UNRESOLVED GC
2193 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
2194 for the use of a test machine, and thanks to Stefan Jahn for ia64
2197 ** New functions: setitimer and getitimer.
2199 These implement a fairly direct interface to the libc functions of the
2202 ** The #. reader extension is now disabled by default.
2204 For safety reasons, #. evaluation is disabled by default. To
2205 re-enable it, set the fluid read-eval? to #t. For example:
2207 (fluid-set! read-eval? #t)
2209 but make sure you realize the potential security risks involved. With
2210 read-eval? enabled, reading a data file from an untrusted source can
2213 ** New SRFI modules have been added:
2215 SRFI-0 `cond-expand' is now supported in Guile, without requiring
2218 (srfi srfi-1) is a library containing many useful pair- and list-processing
2221 (srfi srfi-2) exports and-let*.
2223 (srfi srfi-4) implements homogeneous numeric vector datatypes.
2225 (srfi srfi-6) is a dummy module for now, since guile already provides
2226 all of the srfi-6 procedures by default: open-input-string,
2227 open-output-string, get-output-string.
2229 (srfi srfi-8) exports receive.
2231 (srfi srfi-9) exports define-record-type.
2233 (srfi srfi-10) exports define-reader-ctor and implements the reader
2236 (srfi srfi-11) exports let-values and let*-values.
2238 (srfi srfi-13) implements the SRFI String Library.
2240 (srfi srfi-14) implements the SRFI Character-Set Library.
2242 (srfi srfi-17) implements setter and getter-with-setter and redefines
2243 some accessor procedures as procedures with getters. (such as car,
2244 cdr, vector-ref etc.)
2246 (srfi srfi-19) implements the SRFI Time/Date Library.
2248 ** New scripts / "executable modules"
2250 Subdirectory "scripts" contains Scheme modules that are packaged to
2251 also be executable as scripts. At this time, these scripts are available:
2260 See README there for more info.
2262 These scripts can be invoked from the shell with the new program
2263 "guile-tools", which keeps track of installation directory for you.
2266 $ guile-tools display-commentary srfi/*.scm
2268 guile-tools is copied to the standard $bindir on "make install".
2270 ** New module (ice-9 stack-catch):
2272 stack-catch is like catch, but saves the current state of the stack in
2273 the fluid the-last-stack. This fluid can be useful when using the
2274 debugger and when re-throwing an error.
2276 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
2278 This has been done to prevent problems on lesser operating systems
2279 that can't tolerate `*'s in file names. The exported macro continues
2280 to be named `and-let*', of course.
2282 On systems that support it, there is also a compatibility module named
2283 (ice-9 and-let*). It will go away in the next release.
2285 ** New modules (oop goops) etc.:
2288 (oop goops describe)
2290 (oop goops active-slot)
2291 (oop goops composite-slot)
2293 The Guile Object Oriented Programming System (GOOPS) has been
2294 integrated into Guile. For further information, consult the GOOPS
2295 manual and tutorial in the `doc' directory.
2297 ** New module (ice-9 rdelim).
2299 This exports the following procedures which were previously defined
2300 in the default environment:
2302 read-line read-line! read-delimited read-delimited! %read-delimited!
2303 %read-line write-line
2305 For backwards compatibility the definitions are still imported into the
2306 default environment in this version of Guile. However you should add:
2308 (use-modules (ice-9 rdelim))
2310 to any program which uses the definitions, since this may change in
2313 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
2314 can be used for similar functionality.
2316 ** New module (ice-9 rw)
2318 This is a subset of the (scsh rw) module from guile-scsh. Currently
2319 it defines two procedures:
2321 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
2323 Read characters from a port or file descriptor into a string STR.
2324 A port must have an underlying file descriptor -- a so-called
2325 fport. This procedure is scsh-compatible and can efficiently read
2328 *** New function: write-string/partial str [port_or_fdes [start [end]]]
2330 Write characters from a string STR to a port or file descriptor.
2331 A port must have an underlying file descriptor -- a so-called
2332 fport. This procedure is mostly compatible and can efficiently
2333 write large strings.
2335 ** New module (ice-9 match)
2337 This module includes Andrew K. Wright's pattern matcher. See
2338 ice-9/match.scm for brief description or
2340 http://www.star-lab.com/wright/code.html
2342 for complete documentation.
2344 ** New module (ice-9 buffered-input)
2346 This module provides procedures to construct an input port from an
2347 underlying source of input that reads and returns its input in chunks.
2348 The underlying input source is a Scheme procedure, specified by the
2349 caller, which the port invokes whenever it needs more input.
2351 This is useful when building an input port whose back end is Readline
2352 or a UI element such as the GtkEntry widget.
2356 The reference and tutorial documentation that was previously
2357 distributed separately, as `guile-doc', is now included in the core
2358 Guile distribution. The documentation consists of the following
2361 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
2364 - The Guile Reference Manual (guile.texi) contains (or is intended to
2365 contain) reference documentation on all aspects of Guile.
2367 - The GOOPS Manual (goops.texi) contains both tutorial-style and
2368 reference documentation for using GOOPS, Guile's Object Oriented
2371 - The Revised^5 Report on the Algorithmic Language Scheme
2374 See the README file in the `doc' directory for more details.
2376 ** There are a couple of examples in the examples/ directory now.
2378 * Changes to the stand-alone interpreter
2380 ** New command line option `--use-srfi'
2382 Using this option, SRFI modules can be loaded on startup and be
2383 available right from the beginning. This makes programming portable
2384 Scheme programs easier.
2386 The option `--use-srfi' expects a comma-separated list of numbers,
2387 each representing a SRFI number to be loaded into the interpreter
2388 before starting evaluating a script file or the REPL. Additionally,
2389 the feature identifier for the loaded SRFIs is recognized by
2390 `cond-expand' when using this option.
2393 $ guile --use-srfi=8,13
2394 guile> (receive (x z) (values 1 2) (+ 1 2))
2396 guile> (string-pad "bla" 20)
2399 ** Guile now always starts up in the `(guile-user)' module.
2401 Previously, scripts executed via the `-s' option would run in the
2402 `(guile)' module and the repl would run in the `(guile-user)' module.
2403 Now every user action takes place in the `(guile-user)' module by
2406 * Changes to Scheme functions and syntax
2408 ** Character classifiers work for non-ASCII characters.
2410 The predicates `char-alphabetic?', `char-numeric?',
2411 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
2412 no longer check whether their arguments are ASCII characters.
2413 Previously, a character would only be considered alphabetic when it
2414 was also ASCII, for example.
2416 ** Previously deprecated Scheme functions have been removed:
2418 tag - no replacement.
2419 fseek - replaced by seek.
2420 list* - replaced by cons*.
2422 ** It's now possible to create modules with controlled environments
2426 (use-modules (ice-9 safe))
2427 (define m (make-safe-module))
2428 ;;; m will now be a module containing only a safe subset of R5RS
2429 (eval '(+ 1 2) m) --> 3
2430 (eval 'load m) --> ERROR: Unbound variable: load
2432 ** Evaluation of "()", the empty list, is now an error.
2434 Previously, the expression "()" evaluated to the empty list. This has
2435 been changed to signal a "missing expression" error. The correct way
2436 to write the empty list as a literal constant is to use quote: "'()".
2438 ** New concept of `Guile Extensions'.
2440 A Guile Extension is just a ordinary shared library that can be linked
2441 at run-time. We found it advantageous to give this simple concept a
2442 dedicated name to distinguish the issues related to shared libraries
2443 from the issues related to the module system.
2445 *** New function: load-extension
2447 Executing (load-extension lib init) is mostly equivalent to
2449 (dynamic-call init (dynamic-link lib))
2451 except when scm_register_extension has been called previously.
2452 Whenever appropriate, you should use `load-extension' instead of
2453 dynamic-link and dynamic-call.
2455 *** New C function: scm_c_register_extension
2457 This function registers a initialization function for use by
2458 `load-extension'. Use it when you don't want specific extensions to
2459 be loaded as shared libraries (for example on platforms that don't
2460 support dynamic linking).
2462 ** Auto-loading of compiled-code modules is deprecated.
2464 Guile used to be able to automatically find and link a shared
2465 library to satisfy requests for a module. For example, the module
2466 `(foo bar)' could be implemented by placing a shared library named
2467 "foo/libbar.so" (or with a different extension) in a directory on the
2470 This has been found to be too tricky, and is no longer supported. The
2471 shared libraries are now called "extensions". You should now write a
2472 small Scheme file that calls `load-extension' to load the shared
2473 library and initialize it explicitly.
2475 The shared libraries themselves should be installed in the usual
2476 places for shared libraries, with names like "libguile-foo-bar".
2478 For example, place this into a file "foo/bar.scm"
2480 (define-module (foo bar))
2482 (load-extension "libguile-foo-bar" "foobar_init")
2484 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
2486 `eval' is now R5RS, that is it takes two arguments.
2487 The second argument is an environment specifier, i.e. either
2489 (scheme-report-environment 5)
2490 (null-environment 5)
2491 (interaction-environment)
2497 ** The module system has been made more disciplined.
2499 The function `eval' will save and restore the current module around
2500 the evaluation of the specified expression. While this expression is
2501 evaluated, `(current-module)' will now return the right module, which
2502 is the module specified as the second argument to `eval'.
2504 A consequence of this change is that `eval' is not particularly
2505 useful when you want allow the evaluated code to change what module is
2506 designated as the current module and have this change persist from one
2507 call to `eval' to the next. The read-eval-print-loop is an example
2508 where `eval' is now inadequate. To compensate, there is a new
2509 function `primitive-eval' that does not take a module specifier and
2510 that does not save/restore the current module. You should use this
2511 function together with `set-current-module', `current-module', etc
2512 when you want to have more control over the state that is carried from
2513 one eval to the next.
2515 Additionally, it has been made sure that forms that are evaluated at
2516 the top level are always evaluated with respect to the current module.
2517 Previously, subforms of top-level forms such as `begin', `case',
2518 etc. did not respect changes to the current module although these
2519 subforms are at the top-level as well.
2521 To prevent strange behavior, the forms `define-module',
2522 `use-modules', `use-syntax', and `export' have been restricted to only
2523 work on the top level. The forms `define-public' and
2524 `defmacro-public' only export the new binding on the top level. They
2525 behave just like `define' and `defmacro', respectively, when they are
2526 used in a lexical environment.
2528 Also, `export' will no longer silently re-export bindings imported
2529 from a used module. It will emit a `deprecation' warning and will
2530 cease to perform any re-export in the next version. If you actually
2531 want to re-export bindings, use the new `re-export' in place of
2532 `export'. The new `re-export' will not make copies of variables when
2533 rexporting them, as `export' did wrongly.
2535 ** Module system now allows selection and renaming of imported bindings
2537 Previously, when using `use-modules' or the `#:use-module' clause in
2538 the `define-module' form, all the bindings (association of symbols to
2539 values) for imported modules were added to the "current module" on an
2540 as-is basis. This has been changed to allow finer control through two
2541 new facilities: selection and renaming.
2543 You can now select which of the imported module's bindings are to be
2544 visible in the current module by using the `:select' clause. This
2545 clause also can be used to rename individual bindings. For example:
2547 ;; import all bindings no questions asked
2548 (use-modules (ice-9 common-list))
2550 ;; import four bindings, renaming two of them;
2551 ;; the current module sees: every some zonk-y zonk-n
2552 (use-modules ((ice-9 common-list)
2554 (remove-if . zonk-y)
2555 (remove-if-not . zonk-n))))
2557 You can also programmatically rename all selected bindings using the
2558 `:renamer' clause, which specifies a proc that takes a symbol and
2559 returns another symbol. Because it is common practice to use a prefix,
2560 we now provide the convenience procedure `symbol-prefix-proc'. For
2563 ;; import four bindings, renaming two of them specifically,
2564 ;; and all four w/ prefix "CL:";
2565 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
2566 (use-modules ((ice-9 common-list)
2568 (remove-if . zonk-y)
2569 (remove-if-not . zonk-n))
2570 :renamer (symbol-prefix-proc 'CL:)))
2572 ;; import four bindings, renaming two of them specifically,
2573 ;; and all four by upcasing.
2574 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
2575 (define (upcase-symbol sym)
2576 (string->symbol (string-upcase (symbol->string sym))))
2578 (use-modules ((ice-9 common-list)
2580 (remove-if . zonk-y)
2581 (remove-if-not . zonk-n))
2582 :renamer upcase-symbol))
2584 Note that programmatic renaming is done *after* individual renaming.
2585 Also, the above examples show `use-modules', but the same facilities are
2586 available for the `#:use-module' clause of `define-module'.
2588 See manual for more info.
2590 ** The semantics of guardians have changed.
2592 The changes are for the most part compatible. An important criterion
2593 was to keep the typical usage of guardians as simple as before, but to
2594 make the semantics safer and (as a result) more useful.
2596 *** All objects returned from guardians are now properly alive.
2598 It is now guaranteed that any object referenced by an object returned
2599 from a guardian is alive. It's now impossible for a guardian to
2600 return a "contained" object before its "containing" object.
2602 One incompatible (but probably not very important) change resulting
2603 from this is that it is no longer possible to guard objects that
2604 indirectly reference themselves (i.e. are parts of cycles). If you do
2605 so accidentally, you'll get a warning.
2607 *** There are now two types of guardians: greedy and sharing.
2609 If you call (make-guardian #t) or just (make-guardian), you'll get a
2610 greedy guardian, and for (make-guardian #f) a sharing guardian.
2612 Greedy guardians are the default because they are more "defensive".
2613 You can only greedily guard an object once. If you guard an object
2614 more than once, once in a greedy guardian and the rest of times in
2615 sharing guardians, then it is guaranteed that the object won't be
2616 returned from sharing guardians as long as it is greedily guarded
2619 Guardians returned by calls to `make-guardian' can now take one more
2620 optional parameter, which says whether to throw an error in case an
2621 attempt is made to greedily guard an object that is already greedily
2622 guarded. The default is true, i.e. throw an error. If the parameter
2623 is false, the guardian invocation returns #t if guarding was
2624 successful and #f if it wasn't.
2626 Also, since greedy guarding is, in effect, a side-effecting operation
2627 on objects, a new function is introduced: `destroy-guardian!'.
2628 Invoking this function on a guardian renders it unoperative and, if
2629 the guardian is greedy, clears the "greedily guarded" property of the
2630 objects that were guarded by it, thus undoing the side effect.
2632 Note that all this hair is hardly very important, since guardian
2633 objects are usually permanent.
2635 ** Continuations created by call-with-current-continuation now accept
2636 any number of arguments, as required by R5RS.
2638 ** New function `issue-deprecation-warning'
2640 This function is used to display the deprecation messages that are
2641 controlled by GUILE_WARN_DEPRECATION as explained in the README.
2644 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
2648 ;; `id' is deprecated. Use `identity' instead.
2653 ** New syntax `begin-deprecated'
2655 When deprecated features are included (as determined by the configure
2656 option --enable-deprecated), `begin-deprecated' is identical to
2657 `begin'. When deprecated features are excluded, it always evaluates
2658 to `#f', ignoring the body forms.
2660 ** New function `make-object-property'
2662 This function returns a new `procedure with setter' P that can be used
2663 to attach a property to objects. When calling P as
2667 where `obj' is any kind of object, it attaches `val' to `obj' in such
2668 a way that it can be retrieved by calling P as
2672 This function will replace procedure properties, symbol properties and
2673 source properties eventually.
2675 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2677 Instead of #&optional, #&key, etc you should now use #:optional,
2678 #:key, etc. Since #:optional is a keyword, you can write it as just
2679 :optional when (read-set! keywords 'prefix) is active.
2681 The old reader syntax `#&' is still supported, but deprecated. It
2682 will be removed in the next release.
2684 ** New define-module option: pure
2686 Tells the module system not to include any bindings from the root
2691 (define-module (totally-empty-module)
2694 ** New define-module option: export NAME1 ...
2696 Export names NAME1 ...
2698 This option is required if you want to be able to export bindings from
2699 a module which doesn't import one of `define-public' or `export'.
2703 (define-module (foo)
2705 :use-module (ice-9 r5rs)
2708 ;;; Note that we're pure R5RS below this point!
2713 ** New function: object->string OBJ
2715 Return a Scheme string obtained by printing a given object.
2717 ** New function: port? X
2719 Returns a boolean indicating whether X is a port. Equivalent to
2720 `(or (input-port? X) (output-port? X))'.
2722 ** New function: file-port?
2724 Determines whether a given object is a port that is related to a file.
2726 ** New function: port-for-each proc
2728 Apply PROC to each port in the Guile port table in turn. The return
2729 value is unspecified. More specifically, PROC is applied exactly once
2730 to every port that exists in the system at the time PORT-FOR-EACH is
2731 invoked. Changes to the port table while PORT-FOR-EACH is running
2732 have no effect as far as PORT-FOR-EACH is concerned.
2734 ** New function: dup2 oldfd newfd
2736 A simple wrapper for the `dup2' system call. Copies the file
2737 descriptor OLDFD to descriptor number NEWFD, replacing the
2738 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2739 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2740 to move away ports which are using NEWFD. The return value is
2743 ** New function: close-fdes fd
2745 A simple wrapper for the `close' system call. Close file
2746 descriptor FD, which must be an integer. Unlike close (*note
2747 close: Ports and File Descriptors.), the file descriptor will be
2748 closed even if a port is using it. The return value is
2751 ** New function: crypt password salt
2753 Encrypts `password' using the standard unix password encryption
2756 ** New function: chroot path
2758 Change the root directory of the running process to `path'.
2760 ** New functions: getlogin, cuserid
2762 Return the login name or the user name of the current effective user
2765 ** New functions: getpriority which who, setpriority which who prio
2767 Get or set the priority of the running process.
2769 ** New function: getpass prompt
2771 Read a password from the terminal, first displaying `prompt' and
2774 ** New function: flock file operation
2776 Set/remove an advisory shared or exclusive lock on `file'.
2778 ** New functions: sethostname name, gethostname
2780 Set or get the hostname of the machine the current process is running
2783 ** New function: mkstemp! tmpl
2785 mkstemp creates a new unique file in the file system and returns a
2786 new buffered port open for reading and writing to the file. TMPL
2787 is a string specifying where the file should be created: it must
2788 end with `XXXXXX' and will be changed in place to return the name
2789 of the temporary file.
2791 ** New function: open-input-string string
2793 Return an input string port which delivers the characters from
2794 `string'. This procedure, together with `open-output-string' and
2795 `get-output-string' implements SRFI-6.
2797 ** New function: open-output-string
2799 Return an output string port which collects all data written to it.
2800 The data can then be retrieved by `get-output-string'.
2802 ** New function: get-output-string
2804 Return the contents of an output string port.
2806 ** New function: identity
2808 Return the argument.
2810 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2811 are represented in Scheme as integers with normal host byte ordering.
2813 ** New function: inet-pton family address
2815 Convert a printable string network address into an integer. Note that
2816 unlike the C version of this function, the result is an integer with
2817 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2820 (inet-pton AF_INET "127.0.0.1") => 2130706433
2821 (inet-pton AF_INET6 "::1") => 1
2823 ** New function: inet-ntop family address
2825 Convert an integer network address into a printable string. Note that
2826 unlike the C version of this function, the input is an integer with
2827 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2830 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2831 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2832 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2836 Use `identity' instead.
2842 ** Deprecated: return-it
2846 ** Deprecated: string-character-length
2848 Use `string-length' instead.
2850 ** Deprecated: flags
2852 Use `logior' instead.
2854 ** Deprecated: close-all-ports-except.
2856 This was intended for closing ports in a child process after a fork,
2857 but it has the undesirable side effect of flushing buffers.
2858 port-for-each is more flexible.
2860 ** The (ice-9 popen) module now attempts to set up file descriptors in
2861 the child process from the current Scheme ports, instead of using the
2862 current values of file descriptors 0, 1, and 2 in the parent process.
2864 ** Removed function: builtin-weak-bindings
2866 There is no such concept as a weak binding any more.
2868 ** Removed constants: bignum-radix, scm-line-incrementors
2870 ** define-method: New syntax mandatory.
2872 The new method syntax is now mandatory:
2874 (define-method (NAME ARG-SPEC ...) BODY ...)
2875 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2877 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2878 REST-ARG ::= ARG-NAME
2880 If you have old code using the old syntax, import
2881 (oop goops old-define-method) before (oop goops) as in:
2883 (use-modules (oop goops old-define-method) (oop goops))
2885 ** Deprecated function: builtin-variable
2886 Removed function: builtin-bindings
2888 There is no longer a distinction between builtin or other variables.
2889 Use module system operations for all variables.
2891 ** Lazy-catch handlers are no longer allowed to return.
2893 That is, a call to `throw', `error', etc is now guaranteed to not
2896 ** Bugfixes for (ice-9 getopt-long)
2898 This module is now tested using test-suite/tests/getopt-long.test.
2899 The following bugs have been fixed:
2901 *** Parsing for options that are specified to have `optional' args now checks
2902 if the next element is an option instead of unconditionally taking it as the
2905 *** An error is now thrown for `--opt=val' when the option description
2906 does not specify `(value #t)' or `(value optional)'. This condition used to
2907 be accepted w/o error, contrary to the documentation.
2909 *** The error message for unrecognized options is now more informative.
2910 It used to be "not a record", an artifact of the implementation.
2912 *** The error message for `--opt' terminating the arg list (no value), when
2913 `(value #t)' is specified, is now more informative. It used to be "not enough
2916 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2917 The expansion used to be like so:
2919 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2921 Note that the "5d" is dropped. Now it is like so:
2923 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2925 This enables single-char options to have adjoining arguments as long as their
2926 constituent characters are not potential single-char options.
2928 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2930 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2931 property `arglist', which can be retrieved by `arity'. The result is that
2932 `arity' can give more detailed information than before:
2936 guile> (use-modules (ice-9 optargs))
2937 guile> (define* (foo #:optional a b c) a)
2939 0 or more arguments in `lambda*:G0'.
2944 3 optional arguments: `a', `b' and `c'.
2945 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2947 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2948 and `d', other keywords allowed.
2949 guile> (define* (baz a b #:optional c #:rest r) a)
2951 2 required arguments: `a' and `b', 1 optional argument: `c',
2954 * Changes to the C interface
2956 ** Types have been renamed from scm_*_t to scm_t_*.
2958 This has been done for POSIX sake. It reserves identifiers ending
2959 with "_t". What a concept.
2961 The old names are still available with status `deprecated'.
2963 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2965 ** Deprecated features have been removed.
2969 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2970 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2972 *** C Functions removed
2974 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2975 scm_fseek - replaced by scm_seek.
2976 gc-thunk - replaced by after-gc-hook.
2977 gh_int2scmb - replaced by gh_bool2scm.
2978 scm_tc_dblr - replaced by scm_tc16_real.
2979 scm_tc_dblc - replaced by scm_tc16_complex.
2980 scm_list_star - replaced by scm_cons_star.
2982 ** Deprecated: scm_makfromstr
2984 Use scm_mem2string instead.
2986 ** Deprecated: scm_make_shared_substring
2988 Explicit shared substrings will disappear from Guile.
2990 Instead, "normal" strings will be implemented using sharing
2991 internally, combined with a copy-on-write strategy.
2993 ** Deprecated: scm_read_only_string_p
2995 The concept of read-only strings will disappear in next release of
2998 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
3000 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
3002 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
3004 Call a procedure with the indicated number of arguments. See "Fly
3005 Evaluation" in the manual.
3007 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
3009 Call a procedure with the indicated number of arguments and a list of
3010 further arguments. See "Fly Evaluation" in the manual.
3012 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
3014 Create a list of the given number of elements. See "List
3015 Constructors" in the manual.
3017 ** Renamed function: scm_listify has been replaced by scm_list_n.
3019 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
3020 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
3022 Use functions scm_list_N instead.
3024 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
3026 Used by an application to read arbitrary number of bytes from a port.
3027 Same semantics as libc read, except that scm_c_read only returns less
3028 than SIZE bytes if at end-of-file.
3030 Warning: Doesn't update port line and column counts!
3032 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
3034 Used by an application to write arbitrary number of bytes to an SCM
3035 port. Similar semantics as libc write. However, unlike libc
3036 write, scm_c_write writes the requested number of bytes and has no
3039 Warning: Doesn't update port line and column counts!
3041 ** New function: scm_init_guile ()
3043 In contrast to scm_boot_guile, scm_init_guile will return normally
3044 after initializing Guile. It is not available on all systems, tho.
3046 ** New functions: scm_str2symbol, scm_mem2symbol
3048 The function scm_str2symbol takes a const char* pointing to a zero-terminated
3049 field of characters and creates a scheme symbol object from that C string.
3050 The function scm_mem2symbol takes a const char* and a number of characters and
3051 creates a symbol from the characters in that memory area.
3053 ** New functions: scm_primitive_make_property
3054 scm_primitive_property_ref
3055 scm_primitive_property_set_x
3056 scm_primitive_property_del_x
3058 These functions implement a new way to deal with object properties.
3059 See libguile/properties.c for their documentation.
3061 ** New function: scm_done_free (long size)
3063 This function is the inverse of scm_done_malloc. Use it to report the
3064 amount of smob memory you free. The previous method, which involved
3065 calling scm_done_malloc with negative argument, was somewhat
3066 unintuitive (and is still available, of course).
3068 ** New function: scm_c_memq (SCM obj, SCM list)
3070 This function provides a fast C level alternative for scm_memq for the case
3071 that the list parameter is known to be a proper list. The function is a
3072 replacement for scm_sloppy_memq, but is stricter in its requirements on its
3073 list input parameter, since for anything else but a proper list the function's
3074 behaviour is undefined - it may even crash or loop endlessly. Further, for
3075 the case that the object is not found in the list, scm_c_memq returns #f which
3076 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
3078 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
3079 scm_remember_upto_here
3081 These functions replace the function scm_remember.
3083 ** Deprecated function: scm_remember
3085 Use one of the new functions scm_remember_upto_here_1,
3086 scm_remember_upto_here_2 or scm_remember_upto_here instead.
3088 ** New function: scm_allocate_string
3090 This function replaces the function scm_makstr.
3092 ** Deprecated function: scm_makstr
3094 Use the new function scm_allocate_string instead.
3096 ** New global variable scm_gc_running_p introduced.
3098 Use this variable to find out if garbage collection is being executed. Up to
3099 now applications have used scm_gc_heap_lock to test if garbage collection was
3100 running, which also works because of the fact that up to know only the garbage
3101 collector has set this variable. But, this is an implementation detail that
3102 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
3103 of this variable is (and has been) not fully safe anyway.
3105 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
3107 Use these instead of SCM_LENGTH_MAX.
3109 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
3110 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
3111 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
3113 Use these instead of SCM_LENGTH.
3115 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
3116 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
3117 SCM_SET_BITVECTOR_LENGTH
3119 Use these instead of SCM_SETLENGTH
3121 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
3122 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
3125 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
3128 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
3129 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
3132 Use these instead of SCM_SETCHARS.
3134 ** New macro: SCM_BITVECTOR_P
3136 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
3138 Use instead of SCM_COERCE_SUBSTR.
3140 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
3142 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
3144 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
3145 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
3146 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
3147 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
3148 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
3149 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
3150 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
3151 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
3152 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
3153 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
3154 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
3155 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
3156 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
3157 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
3158 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
3160 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
3161 Use scm_memory_error instead of SCM_NALLOC.
3162 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
3163 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
3164 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
3165 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
3166 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
3167 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
3168 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
3169 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
3170 Use SCM_STRINGP instead of SCM_RWSTRINGP.
3171 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
3172 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
3173 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
3174 Use a type specific setter macro instead of SCM_SETLENGTH.
3175 Use a type specific setter macro instead of SCM_SETCHARS.
3176 Use a type specific length macro instead of SCM_LENGTH_MAX.
3177 Use SCM_GCMARKP instead of SCM_GC8MARKP.
3178 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
3179 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
3180 Use SCM_TYP16 instead of SCM_GCTYP16.
3181 Use SCM_CDR instead of SCM_GCCDR.
3182 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
3183 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
3184 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
3185 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
3186 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
3187 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
3188 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
3190 ** Removed function: scm_struct_init
3192 ** Removed variable: scm_symhash_dim
3194 ** Renamed function: scm_make_cont has been replaced by
3195 scm_make_continuation, which has a different interface.
3197 ** Deprecated function: scm_call_catching_errors
3199 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
3201 ** Deprecated function: scm_strhash
3203 Use scm_string_hash instead.
3205 ** Deprecated function: scm_vector_set_length_x
3207 Instead, create a fresh vector of the desired size and copy the contents.
3209 ** scm_gensym has changed prototype
3211 scm_gensym now only takes one argument.
3213 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
3216 There is now only a single symbol type scm_tc7_symbol.
3217 The tag scm_tc7_lvector was not used anyway.
3219 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
3221 Use scm_make_smob_type and scm_set_smob_XXX instead.
3223 ** New function scm_set_smob_apply.
3225 This can be used to set an apply function to a smob type.
3227 ** Deprecated function: scm_strprint_obj
3229 Use scm_object_to_string instead.
3231 ** Deprecated function: scm_wta
3233 Use scm_wrong_type_arg, or another appropriate error signalling function
3236 ** Explicit support for obarrays has been deprecated.
3238 Use `scm_str2symbol' and the generic hashtable functions instead.
3240 ** The concept of `vcells' has been deprecated.
3242 The data type `variable' is now used exclusively. `Vcells' have been
3243 a low-level concept so you are likely not affected by this change.
3245 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
3246 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
3248 Use scm_c_define or scm_c_lookup instead, as appropriate.
3250 *** New functions: scm_c_module_lookup, scm_c_lookup,
3251 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
3252 scm_module_define, scm_define.
3254 These functions work with variables instead of with vcells.
3256 ** New functions for creating and defining `subr's and `gsubr's.
3258 The new functions more clearly distinguish between creating a subr (or
3259 gsubr) object and adding it to the current module.
3261 These new functions are available: scm_c_make_subr, scm_c_define_subr,
3262 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
3263 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
3264 scm_c_define_gsubr_with_generic.
3266 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
3267 scm_make_subr_with_generic, scm_make_gsubr,
3268 scm_make_gsubr_with_generic.
3270 Use the new ones from above instead.
3272 ** C interface to the module system has changed.
3274 While we suggest that you avoid as many explicit module system
3275 operations from C as possible for the time being, the C interface has
3276 been made more similar to the high-level Scheme module system.
3278 *** New functions: scm_c_define_module, scm_c_use_module,
3279 scm_c_export, scm_c_resolve_module.
3281 They mostly work like their Scheme namesakes. scm_c_define_module
3282 takes a function that is called a context where the new module is
3285 *** Deprecated functions: scm_the_root_module, scm_make_module,
3286 scm_ensure_user_module, scm_load_scheme_module.
3288 Use the new functions instead.
3290 ** Renamed function: scm_internal_with_fluids becomes
3293 scm_internal_with_fluids is available as a deprecated function.
3295 ** New function: scm_c_with_fluid.
3297 Just like scm_c_with_fluids, but takes one fluid and one value instead
3300 ** Deprecated typedefs: long_long, ulong_long.
3302 They are of questionable utility and they pollute the global
3305 ** Deprecated typedef: scm_sizet
3307 It is of questionable utility now that Guile requires ANSI C, and is
3310 ** Deprecated typedefs: scm_port_rw_active, scm_port,
3311 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
3312 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
3314 Made more compliant with the naming policy by adding a _t at the end.
3316 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
3317 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
3319 With the exception of the mysterious scm_2ulong2big, they are still
3320 available under new names (scm_i_mkbig etc). These functions are not
3321 intended to be used in user code. You should avoid dealing with
3322 bignums directly, and should deal with numbers in general (which can
3325 ** Change in behavior: scm_num2long, scm_num2ulong
3327 The scm_num2[u]long functions don't any longer accept an inexact
3328 argument. This change in behavior is motivated by concordance with
3329 R5RS: It is more common that a primitive doesn't want to accept an
3330 inexact for an exact.
3332 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
3333 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
3334 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
3337 These are conversion functions between the various ANSI C integral
3338 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
3339 accept an inexact argument.
3341 ** New functions: scm_float2num, scm_double2num,
3342 scm_num2float, scm_num2double.
3344 These are conversion functions between the two ANSI C float types and
3347 ** New number validation macros:
3348 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
3352 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
3354 These are just nicer-named old scm_protect_object and
3355 scm_unprotect_object.
3357 ** Deprecated functions: scm_protect_object, scm_unprotect_object
3359 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
3361 These functions can be used to register pointers to locations that
3364 ** Deprecated function: scm_create_hook.
3366 Its sins are: misleading name, non-modularity and lack of general
3370 Changes since Guile 1.3.4:
3372 * Changes to the distribution
3374 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
3376 We've changed the way we handle generated files in the Guile source
3377 repository. As a result, the procedure for building trees obtained
3378 from the nightly FTP snapshots or via CVS has changed:
3379 - You must have appropriate versions of autoconf, automake, and
3380 libtool installed on your system. See README for info on how to
3381 obtain these programs.
3382 - Before configuring the tree, you must first run the script
3383 `autogen.sh' at the top of the source tree.
3385 The Guile repository used to contain not only source files, written by
3386 humans, but also some generated files, like configure scripts and
3387 Makefile.in files. Even though the contents of these files could be
3388 derived mechanically from other files present, we thought it would
3389 make the tree easier to build if we checked them into CVS.
3391 However, this approach means that minor differences between
3392 developer's installed tools and habits affected the whole team.
3393 So we have removed the generated files from the repository, and
3394 added the autogen.sh script, which will reconstruct them
3398 ** configure now has experimental options to remove support for certain
3401 --disable-arrays omit array and uniform array support
3402 --disable-posix omit posix interfaces
3403 --disable-networking omit networking interfaces
3404 --disable-regex omit regular expression interfaces
3406 These are likely to become separate modules some day.
3408 ** New configure option --enable-debug-freelist
3410 This enables a debugging version of SCM_NEWCELL(), and also registers
3411 an extra primitive, the setter `gc-set-debug-check-freelist!'.
3413 Configure with the --enable-debug-freelist option to enable
3414 the gc-set-debug-check-freelist! primitive, and then use:
3416 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
3417 (gc-set-debug-check-freelist! #f) # turn off checking
3419 Checking of the freelist forces a traversal of the freelist and
3420 a garbage collection before each allocation of a cell. This can
3421 slow down the interpreter dramatically, so the setter should be used to
3422 turn on this extra processing only when necessary.
3424 ** New configure option --enable-debug-malloc
3426 Include code for debugging of calls to scm_must_malloc/realloc/free.
3430 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
3431 2. objects reallocated by scm_must_realloc has been allocated by
3433 3. reallocated objects are reallocated with the same what string
3435 But, most importantly, it records the number of allocated objects of
3436 each kind. This is useful when searching for memory leaks.
3438 A Guile compiled with this option provides the primitive
3439 `malloc-stats' which returns an alist with pairs of kind and the
3440 number of objects of that kind.
3442 ** All includes are now referenced relative to the root directory
3444 Since some users have had problems with mixups between Guile and
3445 system headers, we have decided to always refer to Guile headers via
3446 their parent directories. This essentially creates a "private name
3447 space" for Guile headers. This means that the compiler only is given
3448 -I options for the root build and root source directory.
3450 ** Header files kw.h and genio.h have been removed.
3452 ** The module (ice-9 getopt-gnu-style) has been removed.
3454 ** New module (ice-9 documentation)
3456 Implements the interface to documentation strings associated with
3459 ** New module (ice-9 time)
3461 Provides a macro `time', which displays execution time of a given form.
3463 ** New module (ice-9 history)
3465 Loading this module enables value history in the repl.
3467 * Changes to the stand-alone interpreter
3469 ** New command line option --debug
3471 Start Guile with debugging evaluator and backtraces enabled.
3473 This is useful when debugging your .guile init file or scripts.
3475 ** New help facility
3477 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
3478 (help REGEXP) ditto for objects with names matching REGEXP (a string)
3479 (help 'NAME) gives documentation for NAME, even if it is not an object
3480 (help ,EXPR) gives documentation for object returned by EXPR
3481 (help (my module)) gives module commentary for `(my module)'
3482 (help) gives this text
3484 `help' searches among bindings exported from loaded modules, while
3485 `apropos' searches among bindings visible from the "current" module.
3487 Examples: (help help)
3489 (help "output-string")
3491 ** `help' and `apropos' now prints full module names
3493 ** Dynamic linking now uses libltdl from the libtool package.
3495 The old system dependent code for doing dynamic linking has been
3496 replaced with calls to the libltdl functions which do all the hairy
3499 The major improvement is that you can now directly pass libtool
3500 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
3501 will be able to do the best shared library job you can get, via
3504 The way dynamic libraries are found has changed and is not really
3505 portable across platforms, probably. It is therefore recommended to
3506 use absolute filenames when possible.
3508 If you pass a filename without an extension to `dynamic-link', it will
3509 try a few appropriate ones. Thus, the most platform ignorant way is
3510 to specify a name like "libfoo", without any directories and
3513 ** Guile COOP threads are now compatible with LinuxThreads
3515 Previously, COOP threading wasn't possible in applications linked with
3516 Linux POSIX threads due to their use of the stack pointer to find the
3517 thread context. This has now been fixed with a workaround which uses
3518 the pthreads to allocate the stack.
3520 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
3522 ** Positions of erring expression in scripts
3524 With version 1.3.4, the location of the erring expression in Guile
3525 scipts is no longer automatically reported. (This should have been
3526 documented before the 1.3.4 release.)
3528 You can get this information by enabling recording of positions of
3529 source expressions and running the debugging evaluator. Put this at
3530 the top of your script (or in your "site" file):
3532 (read-enable 'positions)
3533 (debug-enable 'debug)
3535 ** Backtraces in scripts
3537 It is now possible to get backtraces in scripts.
3541 (debug-enable 'debug 'backtrace)
3543 at the top of the script.
3545 (The first options enables the debugging evaluator.
3546 The second enables backtraces.)
3548 ** Part of module system symbol lookup now implemented in C
3550 The eval closure of most modules is now implemented in C. Since this
3551 was one of the bottlenecks for loading speed, Guile now loads code
3552 substantially faster than before.
3554 ** Attempting to get the value of an unbound variable now produces
3555 an exception with a key of 'unbound-variable instead of 'misc-error.
3557 ** The initial default output port is now unbuffered if it's using a
3558 tty device. Previously in this situation it was line-buffered.
3560 ** New hook: after-gc-hook
3562 after-gc-hook takes over the role of gc-thunk. This hook is run at
3563 the first SCM_TICK after a GC. (Thus, the code is run at the same
3564 point during evaluation as signal handlers.)
3566 Note that this hook should be used only for diagnostic and debugging
3567 purposes. It is not certain that it will continue to be well-defined
3568 when this hook is run in the future.
3570 C programmers: Note the new C level hooks scm_before_gc_c_hook,
3571 scm_before_sweep_c_hook, scm_after_gc_c_hook.
3573 ** Improvements to garbage collector
3575 Guile 1.4 has a new policy for triggering heap allocation and
3576 determining the sizes of heap segments. It fixes a number of problems
3579 1. The new policy can handle two separate pools of cells
3580 (2-word/4-word) better. (The old policy would run wild, allocating
3581 more and more memory for certain programs.)
3583 2. The old code would sometimes allocate far too much heap so that the
3584 Guile process became gigantic. The new code avoids this.
3586 3. The old code would sometimes allocate too little so that few cells
3587 were freed at GC so that, in turn, too much time was spent in GC.
3589 4. The old code would often trigger heap allocation several times in a
3590 row. (The new scheme predicts how large the segments needs to be
3591 in order not to need further allocation.)
3593 All in all, the new GC policy will make larger applications more
3596 The new GC scheme also is prepared for POSIX threading. Threads can
3597 allocate private pools of cells ("clusters") with just a single
3598 function call. Allocation of single cells from such a cluster can
3599 then proceed without any need of inter-thread synchronization.
3601 ** New environment variables controlling GC parameters
3603 GUILE_MAX_SEGMENT_SIZE Maximal segment size
3606 Allocation of 2-word cell heaps:
3608 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
3611 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
3612 GC in percent of total heap size
3615 Allocation of 4-word cell heaps
3616 (used for real numbers and misc other objects):
3618 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
3620 (See entry "Way for application to customize GC parameters" under
3621 section "Changes to the scm_ interface" below.)
3623 ** Guile now implements reals using 4-word cells
3625 This speeds up computation with reals. (They were earlier allocated
3626 with `malloc'.) There is still some room for optimizations, however.
3628 ** Some further steps toward POSIX thread support have been taken
3630 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
3631 don't have much effect any longer, and many of them will be removed in
3635 are only handled at the top of the evaluator loop, immediately after
3636 I/O, and in scm_equalp.
3638 *** The GC can allocate thread private pools of pairs.
3640 * Changes to Scheme functions and syntax
3642 ** close-input-port and close-output-port are now R5RS
3644 These procedures have been turned into primitives and have R5RS behaviour.
3646 ** New procedure: simple-format PORT MESSAGE ARG1 ...
3648 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
3649 extended by the more sophisticated version in (ice-9 format)
3651 (simple-format port message . args)
3652 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
3653 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
3654 the escapes are replaced with corresponding members of ARGS:
3655 ~A formats using `display' and ~S formats using `write'.
3656 If DESTINATION is #t, then use the `current-output-port',
3657 if DESTINATION is #f, then return a string containing the formatted text.
3658 Does not add a trailing newline."
3660 ** string-ref: the second argument is no longer optional.
3662 ** string, list->string: no longer accept strings in their arguments,
3663 only characters, for compatibility with R5RS.
3665 ** New procedure: port-closed? PORT
3666 Returns #t if PORT is closed or #f if it is open.
3668 ** Deprecated: list*
3670 The list* functionality is now provided by cons* (SRFI-1 compliant)
3672 ** New procedure: cons* ARG1 ARG2 ... ARGn
3674 Like `list', but the last arg provides the tail of the constructed list,
3675 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3677 Requires at least one argument. If given one argument, that argument
3678 is returned as result.
3680 This function is called `list*' in some other Schemes and in Common LISP.
3682 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3684 ** New procedure: object-documentation OBJECT
3686 Returns the documentation string associated with OBJECT. The
3687 procedure uses a caching mechanism so that subsequent lookups are
3690 Exported by (ice-9 documentation).
3692 ** module-name now returns full names of modules
3694 Previously, only the last part of the name was returned (`session' for
3695 `(ice-9 session)'). Ex: `(ice-9 session)'.
3697 * Changes to the gh_ interface
3699 ** Deprecated: gh_int2scmb
3701 Use gh_bool2scm instead.
3703 * Changes to the scm_ interface
3705 ** Guile primitives now carry docstrings!
3707 Thanks to Greg Badros!
3709 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3711 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3712 macros and must contain a docstring that is extracted into foo.doc using a new
3713 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3715 However, a major overhaul of these macros is scheduled for the next release of
3718 ** Guile primitives use a new technique for validation of arguments
3720 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3721 the readability of argument checking.
3723 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3725 ** New macros: SCM_PACK, SCM_UNPACK
3727 Compose/decompose an SCM value.
3729 The SCM type is now treated as an abstract data type and may be defined as a
3730 long, a void* or as a struct, depending on the architecture and compile time
3731 options. This makes it easier to find several types of bugs, for example when
3732 SCM values are treated as integers without conversion. Values of the SCM type
3733 should be treated as "atomic" values. These macros are used when
3734 composing/decomposing an SCM value, either because you want to access
3735 individual bits, or because you want to treat it as an integer value.
3737 E.g., in order to set bit 7 in an SCM value x, use the expression
3739 SCM_PACK (SCM_UNPACK (x) | 0x80)
3741 ** The name property of hooks is deprecated.
3742 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3744 You can emulate this feature by using object properties.
3746 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3747 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3750 These macros will be removed in a future release of Guile.
3752 ** The following types, functions and macros from numbers.h are deprecated:
3753 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3754 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3756 ** Port internals: the rw_random variable in the scm_port structure
3757 must be set to non-zero in any random access port. In recent Guile
3758 releases it was only set for bidirectional random-access ports.
3760 ** Port internals: the seek ptob procedure is now responsible for
3761 resetting the buffers if required. The change was made so that in the
3762 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3763 the fport and strport ptobs can avoid resetting the buffers,
3764 in particular to avoid discarding unread chars. An existing port
3765 type can be fixed by adding something like the following to the
3766 beginning of the ptob seek procedure:
3768 if (pt->rw_active == SCM_PORT_READ)
3769 scm_end_input (object);
3770 else if (pt->rw_active == SCM_PORT_WRITE)
3771 ptob->flush (object);
3773 although to actually avoid resetting the buffers and discard unread
3774 chars requires further hacking that depends on the characteristics
3777 ** Deprecated functions: scm_fseek, scm_tag
3779 These functions are no longer used and will be removed in a future version.
3781 ** The scm_sysmissing procedure is no longer used in libguile.
3782 Unless it turns out to be unexpectedly useful to somebody, it will be
3783 removed in a future version.
3785 ** The format of error message strings has changed
3787 The two C procedures: scm_display_error and scm_error, as well as the
3788 primitive `scm-error', now use scm_simple_format to do their work.
3789 This means that the message strings of all code must be updated to use
3790 ~A where %s was used before, and ~S where %S was used before.
3792 During the period when there still are a lot of old Guiles out there,
3793 you might want to support both old and new versions of Guile.
3795 There are basically two methods to achieve this. Both methods use
3798 AC_CHECK_FUNCS(scm_simple_format)
3800 in your configure.in.
3802 Method 1: Use the string concatenation features of ANSI C's
3807 #ifdef HAVE_SCM_SIMPLE_FORMAT
3813 Then represent each of your error messages using a preprocessor macro:
3815 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3819 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3820 (define make-message string-append)
3822 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3824 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3828 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3833 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3837 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3839 Don't use the functions coop_mutex_init and
3840 coop_condition_variable_init. They will change.
3842 Use scm_mutex_init and scm_cond_init instead.
3844 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3845 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3846 COND, as `scm_cond_wait' does, but it also bounds the duration
3847 of the wait. If COND has not been signaled before time ABSTIME,
3848 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3849 returns the error code `ETIMEDOUT'.
3851 The ABSTIME parameter specifies an absolute time, with the same
3852 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3853 to 00:00:00 GMT, January 1, 1970.
3855 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3856 `scm_cond_broadcast' restarts all the threads that are waiting
3857 on the condition variable COND. Nothing happens if no threads are
3860 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3861 `scm_key_create' allocates a new TSD key. The key is stored in
3862 the location pointed to by KEY. There is no limit on the number
3863 of keys allocated at a given time. The value initially associated
3864 with the returned key is `NULL' in all currently executing threads.
3866 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3867 function associated with the key. When a thread terminates,
3868 DESTR_FUNCTION is called on the value associated with the key in
3869 that thread. The DESTR_FUNCTION is not called if a key is deleted
3870 with `scm_key_delete' or a value is changed with
3871 `scm_setspecific'. The order in which destructor functions are
3872 called at thread termination time is unspecified.
3874 Destructors are not yet implemented.
3876 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3877 `scm_setspecific' changes the value associated with KEY in the
3878 calling thread, storing the given POINTER instead.
3880 ** New function: scm_getspecific (scm_key_t KEY)
3881 `scm_getspecific' returns the value currently associated with
3882 KEY in the calling thread.
3884 ** New function: scm_key_delete (scm_key_t KEY)
3885 `scm_key_delete' deallocates a TSD key. It does not check
3886 whether non-`NULL' values are associated with that key in the
3887 currently executing threads, nor call the destructor function
3888 associated with the key.
3890 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3892 Initialize a C level hook HOOK with associated HOOK_DATA and type
3893 TYPE. (See scm_c_hook_run ().)
3895 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3897 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3898 is true, add it last, otherwise first. The same FUNC can be added
3899 multiple times if FUNC_DATA differ and vice versa.
3901 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3903 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3904 function is only removed if both FUNC and FUNC_DATA matches.
3906 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3908 Run hook HOOK passing DATA to the hook functions.
3910 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3911 returned is undefined.
3913 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3914 returns a non-NULL value. This value is returned as the result of
3915 scm_c_hook_run. If all functions return NULL, NULL is returned.
3917 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3918 returns a NULL value, and NULL is returned. If all functions returns
3919 a non-NULL value, the last value is returned.
3921 ** New C level GC hooks
3923 Five new C level hooks has been added to the garbage collector.
3925 scm_before_gc_c_hook
3928 are run before locking and after unlocking the heap. The system is
3929 thus in a mode where evaluation can take place. (Except that
3930 scm_before_gc_c_hook must not allocate new cells.)
3932 scm_before_mark_c_hook
3933 scm_before_sweep_c_hook
3934 scm_after_sweep_c_hook
3936 are run when the heap is locked. These are intended for extension of
3937 the GC in a modular fashion. Examples are the weaks and guardians
3940 ** Way for application to customize GC parameters
3942 The application can set up other default values for the GC heap
3943 allocation parameters
3945 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3946 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3947 GUILE_MAX_SEGMENT_SIZE,
3951 scm_default_init_heap_size_1, scm_default_min_yield_1,
3952 scm_default_init_heap_size_2, scm_default_min_yield_2,
3953 scm_default_max_segment_size
3955 respectively before callong scm_boot_guile.
3957 (See entry "New environment variables ..." in section
3958 "Changes to the stand-alone interpreter" above.)
3960 ** scm_protect_object/scm_unprotect_object now nest
3962 This means that you can call scm_protect_object multiple times on an
3963 object and count on the object being protected until
3964 scm_unprotect_object has been call the same number of times.
3966 The functions also have better time complexity.
3968 Still, it is usually possible to structure the application in a way
3969 that you don't need to use these functions. For example, if you use a
3970 protected standard Guile list to keep track of live objects rather
3971 than some custom data type, objects will die a natural death when they
3972 are no longer needed.
3974 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3976 Guile does not provide the float representation for inexact real numbers any
3977 more. Now, only doubles are used to represent inexact real numbers. Further,
3978 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3979 and scm_tc16_complex, respectively.
3981 ** Removed deprecated type scm_smobfuns
3983 ** Removed deprecated function scm_newsmob
3985 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3987 There is an ongoing discussion among the developers whether to
3988 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3989 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3990 until this issue has been settled.
3992 ** Removed deprecated type tag scm_tc16_kw
3994 ** Added type tag scm_tc16_keyword
3996 (This was introduced already in release 1.3.4 but was not documented
3999 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
4001 * Changes to system call interfaces:
4003 ** The "select" procedure now tests port buffers for the ability to
4004 provide input or accept output. Previously only the underlying file
4005 descriptors were checked.
4007 ** New variable PIPE_BUF: the maximum number of bytes that can be
4008 atomically written to a pipe.
4010 ** If a facility is not available on the system when Guile is
4011 compiled, the corresponding primitive procedure will not be defined.
4012 Previously it would have been defined but would throw a system-error
4013 exception if called. Exception handlers which catch this case may
4014 need minor modification: an error will be thrown with key
4015 'unbound-variable instead of 'system-error. Alternatively it's
4016 now possible to use `defined?' to check whether the facility is
4019 ** Procedures which depend on the timezone should now give the correct
4020 result on systems which cache the TZ environment variable, even if TZ
4021 is changed without calling tzset.
4023 * Changes to the networking interfaces:
4025 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
4026 long integers between network and host format. For now, it's not
4027 particularly convenient to do this kind of thing, but consider:
4029 (define write-network-long
4030 (lambda (value port)
4031 (let ((v (make-uniform-vector 1 1 0)))
4032 (uniform-vector-set! v 0 (htonl value))
4033 (uniform-vector-write v port))))
4035 (define read-network-long
4037 (let ((v (make-uniform-vector 1 1 0)))
4038 (uniform-vector-read! v port)
4039 (ntohl (uniform-vector-ref v 0)))))
4041 ** If inet-aton fails, it now throws an error with key 'misc-error
4042 instead of 'system-error, since errno is not relevant.
4044 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
4045 specific keys instead of 'system-error. The latter is inappropriate
4046 since errno will not have been set. The keys are:
4047 'host-not-found, 'try-again, 'no-recovery and 'no-data.
4049 ** sethostent, setnetent, setprotoent, setservent: now take an
4050 optional argument STAYOPEN, which specifies whether the database
4051 remains open after a database entry is accessed randomly (e.g., using
4052 gethostbyname for the hosts database.) The default is #f. Previously
4056 Changes since Guile 1.3.2:
4058 * Changes to the stand-alone interpreter
4062 An initial version of the Guile debugger written by Chris Hanson has
4063 been added. The debugger is still under development but is included
4064 in the distribution anyway since it is already quite useful.
4070 after an error to enter the debugger. Type `help' inside the debugger
4071 for a description of available commands.
4073 If you prefer to have stack frames numbered and printed in
4074 anti-chronological order and prefer up in the stack to be down on the
4075 screen as is the case in gdb, you can put
4077 (debug-enable 'backwards)
4079 in your .guile startup file. (However, this means that Guile can't
4080 use indentation to indicate stack level.)
4082 The debugger is autoloaded into Guile at the first use.
4084 ** Further enhancements to backtraces
4086 There is a new debug option `width' which controls the maximum width
4087 on the screen of printed stack frames. Fancy printing parameters
4088 ("level" and "length" as in Common LISP) are adaptively adjusted for
4089 each stack frame to give maximum information while still fitting
4090 within the bounds. If the stack frame can't be made to fit by
4091 adjusting parameters, it is simply cut off at the end. This is marked
4094 ** Some modules are now only loaded when the repl is started
4096 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
4097 regex) are now loaded into (guile-user) only if the repl has been
4098 started. The effect is that the startup time for scripts has been
4099 reduced to 30% of what it was previously.
4101 Correctly written scripts load the modules they require at the top of
4102 the file and should not be affected by this change.
4104 ** Hooks are now represented as smobs
4106 * Changes to Scheme functions and syntax
4108 ** Readline support has changed again.
4110 The old (readline-activator) module is gone. Use (ice-9 readline)
4111 instead, which now contains all readline functionality. So the code
4112 to activate readline is now
4114 (use-modules (ice-9 readline))
4117 This should work at any time, including from the guile prompt.
4119 To avoid confusion about the terms of Guile's license, please only
4120 enable readline for your personal use; please don't make it the
4121 default for others. Here is why we make this rather odd-sounding
4124 Guile is normally licensed under a weakened form of the GNU General
4125 Public License, which allows you to link code with Guile without
4126 placing that code under the GPL. This exception is important to some
4129 However, since readline is distributed under the GNU General Public
4130 License, when you link Guile with readline, either statically or
4131 dynamically, you effectively change Guile's license to the strict GPL.
4132 Whenever you link any strictly GPL'd code into Guile, uses of Guile
4133 which are normally permitted become forbidden. This is a rather
4134 non-obvious consequence of the licensing terms.
4136 So, to make sure things remain clear, please let people choose for
4137 themselves whether to link GPL'd libraries like readline with Guile.
4139 ** regexp-substitute/global has changed slightly, but incompatibly.
4141 If you include a function in the item list, the string of the match
4142 object it receives is the same string passed to
4143 regexp-substitute/global, not some suffix of that string.
4144 Correspondingly, the match's positions are relative to the entire
4145 string, not the suffix.
4147 If the regexp can match the empty string, the way matches are chosen
4148 from the string has changed. regexp-substitute/global recognizes the
4149 same set of matches that list-matches does; see below.
4151 ** New function: list-matches REGEXP STRING [FLAGS]
4153 Return a list of match objects, one for every non-overlapping, maximal
4154 match of REGEXP in STRING. The matches appear in left-to-right order.
4155 list-matches only reports matches of the empty string if there are no
4156 other matches which begin on, end at, or include the empty match's
4159 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4161 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
4163 For each match of REGEXP in STRING, apply PROC to the match object,
4164 and the last value PROC returned, or INIT for the first call. Return
4165 the last value returned by PROC. We apply PROC to the matches as they
4166 appear from left to right.
4168 This function recognizes matches according to the same criteria as
4171 Thus, you could define list-matches like this:
4173 (define (list-matches regexp string . flags)
4174 (reverse! (apply fold-matches regexp string '() cons flags)))
4176 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4180 *** New function: hook? OBJ
4182 Return #t if OBJ is a hook, otherwise #f.
4184 *** New function: make-hook-with-name NAME [ARITY]
4186 Return a hook with name NAME and arity ARITY. The default value for
4187 ARITY is 0. The only effect of NAME is that it will appear when the
4188 hook object is printed to ease debugging.
4190 *** New function: hook-empty? HOOK
4192 Return #t if HOOK doesn't contain any procedures, otherwise #f.
4194 *** New function: hook->list HOOK
4196 Return a list of the procedures that are called when run-hook is
4199 ** `map' signals an error if its argument lists are not all the same length.
4201 This is the behavior required by R5RS, so this change is really a bug
4202 fix. But it seems to affect a lot of people's code, so we're
4203 mentioning it here anyway.
4205 ** Print-state handling has been made more transparent
4207 Under certain circumstances, ports are represented as a port with an
4208 associated print state. Earlier, this pair was represented as a pair
4209 (see "Some magic has been added to the printer" below). It is now
4210 indistinguishable (almost; see `get-print-state') from a port on the
4213 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
4215 Return a new port with the associated print state PRINT-STATE.
4217 *** New function: get-print-state OUTPUT-PORT
4219 Return the print state associated with this port if it exists,
4220 otherwise return #f.
4222 *** New function: directory-stream? OBJECT
4224 Returns true iff OBJECT is a directory stream --- the sort of object
4225 returned by `opendir'.
4227 ** New function: using-readline?
4229 Return #t if readline is in use in the current repl.
4231 ** structs will be removed in 1.4
4233 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
4234 and use GOOPS objects as the fundamental record type.
4236 * Changes to the scm_ interface
4238 ** structs will be removed in 1.4
4240 The entire current struct interface (struct.c, struct.h) will be
4241 replaced in Guile 1.4. We will merge GOOPS into libguile and use
4242 GOOPS objects as the fundamental record type.
4244 ** The internal representation of subr's has changed
4246 Instead of giving a hint to the subr name, the CAR field of the subr
4247 now contains an index to a subr entry in scm_subr_table.
4249 *** New variable: scm_subr_table
4251 An array of subr entries. A subr entry contains the name, properties
4252 and documentation associated with the subr. The properties and
4253 documentation slots are not yet used.
4255 ** A new scheme for "forwarding" calls to a builtin to a generic function
4257 It is now possible to extend the functionality of some Guile
4258 primitives by letting them defer a call to a GOOPS generic function on
4259 argument mismatch. This means that there is no loss of efficiency in
4264 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
4265 (define-method + ((x <string>) (y <string>))
4266 (string-append x y))
4268 + will still be as efficient as usual in numerical calculations, but
4269 can also be used for concatenating strings.
4271 Who will be the first one to extend Guile's numerical tower to
4272 rationals? :) [OK, there a few other things to fix before this can
4273 be made in a clean way.]
4275 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
4277 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
4279 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
4281 These do the same job as SCM_PROC and SCM_PROC1, but they also define
4282 a variable GENERIC which can be used by the dispatch macros below.
4284 [This is experimental code which may change soon.]
4286 *** New macros for forwarding control to a generic on arg type error
4288 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
4290 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
4292 These correspond to the scm_wta function call, and have the same
4293 behaviour until the user has called the GOOPS primitive
4294 `enable-primitive-generic!'. After that, these macros will apply the
4295 generic function GENERIC to the argument(s) instead of calling
4298 [This is experimental code which may change soon.]
4300 *** New macros for argument testing with generic dispatch
4302 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
4304 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
4306 These correspond to the SCM_ASSERT macro, but will defer control to
4307 GENERIC on error after `enable-primitive-generic!' has been called.
4309 [This is experimental code which may change soon.]
4311 ** New function: SCM scm_eval_body (SCM body, SCM env)
4313 Evaluates the body of a special form.
4315 ** The internal representation of struct's has changed
4317 Previously, four slots were allocated for the procedure(s) of entities
4318 and operators. The motivation for this representation had to do with
4319 the structure of the evaluator, the wish to support tail-recursive
4320 generic functions, and efficiency. Since the generic function
4321 dispatch mechanism has changed, there is no longer a need for such an
4322 expensive representation, and the representation has been simplified.
4324 This should not make any difference for most users.
4326 ** GOOPS support has been cleaned up.
4328 Some code has been moved from eval.c to objects.c and code in both of
4329 these compilation units has been cleaned up and better structured.
4331 *** New functions for applying generic functions
4333 New function: SCM scm_apply_generic (GENERIC, ARGS)
4334 New function: SCM scm_call_generic_0 (GENERIC)
4335 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
4336 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
4337 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
4339 ** Deprecated function: scm_make_named_hook
4341 It is now replaced by:
4343 ** New function: SCM scm_create_hook (const char *name, int arity)
4345 Creates a hook in the same way as make-hook above but also
4346 binds a variable named NAME to it.
4348 This is the typical way of creating a hook from C code.
4350 Currently, the variable is created in the "current" module.
4351 This might change when we get the new module system.
4353 [The behaviour is identical to scm_make_named_hook.]
4357 Changes since Guile 1.3:
4359 * Changes to mailing lists
4361 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
4363 See the README file to find current addresses for all the Guile
4366 * Changes to the distribution
4368 ** Readline support is no longer included with Guile by default.
4370 Based on the different license terms of Guile and Readline, we
4371 concluded that Guile should not *by default* cause the linking of
4372 Readline into an application program. Readline support is now offered
4373 as a separate module, which is linked into an application only when
4374 you explicitly specify it.
4376 Although Guile is GNU software, its distribution terms add a special
4377 exception to the usual GNU General Public License (GPL). Guile's
4378 license includes a clause that allows you to link Guile with non-free
4379 programs. We add this exception so as not to put Guile at a
4380 disadvantage vis-a-vis other extensibility packages that support other
4383 In contrast, the GNU Readline library is distributed under the GNU
4384 General Public License pure and simple. This means that you may not
4385 link Readline, even dynamically, into an application unless it is
4386 distributed under a free software license that is compatible the GPL.
4388 Because of this difference in distribution terms, an application that
4389 can use Guile may not be able to use Readline. Now users will be
4390 explicitly offered two independent decisions about the use of these
4393 You can activate the readline support by issuing
4395 (use-modules (readline-activator))
4398 from your ".guile" file, for example.
4400 * Changes to the stand-alone interpreter
4402 ** All builtins now print as primitives.
4403 Previously builtin procedures not belonging to the fundamental subr
4404 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
4405 Now, they print as #<primitive-procedure NAME>.
4407 ** Backtraces slightly more intelligible.
4408 gsubr-apply and macro transformer application frames no longer appear
4411 * Changes to Scheme functions and syntax
4413 ** Guile now correctly handles internal defines by rewriting them into
4414 their equivalent letrec. Previously, internal defines would
4415 incrementally add to the innermost environment, without checking
4416 whether the restrictions specified in RnRS were met. This lead to the
4417 correct behaviour when these restriction actually were met, but didn't
4418 catch all illegal uses. Such an illegal use could lead to crashes of
4419 the Guile interpreter or or other unwanted results. An example of
4420 incorrect internal defines that made Guile behave erratically:
4432 The problem with this example is that the definition of `c' uses the
4433 value of `b' directly. This confuses the meoization machine of Guile
4434 so that the second call of `b' (this time in a larger environment that
4435 also contains bindings for `c' and `d') refers to the binding of `c'
4436 instead of `a'. You could also make Guile crash with a variation on
4441 (define (b flag) (if flag a 1))
4442 (define c (1+ (b flag)))
4450 From now on, Guile will issue an `Unbound variable: b' error message
4455 A hook contains a list of functions which should be called on
4456 particular occasions in an existing program. Hooks are used for
4459 A window manager might have a hook before-window-map-hook. The window
4460 manager uses the function run-hooks to call all functions stored in
4461 before-window-map-hook each time a window is mapped. The user can
4462 store functions in the hook using add-hook!.
4464 In Guile, hooks are first class objects.
4466 *** New function: make-hook [N_ARGS]
4468 Return a hook for hook functions which can take N_ARGS arguments.
4469 The default value for N_ARGS is 0.
4471 (See also scm_make_named_hook below.)
4473 *** New function: add-hook! HOOK PROC [APPEND_P]
4475 Put PROC at the beginning of the list of functions stored in HOOK.
4476 If APPEND_P is supplied, and non-false, put PROC at the end instead.
4478 PROC must be able to take the number of arguments specified when the
4481 If PROC already exists in HOOK, then remove it first.
4483 *** New function: remove-hook! HOOK PROC
4485 Remove PROC from the list of functions in HOOK.
4487 *** New function: reset-hook! HOOK
4489 Clear the list of hook functions stored in HOOK.
4491 *** New function: run-hook HOOK ARG1 ...
4493 Run all hook functions stored in HOOK with arguments ARG1 ... .
4494 The number of arguments supplied must correspond to the number given
4495 when the hook was created.
4497 ** The function `dynamic-link' now takes optional keyword arguments.
4498 The only keyword argument that is currently defined is `:global
4499 BOOL'. With it, you can control whether the shared library will be
4500 linked in global mode or not. In global mode, the symbols from the
4501 linked library can be used to resolve references from other
4502 dynamically linked libraries. In non-global mode, the linked
4503 library is essentially invisible and can only be accessed via
4504 `dynamic-func', etc. The default is now to link in global mode.
4505 Previously, the default has been non-global mode.
4507 The `#:global' keyword is only effective on platforms that support
4508 the dlopen family of functions.
4510 ** New function `provided?'
4512 - Function: provided? FEATURE
4513 Return true iff FEATURE is supported by this installation of
4514 Guile. FEATURE must be a symbol naming a feature; the global
4515 variable `*features*' is a list of available features.
4517 ** Changes to the module (ice-9 expect):
4519 *** The expect-strings macro now matches `$' in a regular expression
4520 only at a line-break or end-of-file by default. Previously it would
4521 match the end of the string accumulated so far. The old behaviour
4522 can be obtained by setting the variable `expect-strings-exec-flags'
4525 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
4526 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
4527 in a regular expression will still match before a line-break or
4528 end-of-file. The default is `regexp/noteol'.
4530 *** The expect-strings macro now uses a variable
4531 `expect-strings-compile-flags' for the flags to be supplied to
4532 `make-regexp'. The default is `regexp/newline', which was previously
4535 *** The expect macro now supplies two arguments to a match procedure:
4536 the current accumulated string and a flag to indicate whether
4537 end-of-file has been reached. Previously only the string was supplied.
4538 If end-of-file is reached, the match procedure will be called an
4539 additional time with the same accumulated string as the previous call
4540 but with the flag set.
4542 ** New module (ice-9 format), implementing the Common Lisp `format' function.
4544 This code, and the documentation for it that appears here, was
4545 borrowed from SLIB, with minor adaptations for Guile.
4547 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
4548 An almost complete implementation of Common LISP format description
4549 according to the CL reference book `Common LISP' from Guy L.
4550 Steele, Digital Press. Backward compatible to most of the
4551 available Scheme format implementations.
4553 Returns `#t', `#f' or a string; has side effect of printing
4554 according to FORMAT-STRING. If DESTINATION is `#t', the output is
4555 to the current output port and `#t' is returned. If DESTINATION
4556 is `#f', a formatted string is returned as the result of the call.
4557 NEW: If DESTINATION is a string, DESTINATION is regarded as the
4558 format string; FORMAT-STRING is then the first argument and the
4559 output is returned as a string. If DESTINATION is a number, the
4560 output is to the current error port if available by the
4561 implementation. Otherwise DESTINATION must be an output port and
4564 FORMAT-STRING must be a string. In case of a formatting error
4565 format returns `#f' and prints a message on the current output or
4566 error port. Characters are output as if the string were output by
4567 the `display' function with the exception of those prefixed by a
4568 tilde (~). For a detailed description of the FORMAT-STRING syntax
4569 please consult a Common LISP format reference manual. For a test
4570 suite to verify this format implementation load `formatst.scm'.
4571 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
4573 Note: `format' is not reentrant, i.e. only one `format'-call may
4574 be executed at a time.
4577 *** Format Specification (Format version 3.0)
4579 Please consult a Common LISP format reference manual for a detailed
4580 description of the format string syntax. For a demonstration of the
4581 implemented directives see `formatst.scm'.
4583 This implementation supports directive parameters and modifiers (`:'
4584 and `@' characters). Multiple parameters must be separated by a comma
4585 (`,'). Parameters can be numerical parameters (positive or negative),
4586 character parameters (prefixed by a quote character (`''), variable
4587 parameters (`v'), number of rest arguments parameter (`#'), empty and
4588 default parameters. Directive characters are case independent. The
4589 general form of a directive is:
4591 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
4593 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
4595 *** Implemented CL Format Control Directives
4597 Documentation syntax: Uppercase characters represent the
4598 corresponding control directive characters. Lowercase characters
4599 represent control directive parameter descriptions.
4602 Any (print as `display' does).
4606 `~MINCOL,COLINC,MINPAD,PADCHARA'
4610 S-expression (print as `write' does).
4614 `~MINCOL,COLINC,MINPAD,PADCHARS'
4620 print number sign always.
4623 print comma separated.
4625 `~MINCOL,PADCHAR,COMMACHARD'
4631 print number sign always.
4634 print comma separated.
4636 `~MINCOL,PADCHAR,COMMACHARX'
4642 print number sign always.
4645 print comma separated.
4647 `~MINCOL,PADCHAR,COMMACHARO'
4653 print number sign always.
4656 print comma separated.
4658 `~MINCOL,PADCHAR,COMMACHARB'
4663 `~N,MINCOL,PADCHAR,COMMACHARR'
4667 print a number as a Roman numeral.
4670 print a number as an "old fashioned" Roman numeral.
4673 print a number as an ordinal English number.
4676 print a number as a cardinal English number.
4681 prints `y' and `ies'.
4684 as `~P but jumps 1 argument backward.'
4687 as `~@P but jumps 1 argument backward.'
4692 prints a character as the reader can understand it (i.e. `#\'
4696 prints a character as emacs does (eg. `^C' for ASCII 03).
4699 Fixed-format floating-point (prints a flonum like MMM.NNN).
4700 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4702 If the number is positive a plus sign is printed.
4705 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4706 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4708 If the number is positive a plus sign is printed.
4711 General floating-point (prints a flonum either fixed or
4713 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4715 If the number is positive a plus sign is printed.
4718 Dollars floating-point (prints a flonum in fixed with signs
4720 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4722 If the number is positive a plus sign is printed.
4725 A sign is always printed and appears before the padding.
4728 The sign appears before the padding.
4736 print newline if not at the beginning of the output line.
4738 prints `~&' and then N-1 newlines.
4743 print N page separators.
4753 newline is ignored, white space left.
4756 newline is left, white space ignored.
4761 relative tabulation.
4767 Indirection (expects indirect arguments as a list).
4769 extracts indirect arguments from format arguments.
4772 Case conversion (converts by `string-downcase').
4774 converts by `string-capitalize'.
4777 converts by `string-capitalize-first'.
4780 converts by `string-upcase'.
4783 Argument Jumping (jumps 1 argument forward).
4785 jumps N arguments forward.
4788 jumps 1 argument backward.
4791 jumps N arguments backward.
4794 jumps to the 0th argument.
4797 jumps to the Nth argument (beginning from 0)
4799 `~[STR0~;STR1~;...~;STRN~]'
4800 Conditional Expression (numerical clause conditional).
4802 take argument from N.
4805 true test conditional.
4808 if-else-then conditional.
4814 default clause follows.
4817 Iteration (args come from the next argument (a list)).
4819 at most N iterations.
4822 args from next arg (a list of lists).
4825 args from the rest of arguments.
4828 args from the rest args (lists).
4839 aborts if N <= M <= K
4841 *** Not Implemented CL Format Control Directives
4844 print `#f' as an empty list (see below).
4847 print `#f' as an empty list (see below).
4853 (sorry I don't understand its semantics completely)
4855 *** Extended, Replaced and Additional Control Directives
4857 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4858 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4859 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4860 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4861 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4862 COMMAWIDTH is the number of characters between two comma
4866 print a R4RS complex number as `~F~@Fi' with passed parameters for
4870 Pretty print formatting of an argument for scheme code lists.
4876 Flushes the output if format DESTINATION is a port.
4879 Print a `#\space' character
4881 print N `#\space' characters.
4884 Print a `#\tab' character
4886 print N `#\tab' characters.
4889 Takes N as an integer representation for a character. No arguments
4890 are consumed. N is converted to a character by `integer->char'. N
4891 must be a positive decimal number.
4894 Print out readproof. Prints out internal objects represented as
4895 `#<...>' as strings `"#<...>"' so that the format output can always
4896 be processed by `read'.
4899 Print out readproof. Prints out internal objects represented as
4900 `#<...>' as strings `"#<...>"' so that the format output can always
4901 be processed by `read'.
4904 Prints information and a copyright notice on the format
4907 prints format version.
4910 may also print number strings, i.e. passing a number as a string
4911 and format it accordingly.
4913 *** Configuration Variables
4915 The format module exports some configuration variables to suit the
4916 systems and users needs. There should be no modification necessary for
4917 the configuration that comes with Guile. Format detects automatically
4918 if the running scheme system implements floating point numbers and
4921 format:symbol-case-conv
4922 Symbols are converted by `symbol->string' so the case type of the
4923 printed symbols is implementation dependent.
4924 `format:symbol-case-conv' is a one arg closure which is either
4925 `#f' (no conversion), `string-upcase', `string-downcase' or
4926 `string-capitalize'. (default `#f')
4928 format:iobj-case-conv
4929 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4930 implementation internal objects. (default `#f')
4933 The character prefixing the exponent value in `~E' printing.
4936 *** Compatibility With Other Format Implementations
4942 Downward compatible except for padding support and `~A', `~S',
4943 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4944 `printf' padding support which is completely replaced by the CL
4945 `format' padding style.
4948 Downward compatible except for `~', which is not documented
4949 (ignores all characters inside the format string up to a newline
4950 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4951 numerical and variable parameters and `:/@' modifiers in the CL
4955 Downward compatible except for `~A' and `~S' which print in
4956 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4957 directive parameters or modifiers)).
4960 Downward compatible except for an optional destination parameter:
4961 S2C accepts a format call without a destination which returns a
4962 formatted string. This is equivalent to a #f destination in S2C.
4963 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4964 parameters or modifiers)).
4967 ** Changes to string-handling functions.
4969 These functions were added to support the (ice-9 format) module, above.
4971 *** New function: string-upcase STRING
4972 *** New function: string-downcase STRING
4974 These are non-destructive versions of the existing string-upcase! and
4975 string-downcase! functions.
4977 *** New function: string-capitalize! STRING
4978 *** New function: string-capitalize STRING
4980 These functions convert the first letter of each word in the string to
4983 (string-capitalize "howdy there")
4986 As with the other functions, string-capitalize! modifies the string in
4987 place, while string-capitalize returns a modified copy of its argument.
4989 *** New function: string-ci->symbol STRING
4991 Return a symbol whose name is STRING, but having the same case as if
4992 the symbol had be read by `read'.
4994 Guile can be configured to be sensitive or insensitive to case
4995 differences in Scheme identifiers. If Guile is case-insensitive, all
4996 symbols are converted to lower case on input. The `string-ci->symbol'
4997 function returns a symbol whose name in STRING, transformed as Guile
4998 would if STRING were input.
5000 *** New function: substring-move! STRING1 START END STRING2 START
5002 Copy the substring of STRING1 from START (inclusive) to END
5003 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
5004 string, and the source and destination areas may overlap; in all
5005 cases, the function behaves as if all the characters were copied
5008 *** Extended functions: substring-move-left! substring-move-right!
5010 These functions now correctly copy arbitrarily overlapping substrings;
5011 they are both synonyms for substring-move!.
5014 ** New module (ice-9 getopt-long), with the function `getopt-long'.
5016 getopt-long is a function for parsing command-line arguments in a
5017 manner consistent with other GNU programs.
5019 (getopt-long ARGS GRAMMAR)
5020 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
5022 ARGS should be a list of strings. Its first element should be the
5023 name of the program; subsequent elements should be the arguments
5024 that were passed to the program on the command line. The
5025 `program-arguments' procedure returns a list of this form.
5027 GRAMMAR is a list of the form:
5028 ((OPTION (PROPERTY VALUE) ...) ...)
5030 Each OPTION should be a symbol. `getopt-long' will accept a
5031 command-line option named `--OPTION'.
5032 Each option can have the following (PROPERTY VALUE) pairs:
5034 (single-char CHAR) --- Accept `-CHAR' as a single-character
5035 equivalent to `--OPTION'. This is how to specify traditional
5037 (required? BOOL) --- If BOOL is true, the option is required.
5038 getopt-long will raise an error if it is not found in ARGS.
5039 (value BOOL) --- If BOOL is #t, the option accepts a value; if
5040 it is #f, it does not; and if it is the symbol
5041 `optional', the option may appear in ARGS with or
5043 (predicate FUNC) --- If the option accepts a value (i.e. you
5044 specified `(value #t)' for this option), then getopt
5045 will apply FUNC to the value, and throw an exception
5046 if it returns #f. FUNC should be a procedure which
5047 accepts a string and returns a boolean value; you may
5048 need to use quasiquotes to get it into GRAMMAR.
5050 The (PROPERTY VALUE) pairs may occur in any order, but each
5051 property may occur only once. By default, options do not have
5052 single-character equivalents, are not required, and do not take
5055 In ARGS, single-character options may be combined, in the usual
5056 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
5057 accepts values, then it must be the last option in the
5058 combination; the value is the next argument. So, for example, using
5059 the following grammar:
5060 ((apples (single-char #\a))
5061 (blimps (single-char #\b) (value #t))
5062 (catalexis (single-char #\c) (value #t)))
5063 the following argument lists would be acceptable:
5064 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
5065 for "blimps" and "catalexis")
5066 ("-ab" "bang" "-c" "couth") (same)
5067 ("-ac" "couth" "-b" "bang") (same)
5068 ("-abc" "couth" "bang") (an error, since `-b' is not the
5069 last option in its combination)
5071 If an option's value is optional, then `getopt-long' decides
5072 whether it has a value by looking at what follows it in ARGS. If
5073 the next element is a string, and it does not appear to be an
5074 option itself, then that string is the option's value.
5076 The value of a long option can appear as the next element in ARGS,
5077 or it can follow the option name, separated by an `=' character.
5078 Thus, using the same grammar as above, the following argument lists
5080 ("--apples" "Braeburn" "--blimps" "Goodyear")
5081 ("--apples=Braeburn" "--blimps" "Goodyear")
5082 ("--blimps" "Goodyear" "--apples=Braeburn")
5084 If the option "--" appears in ARGS, argument parsing stops there;
5085 subsequent arguments are returned as ordinary arguments, even if
5086 they resemble options. So, in the argument list:
5087 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
5088 `getopt-long' will recognize the `apples' option as having the
5089 value "Granny Smith", but it will not recognize the `blimp'
5090 option; it will return the strings "--blimp" and "Goodyear" as
5091 ordinary argument strings.
5093 The `getopt-long' function returns the parsed argument list as an
5094 assocation list, mapping option names --- the symbols from GRAMMAR
5095 --- onto their values, or #t if the option does not accept a value.
5096 Unused options do not appear in the alist.
5098 All arguments that are not the value of any option are returned
5099 as a list, associated with the empty list.
5101 `getopt-long' throws an exception if:
5102 - it finds an unrecognized option in ARGS
5103 - a required option is omitted
5104 - an option that requires an argument doesn't get one
5105 - an option that doesn't accept an argument does get one (this can
5106 only happen using the long option `--opt=value' syntax)
5107 - an option predicate fails
5112 `((lockfile-dir (required? #t)
5115 (predicate ,file-is-directory?))
5116 (verbose (required? #f)
5119 (x-includes (single-char #\x))
5120 (rnet-server (single-char #\y)
5121 (predicate ,string?))))
5123 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
5124 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
5126 => ((() "foo1" "-fred" "foo2" "foo3")
5127 (rnet-server . "lamprod")
5128 (x-includes . "/usr/include")
5129 (lockfile-dir . "/tmp")
5132 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
5134 It will be removed in a few releases.
5136 ** New syntax: lambda*
5137 ** New syntax: define*
5138 ** New syntax: define*-public
5139 ** New syntax: defmacro*
5140 ** New syntax: defmacro*-public
5141 Guile now supports optional arguments.
5143 `lambda*', `define*', `define*-public', `defmacro*' and
5144 `defmacro*-public' are identical to the non-* versions except that
5145 they use an extended type of parameter list that has the following BNF
5146 syntax (parentheses are literal, square brackets indicate grouping,
5147 and `*', `+' and `?' have the usual meaning):
5149 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
5150 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
5151 [[#&rest identifier]|[. identifier]]? ) | [identifier]
5153 ext-var-decl ::= identifier | ( identifier expression )
5155 The semantics are best illustrated with the following documentation
5156 and examples for `lambda*':
5159 lambda extended for optional and keyword arguments
5161 lambda* creates a procedure that takes optional arguments. These
5162 are specified by putting them inside brackets at the end of the
5163 paramater list, but before any dotted rest argument. For example,
5164 (lambda* (a b #&optional c d . e) '())
5165 creates a procedure with fixed arguments a and b, optional arguments c
5166 and d, and rest argument e. If the optional arguments are omitted
5167 in a call, the variables for them are unbound in the procedure. This
5168 can be checked with the bound? macro.
5170 lambda* can also take keyword arguments. For example, a procedure
5172 (lambda* (#&key xyzzy larch) '())
5173 can be called with any of the argument lists (#:xyzzy 11)
5174 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
5175 are given as keywords are bound to values.
5177 Optional and keyword arguments can also be given default values
5178 which they take on when they are not present in a call, by giving a
5179 two-item list in place of an optional argument, for example in:
5180 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
5181 foo is a fixed argument, bar is an optional argument with default
5182 value 42, and baz is a keyword argument with default value 73.
5183 Default value expressions are not evaluated unless they are needed
5184 and until the procedure is called.
5186 lambda* now supports two more special parameter list keywords.
5188 lambda*-defined procedures now throw an error by default if a
5189 keyword other than one of those specified is found in the actual
5190 passed arguments. However, specifying #&allow-other-keys
5191 immediately after the kyword argument declarations restores the
5192 previous behavior of ignoring unknown keywords. lambda* also now
5193 guarantees that if the same keyword is passed more than once, the
5194 last one passed is the one that takes effect. For example,
5195 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
5196 #:heads 37 #:tails 42 #:heads 99)
5197 would result in (99 47) being displayed.
5199 #&rest is also now provided as a synonym for the dotted syntax rest
5200 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
5201 all respects to lambda*. This is provided for more similarity to DSSSL,
5202 MIT-Scheme and Kawa among others, as well as for refugees from other
5205 Further documentation may be found in the optargs.scm file itself.
5207 The optional argument module also exports the macros `let-optional',
5208 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
5209 are not documented here because they may be removed in the future, but
5210 full documentation is still available in optargs.scm.
5212 ** New syntax: and-let*
5213 Guile now supports the `and-let*' form, described in the draft SRFI-2.
5215 Syntax: (land* (<clause> ...) <body> ...)
5216 Each <clause> should have one of the following forms:
5217 (<variable> <expression>)
5220 Each <variable> or <bound-variable> should be an identifier. Each
5221 <expression> should be a valid expression. The <body> should be a
5222 possibly empty sequence of expressions, like the <body> of a
5225 Semantics: A LAND* expression is evaluated by evaluating the
5226 <expression> or <bound-variable> of each of the <clause>s from
5227 left to right. The value of the first <expression> or
5228 <bound-variable> that evaluates to a false value is returned; the
5229 remaining <expression>s and <bound-variable>s are not evaluated.
5230 The <body> forms are evaluated iff all the <expression>s and
5231 <bound-variable>s evaluate to true values.
5233 The <expression>s and the <body> are evaluated in an environment
5234 binding each <variable> of the preceding (<variable> <expression>)
5235 clauses to the value of the <expression>. Later bindings
5236 shadow earlier bindings.
5238 Guile's and-let* macro was contributed by Michael Livshin.
5240 ** New sorting functions
5242 *** New function: sorted? SEQUENCE LESS?
5243 Returns `#t' when the sequence argument is in non-decreasing order
5244 according to LESS? (that is, there is no adjacent pair `... x y
5245 ...' for which `(less? y x)').
5247 Returns `#f' when the sequence contains at least one out-of-order
5248 pair. It is an error if the sequence is neither a list nor a
5251 *** New function: merge LIST1 LIST2 LESS?
5252 LIST1 and LIST2 are sorted lists.
5253 Returns the sorted list of all elements in LIST1 and LIST2.
5255 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
5256 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
5257 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
5258 (Here "<" should read "comes before".)
5260 *** New procedure: merge! LIST1 LIST2 LESS?
5261 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
5262 the result. If the code is compiled, and LESS? constructs no new
5263 pairs, no pairs at all will be allocated. The first pair of the
5264 result will be either the first pair of LIST1 or the first pair of
5267 *** New function: sort SEQUENCE LESS?
5268 Accepts either a list or a vector, and returns a new sequence
5269 which is sorted. The new sequence is the same type as the input.
5270 Always `(sorted? (sort sequence less?) less?)'. The original
5271 sequence is not altered in any way. The new sequence shares its
5272 elements with the old one; no elements are copied.
5274 *** New procedure: sort! SEQUENCE LESS
5275 Returns its sorted result in the original boxes. No new storage is
5276 allocated at all. Proper usage: (set! slist (sort! slist <))
5278 *** New function: stable-sort SEQUENCE LESS?
5279 Similar to `sort' but stable. That is, if "equal" elements are
5280 ordered a < b in the original sequence, they will have the same order
5283 *** New function: stable-sort! SEQUENCE LESS?
5284 Similar to `sort!' but stable.
5285 Uses temporary storage when sorting vectors.
5287 *** New functions: sort-list, sort-list!
5288 Added for compatibility with scsh.
5290 ** New built-in random number support
5292 *** New function: random N [STATE]
5293 Accepts a positive integer or real N and returns a number of the
5294 same type between zero (inclusive) and N (exclusive). The values
5295 returned have a uniform distribution.
5297 The optional argument STATE must be of the type produced by
5298 `copy-random-state' or `seed->random-state'. It defaults to the value
5299 of the variable `*random-state*'. This object is used to maintain the
5300 state of the pseudo-random-number generator and is altered as a side
5301 effect of the `random' operation.
5303 *** New variable: *random-state*
5304 Holds a data structure that encodes the internal state of the
5305 random-number generator that `random' uses by default. The nature
5306 of this data structure is implementation-dependent. It may be
5307 printed out and successfully read back in, but may or may not
5308 function correctly as a random-number state object in another
5311 *** New function: copy-random-state [STATE]
5312 Returns a new object of type suitable for use as the value of the
5313 variable `*random-state*' and as a second argument to `random'.
5314 If argument STATE is given, a copy of it is returned. Otherwise a
5315 copy of `*random-state*' is returned.
5317 *** New function: seed->random-state SEED
5318 Returns a new object of type suitable for use as the value of the
5319 variable `*random-state*' and as a second argument to `random'.
5320 SEED is a string or a number. A new state is generated and
5321 initialized using SEED.
5323 *** New function: random:uniform [STATE]
5324 Returns an uniformly distributed inexact real random number in the
5325 range between 0 and 1.
5327 *** New procedure: random:solid-sphere! VECT [STATE]
5328 Fills VECT with inexact real random numbers the sum of whose
5329 squares is less than 1.0. Thinking of VECT as coordinates in
5330 space of dimension N = `(vector-length VECT)', the coordinates are
5331 uniformly distributed within the unit N-shere. The sum of the
5332 squares of the numbers is returned. VECT can be either a vector
5333 or a uniform vector of doubles.
5335 *** New procedure: random:hollow-sphere! VECT [STATE]
5336 Fills VECT with inexact real random numbers the sum of whose squares
5337 is equal to 1.0. Thinking of VECT as coordinates in space of
5338 dimension n = `(vector-length VECT)', the coordinates are uniformly
5339 distributed over the surface of the unit n-shere. VECT can be either
5340 a vector or a uniform vector of doubles.
5342 *** New function: random:normal [STATE]
5343 Returns an inexact real in a normal distribution with mean 0 and
5344 standard deviation 1. For a normal distribution with mean M and
5345 standard deviation D use `(+ M (* D (random:normal)))'.
5347 *** New procedure: random:normal-vector! VECT [STATE]
5348 Fills VECT with inexact real random numbers which are independent and
5349 standard normally distributed (i.e., with mean 0 and variance 1).
5350 VECT can be either a vector or a uniform vector of doubles.
5352 *** New function: random:exp STATE
5353 Returns an inexact real in an exponential distribution with mean 1.
5354 For an exponential distribution with mean U use (* U (random:exp)).
5356 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
5358 These functions now operate on numbers in the range of a C unsigned
5361 These functions used to operate on numbers in the range of a C signed
5362 long; however, this seems inappropriate, because Guile integers don't
5365 ** New function: make-guardian
5366 This is an implementation of guardians as described in
5367 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
5368 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
5369 Programming Language Design and Implementation, June 1993
5370 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
5372 ** New functions: delq1!, delv1!, delete1!
5373 These procedures behave similar to delq! and friends but delete only
5374 one object if at all.
5376 ** New function: unread-string STRING PORT
5377 Unread STRING to PORT, that is, push it back onto the port so that
5378 next read operation will work on the pushed back characters.
5380 ** unread-char can now be called multiple times
5381 If unread-char is called multiple times, the unread characters will be
5382 read again in last-in first-out order.
5384 ** the procedures uniform-array-read! and uniform-array-write! now
5385 work on any kind of port, not just ports which are open on a file.
5387 ** Now 'l' in a port mode requests line buffering.
5389 ** The procedure truncate-file now works on string ports as well
5390 as file ports. If the size argument is omitted, the current
5391 file position is used.
5393 ** new procedure: seek PORT/FDES OFFSET WHENCE
5394 The arguments are the same as for the old fseek procedure, but it
5395 works on string ports as well as random-access file ports.
5397 ** the fseek procedure now works on string ports, since it has been
5398 redefined using seek.
5400 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
5401 size is not supplied.
5403 ** the newline procedure no longer flushes the port if it's not
5404 line-buffered: previously it did if it was the current output port.
5406 ** open-pipe and close-pipe are no longer primitive procedures, but
5407 an emulation can be obtained using `(use-modules (ice-9 popen))'.
5409 ** the freopen procedure has been removed.
5411 ** new procedure: drain-input PORT
5412 Drains PORT's read buffers (including any pushed-back characters)
5413 and returns the contents as a single string.
5415 ** New function: map-in-order PROC LIST1 LIST2 ...
5416 Version of `map' which guarantees that the procedure is applied to the
5417 lists in serial order.
5419 ** Renamed `serial-array-copy!' and `serial-array-map!' to
5420 `array-copy-in-order!' and `array-map-in-order!'. The old names are
5421 now obsolete and will go away in release 1.5.
5423 ** New syntax: collect BODY1 ...
5424 Version of `begin' which returns a list of the results of the body
5425 forms instead of the result of the last body form. In contrast to
5426 `begin', `collect' allows an empty body.
5428 ** New functions: read-history FILENAME, write-history FILENAME
5429 Read/write command line history from/to file. Returns #t on success
5430 and #f if an error occured.
5432 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
5434 These procedures return a list of definitions available in the specified
5435 argument, a relative module reference. In the case of no argument,
5436 `(current-module)' is now consulted for definitions to return, instead
5437 of simply returning #f, the former behavior.
5439 ** The #/ syntax for lists is no longer supported.
5441 Earlier versions of Scheme accepted this syntax, but printed a
5444 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
5446 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
5449 * Changes to the gh_ interface
5453 Now takes a second argument which is the result array. If this
5454 pointer is NULL, a new array is malloced (the old behaviour).
5456 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
5457 gh_scm2shorts, gh_scm2longs, gh_scm2floats
5461 * Changes to the scm_ interface
5463 ** Function: scm_make_named_hook (char* name, int n_args)
5465 Creates a hook in the same way as make-hook above but also
5466 binds a variable named NAME to it.
5468 This is the typical way of creating a hook from C code.
5470 Currently, the variable is created in the "current" module. This
5471 might change when we get the new module system.
5473 ** The smob interface
5475 The interface for creating smobs has changed. For documentation, see
5476 data-rep.info (made from guile-core/doc/data-rep.texi).
5478 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
5480 >>> This function will be removed in 1.3.4. <<<
5484 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
5485 This function adds a new smob type, named NAME, with instance size
5486 SIZE to the system. The return value is a tag that is used in
5487 creating instances of the type. If SIZE is 0, then no memory will
5488 be allocated when instances of the smob are created, and nothing
5489 will be freed by the default free function.
5491 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
5492 This function sets the smob marking procedure for the smob type
5493 specified by the tag TC. TC is the tag returned by
5494 `scm_make_smob_type'.
5496 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
5497 This function sets the smob freeing procedure for the smob type
5498 specified by the tag TC. TC is the tag returned by
5499 `scm_make_smob_type'.
5501 *** Function: void scm_set_smob_print (tc, print)
5503 - Function: void scm_set_smob_print (long tc,
5504 scm_sizet (*print) (SCM,
5508 This function sets the smob printing procedure for the smob type
5509 specified by the tag TC. TC is the tag returned by
5510 `scm_make_smob_type'.
5512 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
5513 This function sets the smob equality-testing predicate for the
5514 smob type specified by the tag TC. TC is the tag returned by
5515 `scm_make_smob_type'.
5517 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
5518 Make VALUE contain a smob instance of the type with type code TC and
5519 smob data DATA. VALUE must be previously declared as C type `SCM'.
5521 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
5522 This macro expands to a block of code that creates a smob instance
5523 of the type with type code TC and smob data DATA, and returns that
5524 `SCM' value. It should be the last piece of code in a block.
5526 ** The interfaces for using I/O ports and implementing port types
5527 (ptobs) have changed significantly. The new interface is based on
5528 shared access to buffers and a new set of ptob procedures.
5530 *** scm_newptob has been removed
5534 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
5536 - Function: SCM scm_make_port_type (char *type_name,
5537 int (*fill_buffer) (SCM port),
5538 void (*write_flush) (SCM port));
5540 Similarly to the new smob interface, there is a set of function
5541 setters by which the user can customize the behaviour of his port
5542 type. See ports.h (scm_set_port_XXX).
5544 ** scm_strport_to_string: New function: creates a new string from
5545 a string port's buffer.
5547 ** Plug in interface for random number generators
5548 The variable `scm_the_rng' in random.c contains a value and three
5549 function pointers which together define the current random number
5550 generator being used by the Scheme level interface and the random
5551 number library functions.
5553 The user is free to replace the default generator with the generator
5556 *** Variable: size_t scm_the_rng.rstate_size
5557 The size of the random state type used by the current RNG
5560 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
5561 Given the random STATE, return 32 random bits.
5563 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
5564 Seed random state STATE using string S of length N.
5566 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
5567 Given random state STATE, return a malloced copy.
5570 The default RNG is the MWC (Multiply With Carry) random number
5571 generator described by George Marsaglia at the Department of
5572 Statistics and Supercomputer Computations Research Institute, The
5573 Florida State University (http://stat.fsu.edu/~geo).
5575 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
5576 passes all tests in the DIEHARD test suite
5577 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
5578 costs one multiply and one add on platforms which either supports long
5579 longs (gcc does this on most systems) or have 64 bit longs. The cost
5580 is four multiply on other systems but this can be optimized by writing
5581 scm_i_uniform32 in assembler.
5583 These functions are provided through the scm_the_rng interface for use
5584 by libguile and the application.
5586 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
5587 Given the random STATE, return 32 random bits.
5588 Don't use this function directly. Instead go through the plugin
5589 interface (see "Plug in interface" above).
5591 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
5592 Initialize STATE using SEED of length N.
5594 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
5595 Return a malloc:ed copy of STATE. This function can easily be re-used
5596 in the interfaces to other RNGs.
5598 ** Random number library functions
5599 These functions use the current RNG through the scm_the_rng interface.
5600 It might be a good idea to use these functions from your C code so
5601 that only one random generator is used by all code in your program.
5603 The default random state is stored in:
5605 *** Variable: SCM scm_var_random_state
5606 Contains the vcell of the Scheme variable "*random-state*" which is
5607 used as default state by all random number functions in the Scheme
5612 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
5614 *** Function: scm_rstate *scm_c_default_rstate (void)
5615 This is a convenience function which returns the value of
5616 scm_var_random_state. An error message is generated if this value
5617 isn't a random state.
5619 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
5620 Make a new random state from the string SEED of length LENGTH.
5622 It is generally not a good idea to use multiple random states in a
5623 program. While subsequent random numbers generated from one random
5624 state are guaranteed to be reasonably independent, there is no such
5625 guarantee for numbers generated from different random states.
5627 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
5628 Return 32 random bits.
5630 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
5631 Return a sample from the uniform(0,1) distribution.
5633 *** Function: double scm_c_normal01 (scm_rstate *STATE)
5634 Return a sample from the normal(0,1) distribution.
5636 *** Function: double scm_c_exp1 (scm_rstate *STATE)
5637 Return a sample from the exp(1) distribution.
5639 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
5640 Return a sample from the discrete uniform(0,M) distribution.
5642 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
5643 Return a sample from the discrete uniform(0,M) distribution.
5644 M must be a bignum object. The returned value may be an INUM.
5648 Changes in Guile 1.3 (released Monday, October 19, 1998):
5650 * Changes to the distribution
5652 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
5653 To avoid conflicts, programs should name environment variables after
5654 themselves, except when there's a common practice establishing some
5657 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5658 giving the former precedence, and printing a warning message if the
5659 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5661 ** The header files related to multi-byte characters have been removed.
5662 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5663 which referred to these explicitly will probably need to be rewritten,
5664 since the support for the variant string types has been removed; see
5667 ** The header files append.h and sequences.h have been removed. These
5668 files implemented non-R4RS operations which would encourage
5669 non-portable programming style and less easy-to-read code.
5671 * Changes to the stand-alone interpreter
5673 ** New procedures have been added to implement a "batch mode":
5675 *** Function: batch-mode?
5677 Returns a boolean indicating whether the interpreter is in batch
5680 *** Function: set-batch-mode?! ARG
5682 If ARG is true, switches the interpreter to batch mode. The `#f'
5683 case has not been implemented.
5685 ** Guile now provides full command-line editing, when run interactively.
5686 To use this feature, you must have the readline library installed.
5687 The Guile build process will notice it, and automatically include
5690 The readline library is available via anonymous FTP from any GNU
5691 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5693 ** the-last-stack is now a fluid.
5695 * Changes to the procedure for linking libguile with your programs
5697 ** You can now use the `guile-config' utility to build programs that use Guile.
5699 Guile now includes a command-line utility called `guile-config', which
5700 can provide information about how to compile and link programs that
5703 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5704 You should include this command's output on the command line you use
5705 to compile C or C++ code that #includes the Guile header files. It's
5706 usually just a `-I' flag to help the compiler find the Guile headers.
5709 *** `guile-config link' prints any linker flags necessary to link with Guile.
5711 This command writes to its standard output a list of flags which you
5712 must pass to the linker to link your code against the Guile library.
5713 The flags include '-lguile' itself, any other libraries the Guile
5714 library depends upon, and any `-L' flags needed to help the linker
5715 find those libraries.
5717 For example, here is a Makefile rule that builds a program named 'foo'
5718 from the object files ${FOO_OBJECTS}, and links them against Guile:
5721 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5723 Previous Guile releases recommended that you use autoconf to detect
5724 which of a predefined set of libraries were present on your system.
5725 It is more robust to use `guile-config', since it records exactly which
5726 libraries the installed Guile library requires.
5728 This was originally called `build-guile', but was renamed to
5729 `guile-config' before Guile 1.3 was released, to be consistent with
5730 the analogous script for the GTK+ GUI toolkit, which is called
5734 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5736 If you are using the GNU autoconf package to configure your program,
5737 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5738 (described above) and gather the necessary values for use in your
5741 The GUILE_FLAGS macro expands to configure script code which runs the
5742 `guile-config' script, to find out where Guile's header files and
5743 libraries are installed. It sets two variables, marked for
5744 substitution, as by AC_SUBST.
5746 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5747 code that uses Guile header files. This is almost always just a
5750 GUILE_LDFLAGS --- flags to pass to the linker to link a
5751 program against Guile. This includes `-lguile' for the Guile
5752 library itself, any libraries that Guile itself requires (like
5753 -lqthreads), and so on. It may also include a -L flag to tell the
5754 compiler where to find the libraries.
5756 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5757 directory of the Guile distribution. You can copy it into your
5758 package's aclocal.m4 file, and then use it in your configure.in file.
5760 If you are using the `aclocal' program, distributed with GNU automake,
5761 to maintain your aclocal.m4 file, the Guile installation process
5762 installs guile.m4 where aclocal will find it. All you need to do is
5763 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5764 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5768 * Changes to Scheme functions and syntax
5770 ** Multi-byte strings have been removed, as have multi-byte and wide
5771 ports. We felt that these were the wrong approach to
5772 internationalization support.
5774 ** New function: readline [PROMPT]
5775 Read a line from the terminal, and allow the user to edit it,
5776 prompting with PROMPT. READLINE provides a large set of Emacs-like
5777 editing commands, lets the user recall previously typed lines, and
5778 works on almost every kind of terminal, including dumb terminals.
5780 READLINE assumes that the cursor is at the beginning of the line when
5781 it is invoked. Thus, you can't print a prompt yourself, and then call
5782 READLINE; you need to package up your prompt as a string, pass it to
5783 the function, and let READLINE print the prompt itself. This is
5784 because READLINE needs to know the prompt's screen width.
5786 For Guile to provide this function, you must have the readline
5787 library, version 2.1 or later, installed on your system. Readline is
5788 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5789 any GNU mirror site.
5791 See also ADD-HISTORY function.
5793 ** New function: add-history STRING
5794 Add STRING as the most recent line in the history used by the READLINE
5795 command. READLINE does not add lines to the history itself; you must
5796 call ADD-HISTORY to make previous input available to the user.
5798 ** The behavior of the read-line function has changed.
5800 This function now uses standard C library functions to read the line,
5801 for speed. This means that it doesn not respect the value of
5802 scm-line-incrementors; it assumes that lines are delimited with
5805 (Note that this is read-line, the function that reads a line of text
5806 from a port, not readline, the function that reads a line from a
5807 terminal, providing full editing capabilities.)
5809 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5811 This module provides some simple argument parsing. It exports one
5814 Function: getopt-gnu-style ARG-LS
5815 Parse a list of program arguments into an alist of option
5818 Each item in the list of program arguments is examined to see if
5819 it meets the syntax of a GNU long-named option. An argument like
5820 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5821 returned alist, where MUMBLE is a keyword object with the same
5822 name as the argument. An argument like `--MUMBLE=FROB' produces
5823 an element of the form (MUMBLE . FROB), where FROB is a string.
5825 As a special case, the returned alist also contains a pair whose
5826 car is the symbol `rest'. The cdr of this pair is a list
5827 containing all the items in the argument list that are not options
5828 of the form mentioned above.
5830 The argument `--' is treated specially: all items in the argument
5831 list appearing after such an argument are not examined, and are
5832 returned in the special `rest' list.
5834 This function does not parse normal single-character switches.
5835 You will need to parse them out of the `rest' list yourself.
5837 ** The read syntax for byte vectors and short vectors has changed.
5839 Instead of #bytes(...), write #y(...).
5841 Instead of #short(...), write #h(...).
5843 This may seem nutty, but, like the other uniform vectors, byte vectors
5844 and short vectors want to have the same print and read syntax (and,
5845 more basic, want to have read syntax!). Changing the read syntax to
5846 use multiple characters after the hash sign breaks with the
5847 conventions used in R5RS and the conventions used for the other
5848 uniform vectors. It also introduces complexity in the current reader,
5849 both on the C and Scheme levels. (The Right solution is probably to
5850 change the syntax and prototypes for uniform vectors entirely.)
5853 ** The new module (ice-9 session) provides useful interactive functions.
5855 *** New procedure: (apropos REGEXP OPTION ...)
5857 Display a list of top-level variables whose names match REGEXP, and
5858 the modules they are imported from. Each OPTION should be one of the
5861 value --- Show the value of each matching variable.
5862 shadow --- Show bindings shadowed by subsequently imported modules.
5863 full --- Same as both `shadow' and `value'.
5867 guile> (apropos "trace" 'full)
5868 debug: trace #<procedure trace args>
5869 debug: untrace #<procedure untrace args>
5870 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5871 the-scm-module: before-backtrace-hook ()
5872 the-scm-module: backtrace #<primitive-procedure backtrace>
5873 the-scm-module: after-backtrace-hook ()
5874 the-scm-module: has-shown-backtrace-hint? #f
5877 ** There are new functions and syntax for working with macros.
5879 Guile implements macros as a special object type. Any variable whose
5880 top-level binding is a macro object acts as a macro. The macro object
5881 specifies how the expression should be transformed before evaluation.
5883 *** Macro objects now print in a reasonable way, resembling procedures.
5885 *** New function: (macro? OBJ)
5886 True iff OBJ is a macro object.
5888 *** New function: (primitive-macro? OBJ)
5889 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5890 macro transformers, implemented in eval.c rather than Scheme code.
5892 Why do we have this function?
5893 - For symmetry with procedure? and primitive-procedure?,
5894 - to allow custom print procedures to tell whether a macro is
5895 primitive, and display it differently, and
5896 - to allow compilers and user-written evaluators to distinguish
5897 builtin special forms from user-defined ones, which could be
5900 *** New function: (macro-type OBJ)
5901 Return a value indicating what kind of macro OBJ is. Possible return
5904 The symbol `syntax' --- a macro created by procedure->syntax.
5905 The symbol `macro' --- a macro created by procedure->macro.
5906 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5907 The boolean #f --- if OBJ is not a macro object.
5909 *** New function: (macro-name MACRO)
5910 Return the name of the macro object MACRO's procedure, as returned by
5913 *** New function: (macro-transformer MACRO)
5914 Return the transformer procedure for MACRO.
5916 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5918 Specify a new macro expander to use in the current module. Each
5919 MODULE is a module name, with the same meaning as in the `use-modules'
5920 form; each named module's exported bindings are added to the current
5921 top-level environment. TRANSFORMER is an expression evaluated in the
5922 resulting environment which must yield a procedure to use as the
5923 module's eval transformer: every expression evaluated in this module
5924 is passed to this function, and the result passed to the Guile
5927 *** macro-eval! is removed. Use local-eval instead.
5929 ** Some magic has been added to the printer to better handle user
5930 written printing routines (like record printers, closure printers).
5932 The problem is that these user written routines must have access to
5933 the current `print-state' to be able to handle fancy things like
5934 detection of circular references. These print-states have to be
5935 passed to the builtin printing routines (display, write, etc) to
5936 properly continue the print chain.
5938 We didn't want to change all existing print code so that it
5939 explicitly passes thru a print state in addition to a port. Instead,
5940 we extented the possible values that the builtin printing routines
5941 accept as a `port'. In addition to a normal port, they now also take
5942 a pair of a normal port and a print-state. Printing will go to the
5943 port and the print-state will be used to control the detection of
5944 circular references, etc. If the builtin function does not care for a
5945 print-state, it is simply ignored.
5947 User written callbacks are now called with such a pair as their
5948 `port', but because every function now accepts this pair as a PORT
5949 argument, you don't have to worry about that. In fact, it is probably
5950 safest to not check for these pairs.
5952 However, it is sometimes necessary to continue a print chain on a
5953 different port, for example to get a intermediate string
5954 representation of the printed value, mangle that string somehow, and
5955 then to finally print the mangled string. Use the new function
5957 inherit-print-state OLD-PORT NEW-PORT
5959 for this. It constructs a new `port' that prints to NEW-PORT but
5960 inherits the print-state of OLD-PORT.
5962 ** struct-vtable-offset renamed to vtable-offset-user
5964 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5966 ** There is now a third optional argument to make-vtable-vtable
5967 (and fourth to make-struct) when constructing new types (vtables).
5968 This argument initializes field vtable-index-printer of the vtable.
5970 ** The detection of circular references has been extended to structs.
5971 That is, a structure that -- in the process of being printed -- prints
5972 itself does not lead to infinite recursion.
5974 ** There is now some basic support for fluids. Please read
5975 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5976 the following functions and macros:
5978 Function: make-fluid
5980 Create a new fluid object. Fluids are not special variables or
5981 some other extension to the semantics of Scheme, but rather
5982 ordinary Scheme objects. You can store them into variables (that
5983 are still lexically scoped, of course) or into any other place you
5984 like. Every fluid has a initial value of `#f'.
5986 Function: fluid? OBJ
5988 Test whether OBJ is a fluid.
5990 Function: fluid-ref FLUID
5991 Function: fluid-set! FLUID VAL
5993 Access/modify the fluid FLUID. Modifications are only visible
5994 within the current dynamic root (that includes threads).
5996 Function: with-fluids* FLUIDS VALUES THUNK
5998 FLUIDS is a list of fluids and VALUES a corresponding list of
5999 values for these fluids. Before THUNK gets called the values are
6000 installed in the fluids and the old values of the fluids are
6001 saved in the VALUES list. When the flow of control leaves THUNK
6002 or reenters it, the values get swapped again. You might think of
6003 this as a `safe-fluid-excursion'. Note that the VALUES list is
6004 modified by `with-fluids*'.
6006 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
6008 The same as `with-fluids*' but with a different syntax. It looks
6009 just like `let', but both FLUID and VALUE are evaluated. Remember,
6010 fluids are not special variables but ordinary objects. FLUID
6011 should evaluate to a fluid.
6013 ** Changes to system call interfaces:
6015 *** close-port, close-input-port and close-output-port now return a
6016 boolean instead of an `unspecified' object. #t means that the port
6017 was successfully closed, while #f means it was already closed. It is
6018 also now possible for these procedures to raise an exception if an
6019 error occurs (some errors from write can be delayed until close.)
6021 *** the first argument to chmod, fcntl, ftell and fseek can now be a
6024 *** the third argument to fcntl is now optional.
6026 *** the first argument to chown can now be a file descriptor or a port.
6028 *** the argument to stat can now be a port.
6030 *** The following new procedures have been added (most use scsh
6033 *** procedure: close PORT/FD
6034 Similar to close-port (*note close-port: Closing Ports.), but also
6035 works on file descriptors. A side effect of closing a file
6036 descriptor is that any ports using that file descriptor are moved
6037 to a different file descriptor and have their revealed counts set
6040 *** procedure: port->fdes PORT
6041 Returns the integer file descriptor underlying PORT. As a side
6042 effect the revealed count of PORT is incremented.
6044 *** procedure: fdes->ports FDES
6045 Returns a list of existing ports which have FDES as an underlying
6046 file descriptor, without changing their revealed counts.
6048 *** procedure: fdes->inport FDES
6049 Returns an existing input port which has FDES as its underlying
6050 file descriptor, if one exists, and increments its revealed count.
6051 Otherwise, returns a new input port with a revealed count of 1.
6053 *** procedure: fdes->outport FDES
6054 Returns an existing output port which has FDES as its underlying
6055 file descriptor, if one exists, and increments its revealed count.
6056 Otherwise, returns a new output port with a revealed count of 1.
6058 The next group of procedures perform a `dup2' system call, if NEWFD
6059 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
6060 duplicated can be supplied as an integer or contained in a port. The
6061 type of value returned varies depending on which procedure is used.
6063 All procedures also have the side effect when performing `dup2' that
6064 any ports using NEWFD are moved to a different file descriptor and have
6065 their revealed counts set to zero.
6067 *** procedure: dup->fdes PORT/FD [NEWFD]
6068 Returns an integer file descriptor.
6070 *** procedure: dup->inport PORT/FD [NEWFD]
6071 Returns a new input port using the new file descriptor.
6073 *** procedure: dup->outport PORT/FD [NEWFD]
6074 Returns a new output port using the new file descriptor.
6076 *** procedure: dup PORT/FD [NEWFD]
6077 Returns a new port if PORT/FD is a port, with the same mode as the
6078 supplied port, otherwise returns an integer file descriptor.
6080 *** procedure: dup->port PORT/FD MODE [NEWFD]
6081 Returns a new port using the new file descriptor. MODE supplies a
6082 mode string for the port (*note open-file: File Ports.).
6084 *** procedure: setenv NAME VALUE
6085 Modifies the environment of the current process, which is also the
6086 default environment inherited by child processes.
6088 If VALUE is `#f', then NAME is removed from the environment.
6089 Otherwise, the string NAME=VALUE is added to the environment,
6090 replacing any existing string with name matching NAME.
6092 The return value is unspecified.
6094 *** procedure: truncate-file OBJ SIZE
6095 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
6096 can be a string containing a file name or an integer file
6097 descriptor or port open for output on the file. The underlying
6098 system calls are `truncate' and `ftruncate'.
6100 The return value is unspecified.
6102 *** procedure: setvbuf PORT MODE [SIZE]
6103 Set the buffering mode for PORT. MODE can be:
6111 block buffered, using a newly allocated buffer of SIZE bytes.
6112 However if SIZE is zero or unspecified, the port will be made
6115 This procedure should not be used after I/O has been performed with
6118 Ports are usually block buffered by default, with a default buffer
6119 size. Procedures e.g., *Note open-file: File Ports, which accept a
6120 mode string allow `0' to be added to request an unbuffered port.
6122 *** procedure: fsync PORT/FD
6123 Copies any unwritten data for the specified output file descriptor
6124 to disk. If PORT/FD is a port, its buffer is flushed before the
6125 underlying file descriptor is fsync'd. The return value is
6128 *** procedure: open-fdes PATH FLAGS [MODES]
6129 Similar to `open' but returns a file descriptor instead of a port.
6131 *** procedure: execle PATH ENV [ARG] ...
6132 Similar to `execl', but the environment of the new process is
6133 specified by ENV, which must be a list of strings as returned by
6134 the `environ' procedure.
6136 This procedure is currently implemented using the `execve' system
6137 call, but we call it `execle' because of its Scheme calling
6140 *** procedure: strerror ERRNO
6141 Returns the Unix error message corresponding to ERRNO, an integer.
6143 *** procedure: primitive-exit [STATUS]
6144 Terminate the current process without unwinding the Scheme stack.
6145 This is would typically be useful after a fork. The exit status
6146 is STATUS if supplied, otherwise zero.
6148 *** procedure: times
6149 Returns an object with information about real and processor time.
6150 The following procedures accept such an object as an argument and
6151 return a selected component:
6154 The current real time, expressed as time units relative to an
6158 The CPU time units used by the calling process.
6161 The CPU time units used by the system on behalf of the
6165 The CPU time units used by terminated child processes of the
6166 calling process, whose status has been collected (e.g., using
6170 Similarly, the CPU times units used by the system on behalf of
6171 terminated child processes.
6173 ** Removed: list-length
6174 ** Removed: list-append, list-append!
6175 ** Removed: list-reverse, list-reverse!
6177 ** array-map renamed to array-map!
6179 ** serial-array-map renamed to serial-array-map!
6181 ** catch doesn't take #f as first argument any longer
6183 Previously, it was possible to pass #f instead of a key to `catch'.
6184 That would cause `catch' to pass a jump buffer object to the procedure
6185 passed as second argument. The procedure could then use this jump
6186 buffer objekt as an argument to throw.
6188 This mechanism has been removed since its utility doesn't motivate the
6189 extra complexity it introduces.
6191 ** The `#/' notation for lists now provokes a warning message from Guile.
6192 This syntax will be removed from Guile in the near future.
6194 To disable the warning message, set the GUILE_HUSH environment
6195 variable to any non-empty value.
6197 ** The newline character now prints as `#\newline', following the
6198 normal Scheme notation, not `#\nl'.
6200 * Changes to the gh_ interface
6202 ** The gh_enter function now takes care of loading the Guile startup files.
6203 gh_enter works by calling scm_boot_guile; see the remarks below.
6205 ** Function: void gh_write (SCM x)
6207 Write the printed representation of the scheme object x to the current
6208 output port. Corresponds to the scheme level `write'.
6210 ** gh_list_length renamed to gh_length.
6212 ** vector handling routines
6214 Several major changes. In particular, gh_vector() now resembles
6215 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
6216 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
6217 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
6218 vector-related gh_ functions have been implemented.
6220 ** pair and list routines
6222 Implemented several of the R4RS pair and list functions that were
6225 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
6227 New function. Converts double arrays back and forth between Scheme
6230 * Changes to the scm_ interface
6232 ** The function scm_boot_guile now takes care of loading the startup files.
6234 Guile's primary initialization function, scm_boot_guile, now takes
6235 care of loading `boot-9.scm', in the `ice-9' module, to initialize
6236 Guile, define the module system, and put together some standard
6237 bindings. It also loads `init.scm', which is intended to hold
6238 site-specific initialization code.
6240 Since Guile cannot operate properly until boot-9.scm is loaded, there
6241 is no reason to separate loading boot-9.scm from Guile's other
6242 initialization processes.
6244 This job used to be done by scm_compile_shell_switches, which didn't
6245 make much sense; in particular, it meant that people using Guile for
6246 non-shell-like applications had to jump through hoops to get Guile
6247 initialized properly.
6249 ** The function scm_compile_shell_switches no longer loads the startup files.
6250 Now, Guile always loads the startup files, whenever it is initialized;
6251 see the notes above for scm_boot_guile and scm_load_startup_files.
6253 ** Function: scm_load_startup_files
6254 This new function takes care of loading Guile's initialization file
6255 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
6256 this is always called by the Guile initialization process, it's
6257 probably not too useful to call this yourself, but it's there anyway.
6259 ** The semantics of smob marking have changed slightly.
6261 The smob marking function (the `mark' member of the scm_smobfuns
6262 structure) is no longer responsible for setting the mark bit on the
6263 smob. The generic smob handling code in the garbage collector will
6264 set this bit. The mark function need only ensure that any other
6265 objects the smob refers to get marked.
6267 Note that this change means that the smob's GC8MARK bit is typically
6268 already set upon entry to the mark function. Thus, marking functions
6269 which look like this:
6272 if (SCM_GC8MARKP (ptr))
6274 SCM_SETGC8MARK (ptr);
6275 ... mark objects to which the smob refers ...
6278 are now incorrect, since they will return early, and fail to mark any
6279 other objects the smob refers to. Some code in the Guile library used
6282 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
6284 If you have implemented your own I/O port type, by writing the
6285 functions required by the scm_ptobfuns and then calling scm_newptob,
6286 you will need to change your functions slightly.
6288 The functions in a scm_ptobfuns structure now expect the port itself
6289 as their argument; they used to expect the `stream' member of the
6290 port's scm_port_table structure. This allows functions in an
6291 scm_ptobfuns structure to easily access the port's cell (and any flags
6292 it its CAR), and the port's scm_port_table structure.
6294 Guile now passes the I/O port itself as the `port' argument in the
6295 following scm_ptobfuns functions:
6297 int (*free) (SCM port);
6298 int (*fputc) (int, SCM port);
6299 int (*fputs) (char *, SCM port);
6300 scm_sizet (*fwrite) SCM_P ((char *ptr,
6304 int (*fflush) (SCM port);
6305 int (*fgetc) (SCM port);
6306 int (*fclose) (SCM port);
6308 The interfaces to the `mark', `print', `equalp', and `fgets' methods
6311 If you have existing code which defines its own port types, it is easy
6312 to convert your code to the new interface; simply apply SCM_STREAM to
6313 the port argument to yield the value you code used to expect.
6315 Note that since both the port and the stream have the same type in the
6316 C code --- they are both SCM values --- the C compiler will not remind
6317 you if you forget to update your scm_ptobfuns functions.
6320 ** Function: int scm_internal_select (int fds,
6324 struct timeval *timeout);
6326 This is a replacement for the `select' function provided by the OS.
6327 It enables I/O blocking and sleeping to happen for one cooperative
6328 thread without blocking other threads. It also avoids busy-loops in
6329 these situations. It is intended that all I/O blocking and sleeping
6330 will finally go through this function. Currently, this function is
6331 only available on systems providing `gettimeofday' and `select'.
6333 ** Function: SCM scm_internal_stack_catch (SCM tag,
6334 scm_catch_body_t body,
6336 scm_catch_handler_t handler,
6339 A new sibling to the other two C level `catch' functions
6340 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
6341 the stack to be saved automatically into the variable `the-last-stack'
6342 (scm_the_last_stack_var) on error. This is necessary if you want to
6343 use advanced error reporting, such as calling scm_display_error and
6344 scm_display_backtrace. (They both take a stack object as argument.)
6346 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
6348 scm_catch_handler_t handler,
6351 Spawns a new thread. It does a job similar to
6352 scm_call_with_new_thread but takes arguments more suitable when
6353 spawning threads from application C code.
6355 ** The hook scm_error_callback has been removed. It was originally
6356 intended as a way for the user to install his own error handler. But
6357 that method works badly since it intervenes between throw and catch,
6358 thereby changing the semantics of expressions like (catch #t ...).
6359 The correct way to do it is to use one of the C level catch functions
6360 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
6362 ** Removed functions:
6364 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
6365 scm_list_reverse, scm_list_reverse_x
6367 ** New macros: SCM_LISTn where n is one of the integers 0-9.
6369 These can be used for pretty list creation from C. The idea is taken
6370 from Erick Gallesio's STk.
6372 ** scm_array_map renamed to scm_array_map_x
6374 ** mbstrings are now removed
6376 This means that the type codes scm_tc7_mb_string and
6377 scm_tc7_mb_substring has been removed.
6379 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
6381 Since we no longer support multi-byte strings, these I/O functions
6382 have been simplified, and renamed. Here are their old names, and
6383 their new names and arguments:
6385 scm_gen_putc -> void scm_putc (int c, SCM port);
6386 scm_gen_puts -> void scm_puts (char *s, SCM port);
6387 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
6388 scm_gen_getc -> void scm_getc (SCM port);
6391 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
6393 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
6395 SCM_TYP7S now masks away the bit which distinguishes substrings from
6398 ** scm_catch_body_t: Backward incompatible change!
6400 Body functions to scm_internal_catch and friends do not any longer
6401 take a second argument. This is because it is no longer possible to
6402 pass a #f arg to catch.
6404 ** Calls to scm_protect_object and scm_unprotect now nest properly.
6406 The function scm_protect_object protects its argument from being freed
6407 by the garbage collector. scm_unprotect_object removes that
6410 These functions now nest properly. That is, for every object O, there
6411 is a counter which scm_protect_object(O) increments and
6412 scm_unprotect_object(O) decrements, if the counter is greater than
6413 zero. Every object's counter is zero when it is first created. If an
6414 object's counter is greater than zero, the garbage collector will not
6415 reclaim its storage.
6417 This allows you to use scm_protect_object in your code without
6418 worrying that some other function you call will call
6419 scm_unprotect_object, and allow it to be freed. Assuming that the
6420 functions you call are well-behaved, and unprotect only those objects
6421 they protect, you can follow the same rule and have confidence that
6422 objects will be freed only at appropriate times.
6425 Changes in Guile 1.2 (released Tuesday, June 24 1997):
6427 * Changes to the distribution
6429 ** Nightly snapshots are now available from ftp.red-bean.com.
6430 The old server, ftp.cyclic.com, has been relinquished to its rightful
6433 Nightly snapshots of the Guile development sources are now available via
6434 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
6436 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
6437 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
6439 ** To run Guile without installing it, the procedure has changed a bit.
6441 If you used a separate build directory to compile Guile, you'll need
6442 to include the build directory in SCHEME_LOAD_PATH, as well as the
6443 source directory. See the `INSTALL' file for examples.
6445 * Changes to the procedure for linking libguile with your programs
6447 ** The standard Guile load path for Scheme code now includes
6448 $(datadir)/guile (usually /usr/local/share/guile). This means that
6449 you can install your own Scheme files there, and Guile will find them.
6450 (Previous versions of Guile only checked a directory whose name
6451 contained the Guile version number, so you had to re-install or move
6452 your Scheme sources each time you installed a fresh version of Guile.)
6454 The load path also includes $(datadir)/guile/site; we recommend
6455 putting individual Scheme files there. If you want to install a
6456 package with multiple source files, create a directory for them under
6459 ** Guile 1.2 will now use the Rx regular expression library, if it is
6460 installed on your system. When you are linking libguile into your own
6461 programs, this means you will have to link against -lguile, -lqt (if
6462 you configured Guile with thread support), and -lrx.
6464 If you are using autoconf to generate configuration scripts for your
6465 application, the following lines should suffice to add the appropriate
6466 libraries to your link command:
6468 ### Find Rx, quickthreads and libguile.
6469 AC_CHECK_LIB(rx, main)
6470 AC_CHECK_LIB(qt, main)
6471 AC_CHECK_LIB(guile, scm_shell)
6473 The Guile 1.2 distribution does not contain sources for the Rx
6474 library, as Guile 1.0 did. If you want to use Rx, you'll need to
6475 retrieve it from a GNU FTP site and install it separately.
6477 * Changes to Scheme functions and syntax
6479 ** The dynamic linking features of Guile are now enabled by default.
6480 You can disable them by giving the `--disable-dynamic-linking' option
6483 (dynamic-link FILENAME)
6485 Find the object file denoted by FILENAME (a string) and link it
6486 into the running Guile application. When everything works out,
6487 return a Scheme object suitable for representing the linked object
6488 file. Otherwise an error is thrown. How object files are
6489 searched is system dependent.
6491 (dynamic-object? VAL)
6493 Determine whether VAL represents a dynamically linked object file.
6495 (dynamic-unlink DYNOBJ)
6497 Unlink the indicated object file from the application. DYNOBJ
6498 should be one of the values returned by `dynamic-link'.
6500 (dynamic-func FUNCTION DYNOBJ)
6502 Search the C function indicated by FUNCTION (a string or symbol)
6503 in DYNOBJ and return some Scheme object that can later be used
6504 with `dynamic-call' to actually call this function. Right now,
6505 these Scheme objects are formed by casting the address of the
6506 function to `long' and converting this number to its Scheme
6509 (dynamic-call FUNCTION DYNOBJ)
6511 Call the C function indicated by FUNCTION and DYNOBJ. The
6512 function is passed no arguments and its return value is ignored.
6513 When FUNCTION is something returned by `dynamic-func', call that
6514 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
6515 etc.), look it up in DYNOBJ; this is equivalent to
6517 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
6519 Interrupts are deferred while the C function is executing (with
6520 SCM_DEFER_INTS/SCM_ALLOW_INTS).
6522 (dynamic-args-call FUNCTION DYNOBJ ARGS)
6524 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
6525 some arguments and return its return value. The C function is
6526 expected to take two arguments and return an `int', just like
6529 int c_func (int argc, char **argv);
6531 ARGS must be a list of strings and is converted into an array of
6532 `char *'. The array is passed in ARGV and its size in ARGC. The
6533 return value is converted to a Scheme number and returned from the
6534 call to `dynamic-args-call'.
6536 When dynamic linking is disabled or not supported on your system,
6537 the above functions throw errors, but they are still available.
6539 Here is a small example that works on GNU/Linux:
6541 (define libc-obj (dynamic-link "libc.so"))
6542 (dynamic-args-call 'rand libc-obj '())
6544 See the file `libguile/DYNAMIC-LINKING' for additional comments.
6546 ** The #/ syntax for module names is depreciated, and will be removed
6547 in a future version of Guile. Instead of
6555 The latter syntax is more consistent with existing Lisp practice.
6557 ** Guile now does fancier printing of structures. Structures are the
6558 underlying implementation for records, which in turn are used to
6559 implement modules, so all of these object now print differently and in
6560 a more informative way.
6562 The Scheme printer will examine the builtin variable *struct-printer*
6563 whenever it needs to print a structure object. When this variable is
6564 not `#f' it is deemed to be a procedure and will be applied to the
6565 structure object and the output port. When *struct-printer* is `#f'
6566 or the procedure return `#f' the structure object will be printed in
6567 the boring #<struct 80458270> form.
6569 This hook is used by some routines in ice-9/boot-9.scm to implement
6570 type specific printing routines. Please read the comments there about
6573 One of the more specific uses of structs are records. The printing
6574 procedure that could be passed to MAKE-RECORD-TYPE is now actually
6575 called. It should behave like a *struct-printer* procedure (described
6578 ** Guile now supports a new R4RS-compliant syntax for keywords. A
6579 token of the form #:NAME, where NAME has the same syntax as a Scheme
6580 symbol, is the external representation of the keyword named NAME.
6581 Keyword objects print using this syntax as well, so values containing
6582 keyword objects can be read back into Guile. When used in an
6583 expression, keywords are self-quoting objects.
6585 Guile suports this read syntax, and uses this print syntax, regardless
6586 of the current setting of the `keyword' read option. The `keyword'
6587 read option only controls whether Guile recognizes the `:NAME' syntax,
6588 which is incompatible with R4RS. (R4RS says such token represent
6591 ** Guile has regular expression support again. Guile 1.0 included
6592 functions for matching regular expressions, based on the Rx library.
6593 In Guile 1.1, the Guile/Rx interface was removed to simplify the
6594 distribution, and thus Guile had no regular expression support. Guile
6595 1.2 again supports the most commonly used functions, and supports all
6596 of SCSH's regular expression functions.
6598 If your system does not include a POSIX regular expression library,
6599 and you have not linked Guile with a third-party regexp library such as
6600 Rx, these functions will not be available. You can tell whether your
6601 Guile installation includes regular expression support by checking
6602 whether the `*features*' list includes the `regex' symbol.
6604 *** regexp functions
6606 By default, Guile supports POSIX extended regular expressions. That
6607 means that the characters `(', `)', `+' and `?' are special, and must
6608 be escaped if you wish to match the literal characters.
6610 This regular expression interface was modeled after that implemented
6611 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
6612 with SCSH regular expressions.
6614 **** Function: string-match PATTERN STR [START]
6615 Compile the string PATTERN into a regular expression and compare
6616 it with STR. The optional numeric argument START specifies the
6617 position of STR at which to begin matching.
6619 `string-match' returns a "match structure" which describes what,
6620 if anything, was matched by the regular expression. *Note Match
6621 Structures::. If STR does not match PATTERN at all,
6622 `string-match' returns `#f'.
6624 Each time `string-match' is called, it must compile its PATTERN
6625 argument into a regular expression structure. This operation is
6626 expensive, which makes `string-match' inefficient if the same regular
6627 expression is used several times (for example, in a loop). For better
6628 performance, you can compile a regular expression in advance and then
6629 match strings against the compiled regexp.
6631 **** Function: make-regexp STR [FLAGS]
6632 Compile the regular expression described by STR, and return the
6633 compiled regexp structure. If STR does not describe a legal
6634 regular expression, `make-regexp' throws a
6635 `regular-expression-syntax' error.
6637 FLAGS may be the bitwise-or of one or more of the following:
6639 **** Constant: regexp/extended
6640 Use POSIX Extended Regular Expression syntax when interpreting
6641 STR. If not set, POSIX Basic Regular Expression syntax is used.
6642 If the FLAGS argument is omitted, we assume regexp/extended.
6644 **** Constant: regexp/icase
6645 Do not differentiate case. Subsequent searches using the
6646 returned regular expression will be case insensitive.
6648 **** Constant: regexp/newline
6649 Match-any-character operators don't match a newline.
6651 A non-matching list ([^...]) not containing a newline matches a
6654 Match-beginning-of-line operator (^) matches the empty string
6655 immediately after a newline, regardless of whether the FLAGS
6656 passed to regexp-exec contain regexp/notbol.
6658 Match-end-of-line operator ($) matches the empty string
6659 immediately before a newline, regardless of whether the FLAGS
6660 passed to regexp-exec contain regexp/noteol.
6662 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6663 Match the compiled regular expression REGEXP against `str'. If
6664 the optional integer START argument is provided, begin matching
6665 from that position in the string. Return a match structure
6666 describing the results of the match, or `#f' if no match could be
6669 FLAGS may be the bitwise-or of one or more of the following:
6671 **** Constant: regexp/notbol
6672 The match-beginning-of-line operator always fails to match (but
6673 see the compilation flag regexp/newline above) This flag may be
6674 used when different portions of a string are passed to
6675 regexp-exec and the beginning of the string should not be
6676 interpreted as the beginning of the line.
6678 **** Constant: regexp/noteol
6679 The match-end-of-line operator always fails to match (but see the
6680 compilation flag regexp/newline above)
6682 **** Function: regexp? OBJ
6683 Return `#t' if OBJ is a compiled regular expression, or `#f'
6686 Regular expressions are commonly used to find patterns in one string
6687 and replace them with the contents of another string.
6689 **** Function: regexp-substitute PORT MATCH [ITEM...]
6690 Write to the output port PORT selected contents of the match
6691 structure MATCH. Each ITEM specifies what should be written, and
6692 may be one of the following arguments:
6694 * A string. String arguments are written out verbatim.
6696 * An integer. The submatch with that number is written.
6698 * The symbol `pre'. The portion of the matched string preceding
6699 the regexp match is written.
6701 * The symbol `post'. The portion of the matched string
6702 following the regexp match is written.
6704 PORT may be `#f', in which case nothing is written; instead,
6705 `regexp-substitute' constructs a string from the specified ITEMs
6708 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6709 Similar to `regexp-substitute', but can be used to perform global
6710 substitutions on STR. Instead of taking a match structure as an
6711 argument, `regexp-substitute/global' takes two string arguments: a
6712 REGEXP string describing a regular expression, and a TARGET string
6713 which should be matched against this regular expression.
6715 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6718 * A function may be supplied. When this function is called, it
6719 will be passed one argument: a match structure for a given
6720 regular expression match. It should return a string to be
6721 written out to PORT.
6723 * The `post' symbol causes `regexp-substitute/global' to recurse
6724 on the unmatched portion of STR. This *must* be supplied in
6725 order to perform global search-and-replace on STR; if it is
6726 not present among the ITEMs, then `regexp-substitute/global'
6727 will return after processing a single match.
6729 *** Match Structures
6731 A "match structure" is the object returned by `string-match' and
6732 `regexp-exec'. It describes which portion of a string, if any, matched
6733 the given regular expression. Match structures include: a reference to
6734 the string that was checked for matches; the starting and ending
6735 positions of the regexp match; and, if the regexp included any
6736 parenthesized subexpressions, the starting and ending positions of each
6739 In each of the regexp match functions described below, the `match'
6740 argument must be a match structure returned by a previous call to
6741 `string-match' or `regexp-exec'. Most of these functions return some
6742 information about the original target string that was matched against a
6743 regular expression; we will call that string TARGET for easy reference.
6745 **** Function: regexp-match? OBJ
6746 Return `#t' if OBJ is a match structure returned by a previous
6747 call to `regexp-exec', or `#f' otherwise.
6749 **** Function: match:substring MATCH [N]
6750 Return the portion of TARGET matched by subexpression number N.
6751 Submatch 0 (the default) represents the entire regexp match. If
6752 the regular expression as a whole matched, but the subexpression
6753 number N did not match, return `#f'.
6755 **** Function: match:start MATCH [N]
6756 Return the starting position of submatch number N.
6758 **** Function: match:end MATCH [N]
6759 Return the ending position of submatch number N.
6761 **** Function: match:prefix MATCH
6762 Return the unmatched portion of TARGET preceding the regexp match.
6764 **** Function: match:suffix MATCH
6765 Return the unmatched portion of TARGET following the regexp match.
6767 **** Function: match:count MATCH
6768 Return the number of parenthesized subexpressions from MATCH.
6769 Note that the entire regular expression match itself counts as a
6770 subexpression, and failed submatches are included in the count.
6772 **** Function: match:string MATCH
6773 Return the original TARGET string.
6775 *** Backslash Escapes
6777 Sometimes you will want a regexp to match characters like `*' or `$'
6778 exactly. For example, to check whether a particular string represents
6779 a menu entry from an Info node, it would be useful to match it against
6780 a regexp like `^* [^:]*::'. However, this won't work; because the
6781 asterisk is a metacharacter, it won't match the `*' at the beginning of
6782 the string. In this case, we want to make the first asterisk un-magic.
6784 You can do this by preceding the metacharacter with a backslash
6785 character `\'. (This is also called "quoting" the metacharacter, and
6786 is known as a "backslash escape".) When Guile sees a backslash in a
6787 regular expression, it considers the following glyph to be an ordinary
6788 character, no matter what special meaning it would ordinarily have.
6789 Therefore, we can make the above example work by changing the regexp to
6790 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6791 to match only a single asterisk in the target string.
6793 Since the backslash is itself a metacharacter, you may force a
6794 regexp to match a backslash in the target string by preceding the
6795 backslash with itself. For example, to find variable references in a
6796 TeX program, you might want to find occurrences of the string `\let\'
6797 followed by any number of alphabetic characters. The regular expression
6798 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6799 each match a single backslash in the target string.
6801 **** Function: regexp-quote STR
6802 Quote each special character found in STR with a backslash, and
6803 return the resulting string.
6805 *Very important:* Using backslash escapes in Guile source code (as
6806 in Emacs Lisp or C) can be tricky, because the backslash character has
6807 special meaning for the Guile reader. For example, if Guile encounters
6808 the character sequence `\n' in the middle of a string while processing
6809 Scheme code, it replaces those characters with a newline character.
6810 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6811 Several of these "escape sequences" are processed by the Guile reader
6812 before your code is executed. Unrecognized escape sequences are
6813 ignored: if the characters `\*' appear in a string, they will be
6814 translated to the single character `*'.
6816 This translation is obviously undesirable for regular expressions,
6817 since we want to be able to include backslashes in a string in order to
6818 escape regexp metacharacters. Therefore, to make sure that a backslash
6819 is preserved in a string in your Guile program, you must use *two*
6820 consecutive backslashes:
6822 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6824 The string in this example is preprocessed by the Guile reader before
6825 any code is executed. The resulting argument to `make-regexp' is the
6826 string `^\* [^:]*', which is what we really want.
6828 This also means that in order to write a regular expression that
6829 matches a single backslash character, the regular expression string in
6830 the source code must include *four* backslashes. Each consecutive pair
6831 of backslashes gets translated by the Guile reader to a single
6832 backslash, and the resulting double-backslash is interpreted by the
6833 regexp engine as matching a single backslash character. Hence:
6835 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6837 The reason for the unwieldiness of this syntax is historical. Both
6838 regular expression pattern matchers and Unix string processing systems
6839 have traditionally used backslashes with the special meanings described
6840 above. The POSIX regular expression specification and ANSI C standard
6841 both require these semantics. Attempting to abandon either convention
6842 would cause other kinds of compatibility problems, possibly more severe
6843 ones. Therefore, without extending the Scheme reader to support
6844 strings with different quoting conventions (an ungainly and confusing
6845 extension when implemented in other languages), we must adhere to this
6846 cumbersome escape syntax.
6848 * Changes to the gh_ interface
6850 * Changes to the scm_ interface
6852 * Changes to system call interfaces:
6854 ** The value returned by `raise' is now unspecified. It throws an exception
6857 *** A new procedure `sigaction' can be used to install signal handlers
6859 (sigaction signum [action] [flags])
6861 signum is the signal number, which can be specified using the value
6864 If action is omitted, sigaction returns a pair: the CAR is the current
6865 signal hander, which will be either an integer with the value SIG_DFL
6866 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6867 handles the signal, or #f if a non-Scheme procedure handles the
6868 signal. The CDR contains the current sigaction flags for the handler.
6870 If action is provided, it is installed as the new handler for signum.
6871 action can be a Scheme procedure taking one argument, or the value of
6872 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6873 whatever signal handler was installed before sigaction was first used.
6874 Flags can optionally be specified for the new handler (SA_RESTART is
6875 always used if the system provides it, so need not be specified.) The
6876 return value is a pair with information about the old handler as
6879 This interface does not provide access to the "signal blocking"
6880 facility. Maybe this is not needed, since the thread support may
6881 provide solutions to the problem of consistent access to data
6884 *** A new procedure `flush-all-ports' is equivalent to running
6885 `force-output' on every port open for output.
6887 ** Guile now provides information on how it was built, via the new
6888 global variable, %guile-build-info. This variable records the values
6889 of the standard GNU makefile directory variables as an assocation
6890 list, mapping variable names (symbols) onto directory paths (strings).
6891 For example, to find out where the Guile link libraries were
6892 installed, you can say:
6894 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6897 * Changes to the scm_ interface
6899 ** The new function scm_handle_by_message_noexit is just like the
6900 existing scm_handle_by_message function, except that it doesn't call
6901 exit to terminate the process. Instead, it prints a message and just
6902 returns #f. This might be a more appropriate catch-all handler for
6903 new dynamic roots and threads.
6906 Changes in Guile 1.1 (released Friday, May 16 1997):
6908 * Changes to the distribution.
6910 The Guile 1.0 distribution has been split up into several smaller
6912 guile-core --- the Guile interpreter itself.
6913 guile-tcltk --- the interface between the Guile interpreter and
6914 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6915 is a toolkit for building graphical user interfaces.
6916 guile-rgx-ctax --- the interface between Guile and the Rx regular
6917 expression matcher, and the translator for the Ctax
6918 programming language. These are packaged together because the
6919 Ctax translator uses Rx to parse Ctax source code.
6921 This NEWS file describes the changes made to guile-core since the 1.0
6924 We no longer distribute the documentation, since it was either out of
6925 date, or incomplete. As soon as we have current documentation, we
6930 * Changes to the stand-alone interpreter
6932 ** guile now accepts command-line arguments compatible with SCSH, Olin
6933 Shivers' Scheme Shell.
6935 In general, arguments are evaluated from left to right, but there are
6936 exceptions. The following switches stop argument processing, and
6937 stash all remaining command-line arguments as the value returned by
6938 the (command-line) function.
6939 -s SCRIPT load Scheme source code from FILE, and exit
6940 -c EXPR evalute Scheme expression EXPR, and exit
6941 -- stop scanning arguments; run interactively
6943 The switches below are processed as they are encountered.
6944 -l FILE load Scheme source code from FILE
6945 -e FUNCTION after reading script, apply FUNCTION to
6946 command line arguments
6947 -ds do -s script at this point
6948 --emacs enable Emacs protocol (experimental)
6949 -h, --help display this help and exit
6950 -v, --version display version information and exit
6951 \ read arguments from following script lines
6953 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6954 which re-implements the traditional "echo" command:
6956 #!/usr/local/bin/guile -s
6959 (map (lambda (arg) (display arg) (display " "))
6963 (main (command-line))
6965 Suppose we invoke this script as follows:
6967 ekko a speckled gecko
6969 Through the magic of Unix script processing (triggered by the `#!'
6970 token at the top of the file), /usr/local/bin/guile receives the
6971 following list of command-line arguments:
6973 ("-s" "./ekko" "a" "speckled" "gecko")
6975 Unix inserts the name of the script after the argument specified on
6976 the first line of the file (in this case, "-s"), and then follows that
6977 with the arguments given to the script. Guile loads the script, which
6978 defines the `main' function, and then applies it to the list of
6979 remaining command-line arguments, ("a" "speckled" "gecko").
6981 In Unix, the first line of a script file must take the following form:
6983 #!INTERPRETER ARGUMENT
6985 where INTERPRETER is the absolute filename of the interpreter
6986 executable, and ARGUMENT is a single command-line argument to pass to
6989 You may only pass one argument to the interpreter, and its length is
6990 limited. These restrictions can be annoying to work around, so Guile
6991 provides a general mechanism (borrowed from, and compatible with,
6992 SCSH) for circumventing them.
6994 If the ARGUMENT in a Guile script is a single backslash character,
6995 `\', Guile will open the script file, parse arguments from its second
6996 and subsequent lines, and replace the `\' with them. So, for example,
6997 here is another implementation of the `ekko' script:
6999 #!/usr/local/bin/guile \
7003 (for-each (lambda (arg) (display arg) (display " "))
7007 If the user invokes this script as follows:
7009 ekko a speckled gecko
7011 Unix expands this into
7013 /usr/local/bin/guile \ ekko a speckled gecko
7015 When Guile sees the `\' argument, it replaces it with the arguments
7016 read from the second line of the script, producing:
7018 /usr/local/bin/guile -e main -s ekko a speckled gecko
7020 This tells Guile to load the `ekko' script, and apply the function
7021 `main' to the argument list ("a" "speckled" "gecko").
7023 Here is how Guile parses the command-line arguments:
7024 - Each space character terminates an argument. This means that two
7025 spaces in a row introduce an empty-string argument.
7026 - The tab character is not permitted (unless you quote it with the
7027 backslash character, as described below), to avoid confusion.
7028 - The newline character terminates the sequence of arguments, and will
7029 also terminate a final non-empty argument. (However, a newline
7030 following a space will not introduce a final empty-string argument;
7031 it only terminates the argument list.)
7032 - The backslash character is the escape character. It escapes
7033 backslash, space, tab, and newline. The ANSI C escape sequences
7034 like \n and \t are also supported. These produce argument
7035 constituents; the two-character combination \n doesn't act like a
7036 terminating newline. The escape sequence \NNN for exactly three
7037 octal digits reads as the character whose ASCII code is NNN. As
7038 above, characters produced this way are argument constituents.
7039 Backslash followed by other characters is not allowed.
7041 * Changes to the procedure for linking libguile with your programs
7043 ** Guile now builds and installs a shared guile library, if your
7044 system support shared libraries. (It still builds a static library on
7045 all systems.) Guile automatically detects whether your system
7046 supports shared libraries. To prevent Guile from buildisg shared
7047 libraries, pass the `--disable-shared' flag to the configure script.
7049 Guile takes longer to compile when it builds shared libraries, because
7050 it must compile every file twice --- once to produce position-
7051 independent object code, and once to produce normal object code.
7053 ** The libthreads library has been merged into libguile.
7055 To link a program against Guile, you now need only link against
7056 -lguile and -lqt; -lthreads is no longer needed. If you are using
7057 autoconf to generate configuration scripts for your application, the
7058 following lines should suffice to add the appropriate libraries to
7061 ### Find quickthreads and libguile.
7062 AC_CHECK_LIB(qt, main)
7063 AC_CHECK_LIB(guile, scm_shell)
7065 * Changes to Scheme functions
7067 ** Guile Scheme's special syntax for keyword objects is now optional,
7068 and disabled by default.
7070 The syntax variation from R4RS made it difficult to port some
7071 interesting packages to Guile. The routines which accepted keyword
7072 arguments (mostly in the module system) have been modified to also
7073 accept symbols whose names begin with `:'.
7075 To change the keyword syntax, you must first import the (ice-9 debug)
7077 (use-modules (ice-9 debug))
7079 Then you can enable the keyword syntax as follows:
7080 (read-set! keywords 'prefix)
7082 To disable keyword syntax, do this:
7083 (read-set! keywords #f)
7085 ** Many more primitive functions accept shared substrings as
7086 arguments. In the past, these functions required normal, mutable
7087 strings as arguments, although they never made use of this
7090 ** The uniform array functions now operate on byte vectors. These
7091 functions are `array-fill!', `serial-array-copy!', `array-copy!',
7092 `serial-array-map', `array-map', `array-for-each', and
7095 ** The new functions `trace' and `untrace' implement simple debugging
7096 support for Scheme functions.
7098 The `trace' function accepts any number of procedures as arguments,
7099 and tells the Guile interpreter to display each procedure's name and
7100 arguments each time the procedure is invoked. When invoked with no
7101 arguments, `trace' returns the list of procedures currently being
7104 The `untrace' function accepts any number of procedures as arguments,
7105 and tells the Guile interpreter not to trace them any more. When
7106 invoked with no arguments, `untrace' untraces all curretly traced
7109 The tracing in Guile has an advantage over most other systems: we
7110 don't create new procedure objects, but mark the procedure objects
7111 themselves. This means that anonymous and internal procedures can be
7114 ** The function `assert-repl-prompt' has been renamed to
7115 `set-repl-prompt!'. It takes one argument, PROMPT.
7116 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
7117 - If PROMPT is a string, we use it as a prompt.
7118 - If PROMPT is a procedure accepting no arguments, we call it, and
7119 display the result as a prompt.
7120 - Otherwise, we display "> ".
7122 ** The new function `eval-string' reads Scheme expressions from a
7123 string and evaluates them, returning the value of the last expression
7124 in the string. If the string contains no expressions, it returns an
7127 ** The new function `thunk?' returns true iff its argument is a
7128 procedure of zero arguments.
7130 ** `defined?' is now a builtin function, instead of syntax. This
7131 means that its argument should be quoted. It returns #t iff its
7132 argument is bound in the current module.
7134 ** The new syntax `use-modules' allows you to add new modules to your
7135 environment without re-typing a complete `define-module' form. It
7136 accepts any number of module names as arguments, and imports their
7137 public bindings into the current module.
7139 ** The new function (module-defined? NAME MODULE) returns true iff
7140 NAME, a symbol, is defined in MODULE, a module object.
7142 ** The new function `builtin-bindings' creates and returns a hash
7143 table containing copies of all the root module's bindings.
7145 ** The new function `builtin-weak-bindings' does the same as
7146 `builtin-bindings', but creates a doubly-weak hash table.
7148 ** The `equal?' function now considers variable objects to be
7149 equivalent if they have the same name and the same value.
7151 ** The new function `command-line' returns the command-line arguments
7152 given to Guile, as a list of strings.
7154 When using guile as a script interpreter, `command-line' returns the
7155 script's arguments; those processed by the interpreter (like `-s' or
7156 `-c') are omitted. (In other words, you get the normal, expected
7157 behavior.) Any application that uses scm_shell to process its
7158 command-line arguments gets this behavior as well.
7160 ** The new function `load-user-init' looks for a file called `.guile'
7161 in the user's home directory, and loads it if it exists. This is
7162 mostly for use by the code generated by scm_compile_shell_switches,
7163 but we thought it might also be useful in other circumstances.
7165 ** The new function `log10' returns the base-10 logarithm of its
7168 ** Changes to I/O functions
7170 *** The functions `read', `primitive-load', `read-and-eval!', and
7171 `primitive-load-path' no longer take optional arguments controlling
7172 case insensitivity and a `#' parser.
7174 Case sensitivity is now controlled by a read option called
7175 `case-insensitive'. The user can add new `#' syntaxes with the
7176 `read-hash-extend' function (see below).
7178 *** The new function `read-hash-extend' allows the user to change the
7179 syntax of Guile Scheme in a somewhat controlled way.
7181 (read-hash-extend CHAR PROC)
7182 When parsing S-expressions, if we read a `#' character followed by
7183 the character CHAR, use PROC to parse an object from the stream.
7184 If PROC is #f, remove any parsing procedure registered for CHAR.
7186 The reader applies PROC to two arguments: CHAR and an input port.
7188 *** The new functions read-delimited and read-delimited! provide a
7189 general mechanism for doing delimited input on streams.
7191 (read-delimited DELIMS [PORT HANDLE-DELIM])
7192 Read until we encounter one of the characters in DELIMS (a string),
7193 or end-of-file. PORT is the input port to read from; it defaults to
7194 the current input port. The HANDLE-DELIM parameter determines how
7195 the terminating character is handled; it should be one of the
7198 'trim omit delimiter from result
7199 'peek leave delimiter character in input stream
7200 'concat append delimiter character to returned value
7201 'split return a pair: (RESULT . TERMINATOR)
7203 HANDLE-DELIM defaults to 'peek.
7205 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
7206 A side-effecting variant of `read-delimited'.
7208 The data is written into the string BUF at the indices in the
7209 half-open interval [START, END); the default interval is the whole
7210 string: START = 0 and END = (string-length BUF). The values of
7211 START and END must specify a well-defined interval in BUF, i.e.
7212 0 <= START <= END <= (string-length BUF).
7214 It returns NBYTES, the number of bytes read. If the buffer filled
7215 up without a delimiter character being found, it returns #f. If the
7216 port is at EOF when the read starts, it returns the EOF object.
7218 If an integer is returned (i.e., the read is successfully terminated
7219 by reading a delimiter character), then the HANDLE-DELIM parameter
7220 determines how to handle the terminating character. It is described
7221 above, and defaults to 'peek.
7223 (The descriptions of these functions were borrowed from the SCSH
7224 manual, by Olin Shivers and Brian Carlstrom.)
7226 *** The `%read-delimited!' function is the primitive used to implement
7227 `read-delimited' and `read-delimited!'.
7229 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
7231 This returns a pair of values: (TERMINATOR . NUM-READ).
7232 - TERMINATOR describes why the read was terminated. If it is a
7233 character or the eof object, then that is the value that terminated
7234 the read. If it is #f, the function filled the buffer without finding
7235 a delimiting character.
7236 - NUM-READ is the number of characters read into BUF.
7238 If the read is successfully terminated by reading a delimiter
7239 character, then the gobble? parameter determines what to do with the
7240 terminating character. If true, the character is removed from the
7241 input stream; if false, the character is left in the input stream
7242 where a subsequent read operation will retrieve it. In either case,
7243 the character is also the first value returned by the procedure call.
7245 (The descriptions of this function was borrowed from the SCSH manual,
7246 by Olin Shivers and Brian Carlstrom.)
7248 *** The `read-line' and `read-line!' functions have changed; they now
7249 trim the terminator by default; previously they appended it to the
7250 returned string. For the old behavior, use (read-line PORT 'concat).
7252 *** The functions `uniform-array-read!' and `uniform-array-write!' now
7253 take new optional START and END arguments, specifying the region of
7254 the array to read and write.
7256 *** The `ungetc-char-ready?' function has been removed. We feel it's
7257 inappropriate for an interface to expose implementation details this
7260 ** Changes to the Unix library and system call interface
7262 *** The new fcntl function provides access to the Unix `fcntl' system
7265 (fcntl PORT COMMAND VALUE)
7266 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
7267 Values for COMMAND are:
7269 F_DUPFD duplicate a file descriptor
7270 F_GETFD read the descriptor's close-on-exec flag
7271 F_SETFD set the descriptor's close-on-exec flag to VALUE
7272 F_GETFL read the descriptor's flags, as set on open
7273 F_SETFL set the descriptor's flags, as set on open to VALUE
7274 F_GETOWN return the process ID of a socket's owner, for SIGIO
7275 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
7276 FD_CLOEXEC not sure what this is
7278 For details, see the documentation for the fcntl system call.
7280 *** The arguments to `select' have changed, for compatibility with
7281 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
7282 expected behavior. The MILLISECONDS parameter has been changed to
7283 MICROSECONDS, to more closely resemble the underlying system call.
7284 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
7285 corresponding return set will be the same.
7287 *** The arguments to the `mknod' system call have changed. They are
7290 (mknod PATH TYPE PERMS DEV)
7291 Create a new file (`node') in the file system. PATH is the name of
7292 the file to create. TYPE is the kind of file to create; it should
7293 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
7294 permission bits to give the newly created file. If TYPE is
7295 'block-special or 'char-special, DEV specifies which device the
7296 special file refers to; its interpretation depends on the kind of
7297 special file being created.
7299 *** The `fork' function has been renamed to `primitive-fork', to avoid
7300 clashing with various SCSH forks.
7302 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
7303 and `recvfrom!'. They no longer accept a size for a second argument;
7304 you must pass a string to hold the received value. They no longer
7305 return the buffer. Instead, `recv' returns the length of the message
7306 received, and `recvfrom' returns a pair containing the packet's length
7307 and originating address.
7309 *** The file descriptor datatype has been removed, as have the
7310 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
7311 We plan to replace these functions with a SCSH-compatible interface.
7313 *** The `create' function has been removed; it's just a special case
7316 *** There are new functions to break down process termination status
7317 values. In the descriptions below, STATUS is a value returned by
7320 (status:exit-val STATUS)
7321 If the child process exited normally, this function returns the exit
7322 code for the child process (i.e., the value passed to exit, or
7323 returned from main). If the child process did not exit normally,
7324 this function returns #f.
7326 (status:stop-sig STATUS)
7327 If the child process was suspended by a signal, this function
7328 returns the signal that suspended the child. Otherwise, it returns
7331 (status:term-sig STATUS)
7332 If the child process terminated abnormally, this function returns
7333 the signal that terminated the child. Otherwise, this function
7336 POSIX promises that exactly one of these functions will return true on
7337 a valid STATUS value.
7339 These functions are compatible with SCSH.
7341 *** There are new accessors and setters for the broken-out time vectors
7342 returned by `localtime', `gmtime', and that ilk. They are:
7344 Component Accessor Setter
7345 ========================= ============ ============
7346 seconds tm:sec set-tm:sec
7347 minutes tm:min set-tm:min
7348 hours tm:hour set-tm:hour
7349 day of the month tm:mday set-tm:mday
7350 month tm:mon set-tm:mon
7351 year tm:year set-tm:year
7352 day of the week tm:wday set-tm:wday
7353 day in the year tm:yday set-tm:yday
7354 daylight saving time tm:isdst set-tm:isdst
7355 GMT offset, seconds tm:gmtoff set-tm:gmtoff
7356 name of time zone tm:zone set-tm:zone
7358 *** There are new accessors for the vectors returned by `uname',
7359 describing the host system:
7362 ============================================== ================
7363 name of the operating system implementation utsname:sysname
7364 network name of this machine utsname:nodename
7365 release level of the operating system utsname:release
7366 version level of the operating system utsname:version
7367 machine hardware platform utsname:machine
7369 *** There are new accessors for the vectors returned by `getpw',
7370 `getpwnam', `getpwuid', and `getpwent', describing entries from the
7371 system's user database:
7374 ====================== =================
7375 user name passwd:name
7376 user password passwd:passwd
7379 real name passwd:gecos
7380 home directory passwd:dir
7381 shell program passwd:shell
7383 *** There are new accessors for the vectors returned by `getgr',
7384 `getgrnam', `getgrgid', and `getgrent', describing entries from the
7385 system's group database:
7388 ======================= ============
7389 group name group:name
7390 group password group:passwd
7392 group members group:mem
7394 *** There are new accessors for the vectors returned by `gethost',
7395 `gethostbyaddr', `gethostbyname', and `gethostent', describing
7399 ========================= ===============
7400 official name of host hostent:name
7401 alias list hostent:aliases
7402 host address type hostent:addrtype
7403 length of address hostent:length
7404 list of addresses hostent:addr-list
7406 *** There are new accessors for the vectors returned by `getnet',
7407 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
7411 ========================= ===============
7412 official name of net netent:name
7413 alias list netent:aliases
7414 net number type netent:addrtype
7415 net number netent:net
7417 *** There are new accessors for the vectors returned by `getproto',
7418 `getprotobyname', `getprotobynumber', and `getprotoent', describing
7422 ========================= ===============
7423 official protocol name protoent:name
7424 alias list protoent:aliases
7425 protocol number protoent:proto
7427 *** There are new accessors for the vectors returned by `getserv',
7428 `getservbyname', `getservbyport', and `getservent', describing
7432 ========================= ===============
7433 official service name servent:name
7434 alias list servent:aliases
7435 port number servent:port
7436 protocol to use servent:proto
7438 *** There are new accessors for the sockaddr structures returned by
7439 `accept', `getsockname', `getpeername', `recvfrom!':
7442 ======================================== ===============
7443 address format (`family') sockaddr:fam
7444 path, for file domain addresses sockaddr:path
7445 address, for internet domain addresses sockaddr:addr
7446 TCP or UDP port, for internet sockaddr:port
7448 *** The `getpwent', `getgrent', `gethostent', `getnetent',
7449 `getprotoent', and `getservent' functions now return #f at the end of
7450 the user database. (They used to throw an exception.)
7452 Note that calling MUMBLEent function is equivalent to calling the
7453 corresponding MUMBLE function with no arguments.
7455 *** The `setpwent', `setgrent', `sethostent', `setnetent',
7456 `setprotoent', and `setservent' routines now take no arguments.
7458 *** The `gethost', `getproto', `getnet', and `getserv' functions now
7459 provide more useful information when they throw an exception.
7461 *** The `lnaof' function has been renamed to `inet-lnaof'.
7463 *** Guile now claims to have the `current-time' feature.
7465 *** The `mktime' function now takes an optional second argument ZONE,
7466 giving the time zone to use for the conversion. ZONE should be a
7467 string, in the same format as expected for the "TZ" environment variable.
7469 *** The `strptime' function now returns a pair (TIME . COUNT), where
7470 TIME is the parsed time as a vector, and COUNT is the number of
7471 characters from the string left unparsed. This function used to
7472 return the remaining characters as a string.
7474 *** The `gettimeofday' function has replaced the old `time+ticks' function.
7475 The return value is now (SECONDS . MICROSECONDS); the fractional
7476 component is no longer expressed in "ticks".
7478 *** The `ticks/sec' constant has been removed, in light of the above change.
7480 * Changes to the gh_ interface
7482 ** gh_eval_str() now returns an SCM object which is the result of the
7485 ** gh_scm2str() now copies the Scheme data to a caller-provided C
7488 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
7489 and returns the array
7491 ** gh_scm2str0() is gone: there is no need to distinguish
7492 null-terminated from non-null-terminated, since gh_scm2newstr() allows
7493 the user to interpret the data both ways.
7495 * Changes to the scm_ interface
7497 ** The new function scm_symbol_value0 provides an easy way to get a
7498 symbol's value from C code:
7500 SCM scm_symbol_value0 (char *NAME)
7501 Return the value of the symbol named by the null-terminated string
7502 NAME in the current module. If the symbol named NAME is unbound in
7503 the current module, return SCM_UNDEFINED.
7505 ** The new function scm_sysintern0 creates new top-level variables,
7506 without assigning them a value.
7508 SCM scm_sysintern0 (char *NAME)
7509 Create a new Scheme top-level variable named NAME. NAME is a
7510 null-terminated string. Return the variable's value cell.
7512 ** The function scm_internal_catch is the guts of catch. It handles
7513 all the mechanics of setting up a catch target, invoking the catch
7514 body, and perhaps invoking the handler if the body does a throw.
7516 The function is designed to be usable from C code, but is general
7517 enough to implement all the semantics Guile Scheme expects from throw.
7519 TAG is the catch tag. Typically, this is a symbol, but this function
7520 doesn't actually care about that.
7522 BODY is a pointer to a C function which runs the body of the catch;
7523 this is the code you can throw from. We call it like this:
7524 BODY (BODY_DATA, JMPBUF)
7526 BODY_DATA is just the BODY_DATA argument we received; we pass it
7527 through to BODY as its first argument. The caller can make
7528 BODY_DATA point to anything useful that BODY might need.
7529 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
7530 which we have just created and initialized.
7532 HANDLER is a pointer to a C function to deal with a throw to TAG,
7533 should one occur. We call it like this:
7534 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
7536 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
7537 same idea as BODY_DATA above.
7538 THROWN_TAG is the tag that the user threw to; usually this is
7539 TAG, but it could be something else if TAG was #t (i.e., a
7540 catch-all), or the user threw to a jmpbuf.
7541 THROW_ARGS is the list of arguments the user passed to the THROW
7544 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
7545 is just a pointer we pass through to HANDLER. We don't actually
7546 use either of those pointers otherwise ourselves. The idea is
7547 that, if our caller wants to communicate something to BODY or
7548 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
7549 HANDLER can then use. Think of it as a way to make BODY and
7550 HANDLER closures, not just functions; MUMBLE_DATA points to the
7553 Of course, it's up to the caller to make sure that any data a
7554 MUMBLE_DATA needs is protected from GC. A common way to do this is
7555 to make MUMBLE_DATA a pointer to data stored in an automatic
7556 structure variable; since the collector must scan the stack for
7557 references anyway, this assures that any references in MUMBLE_DATA
7560 ** The new function scm_internal_lazy_catch is exactly like
7561 scm_internal_catch, except:
7563 - It does not unwind the stack (this is the major difference).
7564 - If handler returns, its value is returned from the throw.
7565 - BODY always receives #f as its JMPBUF argument (since there's no
7566 jmpbuf associated with a lazy catch, because we don't unwind the
7569 ** scm_body_thunk is a new body function you can pass to
7570 scm_internal_catch if you want the body to be like Scheme's `catch'
7571 --- a thunk, or a function of one argument if the tag is #f.
7573 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
7574 contains the Scheme procedure to invoke as the body, and the tag
7575 we're catching. If the tag is #f, then we pass JMPBUF (created by
7576 scm_internal_catch) to the body procedure; otherwise, the body gets
7579 ** scm_handle_by_proc is a new handler function you can pass to
7580 scm_internal_catch if you want the handler to act like Scheme's catch
7581 --- call a procedure with the tag and the throw arguments.
7583 If the user does a throw to this catch, this function runs a handler
7584 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
7585 variable holding the Scheme procedure object to invoke. It ought to
7586 be a pointer to an automatic variable (i.e., one living on the stack),
7587 or the procedure object should be otherwise protected from GC.
7589 ** scm_handle_by_message is a new handler function to use with
7590 `scm_internal_catch' if you want Guile to print a message and die.
7591 It's useful for dealing with throws to uncaught keys at the top level.
7593 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
7594 message header to print; if zero, we use "guile" instead. That
7595 text is followed by a colon, then the message described by ARGS.
7597 ** The return type of scm_boot_guile is now void; the function does
7598 not return a value, and indeed, never returns at all.
7600 ** The new function scm_shell makes it easy for user applications to
7601 process command-line arguments in a way that is compatible with the
7602 stand-alone guile interpreter (which is in turn compatible with SCSH,
7605 To use the scm_shell function, first initialize any guile modules
7606 linked into your application, and then call scm_shell with the values
7607 of ARGC and ARGV your `main' function received. scm_shell will add
7608 any SCSH-style meta-arguments from the top of the script file to the
7609 argument vector, and then process the command-line arguments. This
7610 generally means loading a script file or starting up an interactive
7611 command interpreter. For details, see "Changes to the stand-alone
7614 ** The new functions scm_get_meta_args and scm_count_argv help you
7615 implement the SCSH-style meta-argument, `\'.
7617 char **scm_get_meta_args (int ARGC, char **ARGV)
7618 If the second element of ARGV is a string consisting of a single
7619 backslash character (i.e. "\\" in Scheme notation), open the file
7620 named by the following argument, parse arguments from it, and return
7621 the spliced command line. The returned array is terminated by a
7624 For details of argument parsing, see above, under "guile now accepts
7625 command-line arguments compatible with SCSH..."
7627 int scm_count_argv (char **ARGV)
7628 Count the arguments in ARGV, assuming it is terminated by a null
7631 For an example of how these functions might be used, see the source
7632 code for the function scm_shell in libguile/script.c.
7634 You will usually want to use scm_shell instead of calling this
7637 ** The new function scm_compile_shell_switches turns an array of
7638 command-line arguments into Scheme code to carry out the actions they
7639 describe. Given ARGC and ARGV, it returns a Scheme expression to
7640 evaluate, and calls scm_set_program_arguments to make any remaining
7641 command-line arguments available to the Scheme code. For example,
7642 given the following arguments:
7644 -e main -s ekko a speckled gecko
7646 scm_set_program_arguments will return the following expression:
7648 (begin (load "ekko") (main (command-line)) (quit))
7650 You will usually want to use scm_shell instead of calling this
7653 ** The function scm_shell_usage prints a usage message appropriate for
7654 an interpreter that uses scm_compile_shell_switches to handle its
7655 command-line arguments.
7657 void scm_shell_usage (int FATAL, char *MESSAGE)
7658 Print a usage message to the standard error output. If MESSAGE is
7659 non-zero, write it before the usage message, followed by a newline.
7660 If FATAL is non-zero, exit the process, using FATAL as the
7661 termination status. (If you want to be compatible with Guile,
7662 always use 1 as the exit status when terminating due to command-line
7665 You will usually want to use scm_shell instead of calling this
7668 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7669 expressions. It used to return SCM_EOL. Earth-shattering.
7671 ** The macros for declaring scheme objects in C code have been
7672 rearranged slightly. They are now:
7674 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7675 Declare a static SCM variable named C_NAME, and initialize it to
7676 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7677 be a C identifier, and SCHEME_NAME should be a C string.
7679 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7680 Just like SCM_SYMBOL, but make C_NAME globally visible.
7682 SCM_VCELL (C_NAME, SCHEME_NAME)
7683 Create a global variable at the Scheme level named SCHEME_NAME.
7684 Declare a static SCM variable named C_NAME, and initialize it to
7685 point to the Scheme variable's value cell.
7687 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7688 Just like SCM_VCELL, but make C_NAME globally visible.
7690 The `guile-snarf' script writes initialization code for these macros
7691 to its standard output, given C source code as input.
7693 The SCM_GLOBAL macro is gone.
7695 ** The scm_read_line and scm_read_line_x functions have been replaced
7696 by Scheme code based on the %read-delimited! procedure (known to C
7697 code as scm_read_delimited_x). See its description above for more
7700 ** The function scm_sys_open has been renamed to scm_open. It now
7701 returns a port instead of an FD object.
7703 * The dynamic linking support has changed. For more information, see
7704 libguile/DYNAMIC-LINKING.
7709 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7712 * Changes to the 'guile' program:
7714 ** Guile now loads some new files when it starts up. Guile first
7715 searches the load path for init.scm, and loads it if found. Then, if
7716 Guile is not being used to execute a script, and the user's home
7717 directory contains a file named `.guile', Guile loads that.
7719 ** You can now use Guile as a shell script interpreter.
7721 To paraphrase the SCSH manual:
7723 When Unix tries to execute an executable file whose first two
7724 characters are the `#!', it treats the file not as machine code to
7725 be directly executed by the native processor, but as source code
7726 to be executed by some interpreter. The interpreter to use is
7727 specified immediately after the #! sequence on the first line of
7728 the source file. The kernel reads in the name of the interpreter,
7729 and executes that instead. It passes the interpreter the source
7730 filename as its first argument, with the original arguments
7731 following. Consult the Unix man page for the `exec' system call
7732 for more information.
7734 Now you can use Guile as an interpreter, using a mechanism which is a
7735 compatible subset of that provided by SCSH.
7737 Guile now recognizes a '-s' command line switch, whose argument is the
7738 name of a file of Scheme code to load. It also treats the two
7739 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7740 to make a file of Scheme code directly executable by Unix, insert the
7741 following two lines at the top of the file:
7743 #!/usr/local/bin/guile -s
7746 Guile treats the argument of the `-s' command-line switch as the name
7747 of a file of Scheme code to load, and treats the sequence `#!' as the
7748 start of a block comment, terminated by `!#'.
7750 For example, here's a version of 'echo' written in Scheme:
7752 #!/usr/local/bin/guile -s
7754 (let loop ((args (cdr (program-arguments))))
7757 (display (car args))
7758 (if (pair? (cdr args))
7760 (loop (cdr args)))))
7763 Why does `#!' start a block comment terminated by `!#', instead of the
7764 end of the line? That is the notation SCSH uses, and although we
7765 don't yet support the other SCSH features that motivate that choice,
7766 we would like to be backward-compatible with any existing Guile
7767 scripts once we do. Furthermore, if the path to Guile on your system
7768 is too long for your kernel, you can start the script with this
7772 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7775 Note that some very old Unix systems don't support the `#!' syntax.
7778 ** You can now run Guile without installing it.
7780 Previous versions of the interactive Guile interpreter (`guile')
7781 couldn't start up unless Guile's Scheme library had been installed;
7782 they used the value of the environment variable `SCHEME_LOAD_PATH'
7783 later on in the startup process, but not to find the startup code
7784 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7787 To run Guile without installing it, build it in the normal way, and
7788 then set the environment variable `SCHEME_LOAD_PATH' to a
7789 colon-separated list of directories, including the top-level directory
7790 of the Guile sources. For example, if you unpacked Guile so that the
7791 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7794 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7797 ** Guile's read-eval-print loop no longer prints #<unspecified>
7798 results. If the user wants to see this, she can evaluate the
7799 expression (assert-repl-print-unspecified #t), perhaps in her startup
7802 ** Guile no longer shows backtraces by default when an error occurs;
7803 however, it does display a message saying how to get one, and how to
7804 request that they be displayed by default. After an error, evaluate
7806 to see a backtrace, and
7807 (debug-enable 'backtrace)
7808 to see them by default.
7812 * Changes to Guile Scheme:
7814 ** Guile now distinguishes between #f and the empty list.
7816 This is for compatibility with the IEEE standard, the (possibly)
7817 upcoming Revised^5 Report on Scheme, and many extant Scheme
7820 Guile used to have #f and '() denote the same object, to make Scheme's
7821 type system more compatible with Emacs Lisp's. However, the change
7822 caused too much trouble for Scheme programmers, and we found another
7823 way to reconcile Emacs Lisp with Scheme that didn't require this.
7826 ** Guile's delq, delv, delete functions, and their destructive
7827 counterparts, delq!, delv!, and delete!, now remove all matching
7828 elements from the list, not just the first. This matches the behavior
7829 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7830 functions which inspired them.
7832 I recognize that this change may break code in subtle ways, but it
7833 seems best to make the change before the FSF's first Guile release,
7837 ** The compiled-library-path function has been deleted from libguile.
7839 ** The facilities for loading Scheme source files have changed.
7841 *** The variable %load-path now tells Guile which directories to search
7842 for Scheme code. Its value is a list of strings, each of which names
7845 *** The variable %load-extensions now tells Guile which extensions to
7846 try appending to a filename when searching the load path. Its value
7847 is a list of strings. Its default value is ("" ".scm").
7849 *** (%search-load-path FILENAME) searches the directories listed in the
7850 value of the %load-path variable for a Scheme file named FILENAME,
7851 with all the extensions listed in %load-extensions. If it finds a
7852 match, then it returns its full filename. If FILENAME is absolute, it
7853 returns it unchanged. Otherwise, it returns #f.
7855 %search-load-path will not return matches that refer to directories.
7857 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7858 uses %seach-load-path to find a file named FILENAME, and loads it if
7859 it finds it. If it can't read FILENAME for any reason, it throws an
7862 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7865 *** load uses the same searching semantics as primitive-load.
7867 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7868 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7869 path, and load-module-with-path have been deleted. The functions
7870 above should serve their purposes.
7872 *** If the value of the variable %load-hook is a procedure,
7873 `primitive-load' applies its value to the name of the file being
7874 loaded (without the load path directory name prepended). If its value
7875 is #f, it is ignored. Otherwise, an error occurs.
7877 This is mostly useful for printing load notification messages.
7880 ** The function `eval!' is no longer accessible from the scheme level.
7881 We can't allow operations which introduce glocs into the scheme level,
7882 because Guile's type system can't handle these as data. Use `eval' or
7883 `read-and-eval!' (see below) as replacement.
7885 ** The new function read-and-eval! reads an expression from PORT,
7886 evaluates it, and returns the result. This is more efficient than
7887 simply calling `read' and `eval', since it is not necessary to make a
7888 copy of the expression for the evaluator to munge.
7890 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7891 for the `read' function.
7894 ** The function `int?' has been removed; its definition was identical
7895 to that of `integer?'.
7897 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7898 use the R4RS names for these functions.
7900 ** The function object-properties no longer returns the hash handle;
7901 it simply returns the object's property list.
7903 ** Many functions have been changed to throw errors, instead of
7904 returning #f on failure. The point of providing exception handling in
7905 the language is to simplify the logic of user code, but this is less
7906 useful if Guile's primitives don't throw exceptions.
7908 ** The function `fileno' has been renamed from `%fileno'.
7910 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7913 * Changes to Guile's C interface:
7915 ** The library's initialization procedure has been simplified.
7916 scm_boot_guile now has the prototype:
7918 void scm_boot_guile (int ARGC,
7920 void (*main_func) (),
7923 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7924 MAIN_FUNC should do all the work of the program (initializing other
7925 packages, reading user input, etc.) before returning. When MAIN_FUNC
7926 returns, call exit (0); this function never returns. If you want some
7927 other exit value, MAIN_FUNC may call exit itself.
7929 scm_boot_guile arranges for program-arguments to return the strings
7930 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7931 scm_set_program_arguments with the final list, so Scheme code will
7932 know which arguments have been processed.
7934 scm_boot_guile establishes a catch-all catch handler which prints an
7935 error message and exits the process. This means that Guile exits in a
7936 coherent way when system errors occur and the user isn't prepared to
7937 handle it. If the user doesn't like this behavior, they can establish
7938 their own universal catcher in MAIN_FUNC to shadow this one.
7940 Why must the caller do all the real work from MAIN_FUNC? The garbage
7941 collector assumes that all local variables of type SCM will be above
7942 scm_boot_guile's stack frame on the stack. If you try to manipulate
7943 SCM values after this function returns, it's the luck of the draw
7944 whether the GC will be able to find the objects you allocate. So,
7945 scm_boot_guile function exits, rather than returning, to discourage
7946 people from making that mistake.
7948 The IN, OUT, and ERR arguments were removed; there are other
7949 convenient ways to override these when desired.
7951 The RESULT argument was deleted; this function should never return.
7953 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7957 ** Guile's header files should no longer conflict with your system's
7960 In order to compile code which #included <libguile.h>, previous
7961 versions of Guile required you to add a directory containing all the
7962 Guile header files to your #include path. This was a problem, since
7963 Guile's header files have names which conflict with many systems'
7966 Now only <libguile.h> need appear in your #include path; you must
7967 refer to all Guile's other header files as <libguile/mumble.h>.
7968 Guile's installation procedure puts libguile.h in $(includedir), and
7969 the rest in $(includedir)/libguile.
7972 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7973 have been added to the Guile library.
7975 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7976 OBJ will not be freed, even if all other references are dropped,
7977 until someone does scm_unprotect_object (OBJ). Both functions
7980 Note that calls to scm_protect_object do not nest. You can call
7981 scm_protect_object any number of times on a given object, and the
7982 next call to scm_unprotect_object will unprotect it completely.
7984 Basically, scm_protect_object and scm_unprotect_object just
7985 maintain a list of references to things. Since the GC knows about
7986 this list, all objects it mentions stay alive. scm_protect_object
7987 adds its argument to the list; scm_unprotect_object remove its
7988 argument from the list.
7991 ** scm_eval_0str now returns the value of the last expression
7994 ** The new function scm_read_0str reads an s-expression from a
7995 null-terminated string, and returns it.
7997 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7998 to a Scheme port object.
8000 ** The new function `scm_set_program_arguments' allows C code to set
8001 the value returned by the Scheme `program-arguments' function.
8006 * Guile no longer includes sophisticated Tcl/Tk support.
8008 The old Tcl/Tk support was unsatisfying to us, because it required the
8009 user to link against the Tcl library, as well as Tk and Guile. The
8010 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
8011 referring to widgets by names, rather than exporting widgets to Scheme
8012 code as a special datatype.
8014 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
8015 maintainers described some very interesting changes in progress to the
8016 Tcl/Tk internals, which would facilitate clean interfaces between lone
8017 Tk and other interpreters --- even for garbage-collected languages
8018 like Scheme. They expected the new Tk to be publicly available in the
8021 Since it seems that Guile might soon have a new, cleaner interface to
8022 lone Tk, and that the old Guile/Tk glue code would probably need to be
8023 completely rewritten, we (Jim Blandy and Richard Stallman) have
8024 decided not to support the old code. We'll spend the time instead on
8025 a good interface to the newer Tk, as soon as it is available.
8027 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
8030 Copyright information:
8032 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
8034 Permission is granted to anyone to make or distribute verbatim copies
8035 of this document as received, in any medium, provided that the
8036 copyright notice and this permission notice are preserved,
8037 thus giving the recipient permission to redistribute in turn.
8039 Permission is granted to distribute modified versions
8040 of this document, or of portions of it,
8041 under the above conditions, provided also that they
8042 carry prominent notices stating who last changed them.
8047 paragraph-separate: "[
\f]*$"