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 1.9.2 (since the 1.9.1 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 Changes in 1.9.x (since the 1.8.x series):
33 * New modules (see the manual for details)
35 ** `(srfi srfi-18)', more sophisticated multithreading support
36 ** `(ice-9 i18n)', internationalization support
37 ** `(rnrs bytevector)', the R6RS bytevector API
38 ** `(rnrs io ports)', a subset of the R6RS I/O port API
39 ** `(system xref)', a cross-referencing facility (FIXME undocumented)
41 * Changes to the stand-alone interpreter
43 ** Guile now can compile Scheme to bytecode for a custom virtual machine.
45 Compiled code loads much faster than Scheme source code, and runs around
46 3 or 4 times as fast, generating much less garbage in the process.
48 ** The stack limit is now initialized from the environment.
50 If getrlimit(2) is available and a stack limit is set, Guile will set
51 its stack limit to 80% of the rlimit. Otherwise the limit is 160000
52 words, a four-fold increase from the earlier default limit.
54 ** New environment variables: GUILE_LOAD_COMPILED_PATH,
55 GUILE_SYSTEM_LOAD_COMPILED_PATH
57 GUILE_LOAD_COMPILED_PATH is for compiled files what GUILE_LOAD_PATH is
58 for source files. It is a different path, however, because compiled
59 files are architecture-specific. GUILE_SYSTEM_LOAD_COMPILED_PATH is like
62 ** New read-eval-print loop (REPL) implementation
64 Running Guile with no arguments drops the user into the new REPL. While
65 it is self-documenting to an extent, the new REPL has not yet been
66 documented in the manual. This will be fixed before 2.0.
68 ** New `guile-tools' commands: `compile', `disassemble'
70 Pass the `--help' command-line option to these commands for more
73 * Changes to Scheme functions and syntax
75 ** Procedure removed: `the-environment'
77 This procedure was part of the interpreter's execution model, and does
78 not apply to the compiler.
80 ** Files loaded with `primitive-load-path' will now be compiled
83 If a compiled .go file corresponding to a .scm file is not found or is
84 not fresh, the .scm file will be compiled on the fly, and the resulting
85 .go file stored away. An advisory note will be printed on the console.
87 Note that this mechanism depends on preservation of the .scm and .go
88 modification times; if the .scm or .go files are moved after
89 installation, care should be taken to preserve their original
92 Autocompiled files will be stored in the $XDG_CACHE_HOME/guile/ccache
93 directory, where $XDG_CACHE_HOME defaults to ~/.cache. This directory
94 will be created if needed.
96 To inhibit autocompilation, set the GUILE_AUTO_COMPILE environment
97 variable to 0, or pass --no-autocompile on the Guile command line.
99 Note that there is currently a bug here: automatic compilation will
100 sometimes be attempted when it shouldn't.
102 For example, the old (lang elisp) modules are meant to be interpreted,
103 not compiled. This bug will be fixed before 2.0. FIXME 2.0: Should say
104 something here about module-transformer called for compile.
106 ** New POSIX procedures: `getrlimit' and `setrlimit'
108 Note however that the interface of these functions is likely to change
109 in the next prerelease.
111 ** New procedure in `(oops goops)': `method-formals'
113 ** BUG: (procedure-property func 'arity) does not work on compiled
116 This will be fixed one way or another before 2.0.
118 ** New procedures in (ice-9 session): `add-value-help-handler!',
119 `remove-value-help-handler!', `add-name-help-handler!'
120 `remove-name-help-handler!', `procedure-arguments',
122 The value and name help handlers provide some minimal extensibility to
123 the help interface. Guile-lib's `(texinfo reflection)' uses them, for
124 example, to make stexinfo help documentation available. See those
125 procedures' docstrings for more information.
127 `procedure-arguments' describes the arguments that a procedure can take,
128 combining arity and formals. For example:
130 (procedure-arguments resolve-interface)
131 => ((required . (name)) (rest . args))
133 Additionally, `module-commentary' is now publically exported from
136 ** Deprecated: `procedure->memoizing-macro', `procedure->syntax'
138 These procedures will not work with syncase expansion, and indeed are
139 not used in the normal course of Guile. They are still used by the old
140 Emacs Lisp support, however.
142 ** New language: ECMAScript
144 Guile now ships with one other high-level language supported,
145 ECMAScript. The goal is to support all of version 3.1 of the standard,
146 but not all of the libraries are there yet. This support is not yet
147 documented; ask on the mailing list if you are interested.
149 ** New language: Brainfuck
151 Brainfuck is a toy language that closely models Turing machines. Guile's
152 brainfuck compiler is meant to be an example of implementing other
153 languages. See the manual for details, or
154 http://en.wikipedia.org/wiki/Brainfuck for more information about the
155 Brainfuck language itself.
157 ** Defmacros may now have docstrings.
159 Indeed, any macro may have a docstring. `object-documentation' from
160 `(ice-9 documentation)' may be used to retrieve the docstring, once you
161 have a macro value -- but see the above note about first-class macros.
162 Docstrings are associated with the syntax transformer procedures.
164 ** The psyntax expander now knows how to interpret the @ and @@ special
167 ** The psyntax expander is now hygienic with respect to modules.
169 Free variables in a macro are scoped in the module that the macro was
170 defined in, not in the module the macro is used in. For example, code
173 (define-module (foo) #:export (bar))
174 (define (helper x) ...)
176 (syntax-rules () ((_ x) (helper x))))
178 (define-module (baz) #:use-module (foo))
181 It used to be you had to export `helper' from `(foo)' as well.
182 Thankfully, this has been fixed.
184 ** New function, `procedure-module'
186 While useful on its own, `procedure-module' is used by psyntax on syntax
187 transformers to determine the module in which to scope introduced
190 ** `eval-case' has been deprecated, and replaced by `eval-when'.
192 The semantics of `eval-when' are easier to understand. It is still
193 missing documentation, however.
195 ** Guile is now more strict about prohibiting definitions in expression
198 Although previous versions of Guile accepted it, the following
199 expression is not valid, in R5RS or R6RS:
201 (if test (define foo 'bar) (define foo 'baz))
203 In this specific case, it would be better to do:
205 (define foo (if test 'bar 'baz))
207 It is certainly possible to circumvent this resriction with e.g.
208 `(module-define! (current-module) 'foo 'baz)'. We would appreciate
209 feedback about this change (a consequence of using psyntax as the
210 default expander), and may choose to revisit this situation before 2.0
211 in response to user feedback.
213 ** Defmacros must now produce valid Scheme expressions.
215 It used to be that defmacros could unquote in Scheme values, as a way of
216 supporting partial evaluation, and avoiding some hygiene issues. For
219 (define (helper x) ...)
220 (define-macro (foo bar)
223 Assuming this macro is in the `(baz)' module, the direct translation of
226 (define (helper x) ...)
227 (define-macro (foo bar)
228 `((@@ (baz) helper) ,bar))
230 Of course, one could just use a hygienic macro instead:
234 ((_ bar) (helper bar))))
236 ** Guile's psyntax now supports docstrings and internal definitions.
238 The following Scheme is not strictly legal:
245 However its intent is fairly clear. Guile interprets "bar" to be the
246 docstring of `foo', and the definition of `baz' is still in definition
249 ** Macros need to be defined before their first use.
251 It used to be that with lazy memoization, this might work:
255 (define-macro (ref x) x)
258 But now, the body of `foo' is interpreted to mean a call to the toplevel
259 `ref' function, instead of a macro expansion. The solution is to define
260 macros before code that uses them.
262 ** Functions needed by macros at expand-time need to be present at
265 For example, this code will work at the REPL:
267 (define (double-helper x) (* x x))
268 (define-macro (double-literal x) (double-helper x))
269 (double-literal 2) => 4
271 But it will not work when a file is compiled, because the definition of
272 `double-helper' is not present at expand-time. The solution is to wrap
273 the definition of `double-helper' in `eval-when':
275 (eval-when (load compile eval)
276 (define (double-helper x) (* x x)))
277 (define-macro (double-literal x) (double-helper x))
278 (double-literal 2) => 4
280 See the (currently missing) documentation for eval-when for more
283 ** New variable, %pre-modules-transformer
285 Need to document this one some more.
287 ** Temporarily removed functions: `macroexpand', `macroexpand-1'
289 `macroexpand' will be added back before 2.0. It is unclear how to
290 implement `macroexpand-1' with syntax-case, though PLT Scheme does prove
293 ** New reader macros: #' #` #, #,@
295 These macros translate, respectively, to `syntax', `quasisyntax',
296 `unsyntax', and `unsyntax-splicing'. See the R6RS for more information.
297 These reader macros may be overridden by `read-hash-extend'.
299 ** Incompatible change to #'
301 Guile did have a #' hash-extension, by default, which just returned the
302 subsequent datum: #'foo => foo. In the unlikely event that anyone
303 actually used this, this behavior may be reinstated via the
304 `read-hash-extend' mechanism.
306 ** Scheme expresssions may be commented out with #;
308 #; comments out an entire expression. See SRFI-62 or the R6RS for more
311 ** `make-stack' with a tail-called procedural narrowing argument no longer
312 works (with compiled procedures)
314 It used to be the case that a captured stack could be narrowed to select
315 calls only up to or from a certain procedure, even if that procedure
316 already tail-called another procedure. This was because the debug
317 information from the original procedure was kept on the stack.
319 Now with the new compiler, the stack only contains active frames from
320 the current continuation. A narrow to a procedure that is not in the
321 stack will result in an empty stack. To fix this, narrow to a procedure
322 that is active in the current continuation, or narrow to a specific
323 number of stack frames.
325 ** backtraces through compiled procedures only show procedures that are
326 active in the current continuation
328 Similarly to the previous issue, backtraces in compiled code may be
329 different from backtraces in interpreted code. There are no semantic
330 differences, however. Please mail bug-guile@gnu.org if you see any
331 deficiencies with Guile's backtraces.
333 ** syntax-rules and syntax-case macros now propagate source information
334 through to the expanded code
336 This should result in better backtraces.
338 ** The currying behavior of `define' has been removed.
340 Before, `(define ((f a) b) (* a b))' would translate to
342 (define f (lambda (a) (lambda (b) (* a b))))
344 Now a syntax error is signaled, as this syntax is not supported by
345 default. If there is sufficient demand, this syntax can be supported
348 ** All modules have names now
350 Before, you could have anonymous modules: modules without names. Now,
351 because of hygiene and macros, all modules have names. If a module was
352 created without a name, the first time `module-name' is called on it, a
353 fresh name will be lazily generated for it.
355 ** Many syntax errors have different texts now
357 Syntax errors still throw to the `syntax-error' key, but the arguments
358 are often different now. Perhaps in the future, Guile will switch to
359 using standard SRFI-35 conditions.
361 ** Returning multiple values to compiled code will silently truncate the
362 values to the expected number
364 For example, the interpreter would raise an error evaluating the form,
365 `(+ (values 1 2) (values 3 4))', because it would see the operands as
366 being two compound "values" objects, to which `+' does not apply.
368 The compiler, on the other hand, receives multiple values on the stack,
369 not as a compound object. Given that it must check the number of values
370 anyway, if too many values are provided for a continuation, it chooses
371 to truncate those values, effectively evaluating `(+ 1 3)' instead.
373 The idea is that the semantics that the compiler implements is more
374 intuitive, and the use of the interpreter will fade out with time.
375 This behavior is allowed both by the R5RS and the R6RS.
377 ** Multiple values in compiled code are not represented by compound
380 This change may manifest itself in the following situation:
382 (let ((val (foo))) (do-something) val)
384 In the interpreter, if `foo' returns multiple values, multiple values
385 are produced from the `let' expression. In the compiler, those values
386 are truncated to the first value, and that first value is returned. In
387 the compiler, if `foo' returns no values, an error will be raised, while
388 the interpreter would proceed.
390 Both of these behaviors are allowed by R5RS and R6RS. The compiler's
391 behavior is more correct, however. If you wish to preserve a potentially
392 multiply-valued return, you will need to set up a multiple-value
393 continuation, using `call-with-values'.
395 ** Defmacros are now implemented in terms of syntax-case.
397 The practical ramification of this is that the `defmacro?' predicate has
398 been removed, along with `defmacro-transformer', `macro-table',
399 `xformer-table', `assert-defmacro?!', `set-defmacro-transformer!' and
400 `defmacro:transformer'. This is because defmacros are simply macros. If
401 any of these procedures provided useful facilities to you, we encourage
402 you to contact the Guile developers.
404 ** psyntax is now the default expander
406 Scheme code is now expanded by default by the psyntax hygienic macro
407 expander. Expansion is performed completely before compilation or
410 Notably, syntax errors will be signalled before interpretation begins.
411 In the past, many syntax errors were only detected at runtime if the
412 code in question was memoized.
414 As part of its expansion, psyntax renames all lexically-bound
415 identifiers. Original identifier names are preserved and given to the
416 compiler, but the interpreter will see the renamed variables, e.g.,
417 `x432' instead of `x'.
419 Note that the psyntax that Guile uses is a fork, as Guile already had
420 modules before incompatible modules were added to psyntax -- about 10
421 years ago! Thus there are surely a number of bugs that have been fixed
422 in psyntax since then. If you find one, please notify bug-guile@gnu.org.
424 ** syntax-rules and syntax-case are available by default.
426 There is no longer any need to import the `(ice-9 syncase)' module
427 (which is now deprecated). The expander may be invoked directly via
428 `sc-expand', though it is normally searched for via the current module
431 Also, the helper routines for syntax-case are available in the default
432 environment as well: `syntax->datum', `datum->syntax',
433 `bound-identifier=?', `free-identifier=?', `generate-temporaries',
434 `identifier?', and `syntax-violation'. See the R6RS for documentation.
436 ** Lexical bindings introduced by hygienic macros may not be referenced
437 by nonhygienic macros.
439 If a lexical binding is introduced by a hygienic macro, it may not be
440 referenced by a nonhygienic macro. For example, this works:
443 (define-macro (bind-x val body)
444 `(let ((x ,val)) ,body))
445 (define-macro (ref x)
452 (define-syntax bind-x
454 ((_ val body) (let ((x val)) body))))
455 (define-macro (ref x)
459 It is not normal to run into this situation with existing code. However,
460 as code is ported over from defmacros to syntax-case, it is possible to
461 run into situations like this. In the future, Guile will probably port
462 its `while' macro to syntax-case, which makes this issue one to know
465 ** Macros may no longer be referenced as first-class values.
467 In the past, you could evaluate e.g. `if', and get its macro value. Now,
468 expanding this form raises a syntax error.
470 Macros still /exist/ as first-class values, but they must be
471 /referenced/ via the module system, e.g. `(module-ref (current-module)
474 This decision may be revisited before the 2.0 release. Feedback welcome
475 to guile-devel@gnu.org (subscription required) or bug-guile@gnu.org (no
476 subscription required).
478 ** New macro type: syncase-macro
480 XXX Need to decide whether to document this for 2.0, probably should:
481 make-syncase-macro, make-extended-syncase-macro, macro-type,
482 syncase-macro-type, syncase-macro-binding
484 ** A new `memoize-symbol' evaluator trap has been added.
486 This trap can be used for efficiently implementing a Scheme code
489 ** Duplicate bindings among used modules are resolved lazily.
491 This slightly improves program startup times.
493 ** New thread cancellation and thread cleanup API
495 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
497 ** Fix bad interaction between `false-if-exception' and stack-call.
499 Exceptions thrown by `false-if-exception' were erronously causing the
500 stack to be saved, causing later errors to show the incorrectly-saved
501 backtrace. This has been fixed.
503 ** New global variables: %load-compiled-path, %load-compiled-extensions
505 These are analogous to %load-path and %load-extensions.
507 ** New procedure, `make-promise'
509 `(make-promise (lambda () foo))' is equivalent to `(delay foo)'.
511 ** New entry into %guile-build-info: `ccachedir'
513 ** Fix bug in `module-bound?'.
515 `module-bound?' was returning true if a module did have a local
516 variable, but one that was unbound, but another imported module bound
517 the variable. This was an error, and was fixed.
519 ** `(ice-9 syncase)' has been deprecated.
521 As syntax-case is available by default, importing `(ice-9 syncase)' has
522 no effect, and will trigger a deprecation warning.
524 ** Removed deprecated uniform array procedures:
525 dimensions->uniform-array, list->uniform-array, array-prototype
527 Instead, use make-typed-array, list->typed-array, or array-type,
530 * Changes to the C interface
532 ** The GH interface (deprecated in version 1.6, 2001) was removed.
534 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
536 This makes these internal functions technically not callable from
539 ** Functions for handling `scm_option' now no longer require an argument
540 indicating length of the `scm_t_option' array.
542 ** scm_primitive_load_path has additional argument, exception_on_error
544 ** New C function: scm_module_public_interface
546 This procedure corresponds to Scheme's `module-public-interface'.
548 ** `scm_stat' has an additional argument, `exception_on_error'
549 ** `scm_primitive_load_path' has an additional argument `exception_on_not_found'
551 ** `scm_set_port_seek' and `scm_set_port_truncate' use the `scm_t_off' type
553 Previously they would use the `off_t' type, which is fragile since its
554 definition depends on the application's value for `_FILE_OFFSET_BITS'.
556 ** The `long_long' C type, deprecated in 1.8, has been removed
558 ** Removed deprecated uniform array procedures: scm_make_uve,
559 scm_array_prototype, scm_list_to_uniform_array,
560 scm_dimensions_to_uniform_array, scm_make_ra, scm_shap2ra, scm_cvref,
561 scm_ra_set_contp, scm_aind, scm_raprin1
563 These functions have been deprecated since early 2005.
565 ** scm_array_p has one argument, not two
567 Use of the second argument produced a deprecation warning, so it is
568 unlikely that any code out there actually used this functionality.
570 * Changes to the distribution
572 ** Guile's license is now LGPLv3+
574 In other words the GNU Lesser General Public License, version 3 or
575 later (at the discretion of each person that chooses to redistribute
578 ** `guile-config' will be deprecated in favor of `pkg-config'
580 `guile-config' has been rewritten to get its information from
581 `pkg-config', so this should be a transparent change. Note however that
582 guile.m4 has yet to be modified to call pkg-config instead of
585 ** New installation directory: $(pkglibdir)/1.9/ccache
587 If $(libdir) is /usr/lib, for example, Guile will install its .go files
588 to /usr/lib/guile/1.9/ccache. These files are architecture-specific.
590 ** New dependency: GNU libunistring.
592 See http://www.gnu.org/software/libunistring/. We hope to merge in
593 Unicode support in the next prerelease.
597 Changes in 1.8.7 (since 1.8.6)
599 * New modules (see the manual for details)
601 ** `(srfi srfi-98)', an interface to access environment variables
605 ** Fix compilation with `--disable-deprecated'
606 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
607 ** Fix MinGW build problem caused by HAVE_STRUCT_TIMESPEC confusion
608 ** Fix build problem when scm_t_timespec is different from struct timespec
609 ** Fix build when compiled with -Wundef -Werror
610 ** More build fixes for `alphaev56-dec-osf5.1b' (Tru64)
611 ** Build fixes for `powerpc-ibm-aix5.3.0.0' (AIX 5.3)
612 ** With GCC, always compile with `-mieee' on `alpha*' and `sh*'
613 ** Better diagnose broken `(strftime "%z" ...)' in `time.test' (bug #24130)
614 ** Fix parsing of SRFI-88/postfix keywords longer than 128 characters
615 ** Fix reading of complex numbers where both parts are inexact decimals
617 ** Allow @ macro to work with (ice-9 syncase)
619 Previously, use of the @ macro in a module whose code is being
620 transformed by (ice-9 syncase) would cause an "Invalid syntax" error.
621 Now it works as you would expect (giving the value of the specified
624 ** Have `scm_take_locale_symbol ()' return an interned symbol (bug #25865)
627 Changes in 1.8.6 (since 1.8.5)
629 * New features (see the manual for details)
631 ** New convenience function `scm_c_symbol_length ()'
633 ** Single stepping through code from Emacs
635 When you use GDS to evaluate Scheme code from Emacs, you can now use
636 `C-u' to indicate that you want to single step through that code. See
637 `Evaluating Scheme Code' in the manual for more details.
639 ** New "guile(1)" man page!
641 * Changes to the distribution
643 ** Automake's `AM_MAINTAINER_MODE' is no longer used
645 Thus, the `--enable-maintainer-mode' configure option is no longer
646 available: Guile is now always configured in "maintainer mode".
648 ** `ChangeLog' files are no longer updated
650 Instead, changes are detailed in the version control system's logs. See
651 the top-level `ChangeLog' files for details.
656 ** `symbol->string' now returns a read-only string, as per R5RS
657 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
658 ** `guile-config link' now prints `-L$libdir' before `-lguile'
659 ** Fix memory corruption involving GOOPS' `class-redefinition'
660 ** Fix possible deadlock in `mutex-lock'
661 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
662 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
663 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
664 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
665 ** Fix misleading output from `(help rationalize)'
666 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
667 ** Fix `gcd' when called with a single, negative argument.
668 ** Fix `Stack overflow' errors seen when building on some platforms
669 ** Fix bug when `scm_with_guile ()' was called several times from the
671 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
672 dynamic environment of the call to `raise'
673 ** Fix potential deadlock in `make-struct'
674 ** Fix compilation problem with libltdl from Libtool 2.2.x
675 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
678 Changes in 1.8.5 (since 1.8.4)
680 * Infrastructure changes
682 ** Guile repository switched from CVS to Git
684 The new repository can be accessed using
685 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
686 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
688 ** Add support for `pkg-config'
690 See "Autoconf Support" in the manual for details.
692 * New modules (see the manual for details)
696 * New features (see the manual for details)
698 ** New `postfix' read option, for SRFI-88 keyword syntax
699 ** Some I/O primitives have been inlined, which improves I/O performance
700 ** New object-based traps infrastructure
702 This is a GOOPS-based infrastructure that builds on Guile's low-level
703 evaluator trap calls and facilitates the development of debugging
704 features like single-stepping, breakpoints, tracing and profiling.
705 See the `Traps' node of the manual for details.
707 ** New support for working on Guile code from within Emacs
709 Guile now incorporates the `GDS' library (previously distributed
710 separately) for working on Guile code from within Emacs. See the
711 `Using Guile In Emacs' node of the manual for details.
715 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
716 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
718 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
719 would trigger an unbound variable error for `match:andmap'.
721 ** `(oop goops describe)' now properly provides the `describe' feature
722 ** Fixed `args-fold' from `(srfi srfi-37)'
724 Previously, parsing short option names of argument-less options would
725 lead to a stack overflow.
727 ** `(srfi srfi-35)' is now visible through `cond-expand'
728 ** Fixed type-checking for the second argument of `eval'
729 ** Fixed type-checking for SRFI-1 `partition'
730 ** Fixed `struct-ref' and `struct-set!' on "light structs"
731 ** Honor struct field access rights in GOOPS
732 ** Changed the storage strategy of source properties, which fixes a deadlock
733 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
734 ** Fixed build issue for GNU/Linux on IA64
735 ** Fixed build issues on NetBSD 1.6
736 ** Fixed build issue on Solaris 2.10 x86_64
737 ** Fixed build issue with DEC/Compaq/HP's compiler
738 ** Fixed `scm_from_complex_double' build issue on FreeBSD
739 ** Fixed `alloca' build issue on FreeBSD 6
740 ** Removed use of non-portable makefile constructs
741 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
742 ** Make sure all tests honor `$TMPDIR'
745 Changes in 1.8.4 (since 1.8.3)
749 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
750 ** Fixed a segmentation fault which occurred when displaying the
751 backtrace of a stack with a promise object (made by `delay') in it.
752 ** Make `accept' leave guile mode while blocking
753 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
754 ** Fixed a build problem on AIX (use of func_data identifier)
755 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
756 called with an associator proc that returns neither a pair nor #f.
757 ** Secondary threads now always return a valid module for (current-module).
758 ** Avoid MacOS build problems caused by incorrect combination of "64"
759 system and library calls.
760 ** `guile-snarf' now honors `$TMPDIR'
761 ** `guile-config compile' now reports CPPFLAGS used at compile-time
762 ** Fixed build with Sun Studio (Solaris 9)
763 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
764 uniform vectors on AIX.
765 ** Fixed a deadlock that occurs upon GC with multiple threads.
766 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
767 ** Fixed autotool-derived build problems on AIX 6.1.
768 ** Fixed NetBSD/alpha support
769 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
771 * New modules (see the manual for details)
775 * Documentation fixes and improvements
777 ** Removed premature breakpoint documentation
779 The features described are not available in the series of 1.8.x
780 releases, so the documentation was misleading and has been removed.
782 ** More about Guile's default *random-state* variable
784 ** GOOPS: more about how to use `next-method'
786 * Changes to the distribution
788 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
790 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
791 General Public License, and the few incorrect files have now been
792 fixed to agree with the rest of the Guile distribution.
794 ** Removed unnecessary extra copies of COPYING*
796 The distribution now contains a single COPYING.LESSER at its top level.
799 Changes in 1.8.3 (since 1.8.2)
801 * New modules (see the manual for details)
808 ** The `(ice-9 slib)' module now works as expected
809 ** Expressions like "(set! 'x #t)" no longer yield a crash
810 ** Warnings about duplicate bindings now go to stderr
811 ** A memory leak in `make-socket-address' was fixed
812 ** Alignment issues (e.g., on SPARC) in network routines were fixed
813 ** A threading issue that showed up at least on NetBSD was fixed
814 ** Build problems on Solaris and IRIX fixed
816 * Implementation improvements
818 ** The reader is now faster, which reduces startup time
819 ** Procedures returned by `record-accessor' and `record-modifier' are faster
822 Changes in 1.8.2 (since 1.8.1):
824 * New procedures (see the manual for details)
826 ** set-program-arguments
829 * Incompatible changes
831 ** The body of a top-level `define' no longer sees the binding being created
833 In a top-level `define', the binding being created is no longer visible
834 from the `define' body. This breaks code like
835 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
836 unbound in the body. However, such code was not R5RS-compliant anyway,
841 ** Fractions were not `equal?' if stored in unreduced form.
842 (A subtle problem, since printing a value reduced it, making it work.)
843 ** srfi-60 `copy-bit' failed on 64-bit systems
844 ** "guile --use-srfi" option at the REPL can replace core functions
845 (Programs run with that option were ok, but in the interactive REPL
846 the core bindings got priority, preventing SRFI replacements or
848 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
849 ** `kill' on mingw throws an error for a PID other than oneself
850 ** Procedure names are attached to procedure-with-setters
851 ** Array read syntax works with negative lower bound
852 ** `array-in-bounds?' fix if an array has different lower bounds on each index
853 ** `*' returns exact 0 for "(* inexact 0)"
854 This follows what it always did for "(* 0 inexact)".
855 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
856 ** SRFI-19: `date->julian-day' did not account for timezone offset
857 ** `ttyname' no longer crashes when passed a non-tty argument
858 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
859 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
860 ** GOOPS: Fixed a bug in `method-more-specific?'
861 ** Build problems on Solaris fixed
862 ** Build problems on HP-UX IA64 fixed
863 ** Build problems on MinGW fixed
866 Changes in 1.8.1 (since 1.8.0):
868 * LFS functions are now used to access 64-bit files on 32-bit systems.
870 * New procedures (see the manual for details)
872 ** primitive-_exit - [Scheme] the-root-module
873 ** scm_primitive__exit - [C]
874 ** make-completion-function - [Scheme] (ice-9 readline)
875 ** scm_c_locale_stringn_to_number - [C]
876 ** scm_srfi1_append_reverse [C]
877 ** scm_srfi1_append_reverse_x [C]
885 ** Build problems have been fixed on MacOS, SunOS, and QNX.
887 ** `strftime' fix sign of %z timezone offset.
889 ** A one-dimensional array can now be 'equal?' to a vector.
891 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
893 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
895 ** `record-accessor' and `record-modifier' now have strict type checks.
897 Record accessor and modifier procedures now throw an error if the
898 record type of the record they're given is not the type expected.
899 (Previously accessors returned #f and modifiers silently did nothing).
901 ** It is now OK to use both autoload and use-modules on a given module.
903 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
905 Previously there was no checking on primatives like make-vector that
906 accept "one or two" arguments. Now there is.
908 ** The srfi-1 assoc function now calls its equality predicate properly.
910 Previously srfi-1 assoc would call the equality predicate with the key
911 last. According to the SRFI, the key should be first.
913 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
915 ** The array-set! procedure no longer segfaults when given a bit vector.
917 ** Bugs in make-shared-array have been fixed.
919 ** string<? and friends now follow char<? etc order on 8-bit chars.
921 ** The format procedure now handles inf and nan values for ~f correctly.
923 ** exact->inexact should no longer overflow when given certain large fractions.
925 ** srfi-9 accessor and modifier procedures now have strict record type checks.
927 This matches the srfi-9 specification.
929 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
931 Previously the (ice-9 ftw) procedures would ignore any file that had
932 the same inode number as a file they had already seen, even if that
933 file was on a different device.
936 Changes in 1.8.0 (changes since the 1.6.x series):
938 * Changes to the distribution
940 ** Guile is now licensed with the GNU Lesser General Public License.
942 ** The manual is now licensed with the GNU Free Documentation License.
944 ** Guile now requires GNU MP (http://swox.com/gmp).
946 Guile now uses the GNU MP library for arbitrary precision arithmetic.
948 ** Guile now has separate private and public configuration headers.
950 That is, things like HAVE_STRING_H no longer leak from Guile's
953 ** Guile now provides and uses an "effective" version number.
955 Guile now provides scm_effective_version and effective-version
956 functions which return the "effective" version number. This is just
957 the normal full version string without the final micro-version number,
958 so the current effective-version is "1.8". The effective version
959 should remain unchanged during a stable series, and should be used for
960 items like the versioned share directory name
961 i.e. /usr/share/guile/1.8.
963 Providing an unchanging version number during a stable release for
964 things like the versioned share directory can be particularly
965 important for Guile "add-on" packages, since it provides a directory
966 that they can install to that won't be changed out from under them
967 with each micro release during a stable series.
969 ** Thread implementation has changed.
971 When you configure "--with-threads=null", you will get the usual
972 threading API (call-with-new-thread, make-mutex, etc), but you can't
973 actually create new threads. Also, "--with-threads=no" is now
974 equivalent to "--with-threads=null". This means that the thread API
975 is always present, although you might not be able to create new
978 When you configure "--with-threads=pthreads" or "--with-threads=yes",
979 you will get threads that are implemented with the portable POSIX
980 threads. These threads can run concurrently (unlike the previous
981 "coop" thread implementation), but need to cooperate for things like
984 The default is "pthreads", unless your platform doesn't have pthreads,
985 in which case "null" threads are used.
987 See the manual for details, nodes "Initialization", "Multi-Threading",
988 "Blocking", and others.
990 ** There is the new notion of 'discouraged' features.
992 This is a milder form of deprecation.
994 Things that are discouraged should not be used in new code, but it is
995 OK to leave them in old code for now. When a discouraged feature is
996 used, no warning message is printed like there is for 'deprecated'
997 features. Also, things that are merely discouraged are nevertheless
998 implemented efficiently, while deprecated features can be very slow.
1000 You can omit discouraged features from libguile by configuring it with
1001 the '--disable-discouraged' option.
1003 ** Deprecation warnings can be controlled at run-time.
1005 (debug-enable 'warn-deprecated) switches them on and (debug-disable
1006 'warn-deprecated) switches them off.
1008 ** Support for SRFI 61, extended cond syntax for multiple values has
1011 This SRFI is always available.
1013 ** Support for require-extension, SRFI-55, has been added.
1015 The SRFI-55 special form `require-extension' has been added. It is
1016 available at startup, and provides a portable way to load Scheme
1017 extensions. SRFI-55 only requires support for one type of extension,
1018 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
1021 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
1023 The (srfi srfi-26) module is an implementation of SRFI-26 which
1024 provides the `cut' and `cute' syntax. These may be used to specialize
1025 parameters without currying.
1027 ** New module (srfi srfi-31)
1029 This is an implementation of SRFI-31 which provides a special form
1030 `rec' for recursive evaluation.
1032 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
1033 been merged with the core, making their functionality always
1036 The modules are still available, tho, and you could use them together
1037 with a renaming import, for example.
1039 ** Guile no longer includes its own version of libltdl.
1041 The official version is good enough now.
1043 ** The --enable-htmldoc option has been removed from 'configure'.
1045 Support for translating the documentation into HTML is now always
1046 provided. Use 'make html'.
1048 ** New module (ice-9 serialize):
1050 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
1051 don't trust the thread safety of most of your program, but where you
1052 have some section(s) of code which you consider can run in parallel to
1053 other sections. See ice-9/serialize.scm for more information.
1055 ** The configure option '--disable-arrays' has been removed.
1057 Support for arrays and uniform numeric arrays is now always included
1060 * Changes to the stand-alone interpreter
1062 ** New command line option `-L'.
1064 This option adds a directory to the front of the load path.
1066 ** New command line option `--no-debug'.
1068 Specifying `--no-debug' on the command line will keep the debugging
1069 evaluator turned off, even for interactive sessions.
1071 ** User-init file ~/.guile is now loaded with the debugging evaluator.
1073 Previously, the normal evaluator would have been used. Using the
1074 debugging evaluator gives better error messages.
1076 ** The '-e' option now 'read's its argument.
1078 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
1079 be used with '-e'. For example, you can now write a script like
1082 exec guile -e '(@ (demo) main)' -s "$0" "$@"
1085 (define-module (demo)
1089 (format #t "Demo: ~a~%" args))
1092 * Changes to Scheme functions and syntax
1094 ** Guardians have changed back to their original semantics
1096 Guardians now behave like described in the paper by Dybvig et al. In
1097 particular, they no longer make guarantees about the order in which
1098 they return objects, and they can no longer be greedy.
1100 They no longer drop cyclic data structures.
1102 The C function scm_make_guardian has been changed incompatibly and no
1103 longer takes the 'greedy_p' argument.
1105 ** New function hashx-remove!
1107 This function completes the set of 'hashx' functions.
1109 ** The concept of dynamic roots has been factored into continuation
1110 barriers and dynamic states.
1112 Each thread has a current dynamic state that carries the values of the
1113 fluids. You can create and copy dynamic states and use them as the
1114 second argument for 'eval'. See "Fluids and Dynamic States" in the
1117 To restrict the influence that captured continuations can have on the
1118 control flow, you can errect continuation barriers. See "Continuation
1119 Barriers" in the manual.
1121 The function call-with-dynamic-root now essentially temporarily
1122 installs a new dynamic state and errects a continuation barrier.
1124 ** The default load path no longer includes "." at the end.
1126 Automatically loading modules from the current directory should not
1127 happen by default. If you want to allow it in a more controlled
1128 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
1129 variable %load-path.
1131 ** The uniform vector and array support has been overhauled.
1133 It now complies with SRFI-4 and the weird prototype based uniform
1134 array creation has been deprecated. See the manual for more details.
1136 Some non-compatible changes have been made:
1137 - characters can no longer be stored into byte arrays.
1138 - strings and bit vectors are no longer considered to be uniform numeric
1140 - array-rank throws an error for non-arrays instead of returning zero.
1141 - array-ref does no longer accept non-arrays when no indices are given.
1143 There is the new notion of 'generalized vectors' and corresponding
1144 procedures like 'generalized-vector-ref'. Generalized vectors include
1145 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
1147 Arrays use generalized vectors as their storage, so that you still
1148 have arrays of characters, bits, etc. However, uniform-array-read!
1149 and uniform-array-write can no longer read/write strings and
1152 ** There is now support for copy-on-write substrings, mutation-sharing
1153 substrings and read-only strings.
1155 Three new procedures are related to this: substring/shared,
1156 substring/copy, and substring/read-only. See the manual for more
1159 ** Backtraces will now highlight the value that caused the error.
1161 By default, these values are enclosed in "{...}", such as in this
1170 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
1171 <unnamed port>:1:1: Wrong type (expecting pair): a
1172 ABORT: (wrong-type-arg)
1174 The prefix and suffix used for highlighting can be set via the two new
1175 printer options 'highlight-prefix' and 'highlight-suffix'. For
1176 example, putting this into ~/.guile will output the bad value in bold
1177 on an ANSI terminal:
1179 (print-set! highlight-prefix "\x1b[1m")
1180 (print-set! highlight-suffix "\x1b[22m")
1183 ** 'gettext' support for internationalization has been added.
1185 See the manual for details.
1187 ** New syntax '@' and '@@':
1189 You can now directly refer to variables exported from a module by
1192 (@ MODULE-NAME VARIABLE-NAME)
1194 For example (@ (ice-9 pretty-print) pretty-print) will directly access
1195 the pretty-print variable exported from the (ice-9 pretty-print)
1196 module. You don't need to 'use' that module first. You can also use
1197 '@' as a target of 'set!', as in (set! (@ mod var) val).
1199 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
1200 but it can also access variables that have not been exported. It is
1201 intended only for kluges and temporary fixes and for debugging, not
1204 ** Keyword syntax has been made more disciplined.
1206 Previously, the name of a keyword was read as a 'token' but printed as
1207 a symbol. Now, it is read as a general Scheme datum which must be a
1218 ERROR: In expression (a b c):
1222 ERROR: Unbound variable: foo
1227 ERROR: Wrong type (expecting symbol): 12
1231 ERROR: Wrong type (expecting symbol): (a b c)
1235 ** The printing of symbols that might look like keywords can be
1238 The new printer option 'quote-keywordish-symbols' controls how symbols
1239 are printed that have a colon as their first or last character. The
1240 default now is to only quote a symbol with #{...}# when the read
1241 option 'keywords' is not '#f'. Thus:
1243 guile> (define foo (string->symbol ":foo"))
1244 guile> (read-set! keywords #f)
1247 guile> (read-set! keywords 'prefix)
1250 guile> (print-set! quote-keywordish-symbols #f)
1254 ** 'while' now provides 'break' and 'continue'
1256 break and continue were previously bound in a while loop, but not
1257 documented, and continue didn't quite work properly. The undocumented
1258 parameter to break which gave a return value for the while has been
1261 ** 'call-with-current-continuation' is now also available under the name
1264 ** The module system now checks for duplicate bindings.
1266 The module system now can check for name conflicts among imported
1269 The behavior can be controlled by specifying one or more 'duplicates'
1270 handlers. For example, to make Guile return an error for every name
1273 (define-module (foo)
1278 The new default behavior of the module system when a name collision
1279 has been detected is to
1281 1. Give priority to bindings marked as a replacement.
1282 2. Issue a warning (different warning if overriding core binding).
1283 3. Give priority to the last encountered binding (this corresponds to
1286 If you want the old behavior back without replacements or warnings you
1289 (default-duplicate-binding-handler 'last)
1291 to your .guile init file.
1293 ** New define-module option: :replace
1295 :replace works as :export, but, in addition, marks the binding as a
1298 A typical example is `format' in (ice-9 format) which is a replacement
1299 for the core binding `format'.
1301 ** Adding prefixes to imported bindings in the module system
1303 There is now a new :use-module option :prefix. It can be used to add
1304 a prefix to all imported bindings.
1306 (define-module (foo)
1307 :use-module ((bar) :prefix bar:))
1309 will import all bindings exported from bar, but rename them by adding
1312 ** Conflicting generic functions can be automatically merged.
1314 When two imported bindings conflict and they are both generic
1315 functions, the two functions can now be merged automatically. This is
1316 activated with the 'duplicates' handler 'merge-generics'.
1318 ** New function: effective-version
1320 Returns the "effective" version number. This is just the normal full
1321 version string without the final micro-version number. See "Changes
1322 to the distribution" above.
1324 ** New threading functions: parallel, letpar, par-map, and friends
1326 These are convenient ways to run calculations in parallel in new
1327 threads. See "Parallel forms" in the manual for details.
1329 ** New function 'try-mutex'.
1331 This function will attempt to lock a mutex but will return immediately
1332 instead of blocking and indicate failure.
1334 ** Waiting on a condition variable can have a timeout.
1336 The function 'wait-condition-variable' now takes a third, optional
1337 argument that specifies the point in time where the waiting should be
1340 ** New function 'broadcast-condition-variable'.
1342 ** New functions 'all-threads' and 'current-thread'.
1344 ** Signals and system asyncs work better with threads.
1346 The function 'sigaction' now takes a fourth, optional, argument that
1347 specifies the thread that the handler should run in. When the
1348 argument is omitted, the handler will run in the thread that called
1351 Likewise, 'system-async-mark' takes a second, optional, argument that
1352 specifies the thread that the async should run in. When it is
1353 omitted, the async will run in the thread that called
1354 'system-async-mark'.
1356 C code can use the new functions scm_sigaction_for_thread and
1357 scm_system_async_mark_for_thread to pass the new thread argument.
1359 When a thread blocks on a mutex, a condition variable or is waiting
1360 for IO to be possible, it will still execute system asyncs. This can
1361 be used to interrupt such a thread by making it execute a 'throw', for
1364 ** The function 'system-async' is deprecated.
1366 You can now pass any zero-argument procedure to 'system-async-mark'.
1367 The function 'system-async' will just return its argument unchanged
1370 ** New functions 'call-with-blocked-asyncs' and
1371 'call-with-unblocked-asyncs'
1373 The expression (call-with-blocked-asyncs PROC) will call PROC and will
1374 block execution of system asyncs for the current thread by one level
1375 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
1376 procedure and will unblock the execution of system asyncs by one
1377 level for the current thread.
1379 Only system asyncs are affected by these functions.
1381 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
1383 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
1384 instead. Those functions are easier to use correctly and can be
1387 ** New function 'unsetenv'.
1389 ** New macro 'define-syntax-public'.
1391 It works like 'define-syntax' and also exports the defined macro (but
1394 ** There is support for Infinity and NaNs.
1396 Following PLT Scheme, Guile can now work with infinite numbers, and
1399 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
1400 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
1401 "+nan.0"). These numbers are inexact and have no exact counterpart.
1403 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
1404 sign of the dividend. The infinities are integers, and they answer #t
1405 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
1406 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
1417 ERROR: Numerical overflow
1419 Two new predicates 'inf?' and 'nan?' can be used to test for the
1422 ** Inexact zero can have a sign.
1424 Guile can now distinguish between plus and minus inexact zero, if your
1425 platform supports this, too. The two zeros are equal according to
1426 '=', but not according to 'eqv?'. For example
1437 ** Guile now has exact rationals.
1439 Guile can now represent fractions such as 1/3 exactly. Computing with
1440 them is also done exactly, of course:
1445 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
1446 for exact arguments.
1448 For example: (floor 2) now returns an exact 2 where in the past it
1449 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
1451 ** inexact->exact no longer returns only integers.
1453 Without exact rationals, the closest exact number was always an
1454 integer, but now inexact->exact returns the fraction that is exactly
1455 equal to a floating point number. For example:
1457 (inexact->exact 1.234)
1458 => 694680242521899/562949953421312
1460 When you want the old behavior, use 'round' explicitly:
1462 (inexact->exact (round 1.234))
1465 ** New function 'rationalize'.
1467 This function finds a simple fraction that is close to a given real
1468 number. For example (and compare with inexact->exact above):
1470 (rationalize (inexact->exact 1.234) 1/2000)
1473 Note that, as required by R5RS, rationalize returns only then an exact
1474 result when both its arguments are exact.
1476 ** 'odd?' and 'even?' work also for inexact integers.
1478 Previously, (odd? 1.0) would signal an error since only exact integers
1479 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
1480 returns #f and (odd? 1.5) signals an error.
1482 ** Guile now has uninterned symbols.
1484 The new function 'make-symbol' will return an uninterned symbol. This
1485 is a symbol that is unique and is guaranteed to remain unique.
1486 However, uninterned symbols can not yet be read back in.
1488 Use the new function 'symbol-interned?' to check whether a symbol is
1491 ** pretty-print has more options.
1493 The function pretty-print from the (ice-9 pretty-print) module can now
1494 also be invoked with keyword arguments that control things like
1495 maximum output width. See the manual for details.
1497 ** Variables have no longer a special behavior for `equal?'.
1499 Previously, comparing two variables with `equal?' would recursivly
1500 compare their values. This is no longer done. Variables are now only
1501 `equal?' if they are `eq?'.
1503 ** `(begin)' is now valid.
1505 You can now use an empty `begin' form. It will yield #<unspecified>
1506 when evaluated and simply be ignored in a definition context.
1508 ** Deprecated: procedure->macro
1510 Change your code to use 'define-macro' or r5rs macros. Also, be aware
1511 that macro expansion will not be done during evaluation, but prior to
1514 ** Soft ports now allow a `char-ready?' procedure
1516 The vector argument to `make-soft-port' can now have a length of
1517 either 5 or 6. (Previously the length had to be 5.) The optional 6th
1518 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
1519 that returns the number of characters that can be read immediately
1520 without the soft port blocking.
1522 ** Deprecated: undefine
1524 There is no replacement for undefine.
1526 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
1527 have been discouraged.
1529 They are relics from a time where a keyword like #:foo was used
1530 directly as a Tcl option "-foo" and thus keywords were internally
1531 stored as a symbol with a starting dash. We now store a symbol
1534 Use symbol->keyword and keyword->symbol instead.
1536 ** The `cheap' debug option is now obsolete
1538 Evaluator trap calls are now unconditionally "cheap" - in other words,
1539 they pass a debug object to the trap handler rather than a full
1540 continuation. The trap handler code can capture a full continuation
1541 by using `call-with-current-continuation' in the usual way, if it so
1544 The `cheap' option is retained for now so as not to break existing
1545 code which gets or sets it, but setting it now has no effect. It will
1546 be removed in the next major Guile release.
1548 ** Evaluator trap calls now support `tweaking'
1550 `Tweaking' means that the trap handler code can modify the Scheme
1551 expression that is about to be evaluated (in the case of an
1552 enter-frame trap) or the value that is being returned (in the case of
1553 an exit-frame trap). The trap handler code indicates that it wants to
1554 do this by returning a pair whose car is the symbol 'instead and whose
1555 cdr is the modified expression or return value.
1557 * Changes to the C interface
1559 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
1560 take a 'delete' function argument.
1562 This argument makes no sense since the delete function is used to
1563 remove a pair from an alist, and this must not be configurable.
1565 This is an incompatible change.
1567 ** The GH interface is now subject to the deprecation mechanism
1569 The GH interface has been deprecated for quite some time but now it is
1570 actually removed from Guile when it is configured with
1571 --disable-deprecated.
1573 See the manual "Transitioning away from GH" for more information.
1575 ** A new family of functions for converting between C values and
1576 Scheme values has been added.
1578 These functions follow a common naming scheme and are designed to be
1579 easier to use, thread-safe and more future-proof than the older
1582 - int scm_is_* (...)
1584 These are predicates that return a C boolean: 1 or 0. Instead of
1585 SCM_NFALSEP, you can now use scm_is_true, for example.
1587 - <type> scm_to_<type> (SCM val, ...)
1589 These are functions that convert a Scheme value into an appropriate
1590 C value. For example, you can use scm_to_int to safely convert from
1593 - SCM scm_from_<type> (<type> val, ...)
1595 These functions convert from a C type to a SCM value; for example,
1596 scm_from_int for ints.
1598 There is a huge number of these functions, for numbers, strings,
1599 symbols, vectors, etc. They are documented in the reference manual in
1600 the API section together with the types that they apply to.
1602 ** New functions for dealing with complex numbers in C have been added.
1604 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1605 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1606 They work like scm_make_rectangular etc but take or return doubles
1609 ** The function scm_make_complex has been discouraged.
1611 Use scm_c_make_rectangular instead.
1613 ** The INUM macros have been deprecated.
1615 A lot of code uses these macros to do general integer conversions,
1616 although the macros only work correctly with fixnums. Use the
1617 following alternatives.
1619 SCM_INUMP -> scm_is_integer or similar
1620 SCM_NINUMP -> !scm_is_integer or similar
1621 SCM_MAKINUM -> scm_from_int or similar
1622 SCM_INUM -> scm_to_int or similar
1624 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1625 do the validating for you.
1627 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1628 have been discouraged.
1630 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1631 new code. The functions have been discouraged since they don't fit
1634 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1636 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1637 evaluates its argument twice. Use scm_is_true, etc. instead for new
1640 ** The macro SCM_EQ_P has been discouraged.
1642 Use scm_is_eq for new code, which fits better into the naming
1645 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1648 Use the function scm_is_pair or scm_is_null instead.
1650 ** The functions scm_round and scm_truncate have been deprecated and
1651 are now available as scm_c_round and scm_c_truncate, respectively.
1653 These functions occupy the names that scm_round_number and
1654 scm_truncate_number should have.
1656 ** The functions scm_c_string2str, scm_c_substring2str, and
1657 scm_c_symbol2str have been deprecated.
1659 Use scm_to_locale_stringbuf or similar instead, maybe together with
1662 ** New functions scm_c_make_string, scm_c_string_length,
1663 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1664 scm_c_substring_shared, scm_c_substring_copy.
1666 These are like scm_make_string, scm_length, etc. but are slightly
1667 easier to use from C.
1669 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1670 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1672 They export too many assumptions about the implementation of strings
1673 and symbols that are no longer true in the presence of
1674 mutation-sharing substrings and when Guile switches to some form of
1677 When working with strings, it is often best to use the normal string
1678 functions provided by Guile, such as scm_c_string_ref,
1679 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1680 manual since many more such functions are now provided than
1683 When you want to convert a SCM string to a C string, use the
1684 scm_to_locale_string function or similar instead. For symbols, use
1685 scm_symbol_to_string and then work with that string. Because of the
1686 new string representation, scm_symbol_to_string does not need to copy
1687 and is thus quite efficient.
1689 ** Some string, symbol and keyword functions have been discouraged.
1691 They don't fit into the uniform naming scheme and are not explicit
1692 about the character encoding.
1694 Replace according to the following table:
1696 scm_allocate_string -> scm_c_make_string
1697 scm_take_str -> scm_take_locale_stringn
1698 scm_take0str -> scm_take_locale_string
1699 scm_mem2string -> scm_from_locale_stringn
1700 scm_str2string -> scm_from_locale_string
1701 scm_makfrom0str -> scm_from_locale_string
1702 scm_mem2symbol -> scm_from_locale_symboln
1703 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1704 scm_str2symbol -> scm_from_locale_symbol
1706 SCM_SYMBOL_HASH -> scm_hashq
1707 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1709 scm_c_make_keyword -> scm_from_locale_keyword
1711 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1712 now also available to C code.
1714 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1716 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1717 the latter returns the true name of the keyword, not the 'dash name',
1718 as SCM_KEYWORDSYM used to do.
1720 ** A new way to access arrays in a thread-safe and efficient way has
1723 See the manual, node "Accessing Arrays From C".
1725 ** The old uniform vector and bitvector implementations have been
1726 unceremoniously removed.
1728 This implementation exposed the details of the tagging system of
1729 Guile. Use the new C API explained in the manual in node "Uniform
1730 Numeric Vectors" and "Bit Vectors", respectively.
1732 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1733 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1734 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1735 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1736 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1737 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1740 ** The macros dealing with vectors have been deprecated.
1742 Use the new functions scm_is_vector, scm_vector_elements,
1743 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1744 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1745 manual for more details.
1747 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1748 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1750 The following macros have been removed: SCM_VECTOR_BASE,
1751 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1752 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1754 ** Some C functions and macros related to arrays have been deprecated.
1756 Migrate according to the following table:
1758 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1759 scm_make_ra -> scm_make_array
1760 scm_shap2ra -> scm_make_array
1761 scm_cvref -> scm_c_generalized_vector_ref
1762 scm_ra_set_contp -> do not use
1763 scm_aind -> scm_array_handle_pos
1764 scm_raprin1 -> scm_display or scm_write
1766 SCM_ARRAYP -> scm_is_array
1767 SCM_ARRAY_NDIM -> scm_c_array_rank
1768 SCM_ARRAY_DIMS -> scm_array_handle_dims
1769 SCM_ARRAY_CONTP -> do not use
1770 SCM_ARRAY_MEM -> do not use
1771 SCM_ARRAY_V -> scm_array_handle_elements or similar
1772 SCM_ARRAY_BASE -> do not use
1774 ** SCM_CELL_WORD_LOC has been deprecated.
1776 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1777 to a SCM, as opposed to a pointer to a scm_t_bits.
1779 This was done to allow the correct use of pointers into the Scheme
1780 heap. Previously, the heap words were of type scm_t_bits and local
1781 variables and function arguments were of type SCM, making it
1782 non-standards-conformant to have a pointer that can point to both.
1784 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1786 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1787 second and third words of double smobs. Likewise for
1788 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1790 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1791 used to get and set the 16 exra bits in the zeroth word of a smob.
1793 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1794 accesing the first immediate word of a smob as a SCM value, and there
1795 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1796 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1798 ** New way to deal with non-local exits and re-entries.
1800 There is a new set of functions that essentially do what
1801 scm_internal_dynamic_wind does, but in a way that is more convenient
1802 for C code in some situations. Here is a quick example of how to
1803 prevent a potential memory leak:
1810 scm_dynwind_begin (0);
1812 mem = scm_malloc (100);
1813 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1815 /* MEM would leak if BAR throws an error.
1816 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1823 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1824 SCM_DYNWIND_END as well.
1828 For full documentation, see the node "Dynamic Wind" in the manual.
1830 ** New function scm_dynwind_free
1832 This function calls 'free' on a given pointer when a dynwind context
1833 is left. Thus the call to scm_dynwind_unwind_handler above could be
1834 replaced with simply scm_dynwind_free (mem).
1836 ** New functions scm_c_call_with_blocked_asyncs and
1837 scm_c_call_with_unblocked_asyncs
1839 Like scm_call_with_blocked_asyncs etc. but for C functions.
1841 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1843 In addition to scm_c_call_with_blocked_asyncs you can now also use
1844 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1845 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1847 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1848 SCM_REALLOW_INTS have been deprecated.
1850 They do no longer fulfill their original role of blocking signal
1851 delivery. Depending on what you want to achieve, replace a pair of
1852 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1853 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1856 ** The value 'scm_mask_ints' is no longer writable.
1858 Previously, you could set scm_mask_ints directly. This is no longer
1859 possible. Use scm_c_call_with_blocked_asyncs and
1860 scm_c_call_with_unblocked_asyncs instead.
1862 ** New way to temporarily set the current input, output or error ports
1864 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1865 context' (see above). <foo> is one of "input", "output" or "error".
1867 ** New way to temporarily set fluids
1869 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1870 above) to temporarily set the value of a fluid.
1872 ** New types scm_t_intmax and scm_t_uintmax.
1874 On platforms that have them, these types are identical to intmax_t and
1875 uintmax_t, respectively. On other platforms, they are identical to
1876 the largest integer types that Guile knows about.
1878 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1880 You should not have used them.
1882 ** Many public #defines with generic names have been made private.
1884 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1885 private or renamed with a more suitable public name.
1887 ** The macro SCM_TYP16S has been deprecated.
1889 This macro is not intended for public use.
1891 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1893 Use scm_is_true (scm_inexact_p (...)) instead.
1895 ** The macro SCM_SLOPPY_REALP has been deprecated.
1897 Use scm_is_real instead.
1899 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1901 Use scm_is_complex instead.
1903 ** Some preprocessor defines have been deprecated.
1905 These defines indicated whether a certain feature was present in Guile
1906 or not. Going forward, assume that the features are always present.
1908 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1909 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1911 The following macros have been removed completely: MEMOIZE_LOCALS,
1912 SCM_RECKLESS, SCM_CAUTIOUS.
1914 ** The preprocessor define STACK_DIRECTION has been deprecated.
1916 There should be no need to know about the stack direction for ordinary
1919 ** New function: scm_effective_version
1921 Returns the "effective" version number. This is just the normal full
1922 version string without the final micro-version number. See "Changes
1923 to the distribution" above.
1925 ** The function scm_call_with_new_thread has a new prototype.
1927 Instead of taking a list with the thunk and handler, these two
1928 arguments are now passed directly:
1930 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1932 This is an incompatible change.
1934 ** New snarfer macro SCM_DEFINE_PUBLIC.
1936 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1937 function in the init section.
1939 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1941 ** Garbage collector rewrite.
1943 The garbage collector is cleaned up a lot, and now uses lazy
1944 sweeping. This is reflected in the output of (gc-stats); since cells
1945 are being freed when they are allocated, the cells-allocated field
1946 stays roughly constant.
1948 For malloc related triggers, the behavior is changed. It uses the same
1949 heuristic as the cell-triggered collections. It may be tuned with the
1950 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1951 for minimum yield of malloc related triggers. The default is 40.
1952 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1955 Debugging operations for the freelist have been deprecated, along with
1956 the C variables that control garbage collection. The environment
1957 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1958 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1960 For understanding the memory usage of a GUILE program, the routine
1961 gc-live-object-stats returns an alist containing the number of live
1962 objects for every type.
1965 ** The function scm_definedp has been renamed to scm_defined_p
1967 The name scm_definedp is deprecated.
1969 ** The struct scm_cell type has been renamed to scm_t_cell
1971 This is in accordance to Guile's naming scheme for types. Note that
1972 the name scm_cell is now used for a function that allocates and
1973 initializes a new cell (see below).
1975 ** New functions for memory management
1977 A new set of functions for memory management has been added since the
1978 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1979 indeed, Guile itself contained some long standing bugs that could
1980 cause aborts in long running programs.
1982 The new functions are more symmetrical and do not need cooperation
1983 from smob free routines, among other improvements.
1985 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1986 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1987 scm_gc_free, scm_gc_register_collectable_memory, and
1988 scm_gc_unregister_collectable_memory. Refer to the manual for more
1989 details and for upgrading instructions.
1991 The old functions for memory management have been deprecated. They
1992 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1993 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1995 ** Declarations of exported features are marked with SCM_API.
1997 Every declaration of a feature that belongs to the exported Guile API
1998 has been marked by adding the macro "SCM_API" to the start of the
1999 declaration. This macro can expand into different things, the most
2000 common of which is just "extern" for Unix platforms. On Win32, it can
2001 be used to control which symbols are exported from a DLL.
2003 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
2004 will expand into "__declspec (dllimport) extern", which is needed for
2005 linking to the Guile DLL in Windows.
2007 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
2008 SCM_SRFI4_IMPORT, for the corresponding libraries.
2010 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
2012 Use the new functions scm_cell and scm_double_cell instead. The old
2013 macros had problems because with them allocation and initialization
2014 was separated and the GC could sometimes observe half initialized
2015 cells. Only careful coding by the user of SCM_NEWCELL and
2016 SCM_NEWCELL2 could make this safe and efficient.
2018 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
2020 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
2023 ** SRCBRKP has been deprecated.
2025 Use scm_c_source_property_breakpoint_p instead.
2027 ** Deprecated: scm_makmacro
2029 Change your code to use either scm_makmmacro or to define macros in
2030 Scheme, using 'define-macro'.
2032 ** New function scm_c_port_for_each.
2034 This function is like scm_port_for_each but takes a pointer to a C
2035 function as the callback instead of a SCM value.
2037 ** The names scm_internal_select, scm_thread_sleep, and
2038 scm_thread_usleep have been discouraged.
2040 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
2042 ** The GC can no longer be blocked.
2044 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
2045 The GC can now run (partially) concurrently with other code and thus
2046 blocking it is not well defined.
2048 ** Many definitions have been removed that were previously deprecated.
2050 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
2051 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
2052 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
2053 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
2054 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
2055 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
2056 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
2057 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
2058 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
2059 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
2060 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
2061 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
2062 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
2063 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
2064 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
2065 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
2066 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
2067 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
2068 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
2069 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
2070 scm_vector_set_length_x, scm_contregs, scm_debug_info,
2071 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
2072 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
2073 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
2074 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
2075 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
2076 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
2077 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
2078 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
2079 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
2080 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
2082 * Changes to bundled modules
2086 Using the (ice-9 debug) module no longer automatically switches Guile
2087 to use the debugging evaluator. If you want to switch to the
2088 debugging evaluator (which is needed for backtrace information if you
2089 hit an error), please add an explicit "(debug-enable 'debug)" to your
2090 code just after the code to use (ice-9 debug).
2093 Changes since Guile 1.4:
2095 * Changes to the distribution
2097 ** A top-level TODO file is included.
2099 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
2101 Guile now always uses three numbers to represent the version,
2102 i.e. "1.6.5". The first number, 1, is the major version number, the
2103 second number, 6, is the minor version number, and the third number,
2104 5, is the micro version number. Changes in major version number
2105 indicate major changes in Guile.
2107 Minor version numbers that are even denote stable releases, and odd
2108 minor version numbers denote development versions (which may be
2109 unstable). The micro version number indicates a minor sub-revision of
2110 a given MAJOR.MINOR release.
2112 In keeping with the new scheme, (minor-version) and scm_minor_version
2113 no longer return everything but the major version number. They now
2114 just return the minor version number. Two new functions
2115 (micro-version) and scm_micro_version have been added to report the
2116 micro version number.
2118 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
2120 ** New preprocessor definitions are available for checking versions.
2122 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
2123 SCM_MICRO_VERSION to the appropriate integer values.
2125 ** Guile now actively warns about deprecated features.
2127 The new configure option `--enable-deprecated=LEVEL' and the
2128 environment variable GUILE_WARN_DEPRECATED control this mechanism.
2129 See INSTALL and README for more information.
2131 ** Guile is much more likely to work on 64-bit architectures.
2133 Guile now compiles and passes "make check" with only two UNRESOLVED GC
2134 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
2135 for the use of a test machine, and thanks to Stefan Jahn for ia64
2138 ** New functions: setitimer and getitimer.
2140 These implement a fairly direct interface to the libc functions of the
2143 ** The #. reader extension is now disabled by default.
2145 For safety reasons, #. evaluation is disabled by default. To
2146 re-enable it, set the fluid read-eval? to #t. For example:
2148 (fluid-set! read-eval? #t)
2150 but make sure you realize the potential security risks involved. With
2151 read-eval? enabled, reading a data file from an untrusted source can
2154 ** New SRFI modules have been added:
2156 SRFI-0 `cond-expand' is now supported in Guile, without requiring
2159 (srfi srfi-1) is a library containing many useful pair- and list-processing
2162 (srfi srfi-2) exports and-let*.
2164 (srfi srfi-4) implements homogeneous numeric vector datatypes.
2166 (srfi srfi-6) is a dummy module for now, since guile already provides
2167 all of the srfi-6 procedures by default: open-input-string,
2168 open-output-string, get-output-string.
2170 (srfi srfi-8) exports receive.
2172 (srfi srfi-9) exports define-record-type.
2174 (srfi srfi-10) exports define-reader-ctor and implements the reader
2177 (srfi srfi-11) exports let-values and let*-values.
2179 (srfi srfi-13) implements the SRFI String Library.
2181 (srfi srfi-14) implements the SRFI Character-Set Library.
2183 (srfi srfi-17) implements setter and getter-with-setter and redefines
2184 some accessor procedures as procedures with getters. (such as car,
2185 cdr, vector-ref etc.)
2187 (srfi srfi-19) implements the SRFI Time/Date Library.
2189 ** New scripts / "executable modules"
2191 Subdirectory "scripts" contains Scheme modules that are packaged to
2192 also be executable as scripts. At this time, these scripts are available:
2201 See README there for more info.
2203 These scripts can be invoked from the shell with the new program
2204 "guile-tools", which keeps track of installation directory for you.
2207 $ guile-tools display-commentary srfi/*.scm
2209 guile-tools is copied to the standard $bindir on "make install".
2211 ** New module (ice-9 stack-catch):
2213 stack-catch is like catch, but saves the current state of the stack in
2214 the fluid the-last-stack. This fluid can be useful when using the
2215 debugger and when re-throwing an error.
2217 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
2219 This has been done to prevent problems on lesser operating systems
2220 that can't tolerate `*'s in file names. The exported macro continues
2221 to be named `and-let*', of course.
2223 On systems that support it, there is also a compatibility module named
2224 (ice-9 and-let*). It will go away in the next release.
2226 ** New modules (oop goops) etc.:
2229 (oop goops describe)
2231 (oop goops active-slot)
2232 (oop goops composite-slot)
2234 The Guile Object Oriented Programming System (GOOPS) has been
2235 integrated into Guile. For further information, consult the GOOPS
2236 manual and tutorial in the `doc' directory.
2238 ** New module (ice-9 rdelim).
2240 This exports the following procedures which were previously defined
2241 in the default environment:
2243 read-line read-line! read-delimited read-delimited! %read-delimited!
2244 %read-line write-line
2246 For backwards compatibility the definitions are still imported into the
2247 default environment in this version of Guile. However you should add:
2249 (use-modules (ice-9 rdelim))
2251 to any program which uses the definitions, since this may change in
2254 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
2255 can be used for similar functionality.
2257 ** New module (ice-9 rw)
2259 This is a subset of the (scsh rw) module from guile-scsh. Currently
2260 it defines two procedures:
2262 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
2264 Read characters from a port or file descriptor into a string STR.
2265 A port must have an underlying file descriptor -- a so-called
2266 fport. This procedure is scsh-compatible and can efficiently read
2269 *** New function: write-string/partial str [port_or_fdes [start [end]]]
2271 Write characters from a string STR to a port or file descriptor.
2272 A port must have an underlying file descriptor -- a so-called
2273 fport. This procedure is mostly compatible and can efficiently
2274 write large strings.
2276 ** New module (ice-9 match)
2278 This module includes Andrew K. Wright's pattern matcher. See
2279 ice-9/match.scm for brief description or
2281 http://www.star-lab.com/wright/code.html
2283 for complete documentation.
2285 ** New module (ice-9 buffered-input)
2287 This module provides procedures to construct an input port from an
2288 underlying source of input that reads and returns its input in chunks.
2289 The underlying input source is a Scheme procedure, specified by the
2290 caller, which the port invokes whenever it needs more input.
2292 This is useful when building an input port whose back end is Readline
2293 or a UI element such as the GtkEntry widget.
2297 The reference and tutorial documentation that was previously
2298 distributed separately, as `guile-doc', is now included in the core
2299 Guile distribution. The documentation consists of the following
2302 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
2305 - The Guile Reference Manual (guile.texi) contains (or is intended to
2306 contain) reference documentation on all aspects of Guile.
2308 - The GOOPS Manual (goops.texi) contains both tutorial-style and
2309 reference documentation for using GOOPS, Guile's Object Oriented
2312 - The Revised^5 Report on the Algorithmic Language Scheme
2315 See the README file in the `doc' directory for more details.
2317 ** There are a couple of examples in the examples/ directory now.
2319 * Changes to the stand-alone interpreter
2321 ** New command line option `--use-srfi'
2323 Using this option, SRFI modules can be loaded on startup and be
2324 available right from the beginning. This makes programming portable
2325 Scheme programs easier.
2327 The option `--use-srfi' expects a comma-separated list of numbers,
2328 each representing a SRFI number to be loaded into the interpreter
2329 before starting evaluating a script file or the REPL. Additionally,
2330 the feature identifier for the loaded SRFIs is recognized by
2331 `cond-expand' when using this option.
2334 $ guile --use-srfi=8,13
2335 guile> (receive (x z) (values 1 2) (+ 1 2))
2337 guile> (string-pad "bla" 20)
2340 ** Guile now always starts up in the `(guile-user)' module.
2342 Previously, scripts executed via the `-s' option would run in the
2343 `(guile)' module and the repl would run in the `(guile-user)' module.
2344 Now every user action takes place in the `(guile-user)' module by
2347 * Changes to Scheme functions and syntax
2349 ** Character classifiers work for non-ASCII characters.
2351 The predicates `char-alphabetic?', `char-numeric?',
2352 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
2353 no longer check whether their arguments are ASCII characters.
2354 Previously, a character would only be considered alphabetic when it
2355 was also ASCII, for example.
2357 ** Previously deprecated Scheme functions have been removed:
2359 tag - no replacement.
2360 fseek - replaced by seek.
2361 list* - replaced by cons*.
2363 ** It's now possible to create modules with controlled environments
2367 (use-modules (ice-9 safe))
2368 (define m (make-safe-module))
2369 ;;; m will now be a module containing only a safe subset of R5RS
2370 (eval '(+ 1 2) m) --> 3
2371 (eval 'load m) --> ERROR: Unbound variable: load
2373 ** Evaluation of "()", the empty list, is now an error.
2375 Previously, the expression "()" evaluated to the empty list. This has
2376 been changed to signal a "missing expression" error. The correct way
2377 to write the empty list as a literal constant is to use quote: "'()".
2379 ** New concept of `Guile Extensions'.
2381 A Guile Extension is just a ordinary shared library that can be linked
2382 at run-time. We found it advantageous to give this simple concept a
2383 dedicated name to distinguish the issues related to shared libraries
2384 from the issues related to the module system.
2386 *** New function: load-extension
2388 Executing (load-extension lib init) is mostly equivalent to
2390 (dynamic-call init (dynamic-link lib))
2392 except when scm_register_extension has been called previously.
2393 Whenever appropriate, you should use `load-extension' instead of
2394 dynamic-link and dynamic-call.
2396 *** New C function: scm_c_register_extension
2398 This function registers a initialization function for use by
2399 `load-extension'. Use it when you don't want specific extensions to
2400 be loaded as shared libraries (for example on platforms that don't
2401 support dynamic linking).
2403 ** Auto-loading of compiled-code modules is deprecated.
2405 Guile used to be able to automatically find and link a shared
2406 library to satisfy requests for a module. For example, the module
2407 `(foo bar)' could be implemented by placing a shared library named
2408 "foo/libbar.so" (or with a different extension) in a directory on the
2411 This has been found to be too tricky, and is no longer supported. The
2412 shared libraries are now called "extensions". You should now write a
2413 small Scheme file that calls `load-extension' to load the shared
2414 library and initialize it explicitly.
2416 The shared libraries themselves should be installed in the usual
2417 places for shared libraries, with names like "libguile-foo-bar".
2419 For example, place this into a file "foo/bar.scm"
2421 (define-module (foo bar))
2423 (load-extension "libguile-foo-bar" "foobar_init")
2425 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
2427 `eval' is now R5RS, that is it takes two arguments.
2428 The second argument is an environment specifier, i.e. either
2430 (scheme-report-environment 5)
2431 (null-environment 5)
2432 (interaction-environment)
2438 ** The module system has been made more disciplined.
2440 The function `eval' will save and restore the current module around
2441 the evaluation of the specified expression. While this expression is
2442 evaluated, `(current-module)' will now return the right module, which
2443 is the module specified as the second argument to `eval'.
2445 A consequence of this change is that `eval' is not particularly
2446 useful when you want allow the evaluated code to change what module is
2447 designated as the current module and have this change persist from one
2448 call to `eval' to the next. The read-eval-print-loop is an example
2449 where `eval' is now inadequate. To compensate, there is a new
2450 function `primitive-eval' that does not take a module specifier and
2451 that does not save/restore the current module. You should use this
2452 function together with `set-current-module', `current-module', etc
2453 when you want to have more control over the state that is carried from
2454 one eval to the next.
2456 Additionally, it has been made sure that forms that are evaluated at
2457 the top level are always evaluated with respect to the current module.
2458 Previously, subforms of top-level forms such as `begin', `case',
2459 etc. did not respect changes to the current module although these
2460 subforms are at the top-level as well.
2462 To prevent strange behavior, the forms `define-module',
2463 `use-modules', `use-syntax', and `export' have been restricted to only
2464 work on the top level. The forms `define-public' and
2465 `defmacro-public' only export the new binding on the top level. They
2466 behave just like `define' and `defmacro', respectively, when they are
2467 used in a lexical environment.
2469 Also, `export' will no longer silently re-export bindings imported
2470 from a used module. It will emit a `deprecation' warning and will
2471 cease to perform any re-export in the next version. If you actually
2472 want to re-export bindings, use the new `re-export' in place of
2473 `export'. The new `re-export' will not make copies of variables when
2474 rexporting them, as `export' did wrongly.
2476 ** Module system now allows selection and renaming of imported bindings
2478 Previously, when using `use-modules' or the `#:use-module' clause in
2479 the `define-module' form, all the bindings (association of symbols to
2480 values) for imported modules were added to the "current module" on an
2481 as-is basis. This has been changed to allow finer control through two
2482 new facilities: selection and renaming.
2484 You can now select which of the imported module's bindings are to be
2485 visible in the current module by using the `:select' clause. This
2486 clause also can be used to rename individual bindings. For example:
2488 ;; import all bindings no questions asked
2489 (use-modules (ice-9 common-list))
2491 ;; import four bindings, renaming two of them;
2492 ;; the current module sees: every some zonk-y zonk-n
2493 (use-modules ((ice-9 common-list)
2495 (remove-if . zonk-y)
2496 (remove-if-not . zonk-n))))
2498 You can also programmatically rename all selected bindings using the
2499 `:renamer' clause, which specifies a proc that takes a symbol and
2500 returns another symbol. Because it is common practice to use a prefix,
2501 we now provide the convenience procedure `symbol-prefix-proc'. For
2504 ;; import four bindings, renaming two of them specifically,
2505 ;; and all four w/ prefix "CL:";
2506 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
2507 (use-modules ((ice-9 common-list)
2509 (remove-if . zonk-y)
2510 (remove-if-not . zonk-n))
2511 :renamer (symbol-prefix-proc 'CL:)))
2513 ;; import four bindings, renaming two of them specifically,
2514 ;; and all four by upcasing.
2515 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
2516 (define (upcase-symbol sym)
2517 (string->symbol (string-upcase (symbol->string sym))))
2519 (use-modules ((ice-9 common-list)
2521 (remove-if . zonk-y)
2522 (remove-if-not . zonk-n))
2523 :renamer upcase-symbol))
2525 Note that programmatic renaming is done *after* individual renaming.
2526 Also, the above examples show `use-modules', but the same facilities are
2527 available for the `#:use-module' clause of `define-module'.
2529 See manual for more info.
2531 ** The semantics of guardians have changed.
2533 The changes are for the most part compatible. An important criterion
2534 was to keep the typical usage of guardians as simple as before, but to
2535 make the semantics safer and (as a result) more useful.
2537 *** All objects returned from guardians are now properly alive.
2539 It is now guaranteed that any object referenced by an object returned
2540 from a guardian is alive. It's now impossible for a guardian to
2541 return a "contained" object before its "containing" object.
2543 One incompatible (but probably not very important) change resulting
2544 from this is that it is no longer possible to guard objects that
2545 indirectly reference themselves (i.e. are parts of cycles). If you do
2546 so accidentally, you'll get a warning.
2548 *** There are now two types of guardians: greedy and sharing.
2550 If you call (make-guardian #t) or just (make-guardian), you'll get a
2551 greedy guardian, and for (make-guardian #f) a sharing guardian.
2553 Greedy guardians are the default because they are more "defensive".
2554 You can only greedily guard an object once. If you guard an object
2555 more than once, once in a greedy guardian and the rest of times in
2556 sharing guardians, then it is guaranteed that the object won't be
2557 returned from sharing guardians as long as it is greedily guarded
2560 Guardians returned by calls to `make-guardian' can now take one more
2561 optional parameter, which says whether to throw an error in case an
2562 attempt is made to greedily guard an object that is already greedily
2563 guarded. The default is true, i.e. throw an error. If the parameter
2564 is false, the guardian invocation returns #t if guarding was
2565 successful and #f if it wasn't.
2567 Also, since greedy guarding is, in effect, a side-effecting operation
2568 on objects, a new function is introduced: `destroy-guardian!'.
2569 Invoking this function on a guardian renders it unoperative and, if
2570 the guardian is greedy, clears the "greedily guarded" property of the
2571 objects that were guarded by it, thus undoing the side effect.
2573 Note that all this hair is hardly very important, since guardian
2574 objects are usually permanent.
2576 ** Continuations created by call-with-current-continuation now accept
2577 any number of arguments, as required by R5RS.
2579 ** New function `issue-deprecation-warning'
2581 This function is used to display the deprecation messages that are
2582 controlled by GUILE_WARN_DEPRECATION as explained in the README.
2585 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
2589 ;; `id' is deprecated. Use `identity' instead.
2594 ** New syntax `begin-deprecated'
2596 When deprecated features are included (as determined by the configure
2597 option --enable-deprecated), `begin-deprecated' is identical to
2598 `begin'. When deprecated features are excluded, it always evaluates
2599 to `#f', ignoring the body forms.
2601 ** New function `make-object-property'
2603 This function returns a new `procedure with setter' P that can be used
2604 to attach a property to objects. When calling P as
2608 where `obj' is any kind of object, it attaches `val' to `obj' in such
2609 a way that it can be retrieved by calling P as
2613 This function will replace procedure properties, symbol properties and
2614 source properties eventually.
2616 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2618 Instead of #&optional, #&key, etc you should now use #:optional,
2619 #:key, etc. Since #:optional is a keyword, you can write it as just
2620 :optional when (read-set! keywords 'prefix) is active.
2622 The old reader syntax `#&' is still supported, but deprecated. It
2623 will be removed in the next release.
2625 ** New define-module option: pure
2627 Tells the module system not to include any bindings from the root
2632 (define-module (totally-empty-module)
2635 ** New define-module option: export NAME1 ...
2637 Export names NAME1 ...
2639 This option is required if you want to be able to export bindings from
2640 a module which doesn't import one of `define-public' or `export'.
2644 (define-module (foo)
2646 :use-module (ice-9 r5rs)
2649 ;;; Note that we're pure R5RS below this point!
2654 ** New function: object->string OBJ
2656 Return a Scheme string obtained by printing a given object.
2658 ** New function: port? X
2660 Returns a boolean indicating whether X is a port. Equivalent to
2661 `(or (input-port? X) (output-port? X))'.
2663 ** New function: file-port?
2665 Determines whether a given object is a port that is related to a file.
2667 ** New function: port-for-each proc
2669 Apply PROC to each port in the Guile port table in turn. The return
2670 value is unspecified. More specifically, PROC is applied exactly once
2671 to every port that exists in the system at the time PORT-FOR-EACH is
2672 invoked. Changes to the port table while PORT-FOR-EACH is running
2673 have no effect as far as PORT-FOR-EACH is concerned.
2675 ** New function: dup2 oldfd newfd
2677 A simple wrapper for the `dup2' system call. Copies the file
2678 descriptor OLDFD to descriptor number NEWFD, replacing the
2679 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2680 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2681 to move away ports which are using NEWFD. The return value is
2684 ** New function: close-fdes fd
2686 A simple wrapper for the `close' system call. Close file
2687 descriptor FD, which must be an integer. Unlike close (*note
2688 close: Ports and File Descriptors.), the file descriptor will be
2689 closed even if a port is using it. The return value is
2692 ** New function: crypt password salt
2694 Encrypts `password' using the standard unix password encryption
2697 ** New function: chroot path
2699 Change the root directory of the running process to `path'.
2701 ** New functions: getlogin, cuserid
2703 Return the login name or the user name of the current effective user
2706 ** New functions: getpriority which who, setpriority which who prio
2708 Get or set the priority of the running process.
2710 ** New function: getpass prompt
2712 Read a password from the terminal, first displaying `prompt' and
2715 ** New function: flock file operation
2717 Set/remove an advisory shared or exclusive lock on `file'.
2719 ** New functions: sethostname name, gethostname
2721 Set or get the hostname of the machine the current process is running
2724 ** New function: mkstemp! tmpl
2726 mkstemp creates a new unique file in the file system and returns a
2727 new buffered port open for reading and writing to the file. TMPL
2728 is a string specifying where the file should be created: it must
2729 end with `XXXXXX' and will be changed in place to return the name
2730 of the temporary file.
2732 ** New function: open-input-string string
2734 Return an input string port which delivers the characters from
2735 `string'. This procedure, together with `open-output-string' and
2736 `get-output-string' implements SRFI-6.
2738 ** New function: open-output-string
2740 Return an output string port which collects all data written to it.
2741 The data can then be retrieved by `get-output-string'.
2743 ** New function: get-output-string
2745 Return the contents of an output string port.
2747 ** New function: identity
2749 Return the argument.
2751 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2752 are represented in Scheme as integers with normal host byte ordering.
2754 ** New function: inet-pton family address
2756 Convert a printable string network address into an integer. Note that
2757 unlike the C version of this function, the result is an integer with
2758 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2761 (inet-pton AF_INET "127.0.0.1") => 2130706433
2762 (inet-pton AF_INET6 "::1") => 1
2764 ** New function: inet-ntop family address
2766 Convert an integer network address into a printable string. Note that
2767 unlike the C version of this function, the input is an integer with
2768 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2771 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2772 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2773 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2777 Use `identity' instead.
2783 ** Deprecated: return-it
2787 ** Deprecated: string-character-length
2789 Use `string-length' instead.
2791 ** Deprecated: flags
2793 Use `logior' instead.
2795 ** Deprecated: close-all-ports-except.
2797 This was intended for closing ports in a child process after a fork,
2798 but it has the undesirable side effect of flushing buffers.
2799 port-for-each is more flexible.
2801 ** The (ice-9 popen) module now attempts to set up file descriptors in
2802 the child process from the current Scheme ports, instead of using the
2803 current values of file descriptors 0, 1, and 2 in the parent process.
2805 ** Removed function: builtin-weak-bindings
2807 There is no such concept as a weak binding any more.
2809 ** Removed constants: bignum-radix, scm-line-incrementors
2811 ** define-method: New syntax mandatory.
2813 The new method syntax is now mandatory:
2815 (define-method (NAME ARG-SPEC ...) BODY ...)
2816 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2818 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2819 REST-ARG ::= ARG-NAME
2821 If you have old code using the old syntax, import
2822 (oop goops old-define-method) before (oop goops) as in:
2824 (use-modules (oop goops old-define-method) (oop goops))
2826 ** Deprecated function: builtin-variable
2827 Removed function: builtin-bindings
2829 There is no longer a distinction between builtin or other variables.
2830 Use module system operations for all variables.
2832 ** Lazy-catch handlers are no longer allowed to return.
2834 That is, a call to `throw', `error', etc is now guaranteed to not
2837 ** Bugfixes for (ice-9 getopt-long)
2839 This module is now tested using test-suite/tests/getopt-long.test.
2840 The following bugs have been fixed:
2842 *** Parsing for options that are specified to have `optional' args now checks
2843 if the next element is an option instead of unconditionally taking it as the
2846 *** An error is now thrown for `--opt=val' when the option description
2847 does not specify `(value #t)' or `(value optional)'. This condition used to
2848 be accepted w/o error, contrary to the documentation.
2850 *** The error message for unrecognized options is now more informative.
2851 It used to be "not a record", an artifact of the implementation.
2853 *** The error message for `--opt' terminating the arg list (no value), when
2854 `(value #t)' is specified, is now more informative. It used to be "not enough
2857 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2858 The expansion used to be like so:
2860 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2862 Note that the "5d" is dropped. Now it is like so:
2864 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2866 This enables single-char options to have adjoining arguments as long as their
2867 constituent characters are not potential single-char options.
2869 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2871 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2872 property `arglist', which can be retrieved by `arity'. The result is that
2873 `arity' can give more detailed information than before:
2877 guile> (use-modules (ice-9 optargs))
2878 guile> (define* (foo #:optional a b c) a)
2880 0 or more arguments in `lambda*:G0'.
2885 3 optional arguments: `a', `b' and `c'.
2886 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2888 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2889 and `d', other keywords allowed.
2890 guile> (define* (baz a b #:optional c #:rest r) a)
2892 2 required arguments: `a' and `b', 1 optional argument: `c',
2895 * Changes to the C interface
2897 ** Types have been renamed from scm_*_t to scm_t_*.
2899 This has been done for POSIX sake. It reserves identifiers ending
2900 with "_t". What a concept.
2902 The old names are still available with status `deprecated'.
2904 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2906 ** Deprecated features have been removed.
2910 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2911 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2913 *** C Functions removed
2915 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2916 scm_fseek - replaced by scm_seek.
2917 gc-thunk - replaced by after-gc-hook.
2918 gh_int2scmb - replaced by gh_bool2scm.
2919 scm_tc_dblr - replaced by scm_tc16_real.
2920 scm_tc_dblc - replaced by scm_tc16_complex.
2921 scm_list_star - replaced by scm_cons_star.
2923 ** Deprecated: scm_makfromstr
2925 Use scm_mem2string instead.
2927 ** Deprecated: scm_make_shared_substring
2929 Explicit shared substrings will disappear from Guile.
2931 Instead, "normal" strings will be implemented using sharing
2932 internally, combined with a copy-on-write strategy.
2934 ** Deprecated: scm_read_only_string_p
2936 The concept of read-only strings will disappear in next release of
2939 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2941 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2943 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2945 Call a procedure with the indicated number of arguments. See "Fly
2946 Evaluation" in the manual.
2948 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2950 Call a procedure with the indicated number of arguments and a list of
2951 further arguments. See "Fly Evaluation" in the manual.
2953 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2955 Create a list of the given number of elements. See "List
2956 Constructors" in the manual.
2958 ** Renamed function: scm_listify has been replaced by scm_list_n.
2960 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2961 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2963 Use functions scm_list_N instead.
2965 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2967 Used by an application to read arbitrary number of bytes from a port.
2968 Same semantics as libc read, except that scm_c_read only returns less
2969 than SIZE bytes if at end-of-file.
2971 Warning: Doesn't update port line and column counts!
2973 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2975 Used by an application to write arbitrary number of bytes to an SCM
2976 port. Similar semantics as libc write. However, unlike libc
2977 write, scm_c_write writes the requested number of bytes and has no
2980 Warning: Doesn't update port line and column counts!
2982 ** New function: scm_init_guile ()
2984 In contrast to scm_boot_guile, scm_init_guile will return normally
2985 after initializing Guile. It is not available on all systems, tho.
2987 ** New functions: scm_str2symbol, scm_mem2symbol
2989 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2990 field of characters and creates a scheme symbol object from that C string.
2991 The function scm_mem2symbol takes a const char* and a number of characters and
2992 creates a symbol from the characters in that memory area.
2994 ** New functions: scm_primitive_make_property
2995 scm_primitive_property_ref
2996 scm_primitive_property_set_x
2997 scm_primitive_property_del_x
2999 These functions implement a new way to deal with object properties.
3000 See libguile/properties.c for their documentation.
3002 ** New function: scm_done_free (long size)
3004 This function is the inverse of scm_done_malloc. Use it to report the
3005 amount of smob memory you free. The previous method, which involved
3006 calling scm_done_malloc with negative argument, was somewhat
3007 unintuitive (and is still available, of course).
3009 ** New function: scm_c_memq (SCM obj, SCM list)
3011 This function provides a fast C level alternative for scm_memq for the case
3012 that the list parameter is known to be a proper list. The function is a
3013 replacement for scm_sloppy_memq, but is stricter in its requirements on its
3014 list input parameter, since for anything else but a proper list the function's
3015 behaviour is undefined - it may even crash or loop endlessly. Further, for
3016 the case that the object is not found in the list, scm_c_memq returns #f which
3017 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
3019 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
3020 scm_remember_upto_here
3022 These functions replace the function scm_remember.
3024 ** Deprecated function: scm_remember
3026 Use one of the new functions scm_remember_upto_here_1,
3027 scm_remember_upto_here_2 or scm_remember_upto_here instead.
3029 ** New function: scm_allocate_string
3031 This function replaces the function scm_makstr.
3033 ** Deprecated function: scm_makstr
3035 Use the new function scm_allocate_string instead.
3037 ** New global variable scm_gc_running_p introduced.
3039 Use this variable to find out if garbage collection is being executed. Up to
3040 now applications have used scm_gc_heap_lock to test if garbage collection was
3041 running, which also works because of the fact that up to know only the garbage
3042 collector has set this variable. But, this is an implementation detail that
3043 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
3044 of this variable is (and has been) not fully safe anyway.
3046 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
3048 Use these instead of SCM_LENGTH_MAX.
3050 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
3051 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
3052 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
3054 Use these instead of SCM_LENGTH.
3056 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
3057 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
3058 SCM_SET_BITVECTOR_LENGTH
3060 Use these instead of SCM_SETLENGTH
3062 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
3063 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
3066 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
3069 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
3070 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
3073 Use these instead of SCM_SETCHARS.
3075 ** New macro: SCM_BITVECTOR_P
3077 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
3079 Use instead of SCM_COERCE_SUBSTR.
3081 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
3083 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
3085 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
3086 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
3087 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
3088 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
3089 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
3090 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
3091 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
3092 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
3093 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
3094 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
3095 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
3096 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
3097 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
3098 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
3099 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
3101 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
3102 Use scm_memory_error instead of SCM_NALLOC.
3103 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
3104 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
3105 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
3106 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
3107 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
3108 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
3109 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
3110 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
3111 Use SCM_STRINGP instead of SCM_RWSTRINGP.
3112 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
3113 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
3114 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
3115 Use a type specific setter macro instead of SCM_SETLENGTH.
3116 Use a type specific setter macro instead of SCM_SETCHARS.
3117 Use a type specific length macro instead of SCM_LENGTH_MAX.
3118 Use SCM_GCMARKP instead of SCM_GC8MARKP.
3119 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
3120 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
3121 Use SCM_TYP16 instead of SCM_GCTYP16.
3122 Use SCM_CDR instead of SCM_GCCDR.
3123 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
3124 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
3125 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
3126 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
3127 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
3128 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
3129 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
3131 ** Removed function: scm_struct_init
3133 ** Removed variable: scm_symhash_dim
3135 ** Renamed function: scm_make_cont has been replaced by
3136 scm_make_continuation, which has a different interface.
3138 ** Deprecated function: scm_call_catching_errors
3140 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
3142 ** Deprecated function: scm_strhash
3144 Use scm_string_hash instead.
3146 ** Deprecated function: scm_vector_set_length_x
3148 Instead, create a fresh vector of the desired size and copy the contents.
3150 ** scm_gensym has changed prototype
3152 scm_gensym now only takes one argument.
3154 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
3157 There is now only a single symbol type scm_tc7_symbol.
3158 The tag scm_tc7_lvector was not used anyway.
3160 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
3162 Use scm_make_smob_type and scm_set_smob_XXX instead.
3164 ** New function scm_set_smob_apply.
3166 This can be used to set an apply function to a smob type.
3168 ** Deprecated function: scm_strprint_obj
3170 Use scm_object_to_string instead.
3172 ** Deprecated function: scm_wta
3174 Use scm_wrong_type_arg, or another appropriate error signalling function
3177 ** Explicit support for obarrays has been deprecated.
3179 Use `scm_str2symbol' and the generic hashtable functions instead.
3181 ** The concept of `vcells' has been deprecated.
3183 The data type `variable' is now used exclusively. `Vcells' have been
3184 a low-level concept so you are likely not affected by this change.
3186 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
3187 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
3189 Use scm_c_define or scm_c_lookup instead, as appropriate.
3191 *** New functions: scm_c_module_lookup, scm_c_lookup,
3192 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
3193 scm_module_define, scm_define.
3195 These functions work with variables instead of with vcells.
3197 ** New functions for creating and defining `subr's and `gsubr's.
3199 The new functions more clearly distinguish between creating a subr (or
3200 gsubr) object and adding it to the current module.
3202 These new functions are available: scm_c_make_subr, scm_c_define_subr,
3203 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
3204 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
3205 scm_c_define_gsubr_with_generic.
3207 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
3208 scm_make_subr_with_generic, scm_make_gsubr,
3209 scm_make_gsubr_with_generic.
3211 Use the new ones from above instead.
3213 ** C interface to the module system has changed.
3215 While we suggest that you avoid as many explicit module system
3216 operations from C as possible for the time being, the C interface has
3217 been made more similar to the high-level Scheme module system.
3219 *** New functions: scm_c_define_module, scm_c_use_module,
3220 scm_c_export, scm_c_resolve_module.
3222 They mostly work like their Scheme namesakes. scm_c_define_module
3223 takes a function that is called a context where the new module is
3226 *** Deprecated functions: scm_the_root_module, scm_make_module,
3227 scm_ensure_user_module, scm_load_scheme_module.
3229 Use the new functions instead.
3231 ** Renamed function: scm_internal_with_fluids becomes
3234 scm_internal_with_fluids is available as a deprecated function.
3236 ** New function: scm_c_with_fluid.
3238 Just like scm_c_with_fluids, but takes one fluid and one value instead
3241 ** Deprecated typedefs: long_long, ulong_long.
3243 They are of questionable utility and they pollute the global
3246 ** Deprecated typedef: scm_sizet
3248 It is of questionable utility now that Guile requires ANSI C, and is
3251 ** Deprecated typedefs: scm_port_rw_active, scm_port,
3252 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
3253 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
3255 Made more compliant with the naming policy by adding a _t at the end.
3257 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
3258 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
3260 With the exception of the mysterious scm_2ulong2big, they are still
3261 available under new names (scm_i_mkbig etc). These functions are not
3262 intended to be used in user code. You should avoid dealing with
3263 bignums directly, and should deal with numbers in general (which can
3266 ** Change in behavior: scm_num2long, scm_num2ulong
3268 The scm_num2[u]long functions don't any longer accept an inexact
3269 argument. This change in behavior is motivated by concordance with
3270 R5RS: It is more common that a primitive doesn't want to accept an
3271 inexact for an exact.
3273 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
3274 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
3275 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
3278 These are conversion functions between the various ANSI C integral
3279 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
3280 accept an inexact argument.
3282 ** New functions: scm_float2num, scm_double2num,
3283 scm_num2float, scm_num2double.
3285 These are conversion functions between the two ANSI C float types and
3288 ** New number validation macros:
3289 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
3293 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
3295 These are just nicer-named old scm_protect_object and
3296 scm_unprotect_object.
3298 ** Deprecated functions: scm_protect_object, scm_unprotect_object
3300 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
3302 These functions can be used to register pointers to locations that
3305 ** Deprecated function: scm_create_hook.
3307 Its sins are: misleading name, non-modularity and lack of general
3311 Changes since Guile 1.3.4:
3313 * Changes to the distribution
3315 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
3317 We've changed the way we handle generated files in the Guile source
3318 repository. As a result, the procedure for building trees obtained
3319 from the nightly FTP snapshots or via CVS has changed:
3320 - You must have appropriate versions of autoconf, automake, and
3321 libtool installed on your system. See README for info on how to
3322 obtain these programs.
3323 - Before configuring the tree, you must first run the script
3324 `autogen.sh' at the top of the source tree.
3326 The Guile repository used to contain not only source files, written by
3327 humans, but also some generated files, like configure scripts and
3328 Makefile.in files. Even though the contents of these files could be
3329 derived mechanically from other files present, we thought it would
3330 make the tree easier to build if we checked them into CVS.
3332 However, this approach means that minor differences between
3333 developer's installed tools and habits affected the whole team.
3334 So we have removed the generated files from the repository, and
3335 added the autogen.sh script, which will reconstruct them
3339 ** configure now has experimental options to remove support for certain
3342 --disable-arrays omit array and uniform array support
3343 --disable-posix omit posix interfaces
3344 --disable-networking omit networking interfaces
3345 --disable-regex omit regular expression interfaces
3347 These are likely to become separate modules some day.
3349 ** New configure option --enable-debug-freelist
3351 This enables a debugging version of SCM_NEWCELL(), and also registers
3352 an extra primitive, the setter `gc-set-debug-check-freelist!'.
3354 Configure with the --enable-debug-freelist option to enable
3355 the gc-set-debug-check-freelist! primitive, and then use:
3357 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
3358 (gc-set-debug-check-freelist! #f) # turn off checking
3360 Checking of the freelist forces a traversal of the freelist and
3361 a garbage collection before each allocation of a cell. This can
3362 slow down the interpreter dramatically, so the setter should be used to
3363 turn on this extra processing only when necessary.
3365 ** New configure option --enable-debug-malloc
3367 Include code for debugging of calls to scm_must_malloc/realloc/free.
3371 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
3372 2. objects reallocated by scm_must_realloc has been allocated by
3374 3. reallocated objects are reallocated with the same what string
3376 But, most importantly, it records the number of allocated objects of
3377 each kind. This is useful when searching for memory leaks.
3379 A Guile compiled with this option provides the primitive
3380 `malloc-stats' which returns an alist with pairs of kind and the
3381 number of objects of that kind.
3383 ** All includes are now referenced relative to the root directory
3385 Since some users have had problems with mixups between Guile and
3386 system headers, we have decided to always refer to Guile headers via
3387 their parent directories. This essentially creates a "private name
3388 space" for Guile headers. This means that the compiler only is given
3389 -I options for the root build and root source directory.
3391 ** Header files kw.h and genio.h have been removed.
3393 ** The module (ice-9 getopt-gnu-style) has been removed.
3395 ** New module (ice-9 documentation)
3397 Implements the interface to documentation strings associated with
3400 ** New module (ice-9 time)
3402 Provides a macro `time', which displays execution time of a given form.
3404 ** New module (ice-9 history)
3406 Loading this module enables value history in the repl.
3408 * Changes to the stand-alone interpreter
3410 ** New command line option --debug
3412 Start Guile with debugging evaluator and backtraces enabled.
3414 This is useful when debugging your .guile init file or scripts.
3416 ** New help facility
3418 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
3419 (help REGEXP) ditto for objects with names matching REGEXP (a string)
3420 (help 'NAME) gives documentation for NAME, even if it is not an object
3421 (help ,EXPR) gives documentation for object returned by EXPR
3422 (help (my module)) gives module commentary for `(my module)'
3423 (help) gives this text
3425 `help' searches among bindings exported from loaded modules, while
3426 `apropos' searches among bindings visible from the "current" module.
3428 Examples: (help help)
3430 (help "output-string")
3432 ** `help' and `apropos' now prints full module names
3434 ** Dynamic linking now uses libltdl from the libtool package.
3436 The old system dependent code for doing dynamic linking has been
3437 replaced with calls to the libltdl functions which do all the hairy
3440 The major improvement is that you can now directly pass libtool
3441 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
3442 will be able to do the best shared library job you can get, via
3445 The way dynamic libraries are found has changed and is not really
3446 portable across platforms, probably. It is therefore recommended to
3447 use absolute filenames when possible.
3449 If you pass a filename without an extension to `dynamic-link', it will
3450 try a few appropriate ones. Thus, the most platform ignorant way is
3451 to specify a name like "libfoo", without any directories and
3454 ** Guile COOP threads are now compatible with LinuxThreads
3456 Previously, COOP threading wasn't possible in applications linked with
3457 Linux POSIX threads due to their use of the stack pointer to find the
3458 thread context. This has now been fixed with a workaround which uses
3459 the pthreads to allocate the stack.
3461 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
3463 ** Positions of erring expression in scripts
3465 With version 1.3.4, the location of the erring expression in Guile
3466 scipts is no longer automatically reported. (This should have been
3467 documented before the 1.3.4 release.)
3469 You can get this information by enabling recording of positions of
3470 source expressions and running the debugging evaluator. Put this at
3471 the top of your script (or in your "site" file):
3473 (read-enable 'positions)
3474 (debug-enable 'debug)
3476 ** Backtraces in scripts
3478 It is now possible to get backtraces in scripts.
3482 (debug-enable 'debug 'backtrace)
3484 at the top of the script.
3486 (The first options enables the debugging evaluator.
3487 The second enables backtraces.)
3489 ** Part of module system symbol lookup now implemented in C
3491 The eval closure of most modules is now implemented in C. Since this
3492 was one of the bottlenecks for loading speed, Guile now loads code
3493 substantially faster than before.
3495 ** Attempting to get the value of an unbound variable now produces
3496 an exception with a key of 'unbound-variable instead of 'misc-error.
3498 ** The initial default output port is now unbuffered if it's using a
3499 tty device. Previously in this situation it was line-buffered.
3501 ** New hook: after-gc-hook
3503 after-gc-hook takes over the role of gc-thunk. This hook is run at
3504 the first SCM_TICK after a GC. (Thus, the code is run at the same
3505 point during evaluation as signal handlers.)
3507 Note that this hook should be used only for diagnostic and debugging
3508 purposes. It is not certain that it will continue to be well-defined
3509 when this hook is run in the future.
3511 C programmers: Note the new C level hooks scm_before_gc_c_hook,
3512 scm_before_sweep_c_hook, scm_after_gc_c_hook.
3514 ** Improvements to garbage collector
3516 Guile 1.4 has a new policy for triggering heap allocation and
3517 determining the sizes of heap segments. It fixes a number of problems
3520 1. The new policy can handle two separate pools of cells
3521 (2-word/4-word) better. (The old policy would run wild, allocating
3522 more and more memory for certain programs.)
3524 2. The old code would sometimes allocate far too much heap so that the
3525 Guile process became gigantic. The new code avoids this.
3527 3. The old code would sometimes allocate too little so that few cells
3528 were freed at GC so that, in turn, too much time was spent in GC.
3530 4. The old code would often trigger heap allocation several times in a
3531 row. (The new scheme predicts how large the segments needs to be
3532 in order not to need further allocation.)
3534 All in all, the new GC policy will make larger applications more
3537 The new GC scheme also is prepared for POSIX threading. Threads can
3538 allocate private pools of cells ("clusters") with just a single
3539 function call. Allocation of single cells from such a cluster can
3540 then proceed without any need of inter-thread synchronization.
3542 ** New environment variables controlling GC parameters
3544 GUILE_MAX_SEGMENT_SIZE Maximal segment size
3547 Allocation of 2-word cell heaps:
3549 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
3552 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
3553 GC in percent of total heap size
3556 Allocation of 4-word cell heaps
3557 (used for real numbers and misc other objects):
3559 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
3561 (See entry "Way for application to customize GC parameters" under
3562 section "Changes to the scm_ interface" below.)
3564 ** Guile now implements reals using 4-word cells
3566 This speeds up computation with reals. (They were earlier allocated
3567 with `malloc'.) There is still some room for optimizations, however.
3569 ** Some further steps toward POSIX thread support have been taken
3571 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
3572 don't have much effect any longer, and many of them will be removed in
3576 are only handled at the top of the evaluator loop, immediately after
3577 I/O, and in scm_equalp.
3579 *** The GC can allocate thread private pools of pairs.
3581 * Changes to Scheme functions and syntax
3583 ** close-input-port and close-output-port are now R5RS
3585 These procedures have been turned into primitives and have R5RS behaviour.
3587 ** New procedure: simple-format PORT MESSAGE ARG1 ...
3589 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
3590 extended by the more sophisticated version in (ice-9 format)
3592 (simple-format port message . args)
3593 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
3594 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
3595 the escapes are replaced with corresponding members of ARGS:
3596 ~A formats using `display' and ~S formats using `write'.
3597 If DESTINATION is #t, then use the `current-output-port',
3598 if DESTINATION is #f, then return a string containing the formatted text.
3599 Does not add a trailing newline."
3601 ** string-ref: the second argument is no longer optional.
3603 ** string, list->string: no longer accept strings in their arguments,
3604 only characters, for compatibility with R5RS.
3606 ** New procedure: port-closed? PORT
3607 Returns #t if PORT is closed or #f if it is open.
3609 ** Deprecated: list*
3611 The list* functionality is now provided by cons* (SRFI-1 compliant)
3613 ** New procedure: cons* ARG1 ARG2 ... ARGn
3615 Like `list', but the last arg provides the tail of the constructed list,
3616 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3618 Requires at least one argument. If given one argument, that argument
3619 is returned as result.
3621 This function is called `list*' in some other Schemes and in Common LISP.
3623 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3625 ** New procedure: object-documentation OBJECT
3627 Returns the documentation string associated with OBJECT. The
3628 procedure uses a caching mechanism so that subsequent lookups are
3631 Exported by (ice-9 documentation).
3633 ** module-name now returns full names of modules
3635 Previously, only the last part of the name was returned (`session' for
3636 `(ice-9 session)'). Ex: `(ice-9 session)'.
3638 * Changes to the gh_ interface
3640 ** Deprecated: gh_int2scmb
3642 Use gh_bool2scm instead.
3644 * Changes to the scm_ interface
3646 ** Guile primitives now carry docstrings!
3648 Thanks to Greg Badros!
3650 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3652 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3653 macros and must contain a docstring that is extracted into foo.doc using a new
3654 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3656 However, a major overhaul of these macros is scheduled for the next release of
3659 ** Guile primitives use a new technique for validation of arguments
3661 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3662 the readability of argument checking.
3664 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3666 ** New macros: SCM_PACK, SCM_UNPACK
3668 Compose/decompose an SCM value.
3670 The SCM type is now treated as an abstract data type and may be defined as a
3671 long, a void* or as a struct, depending on the architecture and compile time
3672 options. This makes it easier to find several types of bugs, for example when
3673 SCM values are treated as integers without conversion. Values of the SCM type
3674 should be treated as "atomic" values. These macros are used when
3675 composing/decomposing an SCM value, either because you want to access
3676 individual bits, or because you want to treat it as an integer value.
3678 E.g., in order to set bit 7 in an SCM value x, use the expression
3680 SCM_PACK (SCM_UNPACK (x) | 0x80)
3682 ** The name property of hooks is deprecated.
3683 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3685 You can emulate this feature by using object properties.
3687 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3688 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3691 These macros will be removed in a future release of Guile.
3693 ** The following types, functions and macros from numbers.h are deprecated:
3694 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3695 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3697 ** Port internals: the rw_random variable in the scm_port structure
3698 must be set to non-zero in any random access port. In recent Guile
3699 releases it was only set for bidirectional random-access ports.
3701 ** Port internals: the seek ptob procedure is now responsible for
3702 resetting the buffers if required. The change was made so that in the
3703 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3704 the fport and strport ptobs can avoid resetting the buffers,
3705 in particular to avoid discarding unread chars. An existing port
3706 type can be fixed by adding something like the following to the
3707 beginning of the ptob seek procedure:
3709 if (pt->rw_active == SCM_PORT_READ)
3710 scm_end_input (object);
3711 else if (pt->rw_active == SCM_PORT_WRITE)
3712 ptob->flush (object);
3714 although to actually avoid resetting the buffers and discard unread
3715 chars requires further hacking that depends on the characteristics
3718 ** Deprecated functions: scm_fseek, scm_tag
3720 These functions are no longer used and will be removed in a future version.
3722 ** The scm_sysmissing procedure is no longer used in libguile.
3723 Unless it turns out to be unexpectedly useful to somebody, it will be
3724 removed in a future version.
3726 ** The format of error message strings has changed
3728 The two C procedures: scm_display_error and scm_error, as well as the
3729 primitive `scm-error', now use scm_simple_format to do their work.
3730 This means that the message strings of all code must be updated to use
3731 ~A where %s was used before, and ~S where %S was used before.
3733 During the period when there still are a lot of old Guiles out there,
3734 you might want to support both old and new versions of Guile.
3736 There are basically two methods to achieve this. Both methods use
3739 AC_CHECK_FUNCS(scm_simple_format)
3741 in your configure.in.
3743 Method 1: Use the string concatenation features of ANSI C's
3748 #ifdef HAVE_SCM_SIMPLE_FORMAT
3754 Then represent each of your error messages using a preprocessor macro:
3756 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3760 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3761 (define make-message string-append)
3763 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3765 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3769 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3774 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3778 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3780 Don't use the functions coop_mutex_init and
3781 coop_condition_variable_init. They will change.
3783 Use scm_mutex_init and scm_cond_init instead.
3785 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3786 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3787 COND, as `scm_cond_wait' does, but it also bounds the duration
3788 of the wait. If COND has not been signaled before time ABSTIME,
3789 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3790 returns the error code `ETIMEDOUT'.
3792 The ABSTIME parameter specifies an absolute time, with the same
3793 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3794 to 00:00:00 GMT, January 1, 1970.
3796 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3797 `scm_cond_broadcast' restarts all the threads that are waiting
3798 on the condition variable COND. Nothing happens if no threads are
3801 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3802 `scm_key_create' allocates a new TSD key. The key is stored in
3803 the location pointed to by KEY. There is no limit on the number
3804 of keys allocated at a given time. The value initially associated
3805 with the returned key is `NULL' in all currently executing threads.
3807 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3808 function associated with the key. When a thread terminates,
3809 DESTR_FUNCTION is called on the value associated with the key in
3810 that thread. The DESTR_FUNCTION is not called if a key is deleted
3811 with `scm_key_delete' or a value is changed with
3812 `scm_setspecific'. The order in which destructor functions are
3813 called at thread termination time is unspecified.
3815 Destructors are not yet implemented.
3817 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3818 `scm_setspecific' changes the value associated with KEY in the
3819 calling thread, storing the given POINTER instead.
3821 ** New function: scm_getspecific (scm_key_t KEY)
3822 `scm_getspecific' returns the value currently associated with
3823 KEY in the calling thread.
3825 ** New function: scm_key_delete (scm_key_t KEY)
3826 `scm_key_delete' deallocates a TSD key. It does not check
3827 whether non-`NULL' values are associated with that key in the
3828 currently executing threads, nor call the destructor function
3829 associated with the key.
3831 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3833 Initialize a C level hook HOOK with associated HOOK_DATA and type
3834 TYPE. (See scm_c_hook_run ().)
3836 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3838 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3839 is true, add it last, otherwise first. The same FUNC can be added
3840 multiple times if FUNC_DATA differ and vice versa.
3842 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3844 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3845 function is only removed if both FUNC and FUNC_DATA matches.
3847 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3849 Run hook HOOK passing DATA to the hook functions.
3851 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3852 returned is undefined.
3854 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3855 returns a non-NULL value. This value is returned as the result of
3856 scm_c_hook_run. If all functions return NULL, NULL is returned.
3858 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3859 returns a NULL value, and NULL is returned. If all functions returns
3860 a non-NULL value, the last value is returned.
3862 ** New C level GC hooks
3864 Five new C level hooks has been added to the garbage collector.
3866 scm_before_gc_c_hook
3869 are run before locking and after unlocking the heap. The system is
3870 thus in a mode where evaluation can take place. (Except that
3871 scm_before_gc_c_hook must not allocate new cells.)
3873 scm_before_mark_c_hook
3874 scm_before_sweep_c_hook
3875 scm_after_sweep_c_hook
3877 are run when the heap is locked. These are intended for extension of
3878 the GC in a modular fashion. Examples are the weaks and guardians
3881 ** Way for application to customize GC parameters
3883 The application can set up other default values for the GC heap
3884 allocation parameters
3886 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3887 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3888 GUILE_MAX_SEGMENT_SIZE,
3892 scm_default_init_heap_size_1, scm_default_min_yield_1,
3893 scm_default_init_heap_size_2, scm_default_min_yield_2,
3894 scm_default_max_segment_size
3896 respectively before callong scm_boot_guile.
3898 (See entry "New environment variables ..." in section
3899 "Changes to the stand-alone interpreter" above.)
3901 ** scm_protect_object/scm_unprotect_object now nest
3903 This means that you can call scm_protect_object multiple times on an
3904 object and count on the object being protected until
3905 scm_unprotect_object has been call the same number of times.
3907 The functions also have better time complexity.
3909 Still, it is usually possible to structure the application in a way
3910 that you don't need to use these functions. For example, if you use a
3911 protected standard Guile list to keep track of live objects rather
3912 than some custom data type, objects will die a natural death when they
3913 are no longer needed.
3915 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3917 Guile does not provide the float representation for inexact real numbers any
3918 more. Now, only doubles are used to represent inexact real numbers. Further,
3919 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3920 and scm_tc16_complex, respectively.
3922 ** Removed deprecated type scm_smobfuns
3924 ** Removed deprecated function scm_newsmob
3926 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3928 There is an ongoing discussion among the developers whether to
3929 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3930 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3931 until this issue has been settled.
3933 ** Removed deprecated type tag scm_tc16_kw
3935 ** Added type tag scm_tc16_keyword
3937 (This was introduced already in release 1.3.4 but was not documented
3940 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3942 * Changes to system call interfaces:
3944 ** The "select" procedure now tests port buffers for the ability to
3945 provide input or accept output. Previously only the underlying file
3946 descriptors were checked.
3948 ** New variable PIPE_BUF: the maximum number of bytes that can be
3949 atomically written to a pipe.
3951 ** If a facility is not available on the system when Guile is
3952 compiled, the corresponding primitive procedure will not be defined.
3953 Previously it would have been defined but would throw a system-error
3954 exception if called. Exception handlers which catch this case may
3955 need minor modification: an error will be thrown with key
3956 'unbound-variable instead of 'system-error. Alternatively it's
3957 now possible to use `defined?' to check whether the facility is
3960 ** Procedures which depend on the timezone should now give the correct
3961 result on systems which cache the TZ environment variable, even if TZ
3962 is changed without calling tzset.
3964 * Changes to the networking interfaces:
3966 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3967 long integers between network and host format. For now, it's not
3968 particularly convenient to do this kind of thing, but consider:
3970 (define write-network-long
3971 (lambda (value port)
3972 (let ((v (make-uniform-vector 1 1 0)))
3973 (uniform-vector-set! v 0 (htonl value))
3974 (uniform-vector-write v port))))
3976 (define read-network-long
3978 (let ((v (make-uniform-vector 1 1 0)))
3979 (uniform-vector-read! v port)
3980 (ntohl (uniform-vector-ref v 0)))))
3982 ** If inet-aton fails, it now throws an error with key 'misc-error
3983 instead of 'system-error, since errno is not relevant.
3985 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3986 specific keys instead of 'system-error. The latter is inappropriate
3987 since errno will not have been set. The keys are:
3988 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3990 ** sethostent, setnetent, setprotoent, setservent: now take an
3991 optional argument STAYOPEN, which specifies whether the database
3992 remains open after a database entry is accessed randomly (e.g., using
3993 gethostbyname for the hosts database.) The default is #f. Previously
3997 Changes since Guile 1.3.2:
3999 * Changes to the stand-alone interpreter
4003 An initial version of the Guile debugger written by Chris Hanson has
4004 been added. The debugger is still under development but is included
4005 in the distribution anyway since it is already quite useful.
4011 after an error to enter the debugger. Type `help' inside the debugger
4012 for a description of available commands.
4014 If you prefer to have stack frames numbered and printed in
4015 anti-chronological order and prefer up in the stack to be down on the
4016 screen as is the case in gdb, you can put
4018 (debug-enable 'backwards)
4020 in your .guile startup file. (However, this means that Guile can't
4021 use indentation to indicate stack level.)
4023 The debugger is autoloaded into Guile at the first use.
4025 ** Further enhancements to backtraces
4027 There is a new debug option `width' which controls the maximum width
4028 on the screen of printed stack frames. Fancy printing parameters
4029 ("level" and "length" as in Common LISP) are adaptively adjusted for
4030 each stack frame to give maximum information while still fitting
4031 within the bounds. If the stack frame can't be made to fit by
4032 adjusting parameters, it is simply cut off at the end. This is marked
4035 ** Some modules are now only loaded when the repl is started
4037 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
4038 regex) are now loaded into (guile-user) only if the repl has been
4039 started. The effect is that the startup time for scripts has been
4040 reduced to 30% of what it was previously.
4042 Correctly written scripts load the modules they require at the top of
4043 the file and should not be affected by this change.
4045 ** Hooks are now represented as smobs
4047 * Changes to Scheme functions and syntax
4049 ** Readline support has changed again.
4051 The old (readline-activator) module is gone. Use (ice-9 readline)
4052 instead, which now contains all readline functionality. So the code
4053 to activate readline is now
4055 (use-modules (ice-9 readline))
4058 This should work at any time, including from the guile prompt.
4060 To avoid confusion about the terms of Guile's license, please only
4061 enable readline for your personal use; please don't make it the
4062 default for others. Here is why we make this rather odd-sounding
4065 Guile is normally licensed under a weakened form of the GNU General
4066 Public License, which allows you to link code with Guile without
4067 placing that code under the GPL. This exception is important to some
4070 However, since readline is distributed under the GNU General Public
4071 License, when you link Guile with readline, either statically or
4072 dynamically, you effectively change Guile's license to the strict GPL.
4073 Whenever you link any strictly GPL'd code into Guile, uses of Guile
4074 which are normally permitted become forbidden. This is a rather
4075 non-obvious consequence of the licensing terms.
4077 So, to make sure things remain clear, please let people choose for
4078 themselves whether to link GPL'd libraries like readline with Guile.
4080 ** regexp-substitute/global has changed slightly, but incompatibly.
4082 If you include a function in the item list, the string of the match
4083 object it receives is the same string passed to
4084 regexp-substitute/global, not some suffix of that string.
4085 Correspondingly, the match's positions are relative to the entire
4086 string, not the suffix.
4088 If the regexp can match the empty string, the way matches are chosen
4089 from the string has changed. regexp-substitute/global recognizes the
4090 same set of matches that list-matches does; see below.
4092 ** New function: list-matches REGEXP STRING [FLAGS]
4094 Return a list of match objects, one for every non-overlapping, maximal
4095 match of REGEXP in STRING. The matches appear in left-to-right order.
4096 list-matches only reports matches of the empty string if there are no
4097 other matches which begin on, end at, or include the empty match's
4100 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4102 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
4104 For each match of REGEXP in STRING, apply PROC to the match object,
4105 and the last value PROC returned, or INIT for the first call. Return
4106 the last value returned by PROC. We apply PROC to the matches as they
4107 appear from left to right.
4109 This function recognizes matches according to the same criteria as
4112 Thus, you could define list-matches like this:
4114 (define (list-matches regexp string . flags)
4115 (reverse! (apply fold-matches regexp string '() cons flags)))
4117 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
4121 *** New function: hook? OBJ
4123 Return #t if OBJ is a hook, otherwise #f.
4125 *** New function: make-hook-with-name NAME [ARITY]
4127 Return a hook with name NAME and arity ARITY. The default value for
4128 ARITY is 0. The only effect of NAME is that it will appear when the
4129 hook object is printed to ease debugging.
4131 *** New function: hook-empty? HOOK
4133 Return #t if HOOK doesn't contain any procedures, otherwise #f.
4135 *** New function: hook->list HOOK
4137 Return a list of the procedures that are called when run-hook is
4140 ** `map' signals an error if its argument lists are not all the same length.
4142 This is the behavior required by R5RS, so this change is really a bug
4143 fix. But it seems to affect a lot of people's code, so we're
4144 mentioning it here anyway.
4146 ** Print-state handling has been made more transparent
4148 Under certain circumstances, ports are represented as a port with an
4149 associated print state. Earlier, this pair was represented as a pair
4150 (see "Some magic has been added to the printer" below). It is now
4151 indistinguishable (almost; see `get-print-state') from a port on the
4154 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
4156 Return a new port with the associated print state PRINT-STATE.
4158 *** New function: get-print-state OUTPUT-PORT
4160 Return the print state associated with this port if it exists,
4161 otherwise return #f.
4163 *** New function: directory-stream? OBJECT
4165 Returns true iff OBJECT is a directory stream --- the sort of object
4166 returned by `opendir'.
4168 ** New function: using-readline?
4170 Return #t if readline is in use in the current repl.
4172 ** structs will be removed in 1.4
4174 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
4175 and use GOOPS objects as the fundamental record type.
4177 * Changes to the scm_ interface
4179 ** structs will be removed in 1.4
4181 The entire current struct interface (struct.c, struct.h) will be
4182 replaced in Guile 1.4. We will merge GOOPS into libguile and use
4183 GOOPS objects as the fundamental record type.
4185 ** The internal representation of subr's has changed
4187 Instead of giving a hint to the subr name, the CAR field of the subr
4188 now contains an index to a subr entry in scm_subr_table.
4190 *** New variable: scm_subr_table
4192 An array of subr entries. A subr entry contains the name, properties
4193 and documentation associated with the subr. The properties and
4194 documentation slots are not yet used.
4196 ** A new scheme for "forwarding" calls to a builtin to a generic function
4198 It is now possible to extend the functionality of some Guile
4199 primitives by letting them defer a call to a GOOPS generic function on
4200 argument mismatch. This means that there is no loss of efficiency in
4205 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
4206 (define-method + ((x <string>) (y <string>))
4207 (string-append x y))
4209 + will still be as efficient as usual in numerical calculations, but
4210 can also be used for concatenating strings.
4212 Who will be the first one to extend Guile's numerical tower to
4213 rationals? :) [OK, there a few other things to fix before this can
4214 be made in a clean way.]
4216 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
4218 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
4220 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
4222 These do the same job as SCM_PROC and SCM_PROC1, but they also define
4223 a variable GENERIC which can be used by the dispatch macros below.
4225 [This is experimental code which may change soon.]
4227 *** New macros for forwarding control to a generic on arg type error
4229 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
4231 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
4233 These correspond to the scm_wta function call, and have the same
4234 behaviour until the user has called the GOOPS primitive
4235 `enable-primitive-generic!'. After that, these macros will apply the
4236 generic function GENERIC to the argument(s) instead of calling
4239 [This is experimental code which may change soon.]
4241 *** New macros for argument testing with generic dispatch
4243 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
4245 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
4247 These correspond to the SCM_ASSERT macro, but will defer control to
4248 GENERIC on error after `enable-primitive-generic!' has been called.
4250 [This is experimental code which may change soon.]
4252 ** New function: SCM scm_eval_body (SCM body, SCM env)
4254 Evaluates the body of a special form.
4256 ** The internal representation of struct's has changed
4258 Previously, four slots were allocated for the procedure(s) of entities
4259 and operators. The motivation for this representation had to do with
4260 the structure of the evaluator, the wish to support tail-recursive
4261 generic functions, and efficiency. Since the generic function
4262 dispatch mechanism has changed, there is no longer a need for such an
4263 expensive representation, and the representation has been simplified.
4265 This should not make any difference for most users.
4267 ** GOOPS support has been cleaned up.
4269 Some code has been moved from eval.c to objects.c and code in both of
4270 these compilation units has been cleaned up and better structured.
4272 *** New functions for applying generic functions
4274 New function: SCM scm_apply_generic (GENERIC, ARGS)
4275 New function: SCM scm_call_generic_0 (GENERIC)
4276 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
4277 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
4278 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
4280 ** Deprecated function: scm_make_named_hook
4282 It is now replaced by:
4284 ** New function: SCM scm_create_hook (const char *name, int arity)
4286 Creates a hook in the same way as make-hook above but also
4287 binds a variable named NAME to it.
4289 This is the typical way of creating a hook from C code.
4291 Currently, the variable is created in the "current" module.
4292 This might change when we get the new module system.
4294 [The behaviour is identical to scm_make_named_hook.]
4298 Changes since Guile 1.3:
4300 * Changes to mailing lists
4302 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
4304 See the README file to find current addresses for all the Guile
4307 * Changes to the distribution
4309 ** Readline support is no longer included with Guile by default.
4311 Based on the different license terms of Guile and Readline, we
4312 concluded that Guile should not *by default* cause the linking of
4313 Readline into an application program. Readline support is now offered
4314 as a separate module, which is linked into an application only when
4315 you explicitly specify it.
4317 Although Guile is GNU software, its distribution terms add a special
4318 exception to the usual GNU General Public License (GPL). Guile's
4319 license includes a clause that allows you to link Guile with non-free
4320 programs. We add this exception so as not to put Guile at a
4321 disadvantage vis-a-vis other extensibility packages that support other
4324 In contrast, the GNU Readline library is distributed under the GNU
4325 General Public License pure and simple. This means that you may not
4326 link Readline, even dynamically, into an application unless it is
4327 distributed under a free software license that is compatible the GPL.
4329 Because of this difference in distribution terms, an application that
4330 can use Guile may not be able to use Readline. Now users will be
4331 explicitly offered two independent decisions about the use of these
4334 You can activate the readline support by issuing
4336 (use-modules (readline-activator))
4339 from your ".guile" file, for example.
4341 * Changes to the stand-alone interpreter
4343 ** All builtins now print as primitives.
4344 Previously builtin procedures not belonging to the fundamental subr
4345 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
4346 Now, they print as #<primitive-procedure NAME>.
4348 ** Backtraces slightly more intelligible.
4349 gsubr-apply and macro transformer application frames no longer appear
4352 * Changes to Scheme functions and syntax
4354 ** Guile now correctly handles internal defines by rewriting them into
4355 their equivalent letrec. Previously, internal defines would
4356 incrementally add to the innermost environment, without checking
4357 whether the restrictions specified in RnRS were met. This lead to the
4358 correct behaviour when these restriction actually were met, but didn't
4359 catch all illegal uses. Such an illegal use could lead to crashes of
4360 the Guile interpreter or or other unwanted results. An example of
4361 incorrect internal defines that made Guile behave erratically:
4373 The problem with this example is that the definition of `c' uses the
4374 value of `b' directly. This confuses the meoization machine of Guile
4375 so that the second call of `b' (this time in a larger environment that
4376 also contains bindings for `c' and `d') refers to the binding of `c'
4377 instead of `a'. You could also make Guile crash with a variation on
4382 (define (b flag) (if flag a 1))
4383 (define c (1+ (b flag)))
4391 From now on, Guile will issue an `Unbound variable: b' error message
4396 A hook contains a list of functions which should be called on
4397 particular occasions in an existing program. Hooks are used for
4400 A window manager might have a hook before-window-map-hook. The window
4401 manager uses the function run-hooks to call all functions stored in
4402 before-window-map-hook each time a window is mapped. The user can
4403 store functions in the hook using add-hook!.
4405 In Guile, hooks are first class objects.
4407 *** New function: make-hook [N_ARGS]
4409 Return a hook for hook functions which can take N_ARGS arguments.
4410 The default value for N_ARGS is 0.
4412 (See also scm_make_named_hook below.)
4414 *** New function: add-hook! HOOK PROC [APPEND_P]
4416 Put PROC at the beginning of the list of functions stored in HOOK.
4417 If APPEND_P is supplied, and non-false, put PROC at the end instead.
4419 PROC must be able to take the number of arguments specified when the
4422 If PROC already exists in HOOK, then remove it first.
4424 *** New function: remove-hook! HOOK PROC
4426 Remove PROC from the list of functions in HOOK.
4428 *** New function: reset-hook! HOOK
4430 Clear the list of hook functions stored in HOOK.
4432 *** New function: run-hook HOOK ARG1 ...
4434 Run all hook functions stored in HOOK with arguments ARG1 ... .
4435 The number of arguments supplied must correspond to the number given
4436 when the hook was created.
4438 ** The function `dynamic-link' now takes optional keyword arguments.
4439 The only keyword argument that is currently defined is `:global
4440 BOOL'. With it, you can control whether the shared library will be
4441 linked in global mode or not. In global mode, the symbols from the
4442 linked library can be used to resolve references from other
4443 dynamically linked libraries. In non-global mode, the linked
4444 library is essentially invisible and can only be accessed via
4445 `dynamic-func', etc. The default is now to link in global mode.
4446 Previously, the default has been non-global mode.
4448 The `#:global' keyword is only effective on platforms that support
4449 the dlopen family of functions.
4451 ** New function `provided?'
4453 - Function: provided? FEATURE
4454 Return true iff FEATURE is supported by this installation of
4455 Guile. FEATURE must be a symbol naming a feature; the global
4456 variable `*features*' is a list of available features.
4458 ** Changes to the module (ice-9 expect):
4460 *** The expect-strings macro now matches `$' in a regular expression
4461 only at a line-break or end-of-file by default. Previously it would
4462 match the end of the string accumulated so far. The old behaviour
4463 can be obtained by setting the variable `expect-strings-exec-flags'
4466 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
4467 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
4468 in a regular expression will still match before a line-break or
4469 end-of-file. The default is `regexp/noteol'.
4471 *** The expect-strings macro now uses a variable
4472 `expect-strings-compile-flags' for the flags to be supplied to
4473 `make-regexp'. The default is `regexp/newline', which was previously
4476 *** The expect macro now supplies two arguments to a match procedure:
4477 the current accumulated string and a flag to indicate whether
4478 end-of-file has been reached. Previously only the string was supplied.
4479 If end-of-file is reached, the match procedure will be called an
4480 additional time with the same accumulated string as the previous call
4481 but with the flag set.
4483 ** New module (ice-9 format), implementing the Common Lisp `format' function.
4485 This code, and the documentation for it that appears here, was
4486 borrowed from SLIB, with minor adaptations for Guile.
4488 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
4489 An almost complete implementation of Common LISP format description
4490 according to the CL reference book `Common LISP' from Guy L.
4491 Steele, Digital Press. Backward compatible to most of the
4492 available Scheme format implementations.
4494 Returns `#t', `#f' or a string; has side effect of printing
4495 according to FORMAT-STRING. If DESTINATION is `#t', the output is
4496 to the current output port and `#t' is returned. If DESTINATION
4497 is `#f', a formatted string is returned as the result of the call.
4498 NEW: If DESTINATION is a string, DESTINATION is regarded as the
4499 format string; FORMAT-STRING is then the first argument and the
4500 output is returned as a string. If DESTINATION is a number, the
4501 output is to the current error port if available by the
4502 implementation. Otherwise DESTINATION must be an output port and
4505 FORMAT-STRING must be a string. In case of a formatting error
4506 format returns `#f' and prints a message on the current output or
4507 error port. Characters are output as if the string were output by
4508 the `display' function with the exception of those prefixed by a
4509 tilde (~). For a detailed description of the FORMAT-STRING syntax
4510 please consult a Common LISP format reference manual. For a test
4511 suite to verify this format implementation load `formatst.scm'.
4512 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
4514 Note: `format' is not reentrant, i.e. only one `format'-call may
4515 be executed at a time.
4518 *** Format Specification (Format version 3.0)
4520 Please consult a Common LISP format reference manual for a detailed
4521 description of the format string syntax. For a demonstration of the
4522 implemented directives see `formatst.scm'.
4524 This implementation supports directive parameters and modifiers (`:'
4525 and `@' characters). Multiple parameters must be separated by a comma
4526 (`,'). Parameters can be numerical parameters (positive or negative),
4527 character parameters (prefixed by a quote character (`''), variable
4528 parameters (`v'), number of rest arguments parameter (`#'), empty and
4529 default parameters. Directive characters are case independent. The
4530 general form of a directive is:
4532 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
4534 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
4536 *** Implemented CL Format Control Directives
4538 Documentation syntax: Uppercase characters represent the
4539 corresponding control directive characters. Lowercase characters
4540 represent control directive parameter descriptions.
4543 Any (print as `display' does).
4547 `~MINCOL,COLINC,MINPAD,PADCHARA'
4551 S-expression (print as `write' does).
4555 `~MINCOL,COLINC,MINPAD,PADCHARS'
4561 print number sign always.
4564 print comma separated.
4566 `~MINCOL,PADCHAR,COMMACHARD'
4572 print number sign always.
4575 print comma separated.
4577 `~MINCOL,PADCHAR,COMMACHARX'
4583 print number sign always.
4586 print comma separated.
4588 `~MINCOL,PADCHAR,COMMACHARO'
4594 print number sign always.
4597 print comma separated.
4599 `~MINCOL,PADCHAR,COMMACHARB'
4604 `~N,MINCOL,PADCHAR,COMMACHARR'
4608 print a number as a Roman numeral.
4611 print a number as an "old fashioned" Roman numeral.
4614 print a number as an ordinal English number.
4617 print a number as a cardinal English number.
4622 prints `y' and `ies'.
4625 as `~P but jumps 1 argument backward.'
4628 as `~@P but jumps 1 argument backward.'
4633 prints a character as the reader can understand it (i.e. `#\'
4637 prints a character as emacs does (eg. `^C' for ASCII 03).
4640 Fixed-format floating-point (prints a flonum like MMM.NNN).
4641 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4643 If the number is positive a plus sign is printed.
4646 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4647 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4649 If the number is positive a plus sign is printed.
4652 General floating-point (prints a flonum either fixed or
4654 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4656 If the number is positive a plus sign is printed.
4659 Dollars floating-point (prints a flonum in fixed with signs
4661 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4663 If the number is positive a plus sign is printed.
4666 A sign is always printed and appears before the padding.
4669 The sign appears before the padding.
4677 print newline if not at the beginning of the output line.
4679 prints `~&' and then N-1 newlines.
4684 print N page separators.
4694 newline is ignored, white space left.
4697 newline is left, white space ignored.
4702 relative tabulation.
4708 Indirection (expects indirect arguments as a list).
4710 extracts indirect arguments from format arguments.
4713 Case conversion (converts by `string-downcase').
4715 converts by `string-capitalize'.
4718 converts by `string-capitalize-first'.
4721 converts by `string-upcase'.
4724 Argument Jumping (jumps 1 argument forward).
4726 jumps N arguments forward.
4729 jumps 1 argument backward.
4732 jumps N arguments backward.
4735 jumps to the 0th argument.
4738 jumps to the Nth argument (beginning from 0)
4740 `~[STR0~;STR1~;...~;STRN~]'
4741 Conditional Expression (numerical clause conditional).
4743 take argument from N.
4746 true test conditional.
4749 if-else-then conditional.
4755 default clause follows.
4758 Iteration (args come from the next argument (a list)).
4760 at most N iterations.
4763 args from next arg (a list of lists).
4766 args from the rest of arguments.
4769 args from the rest args (lists).
4780 aborts if N <= M <= K
4782 *** Not Implemented CL Format Control Directives
4785 print `#f' as an empty list (see below).
4788 print `#f' as an empty list (see below).
4794 (sorry I don't understand its semantics completely)
4796 *** Extended, Replaced and Additional Control Directives
4798 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4799 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4800 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4801 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4802 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4803 COMMAWIDTH is the number of characters between two comma
4807 print a R4RS complex number as `~F~@Fi' with passed parameters for
4811 Pretty print formatting of an argument for scheme code lists.
4817 Flushes the output if format DESTINATION is a port.
4820 Print a `#\space' character
4822 print N `#\space' characters.
4825 Print a `#\tab' character
4827 print N `#\tab' characters.
4830 Takes N as an integer representation for a character. No arguments
4831 are consumed. N is converted to a character by `integer->char'. N
4832 must be a positive decimal number.
4835 Print out readproof. Prints out internal objects represented as
4836 `#<...>' as strings `"#<...>"' so that the format output can always
4837 be processed by `read'.
4840 Print out readproof. Prints out internal objects represented as
4841 `#<...>' as strings `"#<...>"' so that the format output can always
4842 be processed by `read'.
4845 Prints information and a copyright notice on the format
4848 prints format version.
4851 may also print number strings, i.e. passing a number as a string
4852 and format it accordingly.
4854 *** Configuration Variables
4856 The format module exports some configuration variables to suit the
4857 systems and users needs. There should be no modification necessary for
4858 the configuration that comes with Guile. Format detects automatically
4859 if the running scheme system implements floating point numbers and
4862 format:symbol-case-conv
4863 Symbols are converted by `symbol->string' so the case type of the
4864 printed symbols is implementation dependent.
4865 `format:symbol-case-conv' is a one arg closure which is either
4866 `#f' (no conversion), `string-upcase', `string-downcase' or
4867 `string-capitalize'. (default `#f')
4869 format:iobj-case-conv
4870 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4871 implementation internal objects. (default `#f')
4874 The character prefixing the exponent value in `~E' printing.
4877 *** Compatibility With Other Format Implementations
4883 Downward compatible except for padding support and `~A', `~S',
4884 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4885 `printf' padding support which is completely replaced by the CL
4886 `format' padding style.
4889 Downward compatible except for `~', which is not documented
4890 (ignores all characters inside the format string up to a newline
4891 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4892 numerical and variable parameters and `:/@' modifiers in the CL
4896 Downward compatible except for `~A' and `~S' which print in
4897 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4898 directive parameters or modifiers)).
4901 Downward compatible except for an optional destination parameter:
4902 S2C accepts a format call without a destination which returns a
4903 formatted string. This is equivalent to a #f destination in S2C.
4904 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4905 parameters or modifiers)).
4908 ** Changes to string-handling functions.
4910 These functions were added to support the (ice-9 format) module, above.
4912 *** New function: string-upcase STRING
4913 *** New function: string-downcase STRING
4915 These are non-destructive versions of the existing string-upcase! and
4916 string-downcase! functions.
4918 *** New function: string-capitalize! STRING
4919 *** New function: string-capitalize STRING
4921 These functions convert the first letter of each word in the string to
4924 (string-capitalize "howdy there")
4927 As with the other functions, string-capitalize! modifies the string in
4928 place, while string-capitalize returns a modified copy of its argument.
4930 *** New function: string-ci->symbol STRING
4932 Return a symbol whose name is STRING, but having the same case as if
4933 the symbol had be read by `read'.
4935 Guile can be configured to be sensitive or insensitive to case
4936 differences in Scheme identifiers. If Guile is case-insensitive, all
4937 symbols are converted to lower case on input. The `string-ci->symbol'
4938 function returns a symbol whose name in STRING, transformed as Guile
4939 would if STRING were input.
4941 *** New function: substring-move! STRING1 START END STRING2 START
4943 Copy the substring of STRING1 from START (inclusive) to END
4944 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4945 string, and the source and destination areas may overlap; in all
4946 cases, the function behaves as if all the characters were copied
4949 *** Extended functions: substring-move-left! substring-move-right!
4951 These functions now correctly copy arbitrarily overlapping substrings;
4952 they are both synonyms for substring-move!.
4955 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4957 getopt-long is a function for parsing command-line arguments in a
4958 manner consistent with other GNU programs.
4960 (getopt-long ARGS GRAMMAR)
4961 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4963 ARGS should be a list of strings. Its first element should be the
4964 name of the program; subsequent elements should be the arguments
4965 that were passed to the program on the command line. The
4966 `program-arguments' procedure returns a list of this form.
4968 GRAMMAR is a list of the form:
4969 ((OPTION (PROPERTY VALUE) ...) ...)
4971 Each OPTION should be a symbol. `getopt-long' will accept a
4972 command-line option named `--OPTION'.
4973 Each option can have the following (PROPERTY VALUE) pairs:
4975 (single-char CHAR) --- Accept `-CHAR' as a single-character
4976 equivalent to `--OPTION'. This is how to specify traditional
4978 (required? BOOL) --- If BOOL is true, the option is required.
4979 getopt-long will raise an error if it is not found in ARGS.
4980 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4981 it is #f, it does not; and if it is the symbol
4982 `optional', the option may appear in ARGS with or
4984 (predicate FUNC) --- If the option accepts a value (i.e. you
4985 specified `(value #t)' for this option), then getopt
4986 will apply FUNC to the value, and throw an exception
4987 if it returns #f. FUNC should be a procedure which
4988 accepts a string and returns a boolean value; you may
4989 need to use quasiquotes to get it into GRAMMAR.
4991 The (PROPERTY VALUE) pairs may occur in any order, but each
4992 property may occur only once. By default, options do not have
4993 single-character equivalents, are not required, and do not take
4996 In ARGS, single-character options may be combined, in the usual
4997 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4998 accepts values, then it must be the last option in the
4999 combination; the value is the next argument. So, for example, using
5000 the following grammar:
5001 ((apples (single-char #\a))
5002 (blimps (single-char #\b) (value #t))
5003 (catalexis (single-char #\c) (value #t)))
5004 the following argument lists would be acceptable:
5005 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
5006 for "blimps" and "catalexis")
5007 ("-ab" "bang" "-c" "couth") (same)
5008 ("-ac" "couth" "-b" "bang") (same)
5009 ("-abc" "couth" "bang") (an error, since `-b' is not the
5010 last option in its combination)
5012 If an option's value is optional, then `getopt-long' decides
5013 whether it has a value by looking at what follows it in ARGS. If
5014 the next element is a string, and it does not appear to be an
5015 option itself, then that string is the option's value.
5017 The value of a long option can appear as the next element in ARGS,
5018 or it can follow the option name, separated by an `=' character.
5019 Thus, using the same grammar as above, the following argument lists
5021 ("--apples" "Braeburn" "--blimps" "Goodyear")
5022 ("--apples=Braeburn" "--blimps" "Goodyear")
5023 ("--blimps" "Goodyear" "--apples=Braeburn")
5025 If the option "--" appears in ARGS, argument parsing stops there;
5026 subsequent arguments are returned as ordinary arguments, even if
5027 they resemble options. So, in the argument list:
5028 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
5029 `getopt-long' will recognize the `apples' option as having the
5030 value "Granny Smith", but it will not recognize the `blimp'
5031 option; it will return the strings "--blimp" and "Goodyear" as
5032 ordinary argument strings.
5034 The `getopt-long' function returns the parsed argument list as an
5035 assocation list, mapping option names --- the symbols from GRAMMAR
5036 --- onto their values, or #t if the option does not accept a value.
5037 Unused options do not appear in the alist.
5039 All arguments that are not the value of any option are returned
5040 as a list, associated with the empty list.
5042 `getopt-long' throws an exception if:
5043 - it finds an unrecognized option in ARGS
5044 - a required option is omitted
5045 - an option that requires an argument doesn't get one
5046 - an option that doesn't accept an argument does get one (this can
5047 only happen using the long option `--opt=value' syntax)
5048 - an option predicate fails
5053 `((lockfile-dir (required? #t)
5056 (predicate ,file-is-directory?))
5057 (verbose (required? #f)
5060 (x-includes (single-char #\x))
5061 (rnet-server (single-char #\y)
5062 (predicate ,string?))))
5064 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
5065 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
5067 => ((() "foo1" "-fred" "foo2" "foo3")
5068 (rnet-server . "lamprod")
5069 (x-includes . "/usr/include")
5070 (lockfile-dir . "/tmp")
5073 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
5075 It will be removed in a few releases.
5077 ** New syntax: lambda*
5078 ** New syntax: define*
5079 ** New syntax: define*-public
5080 ** New syntax: defmacro*
5081 ** New syntax: defmacro*-public
5082 Guile now supports optional arguments.
5084 `lambda*', `define*', `define*-public', `defmacro*' and
5085 `defmacro*-public' are identical to the non-* versions except that
5086 they use an extended type of parameter list that has the following BNF
5087 syntax (parentheses are literal, square brackets indicate grouping,
5088 and `*', `+' and `?' have the usual meaning):
5090 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
5091 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
5092 [[#&rest identifier]|[. identifier]]? ) | [identifier]
5094 ext-var-decl ::= identifier | ( identifier expression )
5096 The semantics are best illustrated with the following documentation
5097 and examples for `lambda*':
5100 lambda extended for optional and keyword arguments
5102 lambda* creates a procedure that takes optional arguments. These
5103 are specified by putting them inside brackets at the end of the
5104 paramater list, but before any dotted rest argument. For example,
5105 (lambda* (a b #&optional c d . e) '())
5106 creates a procedure with fixed arguments a and b, optional arguments c
5107 and d, and rest argument e. If the optional arguments are omitted
5108 in a call, the variables for them are unbound in the procedure. This
5109 can be checked with the bound? macro.
5111 lambda* can also take keyword arguments. For example, a procedure
5113 (lambda* (#&key xyzzy larch) '())
5114 can be called with any of the argument lists (#:xyzzy 11)
5115 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
5116 are given as keywords are bound to values.
5118 Optional and keyword arguments can also be given default values
5119 which they take on when they are not present in a call, by giving a
5120 two-item list in place of an optional argument, for example in:
5121 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
5122 foo is a fixed argument, bar is an optional argument with default
5123 value 42, and baz is a keyword argument with default value 73.
5124 Default value expressions are not evaluated unless they are needed
5125 and until the procedure is called.
5127 lambda* now supports two more special parameter list keywords.
5129 lambda*-defined procedures now throw an error by default if a
5130 keyword other than one of those specified is found in the actual
5131 passed arguments. However, specifying #&allow-other-keys
5132 immediately after the kyword argument declarations restores the
5133 previous behavior of ignoring unknown keywords. lambda* also now
5134 guarantees that if the same keyword is passed more than once, the
5135 last one passed is the one that takes effect. For example,
5136 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
5137 #:heads 37 #:tails 42 #:heads 99)
5138 would result in (99 47) being displayed.
5140 #&rest is also now provided as a synonym for the dotted syntax rest
5141 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
5142 all respects to lambda*. This is provided for more similarity to DSSSL,
5143 MIT-Scheme and Kawa among others, as well as for refugees from other
5146 Further documentation may be found in the optargs.scm file itself.
5148 The optional argument module also exports the macros `let-optional',
5149 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
5150 are not documented here because they may be removed in the future, but
5151 full documentation is still available in optargs.scm.
5153 ** New syntax: and-let*
5154 Guile now supports the `and-let*' form, described in the draft SRFI-2.
5156 Syntax: (land* (<clause> ...) <body> ...)
5157 Each <clause> should have one of the following forms:
5158 (<variable> <expression>)
5161 Each <variable> or <bound-variable> should be an identifier. Each
5162 <expression> should be a valid expression. The <body> should be a
5163 possibly empty sequence of expressions, like the <body> of a
5166 Semantics: A LAND* expression is evaluated by evaluating the
5167 <expression> or <bound-variable> of each of the <clause>s from
5168 left to right. The value of the first <expression> or
5169 <bound-variable> that evaluates to a false value is returned; the
5170 remaining <expression>s and <bound-variable>s are not evaluated.
5171 The <body> forms are evaluated iff all the <expression>s and
5172 <bound-variable>s evaluate to true values.
5174 The <expression>s and the <body> are evaluated in an environment
5175 binding each <variable> of the preceding (<variable> <expression>)
5176 clauses to the value of the <expression>. Later bindings
5177 shadow earlier bindings.
5179 Guile's and-let* macro was contributed by Michael Livshin.
5181 ** New sorting functions
5183 *** New function: sorted? SEQUENCE LESS?
5184 Returns `#t' when the sequence argument is in non-decreasing order
5185 according to LESS? (that is, there is no adjacent pair `... x y
5186 ...' for which `(less? y x)').
5188 Returns `#f' when the sequence contains at least one out-of-order
5189 pair. It is an error if the sequence is neither a list nor a
5192 *** New function: merge LIST1 LIST2 LESS?
5193 LIST1 and LIST2 are sorted lists.
5194 Returns the sorted list of all elements in LIST1 and LIST2.
5196 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
5197 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
5198 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
5199 (Here "<" should read "comes before".)
5201 *** New procedure: merge! LIST1 LIST2 LESS?
5202 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
5203 the result. If the code is compiled, and LESS? constructs no new
5204 pairs, no pairs at all will be allocated. The first pair of the
5205 result will be either the first pair of LIST1 or the first pair of
5208 *** New function: sort SEQUENCE LESS?
5209 Accepts either a list or a vector, and returns a new sequence
5210 which is sorted. The new sequence is the same type as the input.
5211 Always `(sorted? (sort sequence less?) less?)'. The original
5212 sequence is not altered in any way. The new sequence shares its
5213 elements with the old one; no elements are copied.
5215 *** New procedure: sort! SEQUENCE LESS
5216 Returns its sorted result in the original boxes. No new storage is
5217 allocated at all. Proper usage: (set! slist (sort! slist <))
5219 *** New function: stable-sort SEQUENCE LESS?
5220 Similar to `sort' but stable. That is, if "equal" elements are
5221 ordered a < b in the original sequence, they will have the same order
5224 *** New function: stable-sort! SEQUENCE LESS?
5225 Similar to `sort!' but stable.
5226 Uses temporary storage when sorting vectors.
5228 *** New functions: sort-list, sort-list!
5229 Added for compatibility with scsh.
5231 ** New built-in random number support
5233 *** New function: random N [STATE]
5234 Accepts a positive integer or real N and returns a number of the
5235 same type between zero (inclusive) and N (exclusive). The values
5236 returned have a uniform distribution.
5238 The optional argument STATE must be of the type produced by
5239 `copy-random-state' or `seed->random-state'. It defaults to the value
5240 of the variable `*random-state*'. This object is used to maintain the
5241 state of the pseudo-random-number generator and is altered as a side
5242 effect of the `random' operation.
5244 *** New variable: *random-state*
5245 Holds a data structure that encodes the internal state of the
5246 random-number generator that `random' uses by default. The nature
5247 of this data structure is implementation-dependent. It may be
5248 printed out and successfully read back in, but may or may not
5249 function correctly as a random-number state object in another
5252 *** New function: copy-random-state [STATE]
5253 Returns a new object of type suitable for use as the value of the
5254 variable `*random-state*' and as a second argument to `random'.
5255 If argument STATE is given, a copy of it is returned. Otherwise a
5256 copy of `*random-state*' is returned.
5258 *** New function: seed->random-state SEED
5259 Returns a new object of type suitable for use as the value of the
5260 variable `*random-state*' and as a second argument to `random'.
5261 SEED is a string or a number. A new state is generated and
5262 initialized using SEED.
5264 *** New function: random:uniform [STATE]
5265 Returns an uniformly distributed inexact real random number in the
5266 range between 0 and 1.
5268 *** New procedure: random:solid-sphere! VECT [STATE]
5269 Fills VECT with inexact real random numbers the sum of whose
5270 squares is less than 1.0. Thinking of VECT as coordinates in
5271 space of dimension N = `(vector-length VECT)', the coordinates are
5272 uniformly distributed within the unit N-shere. The sum of the
5273 squares of the numbers is returned. VECT can be either a vector
5274 or a uniform vector of doubles.
5276 *** New procedure: random:hollow-sphere! VECT [STATE]
5277 Fills VECT with inexact real random numbers the sum of whose squares
5278 is equal to 1.0. Thinking of VECT as coordinates in space of
5279 dimension n = `(vector-length VECT)', the coordinates are uniformly
5280 distributed over the surface of the unit n-shere. VECT can be either
5281 a vector or a uniform vector of doubles.
5283 *** New function: random:normal [STATE]
5284 Returns an inexact real in a normal distribution with mean 0 and
5285 standard deviation 1. For a normal distribution with mean M and
5286 standard deviation D use `(+ M (* D (random:normal)))'.
5288 *** New procedure: random:normal-vector! VECT [STATE]
5289 Fills VECT with inexact real random numbers which are independent and
5290 standard normally distributed (i.e., with mean 0 and variance 1).
5291 VECT can be either a vector or a uniform vector of doubles.
5293 *** New function: random:exp STATE
5294 Returns an inexact real in an exponential distribution with mean 1.
5295 For an exponential distribution with mean U use (* U (random:exp)).
5297 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
5299 These functions now operate on numbers in the range of a C unsigned
5302 These functions used to operate on numbers in the range of a C signed
5303 long; however, this seems inappropriate, because Guile integers don't
5306 ** New function: make-guardian
5307 This is an implementation of guardians as described in
5308 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
5309 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
5310 Programming Language Design and Implementation, June 1993
5311 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
5313 ** New functions: delq1!, delv1!, delete1!
5314 These procedures behave similar to delq! and friends but delete only
5315 one object if at all.
5317 ** New function: unread-string STRING PORT
5318 Unread STRING to PORT, that is, push it back onto the port so that
5319 next read operation will work on the pushed back characters.
5321 ** unread-char can now be called multiple times
5322 If unread-char is called multiple times, the unread characters will be
5323 read again in last-in first-out order.
5325 ** the procedures uniform-array-read! and uniform-array-write! now
5326 work on any kind of port, not just ports which are open on a file.
5328 ** Now 'l' in a port mode requests line buffering.
5330 ** The procedure truncate-file now works on string ports as well
5331 as file ports. If the size argument is omitted, the current
5332 file position is used.
5334 ** new procedure: seek PORT/FDES OFFSET WHENCE
5335 The arguments are the same as for the old fseek procedure, but it
5336 works on string ports as well as random-access file ports.
5338 ** the fseek procedure now works on string ports, since it has been
5339 redefined using seek.
5341 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
5342 size is not supplied.
5344 ** the newline procedure no longer flushes the port if it's not
5345 line-buffered: previously it did if it was the current output port.
5347 ** open-pipe and close-pipe are no longer primitive procedures, but
5348 an emulation can be obtained using `(use-modules (ice-9 popen))'.
5350 ** the freopen procedure has been removed.
5352 ** new procedure: drain-input PORT
5353 Drains PORT's read buffers (including any pushed-back characters)
5354 and returns the contents as a single string.
5356 ** New function: map-in-order PROC LIST1 LIST2 ...
5357 Version of `map' which guarantees that the procedure is applied to the
5358 lists in serial order.
5360 ** Renamed `serial-array-copy!' and `serial-array-map!' to
5361 `array-copy-in-order!' and `array-map-in-order!'. The old names are
5362 now obsolete and will go away in release 1.5.
5364 ** New syntax: collect BODY1 ...
5365 Version of `begin' which returns a list of the results of the body
5366 forms instead of the result of the last body form. In contrast to
5367 `begin', `collect' allows an empty body.
5369 ** New functions: read-history FILENAME, write-history FILENAME
5370 Read/write command line history from/to file. Returns #t on success
5371 and #f if an error occured.
5373 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
5375 These procedures return a list of definitions available in the specified
5376 argument, a relative module reference. In the case of no argument,
5377 `(current-module)' is now consulted for definitions to return, instead
5378 of simply returning #f, the former behavior.
5380 ** The #/ syntax for lists is no longer supported.
5382 Earlier versions of Scheme accepted this syntax, but printed a
5385 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
5387 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
5390 * Changes to the gh_ interface
5394 Now takes a second argument which is the result array. If this
5395 pointer is NULL, a new array is malloced (the old behaviour).
5397 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
5398 gh_scm2shorts, gh_scm2longs, gh_scm2floats
5402 * Changes to the scm_ interface
5404 ** Function: scm_make_named_hook (char* name, int n_args)
5406 Creates a hook in the same way as make-hook above but also
5407 binds a variable named NAME to it.
5409 This is the typical way of creating a hook from C code.
5411 Currently, the variable is created in the "current" module. This
5412 might change when we get the new module system.
5414 ** The smob interface
5416 The interface for creating smobs has changed. For documentation, see
5417 data-rep.info (made from guile-core/doc/data-rep.texi).
5419 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
5421 >>> This function will be removed in 1.3.4. <<<
5425 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
5426 This function adds a new smob type, named NAME, with instance size
5427 SIZE to the system. The return value is a tag that is used in
5428 creating instances of the type. If SIZE is 0, then no memory will
5429 be allocated when instances of the smob are created, and nothing
5430 will be freed by the default free function.
5432 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
5433 This function sets the smob marking procedure for the smob type
5434 specified by the tag TC. TC is the tag returned by
5435 `scm_make_smob_type'.
5437 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
5438 This function sets the smob freeing procedure for the smob type
5439 specified by the tag TC. TC is the tag returned by
5440 `scm_make_smob_type'.
5442 *** Function: void scm_set_smob_print (tc, print)
5444 - Function: void scm_set_smob_print (long tc,
5445 scm_sizet (*print) (SCM,
5449 This function sets the smob printing procedure for the smob type
5450 specified by the tag TC. TC is the tag returned by
5451 `scm_make_smob_type'.
5453 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
5454 This function sets the smob equality-testing predicate for the
5455 smob type specified by the tag TC. TC is the tag returned by
5456 `scm_make_smob_type'.
5458 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
5459 Make VALUE contain a smob instance of the type with type code TC and
5460 smob data DATA. VALUE must be previously declared as C type `SCM'.
5462 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
5463 This macro expands to a block of code that creates a smob instance
5464 of the type with type code TC and smob data DATA, and returns that
5465 `SCM' value. It should be the last piece of code in a block.
5467 ** The interfaces for using I/O ports and implementing port types
5468 (ptobs) have changed significantly. The new interface is based on
5469 shared access to buffers and a new set of ptob procedures.
5471 *** scm_newptob has been removed
5475 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
5477 - Function: SCM scm_make_port_type (char *type_name,
5478 int (*fill_buffer) (SCM port),
5479 void (*write_flush) (SCM port));
5481 Similarly to the new smob interface, there is a set of function
5482 setters by which the user can customize the behaviour of his port
5483 type. See ports.h (scm_set_port_XXX).
5485 ** scm_strport_to_string: New function: creates a new string from
5486 a string port's buffer.
5488 ** Plug in interface for random number generators
5489 The variable `scm_the_rng' in random.c contains a value and three
5490 function pointers which together define the current random number
5491 generator being used by the Scheme level interface and the random
5492 number library functions.
5494 The user is free to replace the default generator with the generator
5497 *** Variable: size_t scm_the_rng.rstate_size
5498 The size of the random state type used by the current RNG
5501 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
5502 Given the random STATE, return 32 random bits.
5504 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
5505 Seed random state STATE using string S of length N.
5507 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
5508 Given random state STATE, return a malloced copy.
5511 The default RNG is the MWC (Multiply With Carry) random number
5512 generator described by George Marsaglia at the Department of
5513 Statistics and Supercomputer Computations Research Institute, The
5514 Florida State University (http://stat.fsu.edu/~geo).
5516 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
5517 passes all tests in the DIEHARD test suite
5518 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
5519 costs one multiply and one add on platforms which either supports long
5520 longs (gcc does this on most systems) or have 64 bit longs. The cost
5521 is four multiply on other systems but this can be optimized by writing
5522 scm_i_uniform32 in assembler.
5524 These functions are provided through the scm_the_rng interface for use
5525 by libguile and the application.
5527 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
5528 Given the random STATE, return 32 random bits.
5529 Don't use this function directly. Instead go through the plugin
5530 interface (see "Plug in interface" above).
5532 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
5533 Initialize STATE using SEED of length N.
5535 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
5536 Return a malloc:ed copy of STATE. This function can easily be re-used
5537 in the interfaces to other RNGs.
5539 ** Random number library functions
5540 These functions use the current RNG through the scm_the_rng interface.
5541 It might be a good idea to use these functions from your C code so
5542 that only one random generator is used by all code in your program.
5544 The default random state is stored in:
5546 *** Variable: SCM scm_var_random_state
5547 Contains the vcell of the Scheme variable "*random-state*" which is
5548 used as default state by all random number functions in the Scheme
5553 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
5555 *** Function: scm_rstate *scm_c_default_rstate (void)
5556 This is a convenience function which returns the value of
5557 scm_var_random_state. An error message is generated if this value
5558 isn't a random state.
5560 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
5561 Make a new random state from the string SEED of length LENGTH.
5563 It is generally not a good idea to use multiple random states in a
5564 program. While subsequent random numbers generated from one random
5565 state are guaranteed to be reasonably independent, there is no such
5566 guarantee for numbers generated from different random states.
5568 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
5569 Return 32 random bits.
5571 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
5572 Return a sample from the uniform(0,1) distribution.
5574 *** Function: double scm_c_normal01 (scm_rstate *STATE)
5575 Return a sample from the normal(0,1) distribution.
5577 *** Function: double scm_c_exp1 (scm_rstate *STATE)
5578 Return a sample from the exp(1) distribution.
5580 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
5581 Return a sample from the discrete uniform(0,M) distribution.
5583 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
5584 Return a sample from the discrete uniform(0,M) distribution.
5585 M must be a bignum object. The returned value may be an INUM.
5589 Changes in Guile 1.3 (released Monday, October 19, 1998):
5591 * Changes to the distribution
5593 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
5594 To avoid conflicts, programs should name environment variables after
5595 themselves, except when there's a common practice establishing some
5598 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5599 giving the former precedence, and printing a warning message if the
5600 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5602 ** The header files related to multi-byte characters have been removed.
5603 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5604 which referred to these explicitly will probably need to be rewritten,
5605 since the support for the variant string types has been removed; see
5608 ** The header files append.h and sequences.h have been removed. These
5609 files implemented non-R4RS operations which would encourage
5610 non-portable programming style and less easy-to-read code.
5612 * Changes to the stand-alone interpreter
5614 ** New procedures have been added to implement a "batch mode":
5616 *** Function: batch-mode?
5618 Returns a boolean indicating whether the interpreter is in batch
5621 *** Function: set-batch-mode?! ARG
5623 If ARG is true, switches the interpreter to batch mode. The `#f'
5624 case has not been implemented.
5626 ** Guile now provides full command-line editing, when run interactively.
5627 To use this feature, you must have the readline library installed.
5628 The Guile build process will notice it, and automatically include
5631 The readline library is available via anonymous FTP from any GNU
5632 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5634 ** the-last-stack is now a fluid.
5636 * Changes to the procedure for linking libguile with your programs
5638 ** You can now use the `guile-config' utility to build programs that use Guile.
5640 Guile now includes a command-line utility called `guile-config', which
5641 can provide information about how to compile and link programs that
5644 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5645 You should include this command's output on the command line you use
5646 to compile C or C++ code that #includes the Guile header files. It's
5647 usually just a `-I' flag to help the compiler find the Guile headers.
5650 *** `guile-config link' prints any linker flags necessary to link with Guile.
5652 This command writes to its standard output a list of flags which you
5653 must pass to the linker to link your code against the Guile library.
5654 The flags include '-lguile' itself, any other libraries the Guile
5655 library depends upon, and any `-L' flags needed to help the linker
5656 find those libraries.
5658 For example, here is a Makefile rule that builds a program named 'foo'
5659 from the object files ${FOO_OBJECTS}, and links them against Guile:
5662 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5664 Previous Guile releases recommended that you use autoconf to detect
5665 which of a predefined set of libraries were present on your system.
5666 It is more robust to use `guile-config', since it records exactly which
5667 libraries the installed Guile library requires.
5669 This was originally called `build-guile', but was renamed to
5670 `guile-config' before Guile 1.3 was released, to be consistent with
5671 the analogous script for the GTK+ GUI toolkit, which is called
5675 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5677 If you are using the GNU autoconf package to configure your program,
5678 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5679 (described above) and gather the necessary values for use in your
5682 The GUILE_FLAGS macro expands to configure script code which runs the
5683 `guile-config' script, to find out where Guile's header files and
5684 libraries are installed. It sets two variables, marked for
5685 substitution, as by AC_SUBST.
5687 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5688 code that uses Guile header files. This is almost always just a
5691 GUILE_LDFLAGS --- flags to pass to the linker to link a
5692 program against Guile. This includes `-lguile' for the Guile
5693 library itself, any libraries that Guile itself requires (like
5694 -lqthreads), and so on. It may also include a -L flag to tell the
5695 compiler where to find the libraries.
5697 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5698 directory of the Guile distribution. You can copy it into your
5699 package's aclocal.m4 file, and then use it in your configure.in file.
5701 If you are using the `aclocal' program, distributed with GNU automake,
5702 to maintain your aclocal.m4 file, the Guile installation process
5703 installs guile.m4 where aclocal will find it. All you need to do is
5704 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5705 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5709 * Changes to Scheme functions and syntax
5711 ** Multi-byte strings have been removed, as have multi-byte and wide
5712 ports. We felt that these were the wrong approach to
5713 internationalization support.
5715 ** New function: readline [PROMPT]
5716 Read a line from the terminal, and allow the user to edit it,
5717 prompting with PROMPT. READLINE provides a large set of Emacs-like
5718 editing commands, lets the user recall previously typed lines, and
5719 works on almost every kind of terminal, including dumb terminals.
5721 READLINE assumes that the cursor is at the beginning of the line when
5722 it is invoked. Thus, you can't print a prompt yourself, and then call
5723 READLINE; you need to package up your prompt as a string, pass it to
5724 the function, and let READLINE print the prompt itself. This is
5725 because READLINE needs to know the prompt's screen width.
5727 For Guile to provide this function, you must have the readline
5728 library, version 2.1 or later, installed on your system. Readline is
5729 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5730 any GNU mirror site.
5732 See also ADD-HISTORY function.
5734 ** New function: add-history STRING
5735 Add STRING as the most recent line in the history used by the READLINE
5736 command. READLINE does not add lines to the history itself; you must
5737 call ADD-HISTORY to make previous input available to the user.
5739 ** The behavior of the read-line function has changed.
5741 This function now uses standard C library functions to read the line,
5742 for speed. This means that it doesn not respect the value of
5743 scm-line-incrementors; it assumes that lines are delimited with
5746 (Note that this is read-line, the function that reads a line of text
5747 from a port, not readline, the function that reads a line from a
5748 terminal, providing full editing capabilities.)
5750 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5752 This module provides some simple argument parsing. It exports one
5755 Function: getopt-gnu-style ARG-LS
5756 Parse a list of program arguments into an alist of option
5759 Each item in the list of program arguments is examined to see if
5760 it meets the syntax of a GNU long-named option. An argument like
5761 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5762 returned alist, where MUMBLE is a keyword object with the same
5763 name as the argument. An argument like `--MUMBLE=FROB' produces
5764 an element of the form (MUMBLE . FROB), where FROB is a string.
5766 As a special case, the returned alist also contains a pair whose
5767 car is the symbol `rest'. The cdr of this pair is a list
5768 containing all the items in the argument list that are not options
5769 of the form mentioned above.
5771 The argument `--' is treated specially: all items in the argument
5772 list appearing after such an argument are not examined, and are
5773 returned in the special `rest' list.
5775 This function does not parse normal single-character switches.
5776 You will need to parse them out of the `rest' list yourself.
5778 ** The read syntax for byte vectors and short vectors has changed.
5780 Instead of #bytes(...), write #y(...).
5782 Instead of #short(...), write #h(...).
5784 This may seem nutty, but, like the other uniform vectors, byte vectors
5785 and short vectors want to have the same print and read syntax (and,
5786 more basic, want to have read syntax!). Changing the read syntax to
5787 use multiple characters after the hash sign breaks with the
5788 conventions used in R5RS and the conventions used for the other
5789 uniform vectors. It also introduces complexity in the current reader,
5790 both on the C and Scheme levels. (The Right solution is probably to
5791 change the syntax and prototypes for uniform vectors entirely.)
5794 ** The new module (ice-9 session) provides useful interactive functions.
5796 *** New procedure: (apropos REGEXP OPTION ...)
5798 Display a list of top-level variables whose names match REGEXP, and
5799 the modules they are imported from. Each OPTION should be one of the
5802 value --- Show the value of each matching variable.
5803 shadow --- Show bindings shadowed by subsequently imported modules.
5804 full --- Same as both `shadow' and `value'.
5808 guile> (apropos "trace" 'full)
5809 debug: trace #<procedure trace args>
5810 debug: untrace #<procedure untrace args>
5811 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5812 the-scm-module: before-backtrace-hook ()
5813 the-scm-module: backtrace #<primitive-procedure backtrace>
5814 the-scm-module: after-backtrace-hook ()
5815 the-scm-module: has-shown-backtrace-hint? #f
5818 ** There are new functions and syntax for working with macros.
5820 Guile implements macros as a special object type. Any variable whose
5821 top-level binding is a macro object acts as a macro. The macro object
5822 specifies how the expression should be transformed before evaluation.
5824 *** Macro objects now print in a reasonable way, resembling procedures.
5826 *** New function: (macro? OBJ)
5827 True iff OBJ is a macro object.
5829 *** New function: (primitive-macro? OBJ)
5830 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5831 macro transformers, implemented in eval.c rather than Scheme code.
5833 Why do we have this function?
5834 - For symmetry with procedure? and primitive-procedure?,
5835 - to allow custom print procedures to tell whether a macro is
5836 primitive, and display it differently, and
5837 - to allow compilers and user-written evaluators to distinguish
5838 builtin special forms from user-defined ones, which could be
5841 *** New function: (macro-type OBJ)
5842 Return a value indicating what kind of macro OBJ is. Possible return
5845 The symbol `syntax' --- a macro created by procedure->syntax.
5846 The symbol `macro' --- a macro created by procedure->macro.
5847 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5848 The boolean #f --- if OBJ is not a macro object.
5850 *** New function: (macro-name MACRO)
5851 Return the name of the macro object MACRO's procedure, as returned by
5854 *** New function: (macro-transformer MACRO)
5855 Return the transformer procedure for MACRO.
5857 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5859 Specify a new macro expander to use in the current module. Each
5860 MODULE is a module name, with the same meaning as in the `use-modules'
5861 form; each named module's exported bindings are added to the current
5862 top-level environment. TRANSFORMER is an expression evaluated in the
5863 resulting environment which must yield a procedure to use as the
5864 module's eval transformer: every expression evaluated in this module
5865 is passed to this function, and the result passed to the Guile
5868 *** macro-eval! is removed. Use local-eval instead.
5870 ** Some magic has been added to the printer to better handle user
5871 written printing routines (like record printers, closure printers).
5873 The problem is that these user written routines must have access to
5874 the current `print-state' to be able to handle fancy things like
5875 detection of circular references. These print-states have to be
5876 passed to the builtin printing routines (display, write, etc) to
5877 properly continue the print chain.
5879 We didn't want to change all existing print code so that it
5880 explicitly passes thru a print state in addition to a port. Instead,
5881 we extented the possible values that the builtin printing routines
5882 accept as a `port'. In addition to a normal port, they now also take
5883 a pair of a normal port and a print-state. Printing will go to the
5884 port and the print-state will be used to control the detection of
5885 circular references, etc. If the builtin function does not care for a
5886 print-state, it is simply ignored.
5888 User written callbacks are now called with such a pair as their
5889 `port', but because every function now accepts this pair as a PORT
5890 argument, you don't have to worry about that. In fact, it is probably
5891 safest to not check for these pairs.
5893 However, it is sometimes necessary to continue a print chain on a
5894 different port, for example to get a intermediate string
5895 representation of the printed value, mangle that string somehow, and
5896 then to finally print the mangled string. Use the new function
5898 inherit-print-state OLD-PORT NEW-PORT
5900 for this. It constructs a new `port' that prints to NEW-PORT but
5901 inherits the print-state of OLD-PORT.
5903 ** struct-vtable-offset renamed to vtable-offset-user
5905 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5907 ** There is now a third optional argument to make-vtable-vtable
5908 (and fourth to make-struct) when constructing new types (vtables).
5909 This argument initializes field vtable-index-printer of the vtable.
5911 ** The detection of circular references has been extended to structs.
5912 That is, a structure that -- in the process of being printed -- prints
5913 itself does not lead to infinite recursion.
5915 ** There is now some basic support for fluids. Please read
5916 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5917 the following functions and macros:
5919 Function: make-fluid
5921 Create a new fluid object. Fluids are not special variables or
5922 some other extension to the semantics of Scheme, but rather
5923 ordinary Scheme objects. You can store them into variables (that
5924 are still lexically scoped, of course) or into any other place you
5925 like. Every fluid has a initial value of `#f'.
5927 Function: fluid? OBJ
5929 Test whether OBJ is a fluid.
5931 Function: fluid-ref FLUID
5932 Function: fluid-set! FLUID VAL
5934 Access/modify the fluid FLUID. Modifications are only visible
5935 within the current dynamic root (that includes threads).
5937 Function: with-fluids* FLUIDS VALUES THUNK
5939 FLUIDS is a list of fluids and VALUES a corresponding list of
5940 values for these fluids. Before THUNK gets called the values are
5941 installed in the fluids and the old values of the fluids are
5942 saved in the VALUES list. When the flow of control leaves THUNK
5943 or reenters it, the values get swapped again. You might think of
5944 this as a `safe-fluid-excursion'. Note that the VALUES list is
5945 modified by `with-fluids*'.
5947 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5949 The same as `with-fluids*' but with a different syntax. It looks
5950 just like `let', but both FLUID and VALUE are evaluated. Remember,
5951 fluids are not special variables but ordinary objects. FLUID
5952 should evaluate to a fluid.
5954 ** Changes to system call interfaces:
5956 *** close-port, close-input-port and close-output-port now return a
5957 boolean instead of an `unspecified' object. #t means that the port
5958 was successfully closed, while #f means it was already closed. It is
5959 also now possible for these procedures to raise an exception if an
5960 error occurs (some errors from write can be delayed until close.)
5962 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5965 *** the third argument to fcntl is now optional.
5967 *** the first argument to chown can now be a file descriptor or a port.
5969 *** the argument to stat can now be a port.
5971 *** The following new procedures have been added (most use scsh
5974 *** procedure: close PORT/FD
5975 Similar to close-port (*note close-port: Closing Ports.), but also
5976 works on file descriptors. A side effect of closing a file
5977 descriptor is that any ports using that file descriptor are moved
5978 to a different file descriptor and have their revealed counts set
5981 *** procedure: port->fdes PORT
5982 Returns the integer file descriptor underlying PORT. As a side
5983 effect the revealed count of PORT is incremented.
5985 *** procedure: fdes->ports FDES
5986 Returns a list of existing ports which have FDES as an underlying
5987 file descriptor, without changing their revealed counts.
5989 *** procedure: fdes->inport FDES
5990 Returns an existing input port which has FDES as its underlying
5991 file descriptor, if one exists, and increments its revealed count.
5992 Otherwise, returns a new input port with a revealed count of 1.
5994 *** procedure: fdes->outport FDES
5995 Returns an existing output port which has FDES as its underlying
5996 file descriptor, if one exists, and increments its revealed count.
5997 Otherwise, returns a new output port with a revealed count of 1.
5999 The next group of procedures perform a `dup2' system call, if NEWFD
6000 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
6001 duplicated can be supplied as an integer or contained in a port. The
6002 type of value returned varies depending on which procedure is used.
6004 All procedures also have the side effect when performing `dup2' that
6005 any ports using NEWFD are moved to a different file descriptor and have
6006 their revealed counts set to zero.
6008 *** procedure: dup->fdes PORT/FD [NEWFD]
6009 Returns an integer file descriptor.
6011 *** procedure: dup->inport PORT/FD [NEWFD]
6012 Returns a new input port using the new file descriptor.
6014 *** procedure: dup->outport PORT/FD [NEWFD]
6015 Returns a new output port using the new file descriptor.
6017 *** procedure: dup PORT/FD [NEWFD]
6018 Returns a new port if PORT/FD is a port, with the same mode as the
6019 supplied port, otherwise returns an integer file descriptor.
6021 *** procedure: dup->port PORT/FD MODE [NEWFD]
6022 Returns a new port using the new file descriptor. MODE supplies a
6023 mode string for the port (*note open-file: File Ports.).
6025 *** procedure: setenv NAME VALUE
6026 Modifies the environment of the current process, which is also the
6027 default environment inherited by child processes.
6029 If VALUE is `#f', then NAME is removed from the environment.
6030 Otherwise, the string NAME=VALUE is added to the environment,
6031 replacing any existing string with name matching NAME.
6033 The return value is unspecified.
6035 *** procedure: truncate-file OBJ SIZE
6036 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
6037 can be a string containing a file name or an integer file
6038 descriptor or port open for output on the file. The underlying
6039 system calls are `truncate' and `ftruncate'.
6041 The return value is unspecified.
6043 *** procedure: setvbuf PORT MODE [SIZE]
6044 Set the buffering mode for PORT. MODE can be:
6052 block buffered, using a newly allocated buffer of SIZE bytes.
6053 However if SIZE is zero or unspecified, the port will be made
6056 This procedure should not be used after I/O has been performed with
6059 Ports are usually block buffered by default, with a default buffer
6060 size. Procedures e.g., *Note open-file: File Ports, which accept a
6061 mode string allow `0' to be added to request an unbuffered port.
6063 *** procedure: fsync PORT/FD
6064 Copies any unwritten data for the specified output file descriptor
6065 to disk. If PORT/FD is a port, its buffer is flushed before the
6066 underlying file descriptor is fsync'd. The return value is
6069 *** procedure: open-fdes PATH FLAGS [MODES]
6070 Similar to `open' but returns a file descriptor instead of a port.
6072 *** procedure: execle PATH ENV [ARG] ...
6073 Similar to `execl', but the environment of the new process is
6074 specified by ENV, which must be a list of strings as returned by
6075 the `environ' procedure.
6077 This procedure is currently implemented using the `execve' system
6078 call, but we call it `execle' because of its Scheme calling
6081 *** procedure: strerror ERRNO
6082 Returns the Unix error message corresponding to ERRNO, an integer.
6084 *** procedure: primitive-exit [STATUS]
6085 Terminate the current process without unwinding the Scheme stack.
6086 This is would typically be useful after a fork. The exit status
6087 is STATUS if supplied, otherwise zero.
6089 *** procedure: times
6090 Returns an object with information about real and processor time.
6091 The following procedures accept such an object as an argument and
6092 return a selected component:
6095 The current real time, expressed as time units relative to an
6099 The CPU time units used by the calling process.
6102 The CPU time units used by the system on behalf of the
6106 The CPU time units used by terminated child processes of the
6107 calling process, whose status has been collected (e.g., using
6111 Similarly, the CPU times units used by the system on behalf of
6112 terminated child processes.
6114 ** Removed: list-length
6115 ** Removed: list-append, list-append!
6116 ** Removed: list-reverse, list-reverse!
6118 ** array-map renamed to array-map!
6120 ** serial-array-map renamed to serial-array-map!
6122 ** catch doesn't take #f as first argument any longer
6124 Previously, it was possible to pass #f instead of a key to `catch'.
6125 That would cause `catch' to pass a jump buffer object to the procedure
6126 passed as second argument. The procedure could then use this jump
6127 buffer objekt as an argument to throw.
6129 This mechanism has been removed since its utility doesn't motivate the
6130 extra complexity it introduces.
6132 ** The `#/' notation for lists now provokes a warning message from Guile.
6133 This syntax will be removed from Guile in the near future.
6135 To disable the warning message, set the GUILE_HUSH environment
6136 variable to any non-empty value.
6138 ** The newline character now prints as `#\newline', following the
6139 normal Scheme notation, not `#\nl'.
6141 * Changes to the gh_ interface
6143 ** The gh_enter function now takes care of loading the Guile startup files.
6144 gh_enter works by calling scm_boot_guile; see the remarks below.
6146 ** Function: void gh_write (SCM x)
6148 Write the printed representation of the scheme object x to the current
6149 output port. Corresponds to the scheme level `write'.
6151 ** gh_list_length renamed to gh_length.
6153 ** vector handling routines
6155 Several major changes. In particular, gh_vector() now resembles
6156 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
6157 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
6158 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
6159 vector-related gh_ functions have been implemented.
6161 ** pair and list routines
6163 Implemented several of the R4RS pair and list functions that were
6166 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
6168 New function. Converts double arrays back and forth between Scheme
6171 * Changes to the scm_ interface
6173 ** The function scm_boot_guile now takes care of loading the startup files.
6175 Guile's primary initialization function, scm_boot_guile, now takes
6176 care of loading `boot-9.scm', in the `ice-9' module, to initialize
6177 Guile, define the module system, and put together some standard
6178 bindings. It also loads `init.scm', which is intended to hold
6179 site-specific initialization code.
6181 Since Guile cannot operate properly until boot-9.scm is loaded, there
6182 is no reason to separate loading boot-9.scm from Guile's other
6183 initialization processes.
6185 This job used to be done by scm_compile_shell_switches, which didn't
6186 make much sense; in particular, it meant that people using Guile for
6187 non-shell-like applications had to jump through hoops to get Guile
6188 initialized properly.
6190 ** The function scm_compile_shell_switches no longer loads the startup files.
6191 Now, Guile always loads the startup files, whenever it is initialized;
6192 see the notes above for scm_boot_guile and scm_load_startup_files.
6194 ** Function: scm_load_startup_files
6195 This new function takes care of loading Guile's initialization file
6196 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
6197 this is always called by the Guile initialization process, it's
6198 probably not too useful to call this yourself, but it's there anyway.
6200 ** The semantics of smob marking have changed slightly.
6202 The smob marking function (the `mark' member of the scm_smobfuns
6203 structure) is no longer responsible for setting the mark bit on the
6204 smob. The generic smob handling code in the garbage collector will
6205 set this bit. The mark function need only ensure that any other
6206 objects the smob refers to get marked.
6208 Note that this change means that the smob's GC8MARK bit is typically
6209 already set upon entry to the mark function. Thus, marking functions
6210 which look like this:
6213 if (SCM_GC8MARKP (ptr))
6215 SCM_SETGC8MARK (ptr);
6216 ... mark objects to which the smob refers ...
6219 are now incorrect, since they will return early, and fail to mark any
6220 other objects the smob refers to. Some code in the Guile library used
6223 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
6225 If you have implemented your own I/O port type, by writing the
6226 functions required by the scm_ptobfuns and then calling scm_newptob,
6227 you will need to change your functions slightly.
6229 The functions in a scm_ptobfuns structure now expect the port itself
6230 as their argument; they used to expect the `stream' member of the
6231 port's scm_port_table structure. This allows functions in an
6232 scm_ptobfuns structure to easily access the port's cell (and any flags
6233 it its CAR), and the port's scm_port_table structure.
6235 Guile now passes the I/O port itself as the `port' argument in the
6236 following scm_ptobfuns functions:
6238 int (*free) (SCM port);
6239 int (*fputc) (int, SCM port);
6240 int (*fputs) (char *, SCM port);
6241 scm_sizet (*fwrite) SCM_P ((char *ptr,
6245 int (*fflush) (SCM port);
6246 int (*fgetc) (SCM port);
6247 int (*fclose) (SCM port);
6249 The interfaces to the `mark', `print', `equalp', and `fgets' methods
6252 If you have existing code which defines its own port types, it is easy
6253 to convert your code to the new interface; simply apply SCM_STREAM to
6254 the port argument to yield the value you code used to expect.
6256 Note that since both the port and the stream have the same type in the
6257 C code --- they are both SCM values --- the C compiler will not remind
6258 you if you forget to update your scm_ptobfuns functions.
6261 ** Function: int scm_internal_select (int fds,
6265 struct timeval *timeout);
6267 This is a replacement for the `select' function provided by the OS.
6268 It enables I/O blocking and sleeping to happen for one cooperative
6269 thread without blocking other threads. It also avoids busy-loops in
6270 these situations. It is intended that all I/O blocking and sleeping
6271 will finally go through this function. Currently, this function is
6272 only available on systems providing `gettimeofday' and `select'.
6274 ** Function: SCM scm_internal_stack_catch (SCM tag,
6275 scm_catch_body_t body,
6277 scm_catch_handler_t handler,
6280 A new sibling to the other two C level `catch' functions
6281 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
6282 the stack to be saved automatically into the variable `the-last-stack'
6283 (scm_the_last_stack_var) on error. This is necessary if you want to
6284 use advanced error reporting, such as calling scm_display_error and
6285 scm_display_backtrace. (They both take a stack object as argument.)
6287 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
6289 scm_catch_handler_t handler,
6292 Spawns a new thread. It does a job similar to
6293 scm_call_with_new_thread but takes arguments more suitable when
6294 spawning threads from application C code.
6296 ** The hook scm_error_callback has been removed. It was originally
6297 intended as a way for the user to install his own error handler. But
6298 that method works badly since it intervenes between throw and catch,
6299 thereby changing the semantics of expressions like (catch #t ...).
6300 The correct way to do it is to use one of the C level catch functions
6301 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
6303 ** Removed functions:
6305 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
6306 scm_list_reverse, scm_list_reverse_x
6308 ** New macros: SCM_LISTn where n is one of the integers 0-9.
6310 These can be used for pretty list creation from C. The idea is taken
6311 from Erick Gallesio's STk.
6313 ** scm_array_map renamed to scm_array_map_x
6315 ** mbstrings are now removed
6317 This means that the type codes scm_tc7_mb_string and
6318 scm_tc7_mb_substring has been removed.
6320 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
6322 Since we no longer support multi-byte strings, these I/O functions
6323 have been simplified, and renamed. Here are their old names, and
6324 their new names and arguments:
6326 scm_gen_putc -> void scm_putc (int c, SCM port);
6327 scm_gen_puts -> void scm_puts (char *s, SCM port);
6328 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
6329 scm_gen_getc -> void scm_getc (SCM port);
6332 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
6334 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
6336 SCM_TYP7S now masks away the bit which distinguishes substrings from
6339 ** scm_catch_body_t: Backward incompatible change!
6341 Body functions to scm_internal_catch and friends do not any longer
6342 take a second argument. This is because it is no longer possible to
6343 pass a #f arg to catch.
6345 ** Calls to scm_protect_object and scm_unprotect now nest properly.
6347 The function scm_protect_object protects its argument from being freed
6348 by the garbage collector. scm_unprotect_object removes that
6351 These functions now nest properly. That is, for every object O, there
6352 is a counter which scm_protect_object(O) increments and
6353 scm_unprotect_object(O) decrements, if the counter is greater than
6354 zero. Every object's counter is zero when it is first created. If an
6355 object's counter is greater than zero, the garbage collector will not
6356 reclaim its storage.
6358 This allows you to use scm_protect_object in your code without
6359 worrying that some other function you call will call
6360 scm_unprotect_object, and allow it to be freed. Assuming that the
6361 functions you call are well-behaved, and unprotect only those objects
6362 they protect, you can follow the same rule and have confidence that
6363 objects will be freed only at appropriate times.
6366 Changes in Guile 1.2 (released Tuesday, June 24 1997):
6368 * Changes to the distribution
6370 ** Nightly snapshots are now available from ftp.red-bean.com.
6371 The old server, ftp.cyclic.com, has been relinquished to its rightful
6374 Nightly snapshots of the Guile development sources are now available via
6375 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
6377 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
6378 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
6380 ** To run Guile without installing it, the procedure has changed a bit.
6382 If you used a separate build directory to compile Guile, you'll need
6383 to include the build directory in SCHEME_LOAD_PATH, as well as the
6384 source directory. See the `INSTALL' file for examples.
6386 * Changes to the procedure for linking libguile with your programs
6388 ** The standard Guile load path for Scheme code now includes
6389 $(datadir)/guile (usually /usr/local/share/guile). This means that
6390 you can install your own Scheme files there, and Guile will find them.
6391 (Previous versions of Guile only checked a directory whose name
6392 contained the Guile version number, so you had to re-install or move
6393 your Scheme sources each time you installed a fresh version of Guile.)
6395 The load path also includes $(datadir)/guile/site; we recommend
6396 putting individual Scheme files there. If you want to install a
6397 package with multiple source files, create a directory for them under
6400 ** Guile 1.2 will now use the Rx regular expression library, if it is
6401 installed on your system. When you are linking libguile into your own
6402 programs, this means you will have to link against -lguile, -lqt (if
6403 you configured Guile with thread support), and -lrx.
6405 If you are using autoconf to generate configuration scripts for your
6406 application, the following lines should suffice to add the appropriate
6407 libraries to your link command:
6409 ### Find Rx, quickthreads and libguile.
6410 AC_CHECK_LIB(rx, main)
6411 AC_CHECK_LIB(qt, main)
6412 AC_CHECK_LIB(guile, scm_shell)
6414 The Guile 1.2 distribution does not contain sources for the Rx
6415 library, as Guile 1.0 did. If you want to use Rx, you'll need to
6416 retrieve it from a GNU FTP site and install it separately.
6418 * Changes to Scheme functions and syntax
6420 ** The dynamic linking features of Guile are now enabled by default.
6421 You can disable them by giving the `--disable-dynamic-linking' option
6424 (dynamic-link FILENAME)
6426 Find the object file denoted by FILENAME (a string) and link it
6427 into the running Guile application. When everything works out,
6428 return a Scheme object suitable for representing the linked object
6429 file. Otherwise an error is thrown. How object files are
6430 searched is system dependent.
6432 (dynamic-object? VAL)
6434 Determine whether VAL represents a dynamically linked object file.
6436 (dynamic-unlink DYNOBJ)
6438 Unlink the indicated object file from the application. DYNOBJ
6439 should be one of the values returned by `dynamic-link'.
6441 (dynamic-func FUNCTION DYNOBJ)
6443 Search the C function indicated by FUNCTION (a string or symbol)
6444 in DYNOBJ and return some Scheme object that can later be used
6445 with `dynamic-call' to actually call this function. Right now,
6446 these Scheme objects are formed by casting the address of the
6447 function to `long' and converting this number to its Scheme
6450 (dynamic-call FUNCTION DYNOBJ)
6452 Call the C function indicated by FUNCTION and DYNOBJ. The
6453 function is passed no arguments and its return value is ignored.
6454 When FUNCTION is something returned by `dynamic-func', call that
6455 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
6456 etc.), look it up in DYNOBJ; this is equivalent to
6458 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
6460 Interrupts are deferred while the C function is executing (with
6461 SCM_DEFER_INTS/SCM_ALLOW_INTS).
6463 (dynamic-args-call FUNCTION DYNOBJ ARGS)
6465 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
6466 some arguments and return its return value. The C function is
6467 expected to take two arguments and return an `int', just like
6470 int c_func (int argc, char **argv);
6472 ARGS must be a list of strings and is converted into an array of
6473 `char *'. The array is passed in ARGV and its size in ARGC. The
6474 return value is converted to a Scheme number and returned from the
6475 call to `dynamic-args-call'.
6477 When dynamic linking is disabled or not supported on your system,
6478 the above functions throw errors, but they are still available.
6480 Here is a small example that works on GNU/Linux:
6482 (define libc-obj (dynamic-link "libc.so"))
6483 (dynamic-args-call 'rand libc-obj '())
6485 See the file `libguile/DYNAMIC-LINKING' for additional comments.
6487 ** The #/ syntax for module names is depreciated, and will be removed
6488 in a future version of Guile. Instead of
6496 The latter syntax is more consistent with existing Lisp practice.
6498 ** Guile now does fancier printing of structures. Structures are the
6499 underlying implementation for records, which in turn are used to
6500 implement modules, so all of these object now print differently and in
6501 a more informative way.
6503 The Scheme printer will examine the builtin variable *struct-printer*
6504 whenever it needs to print a structure object. When this variable is
6505 not `#f' it is deemed to be a procedure and will be applied to the
6506 structure object and the output port. When *struct-printer* is `#f'
6507 or the procedure return `#f' the structure object will be printed in
6508 the boring #<struct 80458270> form.
6510 This hook is used by some routines in ice-9/boot-9.scm to implement
6511 type specific printing routines. Please read the comments there about
6514 One of the more specific uses of structs are records. The printing
6515 procedure that could be passed to MAKE-RECORD-TYPE is now actually
6516 called. It should behave like a *struct-printer* procedure (described
6519 ** Guile now supports a new R4RS-compliant syntax for keywords. A
6520 token of the form #:NAME, where NAME has the same syntax as a Scheme
6521 symbol, is the external representation of the keyword named NAME.
6522 Keyword objects print using this syntax as well, so values containing
6523 keyword objects can be read back into Guile. When used in an
6524 expression, keywords are self-quoting objects.
6526 Guile suports this read syntax, and uses this print syntax, regardless
6527 of the current setting of the `keyword' read option. The `keyword'
6528 read option only controls whether Guile recognizes the `:NAME' syntax,
6529 which is incompatible with R4RS. (R4RS says such token represent
6532 ** Guile has regular expression support again. Guile 1.0 included
6533 functions for matching regular expressions, based on the Rx library.
6534 In Guile 1.1, the Guile/Rx interface was removed to simplify the
6535 distribution, and thus Guile had no regular expression support. Guile
6536 1.2 again supports the most commonly used functions, and supports all
6537 of SCSH's regular expression functions.
6539 If your system does not include a POSIX regular expression library,
6540 and you have not linked Guile with a third-party regexp library such as
6541 Rx, these functions will not be available. You can tell whether your
6542 Guile installation includes regular expression support by checking
6543 whether the `*features*' list includes the `regex' symbol.
6545 *** regexp functions
6547 By default, Guile supports POSIX extended regular expressions. That
6548 means that the characters `(', `)', `+' and `?' are special, and must
6549 be escaped if you wish to match the literal characters.
6551 This regular expression interface was modeled after that implemented
6552 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
6553 with SCSH regular expressions.
6555 **** Function: string-match PATTERN STR [START]
6556 Compile the string PATTERN into a regular expression and compare
6557 it with STR. The optional numeric argument START specifies the
6558 position of STR at which to begin matching.
6560 `string-match' returns a "match structure" which describes what,
6561 if anything, was matched by the regular expression. *Note Match
6562 Structures::. If STR does not match PATTERN at all,
6563 `string-match' returns `#f'.
6565 Each time `string-match' is called, it must compile its PATTERN
6566 argument into a regular expression structure. This operation is
6567 expensive, which makes `string-match' inefficient if the same regular
6568 expression is used several times (for example, in a loop). For better
6569 performance, you can compile a regular expression in advance and then
6570 match strings against the compiled regexp.
6572 **** Function: make-regexp STR [FLAGS]
6573 Compile the regular expression described by STR, and return the
6574 compiled regexp structure. If STR does not describe a legal
6575 regular expression, `make-regexp' throws a
6576 `regular-expression-syntax' error.
6578 FLAGS may be the bitwise-or of one or more of the following:
6580 **** Constant: regexp/extended
6581 Use POSIX Extended Regular Expression syntax when interpreting
6582 STR. If not set, POSIX Basic Regular Expression syntax is used.
6583 If the FLAGS argument is omitted, we assume regexp/extended.
6585 **** Constant: regexp/icase
6586 Do not differentiate case. Subsequent searches using the
6587 returned regular expression will be case insensitive.
6589 **** Constant: regexp/newline
6590 Match-any-character operators don't match a newline.
6592 A non-matching list ([^...]) not containing a newline matches a
6595 Match-beginning-of-line operator (^) matches the empty string
6596 immediately after a newline, regardless of whether the FLAGS
6597 passed to regexp-exec contain regexp/notbol.
6599 Match-end-of-line operator ($) matches the empty string
6600 immediately before a newline, regardless of whether the FLAGS
6601 passed to regexp-exec contain regexp/noteol.
6603 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6604 Match the compiled regular expression REGEXP against `str'. If
6605 the optional integer START argument is provided, begin matching
6606 from that position in the string. Return a match structure
6607 describing the results of the match, or `#f' if no match could be
6610 FLAGS may be the bitwise-or of one or more of the following:
6612 **** Constant: regexp/notbol
6613 The match-beginning-of-line operator always fails to match (but
6614 see the compilation flag regexp/newline above) This flag may be
6615 used when different portions of a string are passed to
6616 regexp-exec and the beginning of the string should not be
6617 interpreted as the beginning of the line.
6619 **** Constant: regexp/noteol
6620 The match-end-of-line operator always fails to match (but see the
6621 compilation flag regexp/newline above)
6623 **** Function: regexp? OBJ
6624 Return `#t' if OBJ is a compiled regular expression, or `#f'
6627 Regular expressions are commonly used to find patterns in one string
6628 and replace them with the contents of another string.
6630 **** Function: regexp-substitute PORT MATCH [ITEM...]
6631 Write to the output port PORT selected contents of the match
6632 structure MATCH. Each ITEM specifies what should be written, and
6633 may be one of the following arguments:
6635 * A string. String arguments are written out verbatim.
6637 * An integer. The submatch with that number is written.
6639 * The symbol `pre'. The portion of the matched string preceding
6640 the regexp match is written.
6642 * The symbol `post'. The portion of the matched string
6643 following the regexp match is written.
6645 PORT may be `#f', in which case nothing is written; instead,
6646 `regexp-substitute' constructs a string from the specified ITEMs
6649 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6650 Similar to `regexp-substitute', but can be used to perform global
6651 substitutions on STR. Instead of taking a match structure as an
6652 argument, `regexp-substitute/global' takes two string arguments: a
6653 REGEXP string describing a regular expression, and a TARGET string
6654 which should be matched against this regular expression.
6656 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6659 * A function may be supplied. When this function is called, it
6660 will be passed one argument: a match structure for a given
6661 regular expression match. It should return a string to be
6662 written out to PORT.
6664 * The `post' symbol causes `regexp-substitute/global' to recurse
6665 on the unmatched portion of STR. This *must* be supplied in
6666 order to perform global search-and-replace on STR; if it is
6667 not present among the ITEMs, then `regexp-substitute/global'
6668 will return after processing a single match.
6670 *** Match Structures
6672 A "match structure" is the object returned by `string-match' and
6673 `regexp-exec'. It describes which portion of a string, if any, matched
6674 the given regular expression. Match structures include: a reference to
6675 the string that was checked for matches; the starting and ending
6676 positions of the regexp match; and, if the regexp included any
6677 parenthesized subexpressions, the starting and ending positions of each
6680 In each of the regexp match functions described below, the `match'
6681 argument must be a match structure returned by a previous call to
6682 `string-match' or `regexp-exec'. Most of these functions return some
6683 information about the original target string that was matched against a
6684 regular expression; we will call that string TARGET for easy reference.
6686 **** Function: regexp-match? OBJ
6687 Return `#t' if OBJ is a match structure returned by a previous
6688 call to `regexp-exec', or `#f' otherwise.
6690 **** Function: match:substring MATCH [N]
6691 Return the portion of TARGET matched by subexpression number N.
6692 Submatch 0 (the default) represents the entire regexp match. If
6693 the regular expression as a whole matched, but the subexpression
6694 number N did not match, return `#f'.
6696 **** Function: match:start MATCH [N]
6697 Return the starting position of submatch number N.
6699 **** Function: match:end MATCH [N]
6700 Return the ending position of submatch number N.
6702 **** Function: match:prefix MATCH
6703 Return the unmatched portion of TARGET preceding the regexp match.
6705 **** Function: match:suffix MATCH
6706 Return the unmatched portion of TARGET following the regexp match.
6708 **** Function: match:count MATCH
6709 Return the number of parenthesized subexpressions from MATCH.
6710 Note that the entire regular expression match itself counts as a
6711 subexpression, and failed submatches are included in the count.
6713 **** Function: match:string MATCH
6714 Return the original TARGET string.
6716 *** Backslash Escapes
6718 Sometimes you will want a regexp to match characters like `*' or `$'
6719 exactly. For example, to check whether a particular string represents
6720 a menu entry from an Info node, it would be useful to match it against
6721 a regexp like `^* [^:]*::'. However, this won't work; because the
6722 asterisk is a metacharacter, it won't match the `*' at the beginning of
6723 the string. In this case, we want to make the first asterisk un-magic.
6725 You can do this by preceding the metacharacter with a backslash
6726 character `\'. (This is also called "quoting" the metacharacter, and
6727 is known as a "backslash escape".) When Guile sees a backslash in a
6728 regular expression, it considers the following glyph to be an ordinary
6729 character, no matter what special meaning it would ordinarily have.
6730 Therefore, we can make the above example work by changing the regexp to
6731 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6732 to match only a single asterisk in the target string.
6734 Since the backslash is itself a metacharacter, you may force a
6735 regexp to match a backslash in the target string by preceding the
6736 backslash with itself. For example, to find variable references in a
6737 TeX program, you might want to find occurrences of the string `\let\'
6738 followed by any number of alphabetic characters. The regular expression
6739 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6740 each match a single backslash in the target string.
6742 **** Function: regexp-quote STR
6743 Quote each special character found in STR with a backslash, and
6744 return the resulting string.
6746 *Very important:* Using backslash escapes in Guile source code (as
6747 in Emacs Lisp or C) can be tricky, because the backslash character has
6748 special meaning for the Guile reader. For example, if Guile encounters
6749 the character sequence `\n' in the middle of a string while processing
6750 Scheme code, it replaces those characters with a newline character.
6751 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6752 Several of these "escape sequences" are processed by the Guile reader
6753 before your code is executed. Unrecognized escape sequences are
6754 ignored: if the characters `\*' appear in a string, they will be
6755 translated to the single character `*'.
6757 This translation is obviously undesirable for regular expressions,
6758 since we want to be able to include backslashes in a string in order to
6759 escape regexp metacharacters. Therefore, to make sure that a backslash
6760 is preserved in a string in your Guile program, you must use *two*
6761 consecutive backslashes:
6763 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6765 The string in this example is preprocessed by the Guile reader before
6766 any code is executed. The resulting argument to `make-regexp' is the
6767 string `^\* [^:]*', which is what we really want.
6769 This also means that in order to write a regular expression that
6770 matches a single backslash character, the regular expression string in
6771 the source code must include *four* backslashes. Each consecutive pair
6772 of backslashes gets translated by the Guile reader to a single
6773 backslash, and the resulting double-backslash is interpreted by the
6774 regexp engine as matching a single backslash character. Hence:
6776 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6778 The reason for the unwieldiness of this syntax is historical. Both
6779 regular expression pattern matchers and Unix string processing systems
6780 have traditionally used backslashes with the special meanings described
6781 above. The POSIX regular expression specification and ANSI C standard
6782 both require these semantics. Attempting to abandon either convention
6783 would cause other kinds of compatibility problems, possibly more severe
6784 ones. Therefore, without extending the Scheme reader to support
6785 strings with different quoting conventions (an ungainly and confusing
6786 extension when implemented in other languages), we must adhere to this
6787 cumbersome escape syntax.
6789 * Changes to the gh_ interface
6791 * Changes to the scm_ interface
6793 * Changes to system call interfaces:
6795 ** The value returned by `raise' is now unspecified. It throws an exception
6798 *** A new procedure `sigaction' can be used to install signal handlers
6800 (sigaction signum [action] [flags])
6802 signum is the signal number, which can be specified using the value
6805 If action is omitted, sigaction returns a pair: the CAR is the current
6806 signal hander, which will be either an integer with the value SIG_DFL
6807 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6808 handles the signal, or #f if a non-Scheme procedure handles the
6809 signal. The CDR contains the current sigaction flags for the handler.
6811 If action is provided, it is installed as the new handler for signum.
6812 action can be a Scheme procedure taking one argument, or the value of
6813 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6814 whatever signal handler was installed before sigaction was first used.
6815 Flags can optionally be specified for the new handler (SA_RESTART is
6816 always used if the system provides it, so need not be specified.) The
6817 return value is a pair with information about the old handler as
6820 This interface does not provide access to the "signal blocking"
6821 facility. Maybe this is not needed, since the thread support may
6822 provide solutions to the problem of consistent access to data
6825 *** A new procedure `flush-all-ports' is equivalent to running
6826 `force-output' on every port open for output.
6828 ** Guile now provides information on how it was built, via the new
6829 global variable, %guile-build-info. This variable records the values
6830 of the standard GNU makefile directory variables as an assocation
6831 list, mapping variable names (symbols) onto directory paths (strings).
6832 For example, to find out where the Guile link libraries were
6833 installed, you can say:
6835 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6838 * Changes to the scm_ interface
6840 ** The new function scm_handle_by_message_noexit is just like the
6841 existing scm_handle_by_message function, except that it doesn't call
6842 exit to terminate the process. Instead, it prints a message and just
6843 returns #f. This might be a more appropriate catch-all handler for
6844 new dynamic roots and threads.
6847 Changes in Guile 1.1 (released Friday, May 16 1997):
6849 * Changes to the distribution.
6851 The Guile 1.0 distribution has been split up into several smaller
6853 guile-core --- the Guile interpreter itself.
6854 guile-tcltk --- the interface between the Guile interpreter and
6855 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6856 is a toolkit for building graphical user interfaces.
6857 guile-rgx-ctax --- the interface between Guile and the Rx regular
6858 expression matcher, and the translator for the Ctax
6859 programming language. These are packaged together because the
6860 Ctax translator uses Rx to parse Ctax source code.
6862 This NEWS file describes the changes made to guile-core since the 1.0
6865 We no longer distribute the documentation, since it was either out of
6866 date, or incomplete. As soon as we have current documentation, we
6871 * Changes to the stand-alone interpreter
6873 ** guile now accepts command-line arguments compatible with SCSH, Olin
6874 Shivers' Scheme Shell.
6876 In general, arguments are evaluated from left to right, but there are
6877 exceptions. The following switches stop argument processing, and
6878 stash all remaining command-line arguments as the value returned by
6879 the (command-line) function.
6880 -s SCRIPT load Scheme source code from FILE, and exit
6881 -c EXPR evalute Scheme expression EXPR, and exit
6882 -- stop scanning arguments; run interactively
6884 The switches below are processed as they are encountered.
6885 -l FILE load Scheme source code from FILE
6886 -e FUNCTION after reading script, apply FUNCTION to
6887 command line arguments
6888 -ds do -s script at this point
6889 --emacs enable Emacs protocol (experimental)
6890 -h, --help display this help and exit
6891 -v, --version display version information and exit
6892 \ read arguments from following script lines
6894 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6895 which re-implements the traditional "echo" command:
6897 #!/usr/local/bin/guile -s
6900 (map (lambda (arg) (display arg) (display " "))
6904 (main (command-line))
6906 Suppose we invoke this script as follows:
6908 ekko a speckled gecko
6910 Through the magic of Unix script processing (triggered by the `#!'
6911 token at the top of the file), /usr/local/bin/guile receives the
6912 following list of command-line arguments:
6914 ("-s" "./ekko" "a" "speckled" "gecko")
6916 Unix inserts the name of the script after the argument specified on
6917 the first line of the file (in this case, "-s"), and then follows that
6918 with the arguments given to the script. Guile loads the script, which
6919 defines the `main' function, and then applies it to the list of
6920 remaining command-line arguments, ("a" "speckled" "gecko").
6922 In Unix, the first line of a script file must take the following form:
6924 #!INTERPRETER ARGUMENT
6926 where INTERPRETER is the absolute filename of the interpreter
6927 executable, and ARGUMENT is a single command-line argument to pass to
6930 You may only pass one argument to the interpreter, and its length is
6931 limited. These restrictions can be annoying to work around, so Guile
6932 provides a general mechanism (borrowed from, and compatible with,
6933 SCSH) for circumventing them.
6935 If the ARGUMENT in a Guile script is a single backslash character,
6936 `\', Guile will open the script file, parse arguments from its second
6937 and subsequent lines, and replace the `\' with them. So, for example,
6938 here is another implementation of the `ekko' script:
6940 #!/usr/local/bin/guile \
6944 (for-each (lambda (arg) (display arg) (display " "))
6948 If the user invokes this script as follows:
6950 ekko a speckled gecko
6952 Unix expands this into
6954 /usr/local/bin/guile \ ekko a speckled gecko
6956 When Guile sees the `\' argument, it replaces it with the arguments
6957 read from the second line of the script, producing:
6959 /usr/local/bin/guile -e main -s ekko a speckled gecko
6961 This tells Guile to load the `ekko' script, and apply the function
6962 `main' to the argument list ("a" "speckled" "gecko").
6964 Here is how Guile parses the command-line arguments:
6965 - Each space character terminates an argument. This means that two
6966 spaces in a row introduce an empty-string argument.
6967 - The tab character is not permitted (unless you quote it with the
6968 backslash character, as described below), to avoid confusion.
6969 - The newline character terminates the sequence of arguments, and will
6970 also terminate a final non-empty argument. (However, a newline
6971 following a space will not introduce a final empty-string argument;
6972 it only terminates the argument list.)
6973 - The backslash character is the escape character. It escapes
6974 backslash, space, tab, and newline. The ANSI C escape sequences
6975 like \n and \t are also supported. These produce argument
6976 constituents; the two-character combination \n doesn't act like a
6977 terminating newline. The escape sequence \NNN for exactly three
6978 octal digits reads as the character whose ASCII code is NNN. As
6979 above, characters produced this way are argument constituents.
6980 Backslash followed by other characters is not allowed.
6982 * Changes to the procedure for linking libguile with your programs
6984 ** Guile now builds and installs a shared guile library, if your
6985 system support shared libraries. (It still builds a static library on
6986 all systems.) Guile automatically detects whether your system
6987 supports shared libraries. To prevent Guile from buildisg shared
6988 libraries, pass the `--disable-shared' flag to the configure script.
6990 Guile takes longer to compile when it builds shared libraries, because
6991 it must compile every file twice --- once to produce position-
6992 independent object code, and once to produce normal object code.
6994 ** The libthreads library has been merged into libguile.
6996 To link a program against Guile, you now need only link against
6997 -lguile and -lqt; -lthreads is no longer needed. If you are using
6998 autoconf to generate configuration scripts for your application, the
6999 following lines should suffice to add the appropriate libraries to
7002 ### Find quickthreads and libguile.
7003 AC_CHECK_LIB(qt, main)
7004 AC_CHECK_LIB(guile, scm_shell)
7006 * Changes to Scheme functions
7008 ** Guile Scheme's special syntax for keyword objects is now optional,
7009 and disabled by default.
7011 The syntax variation from R4RS made it difficult to port some
7012 interesting packages to Guile. The routines which accepted keyword
7013 arguments (mostly in the module system) have been modified to also
7014 accept symbols whose names begin with `:'.
7016 To change the keyword syntax, you must first import the (ice-9 debug)
7018 (use-modules (ice-9 debug))
7020 Then you can enable the keyword syntax as follows:
7021 (read-set! keywords 'prefix)
7023 To disable keyword syntax, do this:
7024 (read-set! keywords #f)
7026 ** Many more primitive functions accept shared substrings as
7027 arguments. In the past, these functions required normal, mutable
7028 strings as arguments, although they never made use of this
7031 ** The uniform array functions now operate on byte vectors. These
7032 functions are `array-fill!', `serial-array-copy!', `array-copy!',
7033 `serial-array-map', `array-map', `array-for-each', and
7036 ** The new functions `trace' and `untrace' implement simple debugging
7037 support for Scheme functions.
7039 The `trace' function accepts any number of procedures as arguments,
7040 and tells the Guile interpreter to display each procedure's name and
7041 arguments each time the procedure is invoked. When invoked with no
7042 arguments, `trace' returns the list of procedures currently being
7045 The `untrace' function accepts any number of procedures as arguments,
7046 and tells the Guile interpreter not to trace them any more. When
7047 invoked with no arguments, `untrace' untraces all curretly traced
7050 The tracing in Guile has an advantage over most other systems: we
7051 don't create new procedure objects, but mark the procedure objects
7052 themselves. This means that anonymous and internal procedures can be
7055 ** The function `assert-repl-prompt' has been renamed to
7056 `set-repl-prompt!'. It takes one argument, PROMPT.
7057 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
7058 - If PROMPT is a string, we use it as a prompt.
7059 - If PROMPT is a procedure accepting no arguments, we call it, and
7060 display the result as a prompt.
7061 - Otherwise, we display "> ".
7063 ** The new function `eval-string' reads Scheme expressions from a
7064 string and evaluates them, returning the value of the last expression
7065 in the string. If the string contains no expressions, it returns an
7068 ** The new function `thunk?' returns true iff its argument is a
7069 procedure of zero arguments.
7071 ** `defined?' is now a builtin function, instead of syntax. This
7072 means that its argument should be quoted. It returns #t iff its
7073 argument is bound in the current module.
7075 ** The new syntax `use-modules' allows you to add new modules to your
7076 environment without re-typing a complete `define-module' form. It
7077 accepts any number of module names as arguments, and imports their
7078 public bindings into the current module.
7080 ** The new function (module-defined? NAME MODULE) returns true iff
7081 NAME, a symbol, is defined in MODULE, a module object.
7083 ** The new function `builtin-bindings' creates and returns a hash
7084 table containing copies of all the root module's bindings.
7086 ** The new function `builtin-weak-bindings' does the same as
7087 `builtin-bindings', but creates a doubly-weak hash table.
7089 ** The `equal?' function now considers variable objects to be
7090 equivalent if they have the same name and the same value.
7092 ** The new function `command-line' returns the command-line arguments
7093 given to Guile, as a list of strings.
7095 When using guile as a script interpreter, `command-line' returns the
7096 script's arguments; those processed by the interpreter (like `-s' or
7097 `-c') are omitted. (In other words, you get the normal, expected
7098 behavior.) Any application that uses scm_shell to process its
7099 command-line arguments gets this behavior as well.
7101 ** The new function `load-user-init' looks for a file called `.guile'
7102 in the user's home directory, and loads it if it exists. This is
7103 mostly for use by the code generated by scm_compile_shell_switches,
7104 but we thought it might also be useful in other circumstances.
7106 ** The new function `log10' returns the base-10 logarithm of its
7109 ** Changes to I/O functions
7111 *** The functions `read', `primitive-load', `read-and-eval!', and
7112 `primitive-load-path' no longer take optional arguments controlling
7113 case insensitivity and a `#' parser.
7115 Case sensitivity is now controlled by a read option called
7116 `case-insensitive'. The user can add new `#' syntaxes with the
7117 `read-hash-extend' function (see below).
7119 *** The new function `read-hash-extend' allows the user to change the
7120 syntax of Guile Scheme in a somewhat controlled way.
7122 (read-hash-extend CHAR PROC)
7123 When parsing S-expressions, if we read a `#' character followed by
7124 the character CHAR, use PROC to parse an object from the stream.
7125 If PROC is #f, remove any parsing procedure registered for CHAR.
7127 The reader applies PROC to two arguments: CHAR and an input port.
7129 *** The new functions read-delimited and read-delimited! provide a
7130 general mechanism for doing delimited input on streams.
7132 (read-delimited DELIMS [PORT HANDLE-DELIM])
7133 Read until we encounter one of the characters in DELIMS (a string),
7134 or end-of-file. PORT is the input port to read from; it defaults to
7135 the current input port. The HANDLE-DELIM parameter determines how
7136 the terminating character is handled; it should be one of the
7139 'trim omit delimiter from result
7140 'peek leave delimiter character in input stream
7141 'concat append delimiter character to returned value
7142 'split return a pair: (RESULT . TERMINATOR)
7144 HANDLE-DELIM defaults to 'peek.
7146 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
7147 A side-effecting variant of `read-delimited'.
7149 The data is written into the string BUF at the indices in the
7150 half-open interval [START, END); the default interval is the whole
7151 string: START = 0 and END = (string-length BUF). The values of
7152 START and END must specify a well-defined interval in BUF, i.e.
7153 0 <= START <= END <= (string-length BUF).
7155 It returns NBYTES, the number of bytes read. If the buffer filled
7156 up without a delimiter character being found, it returns #f. If the
7157 port is at EOF when the read starts, it returns the EOF object.
7159 If an integer is returned (i.e., the read is successfully terminated
7160 by reading a delimiter character), then the HANDLE-DELIM parameter
7161 determines how to handle the terminating character. It is described
7162 above, and defaults to 'peek.
7164 (The descriptions of these functions were borrowed from the SCSH
7165 manual, by Olin Shivers and Brian Carlstrom.)
7167 *** The `%read-delimited!' function is the primitive used to implement
7168 `read-delimited' and `read-delimited!'.
7170 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
7172 This returns a pair of values: (TERMINATOR . NUM-READ).
7173 - TERMINATOR describes why the read was terminated. If it is a
7174 character or the eof object, then that is the value that terminated
7175 the read. If it is #f, the function filled the buffer without finding
7176 a delimiting character.
7177 - NUM-READ is the number of characters read into BUF.
7179 If the read is successfully terminated by reading a delimiter
7180 character, then the gobble? parameter determines what to do with the
7181 terminating character. If true, the character is removed from the
7182 input stream; if false, the character is left in the input stream
7183 where a subsequent read operation will retrieve it. In either case,
7184 the character is also the first value returned by the procedure call.
7186 (The descriptions of this function was borrowed from the SCSH manual,
7187 by Olin Shivers and Brian Carlstrom.)
7189 *** The `read-line' and `read-line!' functions have changed; they now
7190 trim the terminator by default; previously they appended it to the
7191 returned string. For the old behavior, use (read-line PORT 'concat).
7193 *** The functions `uniform-array-read!' and `uniform-array-write!' now
7194 take new optional START and END arguments, specifying the region of
7195 the array to read and write.
7197 *** The `ungetc-char-ready?' function has been removed. We feel it's
7198 inappropriate for an interface to expose implementation details this
7201 ** Changes to the Unix library and system call interface
7203 *** The new fcntl function provides access to the Unix `fcntl' system
7206 (fcntl PORT COMMAND VALUE)
7207 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
7208 Values for COMMAND are:
7210 F_DUPFD duplicate a file descriptor
7211 F_GETFD read the descriptor's close-on-exec flag
7212 F_SETFD set the descriptor's close-on-exec flag to VALUE
7213 F_GETFL read the descriptor's flags, as set on open
7214 F_SETFL set the descriptor's flags, as set on open to VALUE
7215 F_GETOWN return the process ID of a socket's owner, for SIGIO
7216 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
7217 FD_CLOEXEC not sure what this is
7219 For details, see the documentation for the fcntl system call.
7221 *** The arguments to `select' have changed, for compatibility with
7222 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
7223 expected behavior. The MILLISECONDS parameter has been changed to
7224 MICROSECONDS, to more closely resemble the underlying system call.
7225 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
7226 corresponding return set will be the same.
7228 *** The arguments to the `mknod' system call have changed. They are
7231 (mknod PATH TYPE PERMS DEV)
7232 Create a new file (`node') in the file system. PATH is the name of
7233 the file to create. TYPE is the kind of file to create; it should
7234 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
7235 permission bits to give the newly created file. If TYPE is
7236 'block-special or 'char-special, DEV specifies which device the
7237 special file refers to; its interpretation depends on the kind of
7238 special file being created.
7240 *** The `fork' function has been renamed to `primitive-fork', to avoid
7241 clashing with various SCSH forks.
7243 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
7244 and `recvfrom!'. They no longer accept a size for a second argument;
7245 you must pass a string to hold the received value. They no longer
7246 return the buffer. Instead, `recv' returns the length of the message
7247 received, and `recvfrom' returns a pair containing the packet's length
7248 and originating address.
7250 *** The file descriptor datatype has been removed, as have the
7251 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
7252 We plan to replace these functions with a SCSH-compatible interface.
7254 *** The `create' function has been removed; it's just a special case
7257 *** There are new functions to break down process termination status
7258 values. In the descriptions below, STATUS is a value returned by
7261 (status:exit-val STATUS)
7262 If the child process exited normally, this function returns the exit
7263 code for the child process (i.e., the value passed to exit, or
7264 returned from main). If the child process did not exit normally,
7265 this function returns #f.
7267 (status:stop-sig STATUS)
7268 If the child process was suspended by a signal, this function
7269 returns the signal that suspended the child. Otherwise, it returns
7272 (status:term-sig STATUS)
7273 If the child process terminated abnormally, this function returns
7274 the signal that terminated the child. Otherwise, this function
7277 POSIX promises that exactly one of these functions will return true on
7278 a valid STATUS value.
7280 These functions are compatible with SCSH.
7282 *** There are new accessors and setters for the broken-out time vectors
7283 returned by `localtime', `gmtime', and that ilk. They are:
7285 Component Accessor Setter
7286 ========================= ============ ============
7287 seconds tm:sec set-tm:sec
7288 minutes tm:min set-tm:min
7289 hours tm:hour set-tm:hour
7290 day of the month tm:mday set-tm:mday
7291 month tm:mon set-tm:mon
7292 year tm:year set-tm:year
7293 day of the week tm:wday set-tm:wday
7294 day in the year tm:yday set-tm:yday
7295 daylight saving time tm:isdst set-tm:isdst
7296 GMT offset, seconds tm:gmtoff set-tm:gmtoff
7297 name of time zone tm:zone set-tm:zone
7299 *** There are new accessors for the vectors returned by `uname',
7300 describing the host system:
7303 ============================================== ================
7304 name of the operating system implementation utsname:sysname
7305 network name of this machine utsname:nodename
7306 release level of the operating system utsname:release
7307 version level of the operating system utsname:version
7308 machine hardware platform utsname:machine
7310 *** There are new accessors for the vectors returned by `getpw',
7311 `getpwnam', `getpwuid', and `getpwent', describing entries from the
7312 system's user database:
7315 ====================== =================
7316 user name passwd:name
7317 user password passwd:passwd
7320 real name passwd:gecos
7321 home directory passwd:dir
7322 shell program passwd:shell
7324 *** There are new accessors for the vectors returned by `getgr',
7325 `getgrnam', `getgrgid', and `getgrent', describing entries from the
7326 system's group database:
7329 ======================= ============
7330 group name group:name
7331 group password group:passwd
7333 group members group:mem
7335 *** There are new accessors for the vectors returned by `gethost',
7336 `gethostbyaddr', `gethostbyname', and `gethostent', describing
7340 ========================= ===============
7341 official name of host hostent:name
7342 alias list hostent:aliases
7343 host address type hostent:addrtype
7344 length of address hostent:length
7345 list of addresses hostent:addr-list
7347 *** There are new accessors for the vectors returned by `getnet',
7348 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
7352 ========================= ===============
7353 official name of net netent:name
7354 alias list netent:aliases
7355 net number type netent:addrtype
7356 net number netent:net
7358 *** There are new accessors for the vectors returned by `getproto',
7359 `getprotobyname', `getprotobynumber', and `getprotoent', describing
7363 ========================= ===============
7364 official protocol name protoent:name
7365 alias list protoent:aliases
7366 protocol number protoent:proto
7368 *** There are new accessors for the vectors returned by `getserv',
7369 `getservbyname', `getservbyport', and `getservent', describing
7373 ========================= ===============
7374 official service name servent:name
7375 alias list servent:aliases
7376 port number servent:port
7377 protocol to use servent:proto
7379 *** There are new accessors for the sockaddr structures returned by
7380 `accept', `getsockname', `getpeername', `recvfrom!':
7383 ======================================== ===============
7384 address format (`family') sockaddr:fam
7385 path, for file domain addresses sockaddr:path
7386 address, for internet domain addresses sockaddr:addr
7387 TCP or UDP port, for internet sockaddr:port
7389 *** The `getpwent', `getgrent', `gethostent', `getnetent',
7390 `getprotoent', and `getservent' functions now return #f at the end of
7391 the user database. (They used to throw an exception.)
7393 Note that calling MUMBLEent function is equivalent to calling the
7394 corresponding MUMBLE function with no arguments.
7396 *** The `setpwent', `setgrent', `sethostent', `setnetent',
7397 `setprotoent', and `setservent' routines now take no arguments.
7399 *** The `gethost', `getproto', `getnet', and `getserv' functions now
7400 provide more useful information when they throw an exception.
7402 *** The `lnaof' function has been renamed to `inet-lnaof'.
7404 *** Guile now claims to have the `current-time' feature.
7406 *** The `mktime' function now takes an optional second argument ZONE,
7407 giving the time zone to use for the conversion. ZONE should be a
7408 string, in the same format as expected for the "TZ" environment variable.
7410 *** The `strptime' function now returns a pair (TIME . COUNT), where
7411 TIME is the parsed time as a vector, and COUNT is the number of
7412 characters from the string left unparsed. This function used to
7413 return the remaining characters as a string.
7415 *** The `gettimeofday' function has replaced the old `time+ticks' function.
7416 The return value is now (SECONDS . MICROSECONDS); the fractional
7417 component is no longer expressed in "ticks".
7419 *** The `ticks/sec' constant has been removed, in light of the above change.
7421 * Changes to the gh_ interface
7423 ** gh_eval_str() now returns an SCM object which is the result of the
7426 ** gh_scm2str() now copies the Scheme data to a caller-provided C
7429 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
7430 and returns the array
7432 ** gh_scm2str0() is gone: there is no need to distinguish
7433 null-terminated from non-null-terminated, since gh_scm2newstr() allows
7434 the user to interpret the data both ways.
7436 * Changes to the scm_ interface
7438 ** The new function scm_symbol_value0 provides an easy way to get a
7439 symbol's value from C code:
7441 SCM scm_symbol_value0 (char *NAME)
7442 Return the value of the symbol named by the null-terminated string
7443 NAME in the current module. If the symbol named NAME is unbound in
7444 the current module, return SCM_UNDEFINED.
7446 ** The new function scm_sysintern0 creates new top-level variables,
7447 without assigning them a value.
7449 SCM scm_sysintern0 (char *NAME)
7450 Create a new Scheme top-level variable named NAME. NAME is a
7451 null-terminated string. Return the variable's value cell.
7453 ** The function scm_internal_catch is the guts of catch. It handles
7454 all the mechanics of setting up a catch target, invoking the catch
7455 body, and perhaps invoking the handler if the body does a throw.
7457 The function is designed to be usable from C code, but is general
7458 enough to implement all the semantics Guile Scheme expects from throw.
7460 TAG is the catch tag. Typically, this is a symbol, but this function
7461 doesn't actually care about that.
7463 BODY is a pointer to a C function which runs the body of the catch;
7464 this is the code you can throw from. We call it like this:
7465 BODY (BODY_DATA, JMPBUF)
7467 BODY_DATA is just the BODY_DATA argument we received; we pass it
7468 through to BODY as its first argument. The caller can make
7469 BODY_DATA point to anything useful that BODY might need.
7470 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
7471 which we have just created and initialized.
7473 HANDLER is a pointer to a C function to deal with a throw to TAG,
7474 should one occur. We call it like this:
7475 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
7477 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
7478 same idea as BODY_DATA above.
7479 THROWN_TAG is the tag that the user threw to; usually this is
7480 TAG, but it could be something else if TAG was #t (i.e., a
7481 catch-all), or the user threw to a jmpbuf.
7482 THROW_ARGS is the list of arguments the user passed to the THROW
7485 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
7486 is just a pointer we pass through to HANDLER. We don't actually
7487 use either of those pointers otherwise ourselves. The idea is
7488 that, if our caller wants to communicate something to BODY or
7489 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
7490 HANDLER can then use. Think of it as a way to make BODY and
7491 HANDLER closures, not just functions; MUMBLE_DATA points to the
7494 Of course, it's up to the caller to make sure that any data a
7495 MUMBLE_DATA needs is protected from GC. A common way to do this is
7496 to make MUMBLE_DATA a pointer to data stored in an automatic
7497 structure variable; since the collector must scan the stack for
7498 references anyway, this assures that any references in MUMBLE_DATA
7501 ** The new function scm_internal_lazy_catch is exactly like
7502 scm_internal_catch, except:
7504 - It does not unwind the stack (this is the major difference).
7505 - If handler returns, its value is returned from the throw.
7506 - BODY always receives #f as its JMPBUF argument (since there's no
7507 jmpbuf associated with a lazy catch, because we don't unwind the
7510 ** scm_body_thunk is a new body function you can pass to
7511 scm_internal_catch if you want the body to be like Scheme's `catch'
7512 --- a thunk, or a function of one argument if the tag is #f.
7514 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
7515 contains the Scheme procedure to invoke as the body, and the tag
7516 we're catching. If the tag is #f, then we pass JMPBUF (created by
7517 scm_internal_catch) to the body procedure; otherwise, the body gets
7520 ** scm_handle_by_proc is a new handler function you can pass to
7521 scm_internal_catch if you want the handler to act like Scheme's catch
7522 --- call a procedure with the tag and the throw arguments.
7524 If the user does a throw to this catch, this function runs a handler
7525 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
7526 variable holding the Scheme procedure object to invoke. It ought to
7527 be a pointer to an automatic variable (i.e., one living on the stack),
7528 or the procedure object should be otherwise protected from GC.
7530 ** scm_handle_by_message is a new handler function to use with
7531 `scm_internal_catch' if you want Guile to print a message and die.
7532 It's useful for dealing with throws to uncaught keys at the top level.
7534 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
7535 message header to print; if zero, we use "guile" instead. That
7536 text is followed by a colon, then the message described by ARGS.
7538 ** The return type of scm_boot_guile is now void; the function does
7539 not return a value, and indeed, never returns at all.
7541 ** The new function scm_shell makes it easy for user applications to
7542 process command-line arguments in a way that is compatible with the
7543 stand-alone guile interpreter (which is in turn compatible with SCSH,
7546 To use the scm_shell function, first initialize any guile modules
7547 linked into your application, and then call scm_shell with the values
7548 of ARGC and ARGV your `main' function received. scm_shell will add
7549 any SCSH-style meta-arguments from the top of the script file to the
7550 argument vector, and then process the command-line arguments. This
7551 generally means loading a script file or starting up an interactive
7552 command interpreter. For details, see "Changes to the stand-alone
7555 ** The new functions scm_get_meta_args and scm_count_argv help you
7556 implement the SCSH-style meta-argument, `\'.
7558 char **scm_get_meta_args (int ARGC, char **ARGV)
7559 If the second element of ARGV is a string consisting of a single
7560 backslash character (i.e. "\\" in Scheme notation), open the file
7561 named by the following argument, parse arguments from it, and return
7562 the spliced command line. The returned array is terminated by a
7565 For details of argument parsing, see above, under "guile now accepts
7566 command-line arguments compatible with SCSH..."
7568 int scm_count_argv (char **ARGV)
7569 Count the arguments in ARGV, assuming it is terminated by a null
7572 For an example of how these functions might be used, see the source
7573 code for the function scm_shell in libguile/script.c.
7575 You will usually want to use scm_shell instead of calling this
7578 ** The new function scm_compile_shell_switches turns an array of
7579 command-line arguments into Scheme code to carry out the actions they
7580 describe. Given ARGC and ARGV, it returns a Scheme expression to
7581 evaluate, and calls scm_set_program_arguments to make any remaining
7582 command-line arguments available to the Scheme code. For example,
7583 given the following arguments:
7585 -e main -s ekko a speckled gecko
7587 scm_set_program_arguments will return the following expression:
7589 (begin (load "ekko") (main (command-line)) (quit))
7591 You will usually want to use scm_shell instead of calling this
7594 ** The function scm_shell_usage prints a usage message appropriate for
7595 an interpreter that uses scm_compile_shell_switches to handle its
7596 command-line arguments.
7598 void scm_shell_usage (int FATAL, char *MESSAGE)
7599 Print a usage message to the standard error output. If MESSAGE is
7600 non-zero, write it before the usage message, followed by a newline.
7601 If FATAL is non-zero, exit the process, using FATAL as the
7602 termination status. (If you want to be compatible with Guile,
7603 always use 1 as the exit status when terminating due to command-line
7606 You will usually want to use scm_shell instead of calling this
7609 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7610 expressions. It used to return SCM_EOL. Earth-shattering.
7612 ** The macros for declaring scheme objects in C code have been
7613 rearranged slightly. They are now:
7615 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7616 Declare a static SCM variable named C_NAME, and initialize it to
7617 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7618 be a C identifier, and SCHEME_NAME should be a C string.
7620 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7621 Just like SCM_SYMBOL, but make C_NAME globally visible.
7623 SCM_VCELL (C_NAME, SCHEME_NAME)
7624 Create a global variable at the Scheme level named SCHEME_NAME.
7625 Declare a static SCM variable named C_NAME, and initialize it to
7626 point to the Scheme variable's value cell.
7628 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7629 Just like SCM_VCELL, but make C_NAME globally visible.
7631 The `guile-snarf' script writes initialization code for these macros
7632 to its standard output, given C source code as input.
7634 The SCM_GLOBAL macro is gone.
7636 ** The scm_read_line and scm_read_line_x functions have been replaced
7637 by Scheme code based on the %read-delimited! procedure (known to C
7638 code as scm_read_delimited_x). See its description above for more
7641 ** The function scm_sys_open has been renamed to scm_open. It now
7642 returns a port instead of an FD object.
7644 * The dynamic linking support has changed. For more information, see
7645 libguile/DYNAMIC-LINKING.
7650 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7653 * Changes to the 'guile' program:
7655 ** Guile now loads some new files when it starts up. Guile first
7656 searches the load path for init.scm, and loads it if found. Then, if
7657 Guile is not being used to execute a script, and the user's home
7658 directory contains a file named `.guile', Guile loads that.
7660 ** You can now use Guile as a shell script interpreter.
7662 To paraphrase the SCSH manual:
7664 When Unix tries to execute an executable file whose first two
7665 characters are the `#!', it treats the file not as machine code to
7666 be directly executed by the native processor, but as source code
7667 to be executed by some interpreter. The interpreter to use is
7668 specified immediately after the #! sequence on the first line of
7669 the source file. The kernel reads in the name of the interpreter,
7670 and executes that instead. It passes the interpreter the source
7671 filename as its first argument, with the original arguments
7672 following. Consult the Unix man page for the `exec' system call
7673 for more information.
7675 Now you can use Guile as an interpreter, using a mechanism which is a
7676 compatible subset of that provided by SCSH.
7678 Guile now recognizes a '-s' command line switch, whose argument is the
7679 name of a file of Scheme code to load. It also treats the two
7680 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7681 to make a file of Scheme code directly executable by Unix, insert the
7682 following two lines at the top of the file:
7684 #!/usr/local/bin/guile -s
7687 Guile treats the argument of the `-s' command-line switch as the name
7688 of a file of Scheme code to load, and treats the sequence `#!' as the
7689 start of a block comment, terminated by `!#'.
7691 For example, here's a version of 'echo' written in Scheme:
7693 #!/usr/local/bin/guile -s
7695 (let loop ((args (cdr (program-arguments))))
7698 (display (car args))
7699 (if (pair? (cdr args))
7701 (loop (cdr args)))))
7704 Why does `#!' start a block comment terminated by `!#', instead of the
7705 end of the line? That is the notation SCSH uses, and although we
7706 don't yet support the other SCSH features that motivate that choice,
7707 we would like to be backward-compatible with any existing Guile
7708 scripts once we do. Furthermore, if the path to Guile on your system
7709 is too long for your kernel, you can start the script with this
7713 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7716 Note that some very old Unix systems don't support the `#!' syntax.
7719 ** You can now run Guile without installing it.
7721 Previous versions of the interactive Guile interpreter (`guile')
7722 couldn't start up unless Guile's Scheme library had been installed;
7723 they used the value of the environment variable `SCHEME_LOAD_PATH'
7724 later on in the startup process, but not to find the startup code
7725 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7728 To run Guile without installing it, build it in the normal way, and
7729 then set the environment variable `SCHEME_LOAD_PATH' to a
7730 colon-separated list of directories, including the top-level directory
7731 of the Guile sources. For example, if you unpacked Guile so that the
7732 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7735 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7738 ** Guile's read-eval-print loop no longer prints #<unspecified>
7739 results. If the user wants to see this, she can evaluate the
7740 expression (assert-repl-print-unspecified #t), perhaps in her startup
7743 ** Guile no longer shows backtraces by default when an error occurs;
7744 however, it does display a message saying how to get one, and how to
7745 request that they be displayed by default. After an error, evaluate
7747 to see a backtrace, and
7748 (debug-enable 'backtrace)
7749 to see them by default.
7753 * Changes to Guile Scheme:
7755 ** Guile now distinguishes between #f and the empty list.
7757 This is for compatibility with the IEEE standard, the (possibly)
7758 upcoming Revised^5 Report on Scheme, and many extant Scheme
7761 Guile used to have #f and '() denote the same object, to make Scheme's
7762 type system more compatible with Emacs Lisp's. However, the change
7763 caused too much trouble for Scheme programmers, and we found another
7764 way to reconcile Emacs Lisp with Scheme that didn't require this.
7767 ** Guile's delq, delv, delete functions, and their destructive
7768 counterparts, delq!, delv!, and delete!, now remove all matching
7769 elements from the list, not just the first. This matches the behavior
7770 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7771 functions which inspired them.
7773 I recognize that this change may break code in subtle ways, but it
7774 seems best to make the change before the FSF's first Guile release,
7778 ** The compiled-library-path function has been deleted from libguile.
7780 ** The facilities for loading Scheme source files have changed.
7782 *** The variable %load-path now tells Guile which directories to search
7783 for Scheme code. Its value is a list of strings, each of which names
7786 *** The variable %load-extensions now tells Guile which extensions to
7787 try appending to a filename when searching the load path. Its value
7788 is a list of strings. Its default value is ("" ".scm").
7790 *** (%search-load-path FILENAME) searches the directories listed in the
7791 value of the %load-path variable for a Scheme file named FILENAME,
7792 with all the extensions listed in %load-extensions. If it finds a
7793 match, then it returns its full filename. If FILENAME is absolute, it
7794 returns it unchanged. Otherwise, it returns #f.
7796 %search-load-path will not return matches that refer to directories.
7798 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7799 uses %seach-load-path to find a file named FILENAME, and loads it if
7800 it finds it. If it can't read FILENAME for any reason, it throws an
7803 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7806 *** load uses the same searching semantics as primitive-load.
7808 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7809 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7810 path, and load-module-with-path have been deleted. The functions
7811 above should serve their purposes.
7813 *** If the value of the variable %load-hook is a procedure,
7814 `primitive-load' applies its value to the name of the file being
7815 loaded (without the load path directory name prepended). If its value
7816 is #f, it is ignored. Otherwise, an error occurs.
7818 This is mostly useful for printing load notification messages.
7821 ** The function `eval!' is no longer accessible from the scheme level.
7822 We can't allow operations which introduce glocs into the scheme level,
7823 because Guile's type system can't handle these as data. Use `eval' or
7824 `read-and-eval!' (see below) as replacement.
7826 ** The new function read-and-eval! reads an expression from PORT,
7827 evaluates it, and returns the result. This is more efficient than
7828 simply calling `read' and `eval', since it is not necessary to make a
7829 copy of the expression for the evaluator to munge.
7831 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7832 for the `read' function.
7835 ** The function `int?' has been removed; its definition was identical
7836 to that of `integer?'.
7838 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7839 use the R4RS names for these functions.
7841 ** The function object-properties no longer returns the hash handle;
7842 it simply returns the object's property list.
7844 ** Many functions have been changed to throw errors, instead of
7845 returning #f on failure. The point of providing exception handling in
7846 the language is to simplify the logic of user code, but this is less
7847 useful if Guile's primitives don't throw exceptions.
7849 ** The function `fileno' has been renamed from `%fileno'.
7851 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7854 * Changes to Guile's C interface:
7856 ** The library's initialization procedure has been simplified.
7857 scm_boot_guile now has the prototype:
7859 void scm_boot_guile (int ARGC,
7861 void (*main_func) (),
7864 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7865 MAIN_FUNC should do all the work of the program (initializing other
7866 packages, reading user input, etc.) before returning. When MAIN_FUNC
7867 returns, call exit (0); this function never returns. If you want some
7868 other exit value, MAIN_FUNC may call exit itself.
7870 scm_boot_guile arranges for program-arguments to return the strings
7871 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7872 scm_set_program_arguments with the final list, so Scheme code will
7873 know which arguments have been processed.
7875 scm_boot_guile establishes a catch-all catch handler which prints an
7876 error message and exits the process. This means that Guile exits in a
7877 coherent way when system errors occur and the user isn't prepared to
7878 handle it. If the user doesn't like this behavior, they can establish
7879 their own universal catcher in MAIN_FUNC to shadow this one.
7881 Why must the caller do all the real work from MAIN_FUNC? The garbage
7882 collector assumes that all local variables of type SCM will be above
7883 scm_boot_guile's stack frame on the stack. If you try to manipulate
7884 SCM values after this function returns, it's the luck of the draw
7885 whether the GC will be able to find the objects you allocate. So,
7886 scm_boot_guile function exits, rather than returning, to discourage
7887 people from making that mistake.
7889 The IN, OUT, and ERR arguments were removed; there are other
7890 convenient ways to override these when desired.
7892 The RESULT argument was deleted; this function should never return.
7894 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7898 ** Guile's header files should no longer conflict with your system's
7901 In order to compile code which #included <libguile.h>, previous
7902 versions of Guile required you to add a directory containing all the
7903 Guile header files to your #include path. This was a problem, since
7904 Guile's header files have names which conflict with many systems'
7907 Now only <libguile.h> need appear in your #include path; you must
7908 refer to all Guile's other header files as <libguile/mumble.h>.
7909 Guile's installation procedure puts libguile.h in $(includedir), and
7910 the rest in $(includedir)/libguile.
7913 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7914 have been added to the Guile library.
7916 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7917 OBJ will not be freed, even if all other references are dropped,
7918 until someone does scm_unprotect_object (OBJ). Both functions
7921 Note that calls to scm_protect_object do not nest. You can call
7922 scm_protect_object any number of times on a given object, and the
7923 next call to scm_unprotect_object will unprotect it completely.
7925 Basically, scm_protect_object and scm_unprotect_object just
7926 maintain a list of references to things. Since the GC knows about
7927 this list, all objects it mentions stay alive. scm_protect_object
7928 adds its argument to the list; scm_unprotect_object remove its
7929 argument from the list.
7932 ** scm_eval_0str now returns the value of the last expression
7935 ** The new function scm_read_0str reads an s-expression from a
7936 null-terminated string, and returns it.
7938 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7939 to a Scheme port object.
7941 ** The new function `scm_set_program_arguments' allows C code to set
7942 the value returned by the Scheme `program-arguments' function.
7947 * Guile no longer includes sophisticated Tcl/Tk support.
7949 The old Tcl/Tk support was unsatisfying to us, because it required the
7950 user to link against the Tcl library, as well as Tk and Guile. The
7951 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7952 referring to widgets by names, rather than exporting widgets to Scheme
7953 code as a special datatype.
7955 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7956 maintainers described some very interesting changes in progress to the
7957 Tcl/Tk internals, which would facilitate clean interfaces between lone
7958 Tk and other interpreters --- even for garbage-collected languages
7959 like Scheme. They expected the new Tk to be publicly available in the
7962 Since it seems that Guile might soon have a new, cleaner interface to
7963 lone Tk, and that the old Guile/Tk glue code would probably need to be
7964 completely rewritten, we (Jim Blandy and Richard Stallman) have
7965 decided not to support the old code. We'll spend the time instead on
7966 a good interface to the newer Tk, as soon as it is available.
7968 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7971 Copyright information:
7973 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7975 Permission is granted to anyone to make or distribute verbatim copies
7976 of this document as received, in any medium, provided that the
7977 copyright notice and this permission notice are preserved,
7978 thus giving the recipient permission to redistribute in turn.
7980 Permission is granted to distribute modified versions
7981 of this document, or of portions of it,
7982 under the above conditions, provided also that they
7983 carry prominent notices stating who last changed them.
7988 paragraph-separate: "[
\f]*$"