1 Guile NEWS --- history of user-visible changes. -*- text -*-
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
7 Changes since Guile 1.4:
9 * Changes to the distribution
11 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
13 Guile now always uses three numbers to represent the version,
14 i.e. "1.6.5". The first number, 1, is the major version number, the
15 second number, 6, is the minor version number, and the third number,
16 5, is the micro version number. Changes in major version number
17 indicate major changes in Guile.
19 Minor version numbers that are even denote stable releases, and odd
20 minor version numbers denote development versions (which may be
21 unstable). The micro version number indicates a minor sub-revision of
22 a given MAJOR.MINOR release.
24 In keeping with the new scheme, (minor-version) and scm_minor_version
25 no longer return everything but the major version number. They now
26 just return the minor version number. Two new functions
27 (micro-version) and scm_micro_version have been added to report the
30 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
32 ** Guile now actively warns about deprecated features.
34 The new configure option `--enable-deprecated=LEVEL' and the
35 environment variable GUILE_WARN_DEPRECATED control this mechanism.
36 See INSTALL and README for more information.
38 ** New SRFI modules have been added:
40 SRFI-0 `cond-expand' is now supported in Guile, without requiring
43 (srfi srfi-2) exports and-let*.
45 (srfi srfi-6) is a dummy module for now, since guile already provides
46 all of the srfi-6 procedures by default: open-input-string,
47 open-output-string, get-output-string.
49 (srfi srfi-8) exports receive.
51 (srfi srfi-9) exports define-record-type.
53 (srfi srfi-10) exports define-reader-ctor and implements the reader
56 (srfi srfi-11) exports let-values and let*-values.
58 (srfi srfi-13) implements the SRFI String Library.
60 (srfi srfi-14) implements the SRFI Character-Set Library.
62 (srfi srfi-17) implements setter and getter-with-setter and redefines
63 some accessor procedures as procedures with getters. (such as car,
66 (srfi srfi-19) implements the SRFI Time/Date Library.
68 ** New scripts / "executable modules"
70 Subdirectory "scripts" contains Scheme modules that are packaged to
71 also be executable as scripts. At this time, these scripts are available:
80 See README there for more info.
82 These scripts can be invoked from the shell with the new program
83 "guile-tools", which keeps track of installation directory for you.
86 $ guile-tools display-commentary srfi/*.scm
88 guile-tools is copied to the standard $bindir on "make install".
90 ** New module (ice-9 stack-catch):
92 stack-catch is like catch, but saves the current state of the stack in
93 the fluid the-last-stack. This fluid can be useful when using the
94 debugger and when re-throwing an error.
96 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
98 This has been done to prevent problems on lesser operating systems
99 that can't tolerate `*'s in file names. The exported macro continues
100 to be named `and-let*', of course.
102 On systems that support it, there is also a compatibility module named
103 (ice-9 and-let*). It will go away in the next release.
105 ** New modules (oop goops) etc.:
110 (oop goops active-slot)
111 (oop goops composite-slot)
113 The Guile Object Oriented Programming System (GOOPS) has been
114 integrated into Guile. For further information, consult the GOOPS
115 manual and tutorial in the `doc' directory.
117 ** New module (ice-9 rdelim).
119 This exports the following procedures which were previously defined
120 in the default environment:
122 read-line read-line! read-delimited read-delimited! %read-delimited!
123 %read-line write-line
125 For backwards compatibility the definitions are still imported into the
126 default environment in this version of Guile. However you should add:
128 (use-modules (ice-9 rdelim))
130 to any program which uses the definitions, since this may change in
133 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
134 can be used for similar functionality.
136 ** New module (ice-9 rw)
138 This is a subset of the (scsh rw) module from guile-scsh. Currently
139 it defines a single procedure:
141 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
143 Read characters from an fport or file descriptor into a string
144 STR. This procedure is scsh-compatible and can efficiently read
147 ** New module (ice-9 match)
149 This module includes Andrew K. Wright's pattern matcher. See
150 ice-9/match.scm for brief description or
152 http://www.star-lab.com/wright/code.html
154 for complete documentation.
156 This module requires SLIB to be installed and available from Guile.
158 ** New module (ice-9 buffered-input)
160 This module provides procedures to construct an input port from an
161 underlying source of input that reads and returns its input in chunks.
162 The underlying input source is a Scheme procedure, specified by the
163 caller, which the port invokes whenever it needs more input.
165 This is useful when building an input port whose back end is Readline
166 or a UI element such as the GtkEntry widget.
170 The reference and tutorial documentation that was previously
171 distributed separately, as `guile-doc', is now included in the core
172 Guile distribution. The documentation consists of the following
175 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
178 - The Guile Reference Manual (guile.texi) contains (or is intended to
179 contain) reference documentation on all aspects of Guile.
181 - The GOOPS Manual (goops.texi) contains both tutorial-style and
182 reference documentation for using GOOPS, Guile's Object Oriented
185 - The Revised^5 Report on the Algorithmic Language Scheme
188 See the README file in the `doc' directory for more details.
190 * Changes to the stand-alone interpreter
192 ** New command line option `--use-srfi'
194 Using this option, SRFI modules can be loaded on startup and be
195 available right from the beginning. This makes programming portable
196 Scheme programs easier.
198 The option `--use-srfi' expects a comma-separated list of numbers,
199 each representing a SRFI number to be loaded into the interpreter
200 before starting evaluating a script file or the REPL. Additionally,
201 the feature identifier for the loaded SRFIs is recognized by
202 `cond-expand' when using this option.
205 $ guile --use-srfi=8,13
206 guile> (receive (x z) (values 1 2) (+ 1 2))
208 guile> (string-pad "bla" 20)
212 * Changes to Scheme functions and syntax
214 ** Previously deprecated Scheme functions have been removed:
216 tag - no replacement.
217 fseek - replaced by seek.
218 list* - replaced by cons*.
220 ** It's now possible to create modules with controlled environments
224 (use-modules (ice-9 safe))
225 (define m (make-safe-module))
226 ;;; m will now be a module containing only a safe subset of R5RS
227 (eval '(+ 1 2) m) --> 3
228 (eval 'load m) --> ERROR: Unbound variable: load
230 ** Evaluation of "()", the empty list, is now an error.
232 Previously, the expression "()" evaluated to the empty list. This has
233 been changed to signal a "missing expression" error. The correct way
234 to write the empty list as a literal constant is to use quote: "'()".
236 ** New concept of `Guile Extensions'.
238 A Guile Extension is just a ordinary shared library that can be linked
239 at run-time. We found it advantageous to give this simple concept a
240 dedicated name to distinguish the issues related to shared libraries
241 from the issues related to the module system.
243 *** New function: load-extension
245 Executing (load-extension lib init) is mostly equivalent to
247 (dynamic-call init (dynamic-link lib))
249 except when scm_register_extension has been called previously.
250 Whenever appropriate, you should use `load-extension' instead of
251 dynamic-link and dynamic-call.
253 *** New C function: scm_c_register_extension
255 This function registers a initialization function for use by
256 `load-extension'. Use it when you don't want specific extensions to
257 be loaded as shared libraries (for example on platforms that don't
258 support dynamic linking).
260 ** Auto-loading of compiled-code modules is deprecated.
262 Guile used to be able to automatically find and link a shared
263 library to satisfy requests for a module. For example, the module
264 `(foo bar)' could be implemented by placing a shared library named
265 "foo/libbar.so" (or with a different extension) in a directory on the
268 This has been found to be too tricky, and is no longer supported. The
269 shared libraries are now called "extensions". You should now write a
270 small Scheme file that calls `load-extension' to load the shared
271 library and initialize it explicitely.
273 The shared libraries themselves should be installed in the usual
274 places for shared libraries, with names like "libguile-foo-bar".
276 For example, place this into a file "foo/bar.scm"
278 (define-module (foo bar))
280 (load-extension "libguile-foo-bar" "foobar_init")
282 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
284 `eval' is now R5RS, that is it takes two arguments.
285 The second argument is an environment specifier, i.e. either
287 (scheme-report-environment 5)
289 (interaction-environment)
295 ** The module system has been made more disciplined.
297 The function `eval' will save and restore the current module around
298 the evaluation of the specified expression. While this expression is
299 evaluated, `(current-module)' will now return the right module, which
300 is the module specified as the second argument to `eval'.
302 A consequence of this change is that `eval' is not particularly
303 useful when you want allow the evaluated code to change what module is
304 designated as the current module and have this change persist from one
305 call to `eval' to the next. The read-eval-print-loop is an example
306 where `eval' is now inadequate. To compensate, there is a new
307 function `primitive-eval' that does not take a module specifier and
308 that does not save/restore the current module. You should use this
309 function together with `set-current-module', `current-module', etc
310 when you want to have more control over the state that is carried from
311 one eval to the next.
313 Additionally, it has been made sure that forms that are evaluated at
314 the top level are always evaluated with respect to the current module.
315 Previously, subforms of top-level forms such as `begin', `case',
316 etc. did not respect changes to the current module although these
317 subforms are at the top-level as well.
319 To prevent strange behavior, the forms `define-module',
320 `use-modules', `use-syntax', and `export' have been restricted to only
321 work on the top level. The forms `define-public' and
322 `defmacro-public' only export the new binding on the top level. They
323 behave just like `define' and `defmacro', respectively, when they are
324 used in a lexical environment.
326 ** The semantics of guardians have changed.
328 The changes are for the most part compatible. An important criterion
329 was to keep the typical usage of guardians as simple as before, but to
330 make the semantics safer and (as a result) more useful.
332 *** All objects returned from guardians are now properly alive.
334 It is now guaranteed that any object referenced by an object returned
335 from a guardian is alive. It's now impossible for a guardian to
336 return a "contained" object before its "containing" object.
338 One incompatible (but probably not very important) change resulting
339 from this is that it is no longer possible to guard objects that
340 indirectly reference themselves (i.e. are parts of cycles). If you do
341 so accidentally, you'll get a warning.
343 *** There are now two types of guardians: greedy and sharing.
345 If you call (make-guardian #t) or just (make-guardian), you'll get a
346 greedy guardian, and for (make-guardian #f) a sharing guardian.
348 Greedy guardians are the default because they are more "defensive".
349 You can only greedily guard an object once. If you guard an object
350 more than once, once in a greedy guardian and the rest of times in
351 sharing guardians, then it is guaranteed that the object won't be
352 returned from sharing guardians as long as it is greedily guarded
355 Guardians returned by calls to `make-guardian' can now take one more
356 optional parameter, which says whether to throw an error in case an
357 attempt is made to greedily guard an object that is already greedily
358 guarded. The default is true, i.e. throw an error. If the parameter
359 is false, the guardian invocation returns #t if guarding was
360 successful and #f if it wasn't.
362 Also, since greedy guarding is, in effect, a side-effecting operation
363 on objects, a new function is introduced: `destroy-guardian!'.
364 Invoking this function on a guardian renders it unoperative and, if
365 the guardian is greedy, clears the "greedily guarded" property of the
366 objects that were guarded by it, thus undoing the side effect.
368 Note that all this hair is hardly very important, since guardian
369 objects are usually permanent.
371 ** Continuations created by call-with-current-continuation now accept
372 any number of arguments, as required by R5RS.
374 ** New function `issue-deprecation-warning'
376 This function is used to display the deprecation messages that are
377 controlled by GUILE_WARN_DEPRECATION as explained in the README.
380 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
384 ;; `id' is deprecated. Use `identity' instead.
389 ** New syntax `begin-deprecated'
391 When deprecated features are included (as determined by the configure
392 option --enable-deprecated), `begin-deprecated' is identical to
393 `begin'. When deprecated features are excluded, it always evaluates
394 to `#f', ignoring the body forms.
396 ** New function `make-object-property'
398 This function returns a new `procedure with setter' P that can be used
399 to attach a property to objects. When calling P as
403 where `obj' is any kind of object, it attaches `val' to `obj' in such
404 a way that it can be retrieved by calling P as
408 This function will replace procedure properties, symbol properties and
409 source properties eventually.
411 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
413 Instead of #&optional, #&key, etc you should now use #:optional,
414 #:key, etc. Since #:optional is a keyword, you can write it as just
415 :optional when (read-set! keywords 'prefix) is active.
417 The old reader syntax `#&' is still supported, but deprecated. It
418 will be removed in the next release.
420 ** New define-module option: pure
422 Tells the module system not to include any bindings from the root
427 (define-module (totally-empty-module)
430 ** New define-module option: export NAME1 ...
432 Export names NAME1 ...
434 This option is required if you want to be able to export bindings from
435 a module which doesn't import one of `define-public' or `export'.
441 :use-module (ice-9 r5rs)
444 ;;; Note that we're pure R5RS below this point!
449 ** New function: object->string OBJ
451 Return a Scheme string obtained by printing a given object.
453 ** New function: port? X
455 Returns a boolean indicating whether X is a port. Equivalent to
456 `(or (input-port? X) (output-port? X))'.
458 ** New function: file-port?
460 Determines whether a given object is a port that is related to a file.
462 ** New function: port-for-each proc
464 Apply PROC to each port in the Guile port table in turn. The return
465 value is unspecified. More specifically, PROC is applied exactly once
466 to every port that exists in the system at the time PORT-FOR-EACH is
467 invoked. Changes to the port table while PORT-FOR-EACH is running
468 have no effect as far as PORT-FOR-EACH is concerned.
470 ** New function: dup2 oldfd newfd
472 A simple wrapper for the `dup2' system call. Copies the file
473 descriptor OLDFD to descriptor number NEWFD, replacing the
474 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
475 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
476 to move away ports which are using NEWFD. The return value is
479 ** New function: close-fdes fd
481 A simple wrapper for the `close' system call. Close file
482 descriptor FD, which must be an integer. Unlike close (*note
483 close: Ports and File Descriptors.), the file descriptor will be
484 closed even if a port is using it. The return value is
487 ** New function: crypt password salt
489 Encrypts `password' using the standard unix password encryption
492 ** New function: chroot path
494 Change the root directory of the running process to `path'.
496 ** New functions: getlogin, cuserid
498 Return the login name or the user name of the current effective user
501 ** New functions: getpriority which who, setpriority which who prio
503 Get or set the priority of the running process.
505 ** New function: getpass prompt
507 Read a password from the terminal, first displaying `prompt' and
510 ** New function: flock file operation
512 Set/remove an advisory shared or exclusive lock on `file'.
514 ** New functions: sethostname name, gethostname
516 Set or get the hostname of the machine the current process is running
519 ** New function: mkstemp! tmpl
521 mkstemp creates a new unique file in the file system and returns a
522 new buffered port open for reading and writing to the file. TMPL
523 is a string specifying where the file should be created: it must
524 end with `XXXXXX' and will be changed in place to return the name
525 of the temporary file.
527 ** New function: open-input-string string
529 Return an input string port which delivers the characters from
530 `string'. This procedure, together with `open-output-string' and
531 `get-output-string' implements SRFI-6.
533 ** New function: open-output-string
535 Return an output string port which collects all data written to it.
536 The data can then be retrieved by `get-output-string'.
538 ** New function: get-output-string
540 Return the contents of an output string port.
542 ** New function: identity
546 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
547 are represented in Scheme as integers with normal host byte ordering.
549 ** New function: inet-pton family address
551 Convert a printable string network address into an integer. Note that
552 unlike the C version of this function, the result is an integer with
553 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
556 (inet-pton AF_INET "127.0.0.1") => 2130706433
557 (inet-pton AF_INET6 "::1") => 1
559 ** New function: inet-ntop family address
561 Convert an integer network address into a printable string. Note that
562 unlike the C version of this function, the input is an integer with
563 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
566 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
567 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
568 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
572 Use `identity' instead.
578 ** Deprecated: return-it
582 ** Deprecated: string-character-length
584 Use `string-length' instead.
588 Use `logior' instead.
590 ** Deprecated: close-all-ports-except.
592 This was intended for closing ports in a child process after a fork,
593 but it has the undesirable side effect of flushing buffers.
594 port-for-each is more flexible.
596 ** The (ice-9 popen) module now attempts to set up file descriptors in
597 the child process from the current Scheme ports, instead of using the
598 current values of file descriptors 0, 1, and 2 in the parent process.
600 ** Removed function: builtin-weak-bindings
602 There is no such concept as a weak binding any more.
604 ** Removed constants: bignum-radix, scm-line-incrementors
606 ** define-method: New syntax mandatory.
608 The new method syntax is now mandatory:
610 (define-method (NAME ARG-SPEC ...) BODY ...)
611 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
613 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
614 REST-ARG ::= ARG-NAME
616 If you have old code using the old syntax, import
617 (oop goops old-define-method) before (oop goops) as in:
619 (use-modules (oop goops old-define-method) (oop goops))
621 ** Deprecated function: builtin-variable
622 Removed function: builtin-bindings
624 There is no longer a distinction between builtin or other variables.
625 Use module system operations for all variables.
627 ** Lazy-catch handlers are no longer allowed to return.
629 That is, a call to `throw', `error', etc is now guaranteed to not
632 * Changes to the C interface
634 ** Deprecated feature have been removed.
638 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
639 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
641 *** C Functions removed
643 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
644 scm_fseek - replaced by scm_seek.
645 gc-thunk - replaced by after-gc-hook.
646 gh_int2scmb - replaced by gh_bool2scm.
647 scm_tc_dblr - replaced by scm_tc16_real.
648 scm_tc_dblc - replaced by scm_tc16_complex.
649 scm_list_star - replaced by scm_cons_star.
651 ** Deprecated: scm_make_shared_substring
653 Explicit shared substrings will disappear from Guile.
655 Instead, "normal" strings will be implemented using sharing
656 internally, combined with a copy-on-write strategy.
658 ** Deprecated: scm_read_only_string_p
660 The concept of read-only strings will disappear in next release of
663 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
665 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
667 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
669 Used by an application to read arbitrary number of bytes from a port.
670 Same semantics as libc read, except that scm_c_read only returns less
671 than SIZE bytes if at end-of-file.
673 Warning: Doesn't update port line and column counts!
675 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
677 Used by an application to write arbitrary number of bytes to an SCM
678 port. Similar semantics as libc write. However, unlike libc
679 write, scm_c_write writes the requested number of bytes and has no
682 Warning: Doesn't update port line and column counts!
684 ** New function: scm_init_guile ()
686 In contrast to scm_boot_guile, scm_init_guile will return normally
687 after initializing Guile. It is not available on all systems, tho.
689 ** New functions: scm_str2symbol, scm_mem2symbol
691 The function scm_str2symbol takes a const char* pointing to a zero-terminated
692 field of characters and creates a scheme symbol object from that C string.
693 The function scm_mem2symbol takes a const char* and a number of characters and
694 creates a symbol from the characters in that memory area.
696 ** New functions: scm_primitive_make_property
697 scm_primitive_property_ref
698 scm_primitive_property_set_x
699 scm_primitive_property_del_x
701 These functions implement a new way to deal with object properties.
702 See libguile/properties.c for their documentation.
704 ** New function: scm_done_free (long size)
706 This function is the inverse of scm_done_malloc. Use it to report the
707 amount of smob memory you free. The previous method, which involved
708 calling scm_done_malloc with negative argument, was somewhat
709 unintuitive (and is still available, of course).
711 ** New function: scm_c_memq (SCM obj, SCM list)
713 This function provides a fast C level alternative for scm_memq for the case
714 that the list parameter is known to be a proper list. The function is a
715 replacement for scm_sloppy_memq, but is stricter in its requirements on its
716 list input parameter, since for anything else but a proper list the function's
717 behaviour is undefined - it may even crash or loop endlessly. Further, for
718 the case that the object is not found in the list, scm_c_memq returns #f which
719 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
721 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
722 scm_remember_upto_here
724 These functions replace the function scm_remember.
726 ** Deprecated function: scm_remember
728 Use one of the new functions scm_remember_upto_here_1,
729 scm_remember_upto_here_2 or scm_remember_upto_here instead.
731 ** New function: scm_allocate_string
733 This function replaces the function scm_makstr.
735 ** Deprecated function: scm_makstr
737 Use the new function scm_allocate_string instead.
739 ** New global variable scm_gc_running_p introduced.
741 Use this variable to find out if garbage collection is being executed. Up to
742 now applications have used scm_gc_heap_lock to test if garbage collection was
743 running, which also works because of the fact that up to know only the garbage
744 collector has set this variable. But, this is an implementation detail that
745 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
746 of this variable is (and has been) not fully safe anyway.
748 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
750 Use these instead of SCM_LENGTH_MAX.
752 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
753 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
754 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
756 Use these instead of SCM_LENGTH.
758 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
759 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
760 SCM_SET_BITVECTOR_LENGTH
762 Use these instead of SCM_SETLENGTH
764 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
765 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
768 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
771 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
772 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
775 Use these instead of SCM_SETCHARS.
777 ** New macro: SCM_BITVECTOR_P
779 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
781 Use instead of SCM_COERCE_SUBSTR.
783 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
785 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
787 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
788 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
789 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
790 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
791 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
792 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
793 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
794 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
795 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
796 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
797 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
798 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
799 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
800 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
801 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
803 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
804 Use scm_memory_error instead of SCM_NALLOC.
805 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
806 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
807 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
808 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
809 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
810 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
811 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
812 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
813 Use SCM_STRINGP instead of SCM_RWSTRINGP.
814 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
815 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
816 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
817 Use a type specific setter macro instead of SCM_SETLENGTH.
818 Use a type specific setter macro instead of SCM_SETCHARS.
819 Use a type specific length macro instead of SCM_LENGTH_MAX.
820 Use SCM_GCMARKP instead of SCM_GC8MARKP.
821 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
822 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
823 Use SCM_TYP16 instead of SCM_GCTYP16.
824 Use SCM_CDR instead of SCM_GCCDR.
825 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
826 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
827 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
828 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
829 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
830 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
831 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
833 ** Removed function: scm_struct_init
835 ** Removed variable: scm_symhash_dim
837 ** Renamed function: scm_make_cont has been replaced by
838 scm_make_continuation, which has a different interface.
840 ** Deprecated function: scm_call_catching_errors
842 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
844 ** Deprecated function: scm_strhash
846 Use scm_string_hash instead.
848 ** Deprecated function: scm_vector_set_length_x
850 Instead, create a fresh vector of the desired size and copy the contents.
852 ** scm_gensym has changed prototype
854 scm_gensym now only takes one argument.
856 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
859 There is now only a single symbol type scm_tc7_symbol.
860 The tag scm_tc7_lvector was not used anyway.
862 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
864 Use scm_make_smob_type and scm_set_smob_XXX instead.
866 ** New function scm_set_smob_apply.
868 This can be used to set an apply function to a smob type.
870 ** Deprecated function: scm_strprint_obj
872 Use scm_object_to_string instead.
874 ** Deprecated function: scm_wta
876 Use scm_wrong_type_arg, or another appropriate error signalling function
879 ** Explicit support for obarrays has been deprecated.
881 Use `scm_str2symbol' and the generic hashtable functions instead.
883 ** The concept of `vcells' has been deprecated.
885 The data type `variable' is now used exclusively. `Vcells' have been
886 a low-level concept so you are likely not affected by this change.
888 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
889 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
891 Use scm_c_define or scm_c_lookup instead, as appropriate.
893 *** New functions: scm_c_module_lookup, scm_c_lookup,
894 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
895 scm_module_define, scm_define.
897 These functions work with variables instead of with vcells.
899 ** New functions for creating and defining `subr's and `gsubr's.
901 The new functions more clearly distinguish between creating a subr (or
902 gsubr) object and adding it to the current module.
904 These new functions are available: scm_c_make_subr, scm_c_define_subr,
905 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
906 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
907 scm_c_define_gsubr_with_generic.
909 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
910 scm_make_subr_with_generic, scm_make_gsubr,
911 scm_make_gsubr_with_generic.
913 Use the new ones from above instead.
915 ** C interface to the module system has changed.
917 While we suggest that you avoid as many explicit module system
918 operations from C as possible for the time being, the C interface has
919 been made more similar to the high-level Scheme module system.
921 *** New functions: scm_c_define_module, scm_c_use_module,
922 scm_c_export, scm_c_resolve_module.
924 They mostly work like their Scheme namesakes. scm_c_define_module
925 takes a function that is called a context where the new module is
928 *** Deprecated functions: scm_the_root_module, scm_make_module,
929 scm_ensure_user_module, scm_load_scheme_module.
931 Use the new functions instead.
933 ** Renamed function: scm_internal_with_fluids becomes
936 scm_internal_with_fluids is available as a deprecated function.
938 ** New function: scm_c_with_fluid.
940 Just like scm_c_with_fluids, but takes one fluid and one value instead
943 ** Deprecated typedefs: long_long, ulong_long.
945 They are of questionable utility and they pollute the global
948 ** Deprecated typedef: scm_sizet
950 It is of questionable utility now that Guile requires ANSI C, and is
953 ** Deprecated typedefs: scm_port_rw_active, scm_port,
954 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
955 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
957 Made more compliant with the naming policy by adding a _t at the end.
959 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
960 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
962 With the exception of the misterious scm_2ulong2big, they are still
963 available under new names (scm_i_mkbig etc). These functions are not
964 intended to be used in user code. You should avoid dealing with
965 bignums directly, and should deal with numbers in general (which can
968 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
969 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
970 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
973 These are conversion functions between the various ANSI C integral
974 types and Scheme numbers.
976 ** New number validation macros:
977 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
981 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
983 These are just nicer-named old scm_protect_object and
984 scm_unprotect_object.
986 ** Deprecated functions: scm_protect_object, scm_unprotect_object
988 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
990 These functions can be used to register pointers to locations that
994 Changes since Guile 1.3.4:
996 * Changes to the distribution
998 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
1000 We've changed the way we handle generated files in the Guile source
1001 repository. As a result, the procedure for building trees obtained
1002 from the nightly FTP snapshots or via CVS has changed:
1003 - You must have appropriate versions of autoconf, automake, and
1004 libtool installed on your system. See README for info on how to
1005 obtain these programs.
1006 - Before configuring the tree, you must first run the script
1007 `autogen.sh' at the top of the source tree.
1009 The Guile repository used to contain not only source files, written by
1010 humans, but also some generated files, like configure scripts and
1011 Makefile.in files. Even though the contents of these files could be
1012 derived mechanically from other files present, we thought it would
1013 make the tree easier to build if we checked them into CVS.
1015 However, this approach means that minor differences between
1016 developer's installed tools and habits affected the whole team.
1017 So we have removed the generated files from the repository, and
1018 added the autogen.sh script, which will reconstruct them
1022 ** configure now has experimental options to remove support for certain
1025 --disable-arrays omit array and uniform array support
1026 --disable-posix omit posix interfaces
1027 --disable-networking omit networking interfaces
1028 --disable-regex omit regular expression interfaces
1030 These are likely to become separate modules some day.
1032 ** New configure option --enable-debug-freelist
1034 This enables a debugging version of SCM_NEWCELL(), and also registers
1035 an extra primitive, the setter `gc-set-debug-check-freelist!'.
1037 Configure with the --enable-debug-freelist option to enable
1038 the gc-set-debug-check-freelist! primitive, and then use:
1040 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
1041 (gc-set-debug-check-freelist! #f) # turn off checking
1043 Checking of the freelist forces a traversal of the freelist and
1044 a garbage collection before each allocation of a cell. This can
1045 slow down the interpreter dramatically, so the setter should be used to
1046 turn on this extra processing only when necessary.
1048 ** New configure option --enable-debug-malloc
1050 Include code for debugging of calls to scm_must_malloc/realloc/free.
1054 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
1055 2. objects reallocated by scm_must_realloc has been allocated by
1057 3. reallocated objects are reallocated with the same what string
1059 But, most importantly, it records the number of allocated objects of
1060 each kind. This is useful when searching for memory leaks.
1062 A Guile compiled with this option provides the primitive
1063 `malloc-stats' which returns an alist with pairs of kind and the
1064 number of objects of that kind.
1066 ** All includes are now referenced relative to the root directory
1068 Since some users have had problems with mixups between Guile and
1069 system headers, we have decided to always refer to Guile headers via
1070 their parent directories. This essentially creates a "private name
1071 space" for Guile headers. This means that the compiler only is given
1072 -I options for the root build and root source directory.
1074 ** Header files kw.h and genio.h have been removed.
1076 ** The module (ice-9 getopt-gnu-style) has been removed.
1078 ** New module (ice-9 documentation)
1080 Implements the interface to documentation strings associated with
1083 ** New module (ice-9 time)
1085 Provides a macro `time', which displays execution time of a given form.
1087 ** New module (ice-9 history)
1089 Loading this module enables value history in the repl.
1091 * Changes to the stand-alone interpreter
1093 ** New command line option --debug
1095 Start Guile with debugging evaluator and backtraces enabled.
1097 This is useful when debugging your .guile init file or scripts.
1099 ** New help facility
1101 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
1102 (help REGEXP) ditto for objects with names matching REGEXP (a string)
1103 (help 'NAME) gives documentation for NAME, even if it is not an object
1104 (help ,EXPR) gives documentation for object returned by EXPR
1105 (help (my module)) gives module commentary for `(my module)'
1106 (help) gives this text
1108 `help' searches among bindings exported from loaded modules, while
1109 `apropos' searches among bindings visible from the "current" module.
1111 Examples: (help help)
1113 (help "output-string")
1115 ** `help' and `apropos' now prints full module names
1117 ** Dynamic linking now uses libltdl from the libtool package.
1119 The old system dependent code for doing dynamic linking has been
1120 replaced with calls to the libltdl functions which do all the hairy
1123 The major improvement is that you can now directly pass libtool
1124 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
1125 will be able to do the best shared library job you can get, via
1128 The way dynamic libraries are found has changed and is not really
1129 portable across platforms, probably. It is therefore recommended to
1130 use absolute filenames when possible.
1132 If you pass a filename without an extension to `dynamic-link', it will
1133 try a few appropriate ones. Thus, the most platform ignorant way is
1134 to specify a name like "libfoo", without any directories and
1137 ** Guile COOP threads are now compatible with LinuxThreads
1139 Previously, COOP threading wasn't possible in applications linked with
1140 Linux POSIX threads due to their use of the stack pointer to find the
1141 thread context. This has now been fixed with a workaround which uses
1142 the pthreads to allocate the stack.
1144 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
1146 ** Positions of erring expression in scripts
1148 With version 1.3.4, the location of the erring expression in Guile
1149 scipts is no longer automatically reported. (This should have been
1150 documented before the 1.3.4 release.)
1152 You can get this information by enabling recording of positions of
1153 source expressions and running the debugging evaluator. Put this at
1154 the top of your script (or in your "site" file):
1156 (read-enable 'positions)
1157 (debug-enable 'debug)
1159 ** Backtraces in scripts
1161 It is now possible to get backtraces in scripts.
1165 (debug-enable 'debug 'backtrace)
1167 at the top of the script.
1169 (The first options enables the debugging evaluator.
1170 The second enables backtraces.)
1172 ** Part of module system symbol lookup now implemented in C
1174 The eval closure of most modules is now implemented in C. Since this
1175 was one of the bottlenecks for loading speed, Guile now loads code
1176 substantially faster than before.
1178 ** Attempting to get the value of an unbound variable now produces
1179 an exception with a key of 'unbound-variable instead of 'misc-error.
1181 ** The initial default output port is now unbuffered if it's using a
1182 tty device. Previously in this situation it was line-buffered.
1184 ** New hook: after-gc-hook
1186 after-gc-hook takes over the role of gc-thunk. This hook is run at
1187 the first SCM_TICK after a GC. (Thus, the code is run at the same
1188 point during evaluation as signal handlers.)
1190 Note that this hook should be used only for diagnostic and debugging
1191 purposes. It is not certain that it will continue to be well-defined
1192 when this hook is run in the future.
1194 C programmers: Note the new C level hooks scm_before_gc_c_hook,
1195 scm_before_sweep_c_hook, scm_after_gc_c_hook.
1197 ** Improvements to garbage collector
1199 Guile 1.4 has a new policy for triggering heap allocation and
1200 determining the sizes of heap segments. It fixes a number of problems
1203 1. The new policy can handle two separate pools of cells
1204 (2-word/4-word) better. (The old policy would run wild, allocating
1205 more and more memory for certain programs.)
1207 2. The old code would sometimes allocate far too much heap so that the
1208 Guile process became gigantic. The new code avoids this.
1210 3. The old code would sometimes allocate too little so that few cells
1211 were freed at GC so that, in turn, too much time was spent in GC.
1213 4. The old code would often trigger heap allocation several times in a
1214 row. (The new scheme predicts how large the segments needs to be
1215 in order not to need further allocation.)
1217 All in all, the new GC policy will make larger applications more
1220 The new GC scheme also is prepared for POSIX threading. Threads can
1221 allocate private pools of cells ("clusters") with just a single
1222 function call. Allocation of single cells from such a cluster can
1223 then proceed without any need of inter-thread synchronization.
1225 ** New environment variables controlling GC parameters
1227 GUILE_MAX_SEGMENT_SIZE Maximal segment size
1230 Allocation of 2-word cell heaps:
1232 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
1235 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
1236 GC in percent of total heap size
1239 Allocation of 4-word cell heaps
1240 (used for real numbers and misc other objects):
1242 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
1244 (See entry "Way for application to customize GC parameters" under
1245 section "Changes to the scm_ interface" below.)
1247 ** Guile now implements reals using 4-word cells
1249 This speeds up computation with reals. (They were earlier allocated
1250 with `malloc'.) There is still some room for optimizations, however.
1252 ** Some further steps toward POSIX thread support have been taken
1254 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
1255 don't have much effect any longer, and many of them will be removed in
1259 are only handled at the top of the evaluator loop, immediately after
1260 I/O, and in scm_equalp.
1262 *** The GC can allocate thread private pools of pairs.
1264 * Changes to Scheme functions and syntax
1266 ** close-input-port and close-output-port are now R5RS
1268 These procedures have been turned into primitives and have R5RS behaviour.
1270 ** New procedure: simple-format PORT MESSAGE ARG1 ...
1272 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
1273 extended by the more sophisticated version in (ice-9 format)
1275 (simple-format port message . args)
1276 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
1277 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
1278 the escapes are replaced with corresponding members of ARGS:
1279 ~A formats using `display' and ~S formats using `write'.
1280 If DESTINATION is #t, then use the `current-output-port',
1281 if DESTINATION is #f, then return a string containing the formatted text.
1282 Does not add a trailing newline."
1284 ** string-ref: the second argument is no longer optional.
1286 ** string, list->string: no longer accept strings in their arguments,
1287 only characters, for compatibility with R5RS.
1289 ** New procedure: port-closed? PORT
1290 Returns #t if PORT is closed or #f if it is open.
1292 ** Deprecated: list*
1294 The list* functionality is now provided by cons* (SRFI-1 compliant)
1296 ** New procedure: cons* ARG1 ARG2 ... ARGn
1298 Like `list', but the last arg provides the tail of the constructed list,
1299 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
1301 Requires at least one argument. If given one argument, that argument
1302 is returned as result.
1304 This function is called `list*' in some other Schemes and in Common LISP.
1306 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
1308 ** New procedure: object-documentation OBJECT
1310 Returns the documentation string associated with OBJECT. The
1311 procedure uses a caching mechanism so that subsequent lookups are
1314 Exported by (ice-9 documentation).
1316 ** module-name now returns full names of modules
1318 Previously, only the last part of the name was returned (`session' for
1319 `(ice-9 session)'). Ex: `(ice-9 session)'.
1321 * Changes to the gh_ interface
1323 ** Deprecated: gh_int2scmb
1325 Use gh_bool2scm instead.
1327 * Changes to the scm_ interface
1329 ** Guile primitives now carry docstrings!
1331 Thanks to Greg Badros!
1333 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1335 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1336 macros and must contain a docstring that is extracted into foo.doc using a new
1337 guile-doc-snarf script (that uses guile-doc-snarf.awk).
1339 However, a major overhaul of these macros is scheduled for the next release of
1342 ** Guile primitives use a new technique for validation of arguments
1344 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
1345 the readability of argument checking.
1347 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
1349 ** New macros: SCM_PACK, SCM_UNPACK
1351 Compose/decompose an SCM value.
1353 The SCM type is now treated as an abstract data type and may be defined as a
1354 long, a void* or as a struct, depending on the architecture and compile time
1355 options. This makes it easier to find several types of bugs, for example when
1356 SCM values are treated as integers without conversion. Values of the SCM type
1357 should be treated as "atomic" values. These macros are used when
1358 composing/decomposing an SCM value, either because you want to access
1359 individual bits, or because you want to treat it as an integer value.
1361 E.g., in order to set bit 7 in an SCM value x, use the expression
1363 SCM_PACK (SCM_UNPACK (x) | 0x80)
1365 ** The name property of hooks is deprecated.
1366 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
1368 You can emulate this feature by using object properties.
1370 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
1371 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
1374 These macros will be removed in a future release of Guile.
1376 ** The following types, functions and macros from numbers.h are deprecated:
1377 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
1378 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
1380 Further, it is recommended not to rely on implementation details for guile's
1381 current implementation of bignums. It is planned to replace this
1382 implementation with gmp in the future.
1384 ** Port internals: the rw_random variable in the scm_port structure
1385 must be set to non-zero in any random access port. In recent Guile
1386 releases it was only set for bidirectional random-access ports.
1388 ** Port internals: the seek ptob procedure is now responsible for
1389 resetting the buffers if required. The change was made so that in the
1390 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
1391 the fport and strport ptobs can avoid resetting the buffers,
1392 in particular to avoid discarding unread chars. An existing port
1393 type can be fixed by adding something like the following to the
1394 beginning of the ptob seek procedure:
1396 if (pt->rw_active == SCM_PORT_READ)
1397 scm_end_input (object);
1398 else if (pt->rw_active == SCM_PORT_WRITE)
1399 ptob->flush (object);
1401 although to actually avoid resetting the buffers and discard unread
1402 chars requires further hacking that depends on the characteristics
1405 ** Deprecated functions: scm_fseek, scm_tag
1407 These functions are no longer used and will be removed in a future version.
1409 ** The scm_sysmissing procedure is no longer used in libguile.
1410 Unless it turns out to be unexpectedly useful to somebody, it will be
1411 removed in a future version.
1413 ** The format of error message strings has changed
1415 The two C procedures: scm_display_error and scm_error, as well as the
1416 primitive `scm-error', now use scm_simple_format to do their work.
1417 This means that the message strings of all code must be updated to use
1418 ~A where %s was used before, and ~S where %S was used before.
1420 During the period when there still are a lot of old Guiles out there,
1421 you might want to support both old and new versions of Guile.
1423 There are basically two methods to achieve this. Both methods use
1426 AC_CHECK_FUNCS(scm_simple_format)
1428 in your configure.in.
1430 Method 1: Use the string concatenation features of ANSI C's
1435 #ifdef HAVE_SCM_SIMPLE_FORMAT
1441 Then represent each of your error messages using a preprocessor macro:
1443 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
1447 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
1448 (define make-message string-append)
1450 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
1452 Method 2: Use the oldfmt function found in doc/oldfmt.c.
1456 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
1461 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
1465 ** Deprecated: coop_mutex_init, coop_condition_variable_init
1467 Don't use the functions coop_mutex_init and
1468 coop_condition_variable_init. They will change.
1470 Use scm_mutex_init and scm_cond_init instead.
1472 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
1473 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
1474 COND, as `scm_cond_wait' does, but it also bounds the duration
1475 of the wait. If COND has not been signaled before time ABSTIME,
1476 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
1477 returns the error code `ETIMEDOUT'.
1479 The ABSTIME parameter specifies an absolute time, with the same
1480 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
1481 to 00:00:00 GMT, January 1, 1970.
1483 ** New function: scm_cond_broadcast (scm_cond_t *COND)
1484 `scm_cond_broadcast' restarts all the threads that are waiting
1485 on the condition variable COND. Nothing happens if no threads are
1488 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
1489 `scm_key_create' allocates a new TSD key. The key is stored in
1490 the location pointed to by KEY. There is no limit on the number
1491 of keys allocated at a given time. The value initially associated
1492 with the returned key is `NULL' in all currently executing threads.
1494 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
1495 function associated with the key. When a thread terminates,
1496 DESTR_FUNCTION is called on the value associated with the key in
1497 that thread. The DESTR_FUNCTION is not called if a key is deleted
1498 with `scm_key_delete' or a value is changed with
1499 `scm_setspecific'. The order in which destructor functions are
1500 called at thread termination time is unspecified.
1502 Destructors are not yet implemented.
1504 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
1505 `scm_setspecific' changes the value associated with KEY in the
1506 calling thread, storing the given POINTER instead.
1508 ** New function: scm_getspecific (scm_key_t KEY)
1509 `scm_getspecific' returns the value currently associated with
1510 KEY in the calling thread.
1512 ** New function: scm_key_delete (scm_key_t KEY)
1513 `scm_key_delete' deallocates a TSD key. It does not check
1514 whether non-`NULL' values are associated with that key in the
1515 currently executing threads, nor call the destructor function
1516 associated with the key.
1518 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
1520 Initialize a C level hook HOOK with associated HOOK_DATA and type
1521 TYPE. (See scm_c_hook_run ().)
1523 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
1525 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
1526 is true, add it last, otherwise first. The same FUNC can be added
1527 multiple times if FUNC_DATA differ and vice versa.
1529 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
1531 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
1532 function is only removed if both FUNC and FUNC_DATA matches.
1534 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
1536 Run hook HOOK passing DATA to the hook functions.
1538 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
1539 returned is undefined.
1541 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
1542 returns a non-NULL value. This value is returned as the result of
1543 scm_c_hook_run. If all functions return NULL, NULL is returned.
1545 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
1546 returns a NULL value, and NULL is returned. If all functions returns
1547 a non-NULL value, the last value is returned.
1549 ** New C level GC hooks
1551 Five new C level hooks has been added to the garbage collector.
1553 scm_before_gc_c_hook
1556 are run before locking and after unlocking the heap. The system is
1557 thus in a mode where evaluation can take place. (Except that
1558 scm_before_gc_c_hook must not allocate new cells.)
1560 scm_before_mark_c_hook
1561 scm_before_sweep_c_hook
1562 scm_after_sweep_c_hook
1564 are run when the heap is locked. These are intended for extension of
1565 the GC in a modular fashion. Examples are the weaks and guardians
1568 ** Way for application to customize GC parameters
1570 The application can set up other default values for the GC heap
1571 allocation parameters
1573 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
1574 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
1575 GUILE_MAX_SEGMENT_SIZE,
1579 scm_default_init_heap_size_1, scm_default_min_yield_1,
1580 scm_default_init_heap_size_2, scm_default_min_yield_2,
1581 scm_default_max_segment_size
1583 respectively before callong scm_boot_guile.
1585 (See entry "New environment variables ..." in section
1586 "Changes to the stand-alone interpreter" above.)
1588 ** scm_protect_object/scm_unprotect_object now nest
1590 This means that you can call scm_protect_object multiple times on an
1591 object and count on the object being protected until
1592 scm_unprotect_object has been call the same number of times.
1594 The functions also have better time complexity.
1596 Still, it is usually possible to structure the application in a way
1597 that you don't need to use these functions. For example, if you use a
1598 protected standard Guile list to keep track of live objects rather
1599 than some custom data type, objects will die a natural death when they
1600 are no longer needed.
1602 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
1604 Guile does not provide the float representation for inexact real numbers any
1605 more. Now, only doubles are used to represent inexact real numbers. Further,
1606 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
1607 and scm_tc16_complex, respectively.
1609 ** Removed deprecated type scm_smobfuns
1611 ** Removed deprecated function scm_newsmob
1613 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
1615 There is an ongoing discussion among the developers whether to
1616 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
1617 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
1618 until this issue has been settled.
1620 ** Removed deprecated type tag scm_tc16_kw
1622 ** Added type tag scm_tc16_keyword
1624 (This was introduced already in release 1.3.4 but was not documented
1627 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
1629 * Changes to system call interfaces:
1631 ** The "select" procedure now tests port buffers for the ability to
1632 provide input or accept output. Previously only the underlying file
1633 descriptors were checked.
1635 ** New variable PIPE_BUF: the maximum number of bytes that can be
1636 atomically written to a pipe.
1638 ** If a facility is not available on the system when Guile is
1639 compiled, the corresponding primitive procedure will not be defined.
1640 Previously it would have been defined but would throw a system-error
1641 exception if called. Exception handlers which catch this case may
1642 need minor modification: an error will be thrown with key
1643 'unbound-variable instead of 'system-error. Alternatively it's
1644 now possible to use `defined?' to check whether the facility is
1647 ** Procedures which depend on the timezone should now give the correct
1648 result on systems which cache the TZ environment variable, even if TZ
1649 is changed without calling tzset.
1651 * Changes to the networking interfaces:
1653 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
1654 long integers between network and host format. For now, it's not
1655 particularly convenient to do this kind of thing, but consider:
1657 (define write-network-long
1658 (lambda (value port)
1659 (let ((v (make-uniform-vector 1 1 0)))
1660 (uniform-vector-set! v 0 (htonl value))
1661 (uniform-vector-write v port))))
1663 (define read-network-long
1665 (let ((v (make-uniform-vector 1 1 0)))
1666 (uniform-vector-read! v port)
1667 (ntohl (uniform-vector-ref v 0)))))
1669 ** If inet-aton fails, it now throws an error with key 'misc-error
1670 instead of 'system-error, since errno is not relevant.
1672 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
1673 specific keys instead of 'system-error. The latter is inappropriate
1674 since errno will not have been set. The keys are:
1675 'host-not-found, 'try-again, 'no-recovery and 'no-data.
1677 ** sethostent, setnetent, setprotoent, setservent: now take an
1678 optional argument STAYOPEN, which specifies whether the database
1679 remains open after a database entry is accessed randomly (e.g., using
1680 gethostbyname for the hosts database.) The default is #f. Previously
1684 Changes since Guile 1.3.2:
1686 * Changes to the stand-alone interpreter
1690 An initial version of the Guile debugger written by Chris Hanson has
1691 been added. The debugger is still under development but is included
1692 in the distribution anyway since it is already quite useful.
1698 after an error to enter the debugger. Type `help' inside the debugger
1699 for a description of available commands.
1701 If you prefer to have stack frames numbered and printed in
1702 anti-chronological order and prefer up in the stack to be down on the
1703 screen as is the case in gdb, you can put
1705 (debug-enable 'backwards)
1707 in your .guile startup file. (However, this means that Guile can't
1708 use indentation to indicate stack level.)
1710 The debugger is autoloaded into Guile at the first use.
1712 ** Further enhancements to backtraces
1714 There is a new debug option `width' which controls the maximum width
1715 on the screen of printed stack frames. Fancy printing parameters
1716 ("level" and "length" as in Common LISP) are adaptively adjusted for
1717 each stack frame to give maximum information while still fitting
1718 within the bounds. If the stack frame can't be made to fit by
1719 adjusting parameters, it is simply cut off at the end. This is marked
1722 ** Some modules are now only loaded when the repl is started
1724 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
1725 regex) are now loaded into (guile-user) only if the repl has been
1726 started. The effect is that the startup time for scripts has been
1727 reduced to 30% of what it was previously.
1729 Correctly written scripts load the modules they require at the top of
1730 the file and should not be affected by this change.
1732 ** Hooks are now represented as smobs
1734 * Changes to Scheme functions and syntax
1736 ** Readline support has changed again.
1738 The old (readline-activator) module is gone. Use (ice-9 readline)
1739 instead, which now contains all readline functionality. So the code
1740 to activate readline is now
1742 (use-modules (ice-9 readline))
1745 This should work at any time, including from the guile prompt.
1747 To avoid confusion about the terms of Guile's license, please only
1748 enable readline for your personal use; please don't make it the
1749 default for others. Here is why we make this rather odd-sounding
1752 Guile is normally licensed under a weakened form of the GNU General
1753 Public License, which allows you to link code with Guile without
1754 placing that code under the GPL. This exception is important to some
1757 However, since readline is distributed under the GNU General Public
1758 License, when you link Guile with readline, either statically or
1759 dynamically, you effectively change Guile's license to the strict GPL.
1760 Whenever you link any strictly GPL'd code into Guile, uses of Guile
1761 which are normally permitted become forbidden. This is a rather
1762 non-obvious consequence of the licensing terms.
1764 So, to make sure things remain clear, please let people choose for
1765 themselves whether to link GPL'd libraries like readline with Guile.
1767 ** regexp-substitute/global has changed slightly, but incompatibly.
1769 If you include a function in the item list, the string of the match
1770 object it receives is the same string passed to
1771 regexp-substitute/global, not some suffix of that string.
1772 Correspondingly, the match's positions are relative to the entire
1773 string, not the suffix.
1775 If the regexp can match the empty string, the way matches are chosen
1776 from the string has changed. regexp-substitute/global recognizes the
1777 same set of matches that list-matches does; see below.
1779 ** New function: list-matches REGEXP STRING [FLAGS]
1781 Return a list of match objects, one for every non-overlapping, maximal
1782 match of REGEXP in STRING. The matches appear in left-to-right order.
1783 list-matches only reports matches of the empty string if there are no
1784 other matches which begin on, end at, or include the empty match's
1787 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1789 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
1791 For each match of REGEXP in STRING, apply PROC to the match object,
1792 and the last value PROC returned, or INIT for the first call. Return
1793 the last value returned by PROC. We apply PROC to the matches as they
1794 appear from left to right.
1796 This function recognizes matches according to the same criteria as
1799 Thus, you could define list-matches like this:
1801 (define (list-matches regexp string . flags)
1802 (reverse! (apply fold-matches regexp string '() cons flags)))
1804 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1808 *** New function: hook? OBJ
1810 Return #t if OBJ is a hook, otherwise #f.
1812 *** New function: make-hook-with-name NAME [ARITY]
1814 Return a hook with name NAME and arity ARITY. The default value for
1815 ARITY is 0. The only effect of NAME is that it will appear when the
1816 hook object is printed to ease debugging.
1818 *** New function: hook-empty? HOOK
1820 Return #t if HOOK doesn't contain any procedures, otherwise #f.
1822 *** New function: hook->list HOOK
1824 Return a list of the procedures that are called when run-hook is
1827 ** `map' signals an error if its argument lists are not all the same length.
1829 This is the behavior required by R5RS, so this change is really a bug
1830 fix. But it seems to affect a lot of people's code, so we're
1831 mentioning it here anyway.
1833 ** Print-state handling has been made more transparent
1835 Under certain circumstances, ports are represented as a port with an
1836 associated print state. Earlier, this pair was represented as a pair
1837 (see "Some magic has been added to the printer" below). It is now
1838 indistinguishable (almost; see `get-print-state') from a port on the
1841 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
1843 Return a new port with the associated print state PRINT-STATE.
1845 *** New function: get-print-state OUTPUT-PORT
1847 Return the print state associated with this port if it exists,
1848 otherwise return #f.
1850 *** New function: directory-stream? OBJECT
1852 Returns true iff OBJECT is a directory stream --- the sort of object
1853 returned by `opendir'.
1855 ** New function: using-readline?
1857 Return #t if readline is in use in the current repl.
1859 ** structs will be removed in 1.4
1861 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
1862 and use GOOPS objects as the fundamental record type.
1864 * Changes to the scm_ interface
1866 ** structs will be removed in 1.4
1868 The entire current struct interface (struct.c, struct.h) will be
1869 replaced in Guile 1.4. We will merge GOOPS into libguile and use
1870 GOOPS objects as the fundamental record type.
1872 ** The internal representation of subr's has changed
1874 Instead of giving a hint to the subr name, the CAR field of the subr
1875 now contains an index to a subr entry in scm_subr_table.
1877 *** New variable: scm_subr_table
1879 An array of subr entries. A subr entry contains the name, properties
1880 and documentation associated with the subr. The properties and
1881 documentation slots are not yet used.
1883 ** A new scheme for "forwarding" calls to a builtin to a generic function
1885 It is now possible to extend the functionality of some Guile
1886 primitives by letting them defer a call to a GOOPS generic function on
1887 argument mismatch. This means that there is no loss of efficiency in
1892 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
1893 (define-method + ((x <string>) (y <string>))
1894 (string-append x y))
1896 + will still be as efficient as usual in numerical calculations, but
1897 can also be used for concatenating strings.
1899 Who will be the first one to extend Guile's numerical tower to
1900 rationals? :) [OK, there a few other things to fix before this can
1901 be made in a clean way.]
1903 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
1905 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
1907 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
1909 These do the same job as SCM_PROC and SCM_PROC1, but they also define
1910 a variable GENERIC which can be used by the dispatch macros below.
1912 [This is experimental code which may change soon.]
1914 *** New macros for forwarding control to a generic on arg type error
1916 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
1918 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
1920 These correspond to the scm_wta function call, and have the same
1921 behaviour until the user has called the GOOPS primitive
1922 `enable-primitive-generic!'. After that, these macros will apply the
1923 generic function GENERIC to the argument(s) instead of calling
1926 [This is experimental code which may change soon.]
1928 *** New macros for argument testing with generic dispatch
1930 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
1932 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
1934 These correspond to the SCM_ASSERT macro, but will defer control to
1935 GENERIC on error after `enable-primitive-generic!' has been called.
1937 [This is experimental code which may change soon.]
1939 ** New function: SCM scm_eval_body (SCM body, SCM env)
1941 Evaluates the body of a special form.
1943 ** The internal representation of struct's has changed
1945 Previously, four slots were allocated for the procedure(s) of entities
1946 and operators. The motivation for this representation had to do with
1947 the structure of the evaluator, the wish to support tail-recursive
1948 generic functions, and efficiency. Since the generic function
1949 dispatch mechanism has changed, there is no longer a need for such an
1950 expensive representation, and the representation has been simplified.
1952 This should not make any difference for most users.
1954 ** GOOPS support has been cleaned up.
1956 Some code has been moved from eval.c to objects.c and code in both of
1957 these compilation units has been cleaned up and better structured.
1959 *** New functions for applying generic functions
1961 New function: SCM scm_apply_generic (GENERIC, ARGS)
1962 New function: SCM scm_call_generic_0 (GENERIC)
1963 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
1964 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
1965 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
1967 ** Deprecated function: scm_make_named_hook
1969 It is now replaced by:
1971 ** New function: SCM scm_create_hook (const char *name, int arity)
1973 Creates a hook in the same way as make-hook above but also
1974 binds a variable named NAME to it.
1976 This is the typical way of creating a hook from C code.
1978 Currently, the variable is created in the "current" module.
1979 This might change when we get the new module system.
1981 [The behaviour is identical to scm_make_named_hook.]
1985 Changes since Guile 1.3:
1987 * Changes to mailing lists
1989 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
1991 See the README file to find current addresses for all the Guile
1994 * Changes to the distribution
1996 ** Readline support is no longer included with Guile by default.
1998 Based on the different license terms of Guile and Readline, we
1999 concluded that Guile should not *by default* cause the linking of
2000 Readline into an application program. Readline support is now offered
2001 as a separate module, which is linked into an application only when
2002 you explicitly specify it.
2004 Although Guile is GNU software, its distribution terms add a special
2005 exception to the usual GNU General Public License (GPL). Guile's
2006 license includes a clause that allows you to link Guile with non-free
2007 programs. We add this exception so as not to put Guile at a
2008 disadvantage vis-a-vis other extensibility packages that support other
2011 In contrast, the GNU Readline library is distributed under the GNU
2012 General Public License pure and simple. This means that you may not
2013 link Readline, even dynamically, into an application unless it is
2014 distributed under a free software license that is compatible the GPL.
2016 Because of this difference in distribution terms, an application that
2017 can use Guile may not be able to use Readline. Now users will be
2018 explicitly offered two independent decisions about the use of these
2021 You can activate the readline support by issuing
2023 (use-modules (readline-activator))
2026 from your ".guile" file, for example.
2028 * Changes to the stand-alone interpreter
2030 ** All builtins now print as primitives.
2031 Previously builtin procedures not belonging to the fundamental subr
2032 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
2033 Now, they print as #<primitive-procedure NAME>.
2035 ** Backtraces slightly more intelligible.
2036 gsubr-apply and macro transformer application frames no longer appear
2039 * Changes to Scheme functions and syntax
2041 ** Guile now correctly handles internal defines by rewriting them into
2042 their equivalent letrec. Previously, internal defines would
2043 incrementally add to the innermost environment, without checking
2044 whether the restrictions specified in RnRS were met. This lead to the
2045 correct behaviour when these restriction actually were met, but didn't
2046 catch all illegal uses. Such an illegal use could lead to crashes of
2047 the Guile interpreter or or other unwanted results. An example of
2048 incorrect internal defines that made Guile behave erratically:
2060 The problem with this example is that the definition of `c' uses the
2061 value of `b' directly. This confuses the meoization machine of Guile
2062 so that the second call of `b' (this time in a larger environment that
2063 also contains bindings for `c' and `d') refers to the binding of `c'
2064 instead of `a'. You could also make Guile crash with a variation on
2069 (define (b flag) (if flag a 1))
2070 (define c (1+ (b flag)))
2078 From now on, Guile will issue an `Unbound variable: b' error message
2083 A hook contains a list of functions which should be called on
2084 particular occasions in an existing program. Hooks are used for
2087 A window manager might have a hook before-window-map-hook. The window
2088 manager uses the function run-hooks to call all functions stored in
2089 before-window-map-hook each time a window is mapped. The user can
2090 store functions in the hook using add-hook!.
2092 In Guile, hooks are first class objects.
2094 *** New function: make-hook [N_ARGS]
2096 Return a hook for hook functions which can take N_ARGS arguments.
2097 The default value for N_ARGS is 0.
2099 (See also scm_make_named_hook below.)
2101 *** New function: add-hook! HOOK PROC [APPEND_P]
2103 Put PROC at the beginning of the list of functions stored in HOOK.
2104 If APPEND_P is supplied, and non-false, put PROC at the end instead.
2106 PROC must be able to take the number of arguments specified when the
2109 If PROC already exists in HOOK, then remove it first.
2111 *** New function: remove-hook! HOOK PROC
2113 Remove PROC from the list of functions in HOOK.
2115 *** New function: reset-hook! HOOK
2117 Clear the list of hook functions stored in HOOK.
2119 *** New function: run-hook HOOK ARG1 ...
2121 Run all hook functions stored in HOOK with arguments ARG1 ... .
2122 The number of arguments supplied must correspond to the number given
2123 when the hook was created.
2125 ** The function `dynamic-link' now takes optional keyword arguments.
2126 The only keyword argument that is currently defined is `:global
2127 BOOL'. With it, you can control whether the shared library will be
2128 linked in global mode or not. In global mode, the symbols from the
2129 linked library can be used to resolve references from other
2130 dynamically linked libraries. In non-global mode, the linked
2131 library is essentially invisible and can only be accessed via
2132 `dynamic-func', etc. The default is now to link in global mode.
2133 Previously, the default has been non-global mode.
2135 The `#:global' keyword is only effective on platforms that support
2136 the dlopen family of functions.
2138 ** New function `provided?'
2140 - Function: provided? FEATURE
2141 Return true iff FEATURE is supported by this installation of
2142 Guile. FEATURE must be a symbol naming a feature; the global
2143 variable `*features*' is a list of available features.
2145 ** Changes to the module (ice-9 expect):
2147 *** The expect-strings macro now matches `$' in a regular expression
2148 only at a line-break or end-of-file by default. Previously it would
2149 match the end of the string accumulated so far. The old behaviour
2150 can be obtained by setting the variable `expect-strings-exec-flags'
2153 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
2154 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
2155 in a regular expression will still match before a line-break or
2156 end-of-file. The default is `regexp/noteol'.
2158 *** The expect-strings macro now uses a variable
2159 `expect-strings-compile-flags' for the flags to be supplied to
2160 `make-regexp'. The default is `regexp/newline', which was previously
2163 *** The expect macro now supplies two arguments to a match procedure:
2164 the current accumulated string and a flag to indicate whether
2165 end-of-file has been reached. Previously only the string was supplied.
2166 If end-of-file is reached, the match procedure will be called an
2167 additional time with the same accumulated string as the previous call
2168 but with the flag set.
2170 ** New module (ice-9 format), implementing the Common Lisp `format' function.
2172 This code, and the documentation for it that appears here, was
2173 borrowed from SLIB, with minor adaptations for Guile.
2175 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
2176 An almost complete implementation of Common LISP format description
2177 according to the CL reference book `Common LISP' from Guy L.
2178 Steele, Digital Press. Backward compatible to most of the
2179 available Scheme format implementations.
2181 Returns `#t', `#f' or a string; has side effect of printing
2182 according to FORMAT-STRING. If DESTINATION is `#t', the output is
2183 to the current output port and `#t' is returned. If DESTINATION
2184 is `#f', a formatted string is returned as the result of the call.
2185 NEW: If DESTINATION is a string, DESTINATION is regarded as the
2186 format string; FORMAT-STRING is then the first argument and the
2187 output is returned as a string. If DESTINATION is a number, the
2188 output is to the current error port if available by the
2189 implementation. Otherwise DESTINATION must be an output port and
2192 FORMAT-STRING must be a string. In case of a formatting error
2193 format returns `#f' and prints a message on the current output or
2194 error port. Characters are output as if the string were output by
2195 the `display' function with the exception of those prefixed by a
2196 tilde (~). For a detailed description of the FORMAT-STRING syntax
2197 please consult a Common LISP format reference manual. For a test
2198 suite to verify this format implementation load `formatst.scm'.
2199 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
2201 Note: `format' is not reentrant, i.e. only one `format'-call may
2202 be executed at a time.
2205 *** Format Specification (Format version 3.0)
2207 Please consult a Common LISP format reference manual for a detailed
2208 description of the format string syntax. For a demonstration of the
2209 implemented directives see `formatst.scm'.
2211 This implementation supports directive parameters and modifiers (`:'
2212 and `@' characters). Multiple parameters must be separated by a comma
2213 (`,'). Parameters can be numerical parameters (positive or negative),
2214 character parameters (prefixed by a quote character (`''), variable
2215 parameters (`v'), number of rest arguments parameter (`#'), empty and
2216 default parameters. Directive characters are case independent. The
2217 general form of a directive is:
2219 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
2221 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
2223 *** Implemented CL Format Control Directives
2225 Documentation syntax: Uppercase characters represent the
2226 corresponding control directive characters. Lowercase characters
2227 represent control directive parameter descriptions.
2230 Any (print as `display' does).
2234 `~MINCOL,COLINC,MINPAD,PADCHARA'
2238 S-expression (print as `write' does).
2242 `~MINCOL,COLINC,MINPAD,PADCHARS'
2248 print number sign always.
2251 print comma separated.
2253 `~MINCOL,PADCHAR,COMMACHARD'
2259 print number sign always.
2262 print comma separated.
2264 `~MINCOL,PADCHAR,COMMACHARX'
2270 print number sign always.
2273 print comma separated.
2275 `~MINCOL,PADCHAR,COMMACHARO'
2281 print number sign always.
2284 print comma separated.
2286 `~MINCOL,PADCHAR,COMMACHARB'
2291 `~N,MINCOL,PADCHAR,COMMACHARR'
2295 print a number as a Roman numeral.
2298 print a number as an "old fashioned" Roman numeral.
2301 print a number as an ordinal English number.
2304 print a number as a cardinal English number.
2309 prints `y' and `ies'.
2312 as `~P but jumps 1 argument backward.'
2315 as `~@P but jumps 1 argument backward.'
2320 prints a character as the reader can understand it (i.e. `#\'
2324 prints a character as emacs does (eg. `^C' for ASCII 03).
2327 Fixed-format floating-point (prints a flonum like MMM.NNN).
2328 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
2330 If the number is positive a plus sign is printed.
2333 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
2334 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
2336 If the number is positive a plus sign is printed.
2339 General floating-point (prints a flonum either fixed or
2341 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
2343 If the number is positive a plus sign is printed.
2346 Dollars floating-point (prints a flonum in fixed with signs
2348 `~DIGITS,SCALE,WIDTH,PADCHAR$'
2350 If the number is positive a plus sign is printed.
2353 A sign is always printed and appears before the padding.
2356 The sign appears before the padding.
2364 print newline if not at the beginning of the output line.
2366 prints `~&' and then N-1 newlines.
2371 print N page separators.
2381 newline is ignored, white space left.
2384 newline is left, white space ignored.
2389 relative tabulation.
2395 Indirection (expects indirect arguments as a list).
2397 extracts indirect arguments from format arguments.
2400 Case conversion (converts by `string-downcase').
2402 converts by `string-capitalize'.
2405 converts by `string-capitalize-first'.
2408 converts by `string-upcase'.
2411 Argument Jumping (jumps 1 argument forward).
2413 jumps N arguments forward.
2416 jumps 1 argument backward.
2419 jumps N arguments backward.
2422 jumps to the 0th argument.
2425 jumps to the Nth argument (beginning from 0)
2427 `~[STR0~;STR1~;...~;STRN~]'
2428 Conditional Expression (numerical clause conditional).
2430 take argument from N.
2433 true test conditional.
2436 if-else-then conditional.
2442 default clause follows.
2445 Iteration (args come from the next argument (a list)).
2447 at most N iterations.
2450 args from next arg (a list of lists).
2453 args from the rest of arguments.
2456 args from the rest args (lists).
2467 aborts if N <= M <= K
2469 *** Not Implemented CL Format Control Directives
2472 print `#f' as an empty list (see below).
2475 print `#f' as an empty list (see below).
2481 (sorry I don't understand its semantics completely)
2483 *** Extended, Replaced and Additional Control Directives
2485 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
2486 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
2487 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
2488 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
2489 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
2490 COMMAWIDTH is the number of characters between two comma
2494 print a R4RS complex number as `~F~@Fi' with passed parameters for
2498 Pretty print formatting of an argument for scheme code lists.
2504 Flushes the output if format DESTINATION is a port.
2507 Print a `#\space' character
2509 print N `#\space' characters.
2512 Print a `#\tab' character
2514 print N `#\tab' characters.
2517 Takes N as an integer representation for a character. No arguments
2518 are consumed. N is converted to a character by `integer->char'. N
2519 must be a positive decimal number.
2522 Print out readproof. Prints out internal objects represented as
2523 `#<...>' as strings `"#<...>"' so that the format output can always
2524 be processed by `read'.
2527 Print out readproof. Prints out internal objects represented as
2528 `#<...>' as strings `"#<...>"' so that the format output can always
2529 be processed by `read'.
2532 Prints information and a copyright notice on the format
2535 prints format version.
2538 may also print number strings, i.e. passing a number as a string
2539 and format it accordingly.
2541 *** Configuration Variables
2543 The format module exports some configuration variables to suit the
2544 systems and users needs. There should be no modification necessary for
2545 the configuration that comes with Guile. Format detects automatically
2546 if the running scheme system implements floating point numbers and
2549 format:symbol-case-conv
2550 Symbols are converted by `symbol->string' so the case type of the
2551 printed symbols is implementation dependent.
2552 `format:symbol-case-conv' is a one arg closure which is either
2553 `#f' (no conversion), `string-upcase', `string-downcase' or
2554 `string-capitalize'. (default `#f')
2556 format:iobj-case-conv
2557 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
2558 implementation internal objects. (default `#f')
2561 The character prefixing the exponent value in `~E' printing.
2564 *** Compatibility With Other Format Implementations
2570 Downward compatible except for padding support and `~A', `~S',
2571 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
2572 `printf' padding support which is completely replaced by the CL
2573 `format' padding style.
2576 Downward compatible except for `~', which is not documented
2577 (ignores all characters inside the format string up to a newline
2578 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
2579 numerical and variable parameters and `:/@' modifiers in the CL
2583 Downward compatible except for `~A' and `~S' which print in
2584 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
2585 directive parameters or modifiers)).
2588 Downward compatible except for an optional destination parameter:
2589 S2C accepts a format call without a destination which returns a
2590 formatted string. This is equivalent to a #f destination in S2C.
2591 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
2592 parameters or modifiers)).
2595 ** Changes to string-handling functions.
2597 These functions were added to support the (ice-9 format) module, above.
2599 *** New function: string-upcase STRING
2600 *** New function: string-downcase STRING
2602 These are non-destructive versions of the existing string-upcase! and
2603 string-downcase! functions.
2605 *** New function: string-capitalize! STRING
2606 *** New function: string-capitalize STRING
2608 These functions convert the first letter of each word in the string to
2611 (string-capitalize "howdy there")
2614 As with the other functions, string-capitalize! modifies the string in
2615 place, while string-capitalize returns a modified copy of its argument.
2617 *** New function: string-ci->symbol STRING
2619 Return a symbol whose name is STRING, but having the same case as if
2620 the symbol had be read by `read'.
2622 Guile can be configured to be sensitive or insensitive to case
2623 differences in Scheme identifiers. If Guile is case-insensitive, all
2624 symbols are converted to lower case on input. The `string-ci->symbol'
2625 function returns a symbol whose name in STRING, transformed as Guile
2626 would if STRING were input.
2628 *** New function: substring-move! STRING1 START END STRING2 START
2630 Copy the substring of STRING1 from START (inclusive) to END
2631 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
2632 string, and the source and destination areas may overlap; in all
2633 cases, the function behaves as if all the characters were copied
2636 *** Extended functions: substring-move-left! substring-move-right!
2638 These functions now correctly copy arbitrarily overlapping substrings;
2639 they are both synonyms for substring-move!.
2642 ** New module (ice-9 getopt-long), with the function `getopt-long'.
2644 getopt-long is a function for parsing command-line arguments in a
2645 manner consistent with other GNU programs.
2647 (getopt-long ARGS GRAMMAR)
2648 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
2650 ARGS should be a list of strings. Its first element should be the
2651 name of the program; subsequent elements should be the arguments
2652 that were passed to the program on the command line. The
2653 `program-arguments' procedure returns a list of this form.
2655 GRAMMAR is a list of the form:
2656 ((OPTION (PROPERTY VALUE) ...) ...)
2658 Each OPTION should be a symbol. `getopt-long' will accept a
2659 command-line option named `--OPTION'.
2660 Each option can have the following (PROPERTY VALUE) pairs:
2662 (single-char CHAR) --- Accept `-CHAR' as a single-character
2663 equivalent to `--OPTION'. This is how to specify traditional
2665 (required? BOOL) --- If BOOL is true, the option is required.
2666 getopt-long will raise an error if it is not found in ARGS.
2667 (value BOOL) --- If BOOL is #t, the option accepts a value; if
2668 it is #f, it does not; and if it is the symbol
2669 `optional', the option may appear in ARGS with or
2671 (predicate FUNC) --- If the option accepts a value (i.e. you
2672 specified `(value #t)' for this option), then getopt
2673 will apply FUNC to the value, and throw an exception
2674 if it returns #f. FUNC should be a procedure which
2675 accepts a string and returns a boolean value; you may
2676 need to use quasiquotes to get it into GRAMMAR.
2678 The (PROPERTY VALUE) pairs may occur in any order, but each
2679 property may occur only once. By default, options do not have
2680 single-character equivalents, are not required, and do not take
2683 In ARGS, single-character options may be combined, in the usual
2684 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
2685 accepts values, then it must be the last option in the
2686 combination; the value is the next argument. So, for example, using
2687 the following grammar:
2688 ((apples (single-char #\a))
2689 (blimps (single-char #\b) (value #t))
2690 (catalexis (single-char #\c) (value #t)))
2691 the following argument lists would be acceptable:
2692 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
2693 for "blimps" and "catalexis")
2694 ("-ab" "bang" "-c" "couth") (same)
2695 ("-ac" "couth" "-b" "bang") (same)
2696 ("-abc" "couth" "bang") (an error, since `-b' is not the
2697 last option in its combination)
2699 If an option's value is optional, then `getopt-long' decides
2700 whether it has a value by looking at what follows it in ARGS. If
2701 the next element is a string, and it does not appear to be an
2702 option itself, then that string is the option's value.
2704 The value of a long option can appear as the next element in ARGS,
2705 or it can follow the option name, separated by an `=' character.
2706 Thus, using the same grammar as above, the following argument lists
2708 ("--apples" "Braeburn" "--blimps" "Goodyear")
2709 ("--apples=Braeburn" "--blimps" "Goodyear")
2710 ("--blimps" "Goodyear" "--apples=Braeburn")
2712 If the option "--" appears in ARGS, argument parsing stops there;
2713 subsequent arguments are returned as ordinary arguments, even if
2714 they resemble options. So, in the argument list:
2715 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
2716 `getopt-long' will recognize the `apples' option as having the
2717 value "Granny Smith", but it will not recognize the `blimp'
2718 option; it will return the strings "--blimp" and "Goodyear" as
2719 ordinary argument strings.
2721 The `getopt-long' function returns the parsed argument list as an
2722 assocation list, mapping option names --- the symbols from GRAMMAR
2723 --- onto their values, or #t if the option does not accept a value.
2724 Unused options do not appear in the alist.
2726 All arguments that are not the value of any option are returned
2727 as a list, associated with the empty list.
2729 `getopt-long' throws an exception if:
2730 - it finds an unrecognized option in ARGS
2731 - a required option is omitted
2732 - an option that requires an argument doesn't get one
2733 - an option that doesn't accept an argument does get one (this can
2734 only happen using the long option `--opt=value' syntax)
2735 - an option predicate fails
2740 `((lockfile-dir (required? #t)
2743 (predicate ,file-is-directory?))
2744 (verbose (required? #f)
2747 (x-includes (single-char #\x))
2748 (rnet-server (single-char #\y)
2749 (predicate ,string?))))
2751 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
2752 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
2754 => ((() "foo1" "-fred" "foo2" "foo3")
2755 (rnet-server . "lamprod")
2756 (x-includes . "/usr/include")
2757 (lockfile-dir . "/tmp")
2760 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
2762 It will be removed in a few releases.
2764 ** New syntax: lambda*
2765 ** New syntax: define*
2766 ** New syntax: define*-public
2767 ** New syntax: defmacro*
2768 ** New syntax: defmacro*-public
2769 Guile now supports optional arguments.
2771 `lambda*', `define*', `define*-public', `defmacro*' and
2772 `defmacro*-public' are identical to the non-* versions except that
2773 they use an extended type of parameter list that has the following BNF
2774 syntax (parentheses are literal, square brackets indicate grouping,
2775 and `*', `+' and `?' have the usual meaning):
2777 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
2778 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
2779 [[#&rest identifier]|[. identifier]]? ) | [identifier]
2781 ext-var-decl ::= identifier | ( identifier expression )
2783 The semantics are best illustrated with the following documentation
2784 and examples for `lambda*':
2787 lambda extended for optional and keyword arguments
2789 lambda* creates a procedure that takes optional arguments. These
2790 are specified by putting them inside brackets at the end of the
2791 paramater list, but before any dotted rest argument. For example,
2792 (lambda* (a b #&optional c d . e) '())
2793 creates a procedure with fixed arguments a and b, optional arguments c
2794 and d, and rest argument e. If the optional arguments are omitted
2795 in a call, the variables for them are unbound in the procedure. This
2796 can be checked with the bound? macro.
2798 lambda* can also take keyword arguments. For example, a procedure
2800 (lambda* (#&key xyzzy larch) '())
2801 can be called with any of the argument lists (#:xyzzy 11)
2802 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
2803 are given as keywords are bound to values.
2805 Optional and keyword arguments can also be given default values
2806 which they take on when they are not present in a call, by giving a
2807 two-item list in place of an optional argument, for example in:
2808 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
2809 foo is a fixed argument, bar is an optional argument with default
2810 value 42, and baz is a keyword argument with default value 73.
2811 Default value expressions are not evaluated unless they are needed
2812 and until the procedure is called.
2814 lambda* now supports two more special parameter list keywords.
2816 lambda*-defined procedures now throw an error by default if a
2817 keyword other than one of those specified is found in the actual
2818 passed arguments. However, specifying #&allow-other-keys
2819 immediately after the kyword argument declarations restores the
2820 previous behavior of ignoring unknown keywords. lambda* also now
2821 guarantees that if the same keyword is passed more than once, the
2822 last one passed is the one that takes effect. For example,
2823 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
2824 #:heads 37 #:tails 42 #:heads 99)
2825 would result in (99 47) being displayed.
2827 #&rest is also now provided as a synonym for the dotted syntax rest
2828 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
2829 all respects to lambda*. This is provided for more similarity to DSSSL,
2830 MIT-Scheme and Kawa among others, as well as for refugees from other
2833 Further documentation may be found in the optargs.scm file itself.
2835 The optional argument module also exports the macros `let-optional',
2836 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
2837 are not documented here because they may be removed in the future, but
2838 full documentation is still available in optargs.scm.
2840 ** New syntax: and-let*
2841 Guile now supports the `and-let*' form, described in the draft SRFI-2.
2843 Syntax: (land* (<clause> ...) <body> ...)
2844 Each <clause> should have one of the following forms:
2845 (<variable> <expression>)
2848 Each <variable> or <bound-variable> should be an identifier. Each
2849 <expression> should be a valid expression. The <body> should be a
2850 possibly empty sequence of expressions, like the <body> of a
2853 Semantics: A LAND* expression is evaluated by evaluating the
2854 <expression> or <bound-variable> of each of the <clause>s from
2855 left to right. The value of the first <expression> or
2856 <bound-variable> that evaluates to a false value is returned; the
2857 remaining <expression>s and <bound-variable>s are not evaluated.
2858 The <body> forms are evaluated iff all the <expression>s and
2859 <bound-variable>s evaluate to true values.
2861 The <expression>s and the <body> are evaluated in an environment
2862 binding each <variable> of the preceding (<variable> <expression>)
2863 clauses to the value of the <expression>. Later bindings
2864 shadow earlier bindings.
2866 Guile's and-let* macro was contributed by Michael Livshin.
2868 ** New sorting functions
2870 *** New function: sorted? SEQUENCE LESS?
2871 Returns `#t' when the sequence argument is in non-decreasing order
2872 according to LESS? (that is, there is no adjacent pair `... x y
2873 ...' for which `(less? y x)').
2875 Returns `#f' when the sequence contains at least one out-of-order
2876 pair. It is an error if the sequence is neither a list nor a
2879 *** New function: merge LIST1 LIST2 LESS?
2880 LIST1 and LIST2 are sorted lists.
2881 Returns the sorted list of all elements in LIST1 and LIST2.
2883 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
2884 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
2885 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
2886 (Here "<" should read "comes before".)
2888 *** New procedure: merge! LIST1 LIST2 LESS?
2889 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
2890 the result. If the code is compiled, and LESS? constructs no new
2891 pairs, no pairs at all will be allocated. The first pair of the
2892 result will be either the first pair of LIST1 or the first pair of
2895 *** New function: sort SEQUENCE LESS?
2896 Accepts either a list or a vector, and returns a new sequence
2897 which is sorted. The new sequence is the same type as the input.
2898 Always `(sorted? (sort sequence less?) less?)'. The original
2899 sequence is not altered in any way. The new sequence shares its
2900 elements with the old one; no elements are copied.
2902 *** New procedure: sort! SEQUENCE LESS
2903 Returns its sorted result in the original boxes. No new storage is
2904 allocated at all. Proper usage: (set! slist (sort! slist <))
2906 *** New function: stable-sort SEQUENCE LESS?
2907 Similar to `sort' but stable. That is, if "equal" elements are
2908 ordered a < b in the original sequence, they will have the same order
2911 *** New function: stable-sort! SEQUENCE LESS?
2912 Similar to `sort!' but stable.
2913 Uses temporary storage when sorting vectors.
2915 *** New functions: sort-list, sort-list!
2916 Added for compatibility with scsh.
2918 ** New built-in random number support
2920 *** New function: random N [STATE]
2921 Accepts a positive integer or real N and returns a number of the
2922 same type between zero (inclusive) and N (exclusive). The values
2923 returned have a uniform distribution.
2925 The optional argument STATE must be of the type produced by
2926 `copy-random-state' or `seed->random-state'. It defaults to the value
2927 of the variable `*random-state*'. This object is used to maintain the
2928 state of the pseudo-random-number generator and is altered as a side
2929 effect of the `random' operation.
2931 *** New variable: *random-state*
2932 Holds a data structure that encodes the internal state of the
2933 random-number generator that `random' uses by default. The nature
2934 of this data structure is implementation-dependent. It may be
2935 printed out and successfully read back in, but may or may not
2936 function correctly as a random-number state object in another
2939 *** New function: copy-random-state [STATE]
2940 Returns a new object of type suitable for use as the value of the
2941 variable `*random-state*' and as a second argument to `random'.
2942 If argument STATE is given, a copy of it is returned. Otherwise a
2943 copy of `*random-state*' is returned.
2945 *** New function: seed->random-state SEED
2946 Returns a new object of type suitable for use as the value of the
2947 variable `*random-state*' and as a second argument to `random'.
2948 SEED is a string or a number. A new state is generated and
2949 initialized using SEED.
2951 *** New function: random:uniform [STATE]
2952 Returns an uniformly distributed inexact real random number in the
2953 range between 0 and 1.
2955 *** New procedure: random:solid-sphere! VECT [STATE]
2956 Fills VECT with inexact real random numbers the sum of whose
2957 squares is less than 1.0. Thinking of VECT as coordinates in
2958 space of dimension N = `(vector-length VECT)', the coordinates are
2959 uniformly distributed within the unit N-shere. The sum of the
2960 squares of the numbers is returned. VECT can be either a vector
2961 or a uniform vector of doubles.
2963 *** New procedure: random:hollow-sphere! VECT [STATE]
2964 Fills VECT with inexact real random numbers the sum of whose squares
2965 is equal to 1.0. Thinking of VECT as coordinates in space of
2966 dimension n = `(vector-length VECT)', the coordinates are uniformly
2967 distributed over the surface of the unit n-shere. VECT can be either
2968 a vector or a uniform vector of doubles.
2970 *** New function: random:normal [STATE]
2971 Returns an inexact real in a normal distribution with mean 0 and
2972 standard deviation 1. For a normal distribution with mean M and
2973 standard deviation D use `(+ M (* D (random:normal)))'.
2975 *** New procedure: random:normal-vector! VECT [STATE]
2976 Fills VECT with inexact real random numbers which are independent and
2977 standard normally distributed (i.e., with mean 0 and variance 1).
2978 VECT can be either a vector or a uniform vector of doubles.
2980 *** New function: random:exp STATE
2981 Returns an inexact real in an exponential distribution with mean 1.
2982 For an exponential distribution with mean U use (* U (random:exp)).
2984 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
2986 These functions now operate on numbers in the range of a C unsigned
2989 These functions used to operate on numbers in the range of a C signed
2990 long; however, this seems inappropriate, because Guile integers don't
2993 ** New function: make-guardian
2994 This is an implementation of guardians as described in
2995 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
2996 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
2997 Programming Language Design and Implementation, June 1993
2998 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
3000 ** New functions: delq1!, delv1!, delete1!
3001 These procedures behave similar to delq! and friends but delete only
3002 one object if at all.
3004 ** New function: unread-string STRING PORT
3005 Unread STRING to PORT, that is, push it back onto the port so that
3006 next read operation will work on the pushed back characters.
3008 ** unread-char can now be called multiple times
3009 If unread-char is called multiple times, the unread characters will be
3010 read again in last-in first-out order.
3012 ** the procedures uniform-array-read! and uniform-array-write! now
3013 work on any kind of port, not just ports which are open on a file.
3015 ** Now 'l' in a port mode requests line buffering.
3017 ** The procedure truncate-file now works on string ports as well
3018 as file ports. If the size argument is omitted, the current
3019 file position is used.
3021 ** new procedure: seek PORT/FDES OFFSET WHENCE
3022 The arguments are the same as for the old fseek procedure, but it
3023 works on string ports as well as random-access file ports.
3025 ** the fseek procedure now works on string ports, since it has been
3026 redefined using seek.
3028 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
3029 size is not supplied.
3031 ** the newline procedure no longer flushes the port if it's not
3032 line-buffered: previously it did if it was the current output port.
3034 ** open-pipe and close-pipe are no longer primitive procedures, but
3035 an emulation can be obtained using `(use-modules (ice-9 popen))'.
3037 ** the freopen procedure has been removed.
3039 ** new procedure: drain-input PORT
3040 Drains PORT's read buffers (including any pushed-back characters)
3041 and returns the contents as a single string.
3043 ** New function: map-in-order PROC LIST1 LIST2 ...
3044 Version of `map' which guarantees that the procedure is applied to the
3045 lists in serial order.
3047 ** Renamed `serial-array-copy!' and `serial-array-map!' to
3048 `array-copy-in-order!' and `array-map-in-order!'. The old names are
3049 now obsolete and will go away in release 1.5.
3051 ** New syntax: collect BODY1 ...
3052 Version of `begin' which returns a list of the results of the body
3053 forms instead of the result of the last body form. In contrast to
3054 `begin', `collect' allows an empty body.
3056 ** New functions: read-history FILENAME, write-history FILENAME
3057 Read/write command line history from/to file. Returns #t on success
3058 and #f if an error occured.
3060 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
3062 These procedures return a list of definitions available in the specified
3063 argument, a relative module reference. In the case of no argument,
3064 `(current-module)' is now consulted for definitions to return, instead
3065 of simply returning #f, the former behavior.
3067 ** The #/ syntax for lists is no longer supported.
3069 Earlier versions of Scheme accepted this syntax, but printed a
3072 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
3074 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
3077 * Changes to the gh_ interface
3081 Now takes a second argument which is the result array. If this
3082 pointer is NULL, a new array is malloced (the old behaviour).
3084 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
3085 gh_scm2shorts, gh_scm2longs, gh_scm2floats
3089 * Changes to the scm_ interface
3091 ** Function: scm_make_named_hook (char* name, int n_args)
3093 Creates a hook in the same way as make-hook above but also
3094 binds a variable named NAME to it.
3096 This is the typical way of creating a hook from C code.
3098 Currently, the variable is created in the "current" module. This
3099 might change when we get the new module system.
3101 ** The smob interface
3103 The interface for creating smobs has changed. For documentation, see
3104 data-rep.info (made from guile-core/doc/data-rep.texi).
3106 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
3108 >>> This function will be removed in 1.3.4. <<<
3112 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
3113 This function adds a new smob type, named NAME, with instance size
3114 SIZE to the system. The return value is a tag that is used in
3115 creating instances of the type. If SIZE is 0, then no memory will
3116 be allocated when instances of the smob are created, and nothing
3117 will be freed by the default free function.
3119 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
3120 This function sets the smob marking procedure for the smob type
3121 specified by the tag TC. TC is the tag returned by
3122 `scm_make_smob_type'.
3124 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
3125 This function sets the smob freeing procedure for the smob type
3126 specified by the tag TC. TC is the tag returned by
3127 `scm_make_smob_type'.
3129 *** Function: void scm_set_smob_print (tc, print)
3131 - Function: void scm_set_smob_print (long tc,
3132 scm_sizet (*print) (SCM,
3136 This function sets the smob printing procedure for the smob type
3137 specified by the tag TC. TC is the tag returned by
3138 `scm_make_smob_type'.
3140 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
3141 This function sets the smob equality-testing predicate for the
3142 smob type specified by the tag TC. TC is the tag returned by
3143 `scm_make_smob_type'.
3145 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
3146 Make VALUE contain a smob instance of the type with type code TC and
3147 smob data DATA. VALUE must be previously declared as C type `SCM'.
3149 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
3150 This macro expands to a block of code that creates a smob instance
3151 of the type with type code TC and smob data DATA, and returns that
3152 `SCM' value. It should be the last piece of code in a block.
3154 ** The interfaces for using I/O ports and implementing port types
3155 (ptobs) have changed significantly. The new interface is based on
3156 shared access to buffers and a new set of ptob procedures.
3158 *** scm_newptob has been removed
3162 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
3164 - Function: SCM scm_make_port_type (char *type_name,
3165 int (*fill_buffer) (SCM port),
3166 void (*write_flush) (SCM port));
3168 Similarly to the new smob interface, there is a set of function
3169 setters by which the user can customize the behaviour of his port
3170 type. See ports.h (scm_set_port_XXX).
3172 ** scm_strport_to_string: New function: creates a new string from
3173 a string port's buffer.
3175 ** Plug in interface for random number generators
3176 The variable `scm_the_rng' in random.c contains a value and three
3177 function pointers which together define the current random number
3178 generator being used by the Scheme level interface and the random
3179 number library functions.
3181 The user is free to replace the default generator with the generator
3184 *** Variable: size_t scm_the_rng.rstate_size
3185 The size of the random state type used by the current RNG
3188 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
3189 Given the random STATE, return 32 random bits.
3191 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
3192 Seed random state STATE using string S of length N.
3194 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
3195 Given random state STATE, return a malloced copy.
3198 The default RNG is the MWC (Multiply With Carry) random number
3199 generator described by George Marsaglia at the Department of
3200 Statistics and Supercomputer Computations Research Institute, The
3201 Florida State University (http://stat.fsu.edu/~geo).
3203 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
3204 passes all tests in the DIEHARD test suite
3205 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
3206 costs one multiply and one add on platforms which either supports long
3207 longs (gcc does this on most systems) or have 64 bit longs. The cost
3208 is four multiply on other systems but this can be optimized by writing
3209 scm_i_uniform32 in assembler.
3211 These functions are provided through the scm_the_rng interface for use
3212 by libguile and the application.
3214 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
3215 Given the random STATE, return 32 random bits.
3216 Don't use this function directly. Instead go through the plugin
3217 interface (see "Plug in interface" above).
3219 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
3220 Initialize STATE using SEED of length N.
3222 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
3223 Return a malloc:ed copy of STATE. This function can easily be re-used
3224 in the interfaces to other RNGs.
3226 ** Random number library functions
3227 These functions use the current RNG through the scm_the_rng interface.
3228 It might be a good idea to use these functions from your C code so
3229 that only one random generator is used by all code in your program.
3231 The default random state is stored in:
3233 *** Variable: SCM scm_var_random_state
3234 Contains the vcell of the Scheme variable "*random-state*" which is
3235 used as default state by all random number functions in the Scheme
3240 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
3242 *** Function: scm_rstate *scm_c_default_rstate (void)
3243 This is a convenience function which returns the value of
3244 scm_var_random_state. An error message is generated if this value
3245 isn't a random state.
3247 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
3248 Make a new random state from the string SEED of length LENGTH.
3250 It is generally not a good idea to use multiple random states in a
3251 program. While subsequent random numbers generated from one random
3252 state are guaranteed to be reasonably independent, there is no such
3253 guarantee for numbers generated from different random states.
3255 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
3256 Return 32 random bits.
3258 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
3259 Return a sample from the uniform(0,1) distribution.
3261 *** Function: double scm_c_normal01 (scm_rstate *STATE)
3262 Return a sample from the normal(0,1) distribution.
3264 *** Function: double scm_c_exp1 (scm_rstate *STATE)
3265 Return a sample from the exp(1) distribution.
3267 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
3268 Return a sample from the discrete uniform(0,M) distribution.
3270 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
3271 Return a sample from the discrete uniform(0,M) distribution.
3272 M must be a bignum object. The returned value may be an INUM.
3276 Changes in Guile 1.3 (released Monday, October 19, 1998):
3278 * Changes to the distribution
3280 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
3281 To avoid conflicts, programs should name environment variables after
3282 themselves, except when there's a common practice establishing some
3285 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
3286 giving the former precedence, and printing a warning message if the
3287 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
3289 ** The header files related to multi-byte characters have been removed.
3290 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
3291 which referred to these explicitly will probably need to be rewritten,
3292 since the support for the variant string types has been removed; see
3295 ** The header files append.h and sequences.h have been removed. These
3296 files implemented non-R4RS operations which would encourage
3297 non-portable programming style and less easy-to-read code.
3299 * Changes to the stand-alone interpreter
3301 ** New procedures have been added to implement a "batch mode":
3303 *** Function: batch-mode?
3305 Returns a boolean indicating whether the interpreter is in batch
3308 *** Function: set-batch-mode?! ARG
3310 If ARG is true, switches the interpreter to batch mode. The `#f'
3311 case has not been implemented.
3313 ** Guile now provides full command-line editing, when run interactively.
3314 To use this feature, you must have the readline library installed.
3315 The Guile build process will notice it, and automatically include
3318 The readline library is available via anonymous FTP from any GNU
3319 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
3321 ** the-last-stack is now a fluid.
3323 * Changes to the procedure for linking libguile with your programs
3325 ** You can now use the `guile-config' utility to build programs that use Guile.
3327 Guile now includes a command-line utility called `guile-config', which
3328 can provide information about how to compile and link programs that
3331 *** `guile-config compile' prints any C compiler flags needed to use Guile.
3332 You should include this command's output on the command line you use
3333 to compile C or C++ code that #includes the Guile header files. It's
3334 usually just a `-I' flag to help the compiler find the Guile headers.
3337 *** `guile-config link' prints any linker flags necessary to link with Guile.
3339 This command writes to its standard output a list of flags which you
3340 must pass to the linker to link your code against the Guile library.
3341 The flags include '-lguile' itself, any other libraries the Guile
3342 library depends upon, and any `-L' flags needed to help the linker
3343 find those libraries.
3345 For example, here is a Makefile rule that builds a program named 'foo'
3346 from the object files ${FOO_OBJECTS}, and links them against Guile:
3349 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
3351 Previous Guile releases recommended that you use autoconf to detect
3352 which of a predefined set of libraries were present on your system.
3353 It is more robust to use `guile-config', since it records exactly which
3354 libraries the installed Guile library requires.
3356 This was originally called `build-guile', but was renamed to
3357 `guile-config' before Guile 1.3 was released, to be consistent with
3358 the analogous script for the GTK+ GUI toolkit, which is called
3362 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
3364 If you are using the GNU autoconf package to configure your program,
3365 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
3366 (described above) and gather the necessary values for use in your
3369 The GUILE_FLAGS macro expands to configure script code which runs the
3370 `guile-config' script, to find out where Guile's header files and
3371 libraries are installed. It sets two variables, marked for
3372 substitution, as by AC_SUBST.
3374 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
3375 code that uses Guile header files. This is almost always just a
3378 GUILE_LDFLAGS --- flags to pass to the linker to link a
3379 program against Guile. This includes `-lguile' for the Guile
3380 library itself, any libraries that Guile itself requires (like
3381 -lqthreads), and so on. It may also include a -L flag to tell the
3382 compiler where to find the libraries.
3384 GUILE_FLAGS is defined in the file guile.m4, in the top-level
3385 directory of the Guile distribution. You can copy it into your
3386 package's aclocal.m4 file, and then use it in your configure.in file.
3388 If you are using the `aclocal' program, distributed with GNU automake,
3389 to maintain your aclocal.m4 file, the Guile installation process
3390 installs guile.m4 where aclocal will find it. All you need to do is
3391 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
3392 this will copy the definition of GUILE_FLAGS into your aclocal.m4
3396 * Changes to Scheme functions and syntax
3398 ** Multi-byte strings have been removed, as have multi-byte and wide
3399 ports. We felt that these were the wrong approach to
3400 internationalization support.
3402 ** New function: readline [PROMPT]
3403 Read a line from the terminal, and allow the user to edit it,
3404 prompting with PROMPT. READLINE provides a large set of Emacs-like
3405 editing commands, lets the user recall previously typed lines, and
3406 works on almost every kind of terminal, including dumb terminals.
3408 READLINE assumes that the cursor is at the beginning of the line when
3409 it is invoked. Thus, you can't print a prompt yourself, and then call
3410 READLINE; you need to package up your prompt as a string, pass it to
3411 the function, and let READLINE print the prompt itself. This is
3412 because READLINE needs to know the prompt's screen width.
3414 For Guile to provide this function, you must have the readline
3415 library, version 2.1 or later, installed on your system. Readline is
3416 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
3417 any GNU mirror site.
3419 See also ADD-HISTORY function.
3421 ** New function: add-history STRING
3422 Add STRING as the most recent line in the history used by the READLINE
3423 command. READLINE does not add lines to the history itself; you must
3424 call ADD-HISTORY to make previous input available to the user.
3426 ** The behavior of the read-line function has changed.
3428 This function now uses standard C library functions to read the line,
3429 for speed. This means that it doesn not respect the value of
3430 scm-line-incrementors; it assumes that lines are delimited with
3433 (Note that this is read-line, the function that reads a line of text
3434 from a port, not readline, the function that reads a line from a
3435 terminal, providing full editing capabilities.)
3437 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
3439 This module provides some simple argument parsing. It exports one
3442 Function: getopt-gnu-style ARG-LS
3443 Parse a list of program arguments into an alist of option
3446 Each item in the list of program arguments is examined to see if
3447 it meets the syntax of a GNU long-named option. An argument like
3448 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
3449 returned alist, where MUMBLE is a keyword object with the same
3450 name as the argument. An argument like `--MUMBLE=FROB' produces
3451 an element of the form (MUMBLE . FROB), where FROB is a string.
3453 As a special case, the returned alist also contains a pair whose
3454 car is the symbol `rest'. The cdr of this pair is a list
3455 containing all the items in the argument list that are not options
3456 of the form mentioned above.
3458 The argument `--' is treated specially: all items in the argument
3459 list appearing after such an argument are not examined, and are
3460 returned in the special `rest' list.
3462 This function does not parse normal single-character switches.
3463 You will need to parse them out of the `rest' list yourself.
3465 ** The read syntax for byte vectors and short vectors has changed.
3467 Instead of #bytes(...), write #y(...).
3469 Instead of #short(...), write #h(...).
3471 This may seem nutty, but, like the other uniform vectors, byte vectors
3472 and short vectors want to have the same print and read syntax (and,
3473 more basic, want to have read syntax!). Changing the read syntax to
3474 use multiple characters after the hash sign breaks with the
3475 conventions used in R5RS and the conventions used for the other
3476 uniform vectors. It also introduces complexity in the current reader,
3477 both on the C and Scheme levels. (The Right solution is probably to
3478 change the syntax and prototypes for uniform vectors entirely.)
3481 ** The new module (ice-9 session) provides useful interactive functions.
3483 *** New procedure: (apropos REGEXP OPTION ...)
3485 Display a list of top-level variables whose names match REGEXP, and
3486 the modules they are imported from. Each OPTION should be one of the
3489 value --- Show the value of each matching variable.
3490 shadow --- Show bindings shadowed by subsequently imported modules.
3491 full --- Same as both `shadow' and `value'.
3495 guile> (apropos "trace" 'full)
3496 debug: trace #<procedure trace args>
3497 debug: untrace #<procedure untrace args>
3498 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
3499 the-scm-module: before-backtrace-hook ()
3500 the-scm-module: backtrace #<primitive-procedure backtrace>
3501 the-scm-module: after-backtrace-hook ()
3502 the-scm-module: has-shown-backtrace-hint? #f
3505 ** There are new functions and syntax for working with macros.
3507 Guile implements macros as a special object type. Any variable whose
3508 top-level binding is a macro object acts as a macro. The macro object
3509 specifies how the expression should be transformed before evaluation.
3511 *** Macro objects now print in a reasonable way, resembling procedures.
3513 *** New function: (macro? OBJ)
3514 True iff OBJ is a macro object.
3516 *** New function: (primitive-macro? OBJ)
3517 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
3518 macro transformers, implemented in eval.c rather than Scheme code.
3520 Why do we have this function?
3521 - For symmetry with procedure? and primitive-procedure?,
3522 - to allow custom print procedures to tell whether a macro is
3523 primitive, and display it differently, and
3524 - to allow compilers and user-written evaluators to distinguish
3525 builtin special forms from user-defined ones, which could be
3528 *** New function: (macro-type OBJ)
3529 Return a value indicating what kind of macro OBJ is. Possible return
3532 The symbol `syntax' --- a macro created by procedure->syntax.
3533 The symbol `macro' --- a macro created by procedure->macro.
3534 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
3535 The boolean #f --- if OBJ is not a macro object.
3537 *** New function: (macro-name MACRO)
3538 Return the name of the macro object MACRO's procedure, as returned by
3541 *** New function: (macro-transformer MACRO)
3542 Return the transformer procedure for MACRO.
3544 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
3546 Specify a new macro expander to use in the current module. Each
3547 MODULE is a module name, with the same meaning as in the `use-modules'
3548 form; each named module's exported bindings are added to the current
3549 top-level environment. TRANSFORMER is an expression evaluated in the
3550 resulting environment which must yield a procedure to use as the
3551 module's eval transformer: every expression evaluated in this module
3552 is passed to this function, and the result passed to the Guile
3555 *** macro-eval! is removed. Use local-eval instead.
3557 ** Some magic has been added to the printer to better handle user
3558 written printing routines (like record printers, closure printers).
3560 The problem is that these user written routines must have access to
3561 the current `print-state' to be able to handle fancy things like
3562 detection of circular references. These print-states have to be
3563 passed to the builtin printing routines (display, write, etc) to
3564 properly continue the print chain.
3566 We didn't want to change all existing print code so that it
3567 explicitly passes thru a print state in addition to a port. Instead,
3568 we extented the possible values that the builtin printing routines
3569 accept as a `port'. In addition to a normal port, they now also take
3570 a pair of a normal port and a print-state. Printing will go to the
3571 port and the print-state will be used to control the detection of
3572 circular references, etc. If the builtin function does not care for a
3573 print-state, it is simply ignored.
3575 User written callbacks are now called with such a pair as their
3576 `port', but because every function now accepts this pair as a PORT
3577 argument, you don't have to worry about that. In fact, it is probably
3578 safest to not check for these pairs.
3580 However, it is sometimes necessary to continue a print chain on a
3581 different port, for example to get a intermediate string
3582 representation of the printed value, mangle that string somehow, and
3583 then to finally print the mangled string. Use the new function
3585 inherit-print-state OLD-PORT NEW-PORT
3587 for this. It constructs a new `port' that prints to NEW-PORT but
3588 inherits the print-state of OLD-PORT.
3590 ** struct-vtable-offset renamed to vtable-offset-user
3592 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
3594 ** There is now a third optional argument to make-vtable-vtable
3595 (and fourth to make-struct) when constructing new types (vtables).
3596 This argument initializes field vtable-index-printer of the vtable.
3598 ** The detection of circular references has been extended to structs.
3599 That is, a structure that -- in the process of being printed -- prints
3600 itself does not lead to infinite recursion.
3602 ** There is now some basic support for fluids. Please read
3603 "libguile/fluid.h" to find out more. It is accessible from Scheme with
3604 the following functions and macros:
3606 Function: make-fluid
3608 Create a new fluid object. Fluids are not special variables or
3609 some other extension to the semantics of Scheme, but rather
3610 ordinary Scheme objects. You can store them into variables (that
3611 are still lexically scoped, of course) or into any other place you
3612 like. Every fluid has a initial value of `#f'.
3614 Function: fluid? OBJ
3616 Test whether OBJ is a fluid.
3618 Function: fluid-ref FLUID
3619 Function: fluid-set! FLUID VAL
3621 Access/modify the fluid FLUID. Modifications are only visible
3622 within the current dynamic root (that includes threads).
3624 Function: with-fluids* FLUIDS VALUES THUNK
3626 FLUIDS is a list of fluids and VALUES a corresponding list of
3627 values for these fluids. Before THUNK gets called the values are
3628 installed in the fluids and the old values of the fluids are
3629 saved in the VALUES list. When the flow of control leaves THUNK
3630 or reenters it, the values get swapped again. You might think of
3631 this as a `safe-fluid-excursion'. Note that the VALUES list is
3632 modified by `with-fluids*'.
3634 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
3636 The same as `with-fluids*' but with a different syntax. It looks
3637 just like `let', but both FLUID and VALUE are evaluated. Remember,
3638 fluids are not special variables but ordinary objects. FLUID
3639 should evaluate to a fluid.
3641 ** Changes to system call interfaces:
3643 *** close-port, close-input-port and close-output-port now return a
3644 boolean instead of an `unspecified' object. #t means that the port
3645 was successfully closed, while #f means it was already closed. It is
3646 also now possible for these procedures to raise an exception if an
3647 error occurs (some errors from write can be delayed until close.)
3649 *** the first argument to chmod, fcntl, ftell and fseek can now be a
3652 *** the third argument to fcntl is now optional.
3654 *** the first argument to chown can now be a file descriptor or a port.
3656 *** the argument to stat can now be a port.
3658 *** The following new procedures have been added (most use scsh
3661 *** procedure: close PORT/FD
3662 Similar to close-port (*note close-port: Closing Ports.), but also
3663 works on file descriptors. A side effect of closing a file
3664 descriptor is that any ports using that file descriptor are moved
3665 to a different file descriptor and have their revealed counts set
3668 *** procedure: port->fdes PORT
3669 Returns the integer file descriptor underlying PORT. As a side
3670 effect the revealed count of PORT is incremented.
3672 *** procedure: fdes->ports FDES
3673 Returns a list of existing ports which have FDES as an underlying
3674 file descriptor, without changing their revealed counts.
3676 *** procedure: fdes->inport FDES
3677 Returns an existing input port which has FDES as its underlying
3678 file descriptor, if one exists, and increments its revealed count.
3679 Otherwise, returns a new input port with a revealed count of 1.
3681 *** procedure: fdes->outport FDES
3682 Returns an existing output port which has FDES as its underlying
3683 file descriptor, if one exists, and increments its revealed count.
3684 Otherwise, returns a new output port with a revealed count of 1.
3686 The next group of procedures perform a `dup2' system call, if NEWFD
3687 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
3688 duplicated can be supplied as an integer or contained in a port. The
3689 type of value returned varies depending on which procedure is used.
3691 All procedures also have the side effect when performing `dup2' that
3692 any ports using NEWFD are moved to a different file descriptor and have
3693 their revealed counts set to zero.
3695 *** procedure: dup->fdes PORT/FD [NEWFD]
3696 Returns an integer file descriptor.
3698 *** procedure: dup->inport PORT/FD [NEWFD]
3699 Returns a new input port using the new file descriptor.
3701 *** procedure: dup->outport PORT/FD [NEWFD]
3702 Returns a new output port using the new file descriptor.
3704 *** procedure: dup PORT/FD [NEWFD]
3705 Returns a new port if PORT/FD is a port, with the same mode as the
3706 supplied port, otherwise returns an integer file descriptor.
3708 *** procedure: dup->port PORT/FD MODE [NEWFD]
3709 Returns a new port using the new file descriptor. MODE supplies a
3710 mode string for the port (*note open-file: File Ports.).
3712 *** procedure: setenv NAME VALUE
3713 Modifies the environment of the current process, which is also the
3714 default environment inherited by child processes.
3716 If VALUE is `#f', then NAME is removed from the environment.
3717 Otherwise, the string NAME=VALUE is added to the environment,
3718 replacing any existing string with name matching NAME.
3720 The return value is unspecified.
3722 *** procedure: truncate-file OBJ SIZE
3723 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
3724 can be a string containing a file name or an integer file
3725 descriptor or port open for output on the file. The underlying
3726 system calls are `truncate' and `ftruncate'.
3728 The return value is unspecified.
3730 *** procedure: setvbuf PORT MODE [SIZE]
3731 Set the buffering mode for PORT. MODE can be:
3739 block buffered, using a newly allocated buffer of SIZE bytes.
3740 However if SIZE is zero or unspecified, the port will be made
3743 This procedure should not be used after I/O has been performed with
3746 Ports are usually block buffered by default, with a default buffer
3747 size. Procedures e.g., *Note open-file: File Ports, which accept a
3748 mode string allow `0' to be added to request an unbuffered port.
3750 *** procedure: fsync PORT/FD
3751 Copies any unwritten data for the specified output file descriptor
3752 to disk. If PORT/FD is a port, its buffer is flushed before the
3753 underlying file descriptor is fsync'd. The return value is
3756 *** procedure: open-fdes PATH FLAGS [MODES]
3757 Similar to `open' but returns a file descriptor instead of a port.
3759 *** procedure: execle PATH ENV [ARG] ...
3760 Similar to `execl', but the environment of the new process is
3761 specified by ENV, which must be a list of strings as returned by
3762 the `environ' procedure.
3764 This procedure is currently implemented using the `execve' system
3765 call, but we call it `execle' because of its Scheme calling
3768 *** procedure: strerror ERRNO
3769 Returns the Unix error message corresponding to ERRNO, an integer.
3771 *** procedure: primitive-exit [STATUS]
3772 Terminate the current process without unwinding the Scheme stack.
3773 This is would typically be useful after a fork. The exit status
3774 is STATUS if supplied, otherwise zero.
3776 *** procedure: times
3777 Returns an object with information about real and processor time.
3778 The following procedures accept such an object as an argument and
3779 return a selected component:
3782 The current real time, expressed as time units relative to an
3786 The CPU time units used by the calling process.
3789 The CPU time units used by the system on behalf of the
3793 The CPU time units used by terminated child processes of the
3794 calling process, whose status has been collected (e.g., using
3798 Similarly, the CPU times units used by the system on behalf of
3799 terminated child processes.
3801 ** Removed: list-length
3802 ** Removed: list-append, list-append!
3803 ** Removed: list-reverse, list-reverse!
3805 ** array-map renamed to array-map!
3807 ** serial-array-map renamed to serial-array-map!
3809 ** catch doesn't take #f as first argument any longer
3811 Previously, it was possible to pass #f instead of a key to `catch'.
3812 That would cause `catch' to pass a jump buffer object to the procedure
3813 passed as second argument. The procedure could then use this jump
3814 buffer objekt as an argument to throw.
3816 This mechanism has been removed since its utility doesn't motivate the
3817 extra complexity it introduces.
3819 ** The `#/' notation for lists now provokes a warning message from Guile.
3820 This syntax will be removed from Guile in the near future.
3822 To disable the warning message, set the GUILE_HUSH environment
3823 variable to any non-empty value.
3825 ** The newline character now prints as `#\newline', following the
3826 normal Scheme notation, not `#\nl'.
3828 * Changes to the gh_ interface
3830 ** The gh_enter function now takes care of loading the Guile startup files.
3831 gh_enter works by calling scm_boot_guile; see the remarks below.
3833 ** Function: void gh_write (SCM x)
3835 Write the printed representation of the scheme object x to the current
3836 output port. Corresponds to the scheme level `write'.
3838 ** gh_list_length renamed to gh_length.
3840 ** vector handling routines
3842 Several major changes. In particular, gh_vector() now resembles
3843 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
3844 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
3845 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
3846 vector-related gh_ functions have been implemented.
3848 ** pair and list routines
3850 Implemented several of the R4RS pair and list functions that were
3853 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
3855 New function. Converts double arrays back and forth between Scheme
3858 * Changes to the scm_ interface
3860 ** The function scm_boot_guile now takes care of loading the startup files.
3862 Guile's primary initialization function, scm_boot_guile, now takes
3863 care of loading `boot-9.scm', in the `ice-9' module, to initialize
3864 Guile, define the module system, and put together some standard
3865 bindings. It also loads `init.scm', which is intended to hold
3866 site-specific initialization code.
3868 Since Guile cannot operate properly until boot-9.scm is loaded, there
3869 is no reason to separate loading boot-9.scm from Guile's other
3870 initialization processes.
3872 This job used to be done by scm_compile_shell_switches, which didn't
3873 make much sense; in particular, it meant that people using Guile for
3874 non-shell-like applications had to jump through hoops to get Guile
3875 initialized properly.
3877 ** The function scm_compile_shell_switches no longer loads the startup files.
3878 Now, Guile always loads the startup files, whenever it is initialized;
3879 see the notes above for scm_boot_guile and scm_load_startup_files.
3881 ** Function: scm_load_startup_files
3882 This new function takes care of loading Guile's initialization file
3883 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
3884 this is always called by the Guile initialization process, it's
3885 probably not too useful to call this yourself, but it's there anyway.
3887 ** The semantics of smob marking have changed slightly.
3889 The smob marking function (the `mark' member of the scm_smobfuns
3890 structure) is no longer responsible for setting the mark bit on the
3891 smob. The generic smob handling code in the garbage collector will
3892 set this bit. The mark function need only ensure that any other
3893 objects the smob refers to get marked.
3895 Note that this change means that the smob's GC8MARK bit is typically
3896 already set upon entry to the mark function. Thus, marking functions
3897 which look like this:
3900 if (SCM_GC8MARKP (ptr))
3902 SCM_SETGC8MARK (ptr);
3903 ... mark objects to which the smob refers ...
3906 are now incorrect, since they will return early, and fail to mark any
3907 other objects the smob refers to. Some code in the Guile library used
3910 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
3912 If you have implemented your own I/O port type, by writing the
3913 functions required by the scm_ptobfuns and then calling scm_newptob,
3914 you will need to change your functions slightly.
3916 The functions in a scm_ptobfuns structure now expect the port itself
3917 as their argument; they used to expect the `stream' member of the
3918 port's scm_port_table structure. This allows functions in an
3919 scm_ptobfuns structure to easily access the port's cell (and any flags
3920 it its CAR), and the port's scm_port_table structure.
3922 Guile now passes the I/O port itself as the `port' argument in the
3923 following scm_ptobfuns functions:
3925 int (*free) (SCM port);
3926 int (*fputc) (int, SCM port);
3927 int (*fputs) (char *, SCM port);
3928 scm_sizet (*fwrite) SCM_P ((char *ptr,
3932 int (*fflush) (SCM port);
3933 int (*fgetc) (SCM port);
3934 int (*fclose) (SCM port);
3936 The interfaces to the `mark', `print', `equalp', and `fgets' methods
3939 If you have existing code which defines its own port types, it is easy
3940 to convert your code to the new interface; simply apply SCM_STREAM to
3941 the port argument to yield the value you code used to expect.
3943 Note that since both the port and the stream have the same type in the
3944 C code --- they are both SCM values --- the C compiler will not remind
3945 you if you forget to update your scm_ptobfuns functions.
3948 ** Function: int scm_internal_select (int fds,
3952 struct timeval *timeout);
3954 This is a replacement for the `select' function provided by the OS.
3955 It enables I/O blocking and sleeping to happen for one cooperative
3956 thread without blocking other threads. It also avoids busy-loops in
3957 these situations. It is intended that all I/O blocking and sleeping
3958 will finally go through this function. Currently, this function is
3959 only available on systems providing `gettimeofday' and `select'.
3961 ** Function: SCM scm_internal_stack_catch (SCM tag,
3962 scm_catch_body_t body,
3964 scm_catch_handler_t handler,
3967 A new sibling to the other two C level `catch' functions
3968 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
3969 the stack to be saved automatically into the variable `the-last-stack'
3970 (scm_the_last_stack_var) on error. This is necessary if you want to
3971 use advanced error reporting, such as calling scm_display_error and
3972 scm_display_backtrace. (They both take a stack object as argument.)
3974 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
3976 scm_catch_handler_t handler,
3979 Spawns a new thread. It does a job similar to
3980 scm_call_with_new_thread but takes arguments more suitable when
3981 spawning threads from application C code.
3983 ** The hook scm_error_callback has been removed. It was originally
3984 intended as a way for the user to install his own error handler. But
3985 that method works badly since it intervenes between throw and catch,
3986 thereby changing the semantics of expressions like (catch #t ...).
3987 The correct way to do it is to use one of the C level catch functions
3988 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
3990 ** Removed functions:
3992 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
3993 scm_list_reverse, scm_list_reverse_x
3995 ** New macros: SCM_LISTn where n is one of the integers 0-9.
3997 These can be used for pretty list creation from C. The idea is taken
3998 from Erick Gallesio's STk.
4000 ** scm_array_map renamed to scm_array_map_x
4002 ** mbstrings are now removed
4004 This means that the type codes scm_tc7_mb_string and
4005 scm_tc7_mb_substring has been removed.
4007 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
4009 Since we no longer support multi-byte strings, these I/O functions
4010 have been simplified, and renamed. Here are their old names, and
4011 their new names and arguments:
4013 scm_gen_putc -> void scm_putc (int c, SCM port);
4014 scm_gen_puts -> void scm_puts (char *s, SCM port);
4015 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
4016 scm_gen_getc -> void scm_getc (SCM port);
4019 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
4021 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
4023 SCM_TYP7S now masks away the bit which distinguishes substrings from
4026 ** scm_catch_body_t: Backward incompatible change!
4028 Body functions to scm_internal_catch and friends do not any longer
4029 take a second argument. This is because it is no longer possible to
4030 pass a #f arg to catch.
4032 ** Calls to scm_protect_object and scm_unprotect now nest properly.
4034 The function scm_protect_object protects its argument from being freed
4035 by the garbage collector. scm_unprotect_object removes that
4038 These functions now nest properly. That is, for every object O, there
4039 is a counter which scm_protect_object(O) increments and
4040 scm_unprotect_object(O) decrements, if the counter is greater than
4041 zero. Every object's counter is zero when it is first created. If an
4042 object's counter is greater than zero, the garbage collector will not
4043 reclaim its storage.
4045 This allows you to use scm_protect_object in your code without
4046 worrying that some other function you call will call
4047 scm_unprotect_object, and allow it to be freed. Assuming that the
4048 functions you call are well-behaved, and unprotect only those objects
4049 they protect, you can follow the same rule and have confidence that
4050 objects will be freed only at appropriate times.
4053 Changes in Guile 1.2 (released Tuesday, June 24 1997):
4055 * Changes to the distribution
4057 ** Nightly snapshots are now available from ftp.red-bean.com.
4058 The old server, ftp.cyclic.com, has been relinquished to its rightful
4061 Nightly snapshots of the Guile development sources are now available via
4062 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
4064 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
4065 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
4067 ** To run Guile without installing it, the procedure has changed a bit.
4069 If you used a separate build directory to compile Guile, you'll need
4070 to include the build directory in SCHEME_LOAD_PATH, as well as the
4071 source directory. See the `INSTALL' file for examples.
4073 * Changes to the procedure for linking libguile with your programs
4075 ** The standard Guile load path for Scheme code now includes
4076 $(datadir)/guile (usually /usr/local/share/guile). This means that
4077 you can install your own Scheme files there, and Guile will find them.
4078 (Previous versions of Guile only checked a directory whose name
4079 contained the Guile version number, so you had to re-install or move
4080 your Scheme sources each time you installed a fresh version of Guile.)
4082 The load path also includes $(datadir)/guile/site; we recommend
4083 putting individual Scheme files there. If you want to install a
4084 package with multiple source files, create a directory for them under
4087 ** Guile 1.2 will now use the Rx regular expression library, if it is
4088 installed on your system. When you are linking libguile into your own
4089 programs, this means you will have to link against -lguile, -lqt (if
4090 you configured Guile with thread support), and -lrx.
4092 If you are using autoconf to generate configuration scripts for your
4093 application, the following lines should suffice to add the appropriate
4094 libraries to your link command:
4096 ### Find Rx, quickthreads and libguile.
4097 AC_CHECK_LIB(rx, main)
4098 AC_CHECK_LIB(qt, main)
4099 AC_CHECK_LIB(guile, scm_shell)
4101 The Guile 1.2 distribution does not contain sources for the Rx
4102 library, as Guile 1.0 did. If you want to use Rx, you'll need to
4103 retrieve it from a GNU FTP site and install it separately.
4105 * Changes to Scheme functions and syntax
4107 ** The dynamic linking features of Guile are now enabled by default.
4108 You can disable them by giving the `--disable-dynamic-linking' option
4111 (dynamic-link FILENAME)
4113 Find the object file denoted by FILENAME (a string) and link it
4114 into the running Guile application. When everything works out,
4115 return a Scheme object suitable for representing the linked object
4116 file. Otherwise an error is thrown. How object files are
4117 searched is system dependent.
4119 (dynamic-object? VAL)
4121 Determine whether VAL represents a dynamically linked object file.
4123 (dynamic-unlink DYNOBJ)
4125 Unlink the indicated object file from the application. DYNOBJ
4126 should be one of the values returned by `dynamic-link'.
4128 (dynamic-func FUNCTION DYNOBJ)
4130 Search the C function indicated by FUNCTION (a string or symbol)
4131 in DYNOBJ and return some Scheme object that can later be used
4132 with `dynamic-call' to actually call this function. Right now,
4133 these Scheme objects are formed by casting the address of the
4134 function to `long' and converting this number to its Scheme
4137 (dynamic-call FUNCTION DYNOBJ)
4139 Call the C function indicated by FUNCTION and DYNOBJ. The
4140 function is passed no arguments and its return value is ignored.
4141 When FUNCTION is something returned by `dynamic-func', call that
4142 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
4143 etc.), look it up in DYNOBJ; this is equivalent to
4145 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
4147 Interrupts are deferred while the C function is executing (with
4148 SCM_DEFER_INTS/SCM_ALLOW_INTS).
4150 (dynamic-args-call FUNCTION DYNOBJ ARGS)
4152 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
4153 some arguments and return its return value. The C function is
4154 expected to take two arguments and return an `int', just like
4157 int c_func (int argc, char **argv);
4159 ARGS must be a list of strings and is converted into an array of
4160 `char *'. The array is passed in ARGV and its size in ARGC. The
4161 return value is converted to a Scheme number and returned from the
4162 call to `dynamic-args-call'.
4164 When dynamic linking is disabled or not supported on your system,
4165 the above functions throw errors, but they are still available.
4167 Here is a small example that works on GNU/Linux:
4169 (define libc-obj (dynamic-link "libc.so"))
4170 (dynamic-args-call 'rand libc-obj '())
4172 See the file `libguile/DYNAMIC-LINKING' for additional comments.
4174 ** The #/ syntax for module names is depreciated, and will be removed
4175 in a future version of Guile. Instead of
4183 The latter syntax is more consistent with existing Lisp practice.
4185 ** Guile now does fancier printing of structures. Structures are the
4186 underlying implementation for records, which in turn are used to
4187 implement modules, so all of these object now print differently and in
4188 a more informative way.
4190 The Scheme printer will examine the builtin variable *struct-printer*
4191 whenever it needs to print a structure object. When this variable is
4192 not `#f' it is deemed to be a procedure and will be applied to the
4193 structure object and the output port. When *struct-printer* is `#f'
4194 or the procedure return `#f' the structure object will be printed in
4195 the boring #<struct 80458270> form.
4197 This hook is used by some routines in ice-9/boot-9.scm to implement
4198 type specific printing routines. Please read the comments there about
4201 One of the more specific uses of structs are records. The printing
4202 procedure that could be passed to MAKE-RECORD-TYPE is now actually
4203 called. It should behave like a *struct-printer* procedure (described
4206 ** Guile now supports a new R4RS-compliant syntax for keywords. A
4207 token of the form #:NAME, where NAME has the same syntax as a Scheme
4208 symbol, is the external representation of the keyword named NAME.
4209 Keyword objects print using this syntax as well, so values containing
4210 keyword objects can be read back into Guile. When used in an
4211 expression, keywords are self-quoting objects.
4213 Guile suports this read syntax, and uses this print syntax, regardless
4214 of the current setting of the `keyword' read option. The `keyword'
4215 read option only controls whether Guile recognizes the `:NAME' syntax,
4216 which is incompatible with R4RS. (R4RS says such token represent
4219 ** Guile has regular expression support again. Guile 1.0 included
4220 functions for matching regular expressions, based on the Rx library.
4221 In Guile 1.1, the Guile/Rx interface was removed to simplify the
4222 distribution, and thus Guile had no regular expression support. Guile
4223 1.2 again supports the most commonly used functions, and supports all
4224 of SCSH's regular expression functions.
4226 If your system does not include a POSIX regular expression library,
4227 and you have not linked Guile with a third-party regexp library such as
4228 Rx, these functions will not be available. You can tell whether your
4229 Guile installation includes regular expression support by checking
4230 whether the `*features*' list includes the `regex' symbol.
4232 *** regexp functions
4234 By default, Guile supports POSIX extended regular expressions. That
4235 means that the characters `(', `)', `+' and `?' are special, and must
4236 be escaped if you wish to match the literal characters.
4238 This regular expression interface was modeled after that implemented
4239 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
4240 with SCSH regular expressions.
4242 **** Function: string-match PATTERN STR [START]
4243 Compile the string PATTERN into a regular expression and compare
4244 it with STR. The optional numeric argument START specifies the
4245 position of STR at which to begin matching.
4247 `string-match' returns a "match structure" which describes what,
4248 if anything, was matched by the regular expression. *Note Match
4249 Structures::. If STR does not match PATTERN at all,
4250 `string-match' returns `#f'.
4252 Each time `string-match' is called, it must compile its PATTERN
4253 argument into a regular expression structure. This operation is
4254 expensive, which makes `string-match' inefficient if the same regular
4255 expression is used several times (for example, in a loop). For better
4256 performance, you can compile a regular expression in advance and then
4257 match strings against the compiled regexp.
4259 **** Function: make-regexp STR [FLAGS]
4260 Compile the regular expression described by STR, and return the
4261 compiled regexp structure. If STR does not describe a legal
4262 regular expression, `make-regexp' throws a
4263 `regular-expression-syntax' error.
4265 FLAGS may be the bitwise-or of one or more of the following:
4267 **** Constant: regexp/extended
4268 Use POSIX Extended Regular Expression syntax when interpreting
4269 STR. If not set, POSIX Basic Regular Expression syntax is used.
4270 If the FLAGS argument is omitted, we assume regexp/extended.
4272 **** Constant: regexp/icase
4273 Do not differentiate case. Subsequent searches using the
4274 returned regular expression will be case insensitive.
4276 **** Constant: regexp/newline
4277 Match-any-character operators don't match a newline.
4279 A non-matching list ([^...]) not containing a newline matches a
4282 Match-beginning-of-line operator (^) matches the empty string
4283 immediately after a newline, regardless of whether the FLAGS
4284 passed to regexp-exec contain regexp/notbol.
4286 Match-end-of-line operator ($) matches the empty string
4287 immediately before a newline, regardless of whether the FLAGS
4288 passed to regexp-exec contain regexp/noteol.
4290 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
4291 Match the compiled regular expression REGEXP against `str'. If
4292 the optional integer START argument is provided, begin matching
4293 from that position in the string. Return a match structure
4294 describing the results of the match, or `#f' if no match could be
4297 FLAGS may be the bitwise-or of one or more of the following:
4299 **** Constant: regexp/notbol
4300 The match-beginning-of-line operator always fails to match (but
4301 see the compilation flag regexp/newline above) This flag may be
4302 used when different portions of a string are passed to
4303 regexp-exec and the beginning of the string should not be
4304 interpreted as the beginning of the line.
4306 **** Constant: regexp/noteol
4307 The match-end-of-line operator always fails to match (but see the
4308 compilation flag regexp/newline above)
4310 **** Function: regexp? OBJ
4311 Return `#t' if OBJ is a compiled regular expression, or `#f'
4314 Regular expressions are commonly used to find patterns in one string
4315 and replace them with the contents of another string.
4317 **** Function: regexp-substitute PORT MATCH [ITEM...]
4318 Write to the output port PORT selected contents of the match
4319 structure MATCH. Each ITEM specifies what should be written, and
4320 may be one of the following arguments:
4322 * A string. String arguments are written out verbatim.
4324 * An integer. The submatch with that number is written.
4326 * The symbol `pre'. The portion of the matched string preceding
4327 the regexp match is written.
4329 * The symbol `post'. The portion of the matched string
4330 following the regexp match is written.
4332 PORT may be `#f', in which case nothing is written; instead,
4333 `regexp-substitute' constructs a string from the specified ITEMs
4336 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
4337 Similar to `regexp-substitute', but can be used to perform global
4338 substitutions on STR. Instead of taking a match structure as an
4339 argument, `regexp-substitute/global' takes two string arguments: a
4340 REGEXP string describing a regular expression, and a TARGET string
4341 which should be matched against this regular expression.
4343 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
4346 * A function may be supplied. When this function is called, it
4347 will be passed one argument: a match structure for a given
4348 regular expression match. It should return a string to be
4349 written out to PORT.
4351 * The `post' symbol causes `regexp-substitute/global' to recurse
4352 on the unmatched portion of STR. This *must* be supplied in
4353 order to perform global search-and-replace on STR; if it is
4354 not present among the ITEMs, then `regexp-substitute/global'
4355 will return after processing a single match.
4357 *** Match Structures
4359 A "match structure" is the object returned by `string-match' and
4360 `regexp-exec'. It describes which portion of a string, if any, matched
4361 the given regular expression. Match structures include: a reference to
4362 the string that was checked for matches; the starting and ending
4363 positions of the regexp match; and, if the regexp included any
4364 parenthesized subexpressions, the starting and ending positions of each
4367 In each of the regexp match functions described below, the `match'
4368 argument must be a match structure returned by a previous call to
4369 `string-match' or `regexp-exec'. Most of these functions return some
4370 information about the original target string that was matched against a
4371 regular expression; we will call that string TARGET for easy reference.
4373 **** Function: regexp-match? OBJ
4374 Return `#t' if OBJ is a match structure returned by a previous
4375 call to `regexp-exec', or `#f' otherwise.
4377 **** Function: match:substring MATCH [N]
4378 Return the portion of TARGET matched by subexpression number N.
4379 Submatch 0 (the default) represents the entire regexp match. If
4380 the regular expression as a whole matched, but the subexpression
4381 number N did not match, return `#f'.
4383 **** Function: match:start MATCH [N]
4384 Return the starting position of submatch number N.
4386 **** Function: match:end MATCH [N]
4387 Return the ending position of submatch number N.
4389 **** Function: match:prefix MATCH
4390 Return the unmatched portion of TARGET preceding the regexp match.
4392 **** Function: match:suffix MATCH
4393 Return the unmatched portion of TARGET following the regexp match.
4395 **** Function: match:count MATCH
4396 Return the number of parenthesized subexpressions from MATCH.
4397 Note that the entire regular expression match itself counts as a
4398 subexpression, and failed submatches are included in the count.
4400 **** Function: match:string MATCH
4401 Return the original TARGET string.
4403 *** Backslash Escapes
4405 Sometimes you will want a regexp to match characters like `*' or `$'
4406 exactly. For example, to check whether a particular string represents
4407 a menu entry from an Info node, it would be useful to match it against
4408 a regexp like `^* [^:]*::'. However, this won't work; because the
4409 asterisk is a metacharacter, it won't match the `*' at the beginning of
4410 the string. In this case, we want to make the first asterisk un-magic.
4412 You can do this by preceding the metacharacter with a backslash
4413 character `\'. (This is also called "quoting" the metacharacter, and
4414 is known as a "backslash escape".) When Guile sees a backslash in a
4415 regular expression, it considers the following glyph to be an ordinary
4416 character, no matter what special meaning it would ordinarily have.
4417 Therefore, we can make the above example work by changing the regexp to
4418 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
4419 to match only a single asterisk in the target string.
4421 Since the backslash is itself a metacharacter, you may force a
4422 regexp to match a backslash in the target string by preceding the
4423 backslash with itself. For example, to find variable references in a
4424 TeX program, you might want to find occurrences of the string `\let\'
4425 followed by any number of alphabetic characters. The regular expression
4426 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
4427 each match a single backslash in the target string.
4429 **** Function: regexp-quote STR
4430 Quote each special character found in STR with a backslash, and
4431 return the resulting string.
4433 *Very important:* Using backslash escapes in Guile source code (as
4434 in Emacs Lisp or C) can be tricky, because the backslash character has
4435 special meaning for the Guile reader. For example, if Guile encounters
4436 the character sequence `\n' in the middle of a string while processing
4437 Scheme code, it replaces those characters with a newline character.
4438 Similarly, the character sequence `\t' is replaced by a horizontal tab.
4439 Several of these "escape sequences" are processed by the Guile reader
4440 before your code is executed. Unrecognized escape sequences are
4441 ignored: if the characters `\*' appear in a string, they will be
4442 translated to the single character `*'.
4444 This translation is obviously undesirable for regular expressions,
4445 since we want to be able to include backslashes in a string in order to
4446 escape regexp metacharacters. Therefore, to make sure that a backslash
4447 is preserved in a string in your Guile program, you must use *two*
4448 consecutive backslashes:
4450 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
4452 The string in this example is preprocessed by the Guile reader before
4453 any code is executed. The resulting argument to `make-regexp' is the
4454 string `^\* [^:]*', which is what we really want.
4456 This also means that in order to write a regular expression that
4457 matches a single backslash character, the regular expression string in
4458 the source code must include *four* backslashes. Each consecutive pair
4459 of backslashes gets translated by the Guile reader to a single
4460 backslash, and the resulting double-backslash is interpreted by the
4461 regexp engine as matching a single backslash character. Hence:
4463 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
4465 The reason for the unwieldiness of this syntax is historical. Both
4466 regular expression pattern matchers and Unix string processing systems
4467 have traditionally used backslashes with the special meanings described
4468 above. The POSIX regular expression specification and ANSI C standard
4469 both require these semantics. Attempting to abandon either convention
4470 would cause other kinds of compatibility problems, possibly more severe
4471 ones. Therefore, without extending the Scheme reader to support
4472 strings with different quoting conventions (an ungainly and confusing
4473 extension when implemented in other languages), we must adhere to this
4474 cumbersome escape syntax.
4476 * Changes to the gh_ interface
4478 * Changes to the scm_ interface
4480 * Changes to system call interfaces:
4482 ** The value returned by `raise' is now unspecified. It throws an exception
4485 *** A new procedure `sigaction' can be used to install signal handlers
4487 (sigaction signum [action] [flags])
4489 signum is the signal number, which can be specified using the value
4492 If action is omitted, sigaction returns a pair: the CAR is the current
4493 signal hander, which will be either an integer with the value SIG_DFL
4494 (default action) or SIG_IGN (ignore), or the Scheme procedure which
4495 handles the signal, or #f if a non-Scheme procedure handles the
4496 signal. The CDR contains the current sigaction flags for the handler.
4498 If action is provided, it is installed as the new handler for signum.
4499 action can be a Scheme procedure taking one argument, or the value of
4500 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
4501 whatever signal handler was installed before sigaction was first used.
4502 Flags can optionally be specified for the new handler (SA_RESTART is
4503 always used if the system provides it, so need not be specified.) The
4504 return value is a pair with information about the old handler as
4507 This interface does not provide access to the "signal blocking"
4508 facility. Maybe this is not needed, since the thread support may
4509 provide solutions to the problem of consistent access to data
4512 *** A new procedure `flush-all-ports' is equivalent to running
4513 `force-output' on every port open for output.
4515 ** Guile now provides information on how it was built, via the new
4516 global variable, %guile-build-info. This variable records the values
4517 of the standard GNU makefile directory variables as an assocation
4518 list, mapping variable names (symbols) onto directory paths (strings).
4519 For example, to find out where the Guile link libraries were
4520 installed, you can say:
4522 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
4525 * Changes to the scm_ interface
4527 ** The new function scm_handle_by_message_noexit is just like the
4528 existing scm_handle_by_message function, except that it doesn't call
4529 exit to terminate the process. Instead, it prints a message and just
4530 returns #f. This might be a more appropriate catch-all handler for
4531 new dynamic roots and threads.
4534 Changes in Guile 1.1 (released Friday, May 16 1997):
4536 * Changes to the distribution.
4538 The Guile 1.0 distribution has been split up into several smaller
4540 guile-core --- the Guile interpreter itself.
4541 guile-tcltk --- the interface between the Guile interpreter and
4542 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
4543 is a toolkit for building graphical user interfaces.
4544 guile-rgx-ctax --- the interface between Guile and the Rx regular
4545 expression matcher, and the translator for the Ctax
4546 programming language. These are packaged together because the
4547 Ctax translator uses Rx to parse Ctax source code.
4549 This NEWS file describes the changes made to guile-core since the 1.0
4552 We no longer distribute the documentation, since it was either out of
4553 date, or incomplete. As soon as we have current documentation, we
4558 * Changes to the stand-alone interpreter
4560 ** guile now accepts command-line arguments compatible with SCSH, Olin
4561 Shivers' Scheme Shell.
4563 In general, arguments are evaluated from left to right, but there are
4564 exceptions. The following switches stop argument processing, and
4565 stash all remaining command-line arguments as the value returned by
4566 the (command-line) function.
4567 -s SCRIPT load Scheme source code from FILE, and exit
4568 -c EXPR evalute Scheme expression EXPR, and exit
4569 -- stop scanning arguments; run interactively
4571 The switches below are processed as they are encountered.
4572 -l FILE load Scheme source code from FILE
4573 -e FUNCTION after reading script, apply FUNCTION to
4574 command line arguments
4575 -ds do -s script at this point
4576 --emacs enable Emacs protocol (experimental)
4577 -h, --help display this help and exit
4578 -v, --version display version information and exit
4579 \ read arguments from following script lines
4581 So, for example, here is a Guile script named `ekko' (thanks, Olin)
4582 which re-implements the traditional "echo" command:
4584 #!/usr/local/bin/guile -s
4587 (map (lambda (arg) (display arg) (display " "))
4591 (main (command-line))
4593 Suppose we invoke this script as follows:
4595 ekko a speckled gecko
4597 Through the magic of Unix script processing (triggered by the `#!'
4598 token at the top of the file), /usr/local/bin/guile receives the
4599 following list of command-line arguments:
4601 ("-s" "./ekko" "a" "speckled" "gecko")
4603 Unix inserts the name of the script after the argument specified on
4604 the first line of the file (in this case, "-s"), and then follows that
4605 with the arguments given to the script. Guile loads the script, which
4606 defines the `main' function, and then applies it to the list of
4607 remaining command-line arguments, ("a" "speckled" "gecko").
4609 In Unix, the first line of a script file must take the following form:
4611 #!INTERPRETER ARGUMENT
4613 where INTERPRETER is the absolute filename of the interpreter
4614 executable, and ARGUMENT is a single command-line argument to pass to
4617 You may only pass one argument to the interpreter, and its length is
4618 limited. These restrictions can be annoying to work around, so Guile
4619 provides a general mechanism (borrowed from, and compatible with,
4620 SCSH) for circumventing them.
4622 If the ARGUMENT in a Guile script is a single backslash character,
4623 `\', Guile will open the script file, parse arguments from its second
4624 and subsequent lines, and replace the `\' with them. So, for example,
4625 here is another implementation of the `ekko' script:
4627 #!/usr/local/bin/guile \
4631 (for-each (lambda (arg) (display arg) (display " "))
4635 If the user invokes this script as follows:
4637 ekko a speckled gecko
4639 Unix expands this into
4641 /usr/local/bin/guile \ ekko a speckled gecko
4643 When Guile sees the `\' argument, it replaces it with the arguments
4644 read from the second line of the script, producing:
4646 /usr/local/bin/guile -e main -s ekko a speckled gecko
4648 This tells Guile to load the `ekko' script, and apply the function
4649 `main' to the argument list ("a" "speckled" "gecko").
4651 Here is how Guile parses the command-line arguments:
4652 - Each space character terminates an argument. This means that two
4653 spaces in a row introduce an empty-string argument.
4654 - The tab character is not permitted (unless you quote it with the
4655 backslash character, as described below), to avoid confusion.
4656 - The newline character terminates the sequence of arguments, and will
4657 also terminate a final non-empty argument. (However, a newline
4658 following a space will not introduce a final empty-string argument;
4659 it only terminates the argument list.)
4660 - The backslash character is the escape character. It escapes
4661 backslash, space, tab, and newline. The ANSI C escape sequences
4662 like \n and \t are also supported. These produce argument
4663 constituents; the two-character combination \n doesn't act like a
4664 terminating newline. The escape sequence \NNN for exactly three
4665 octal digits reads as the character whose ASCII code is NNN. As
4666 above, characters produced this way are argument constituents.
4667 Backslash followed by other characters is not allowed.
4669 * Changes to the procedure for linking libguile with your programs
4671 ** Guile now builds and installs a shared guile library, if your
4672 system support shared libraries. (It still builds a static library on
4673 all systems.) Guile automatically detects whether your system
4674 supports shared libraries. To prevent Guile from buildisg shared
4675 libraries, pass the `--disable-shared' flag to the configure script.
4677 Guile takes longer to compile when it builds shared libraries, because
4678 it must compile every file twice --- once to produce position-
4679 independent object code, and once to produce normal object code.
4681 ** The libthreads library has been merged into libguile.
4683 To link a program against Guile, you now need only link against
4684 -lguile and -lqt; -lthreads is no longer needed. If you are using
4685 autoconf to generate configuration scripts for your application, the
4686 following lines should suffice to add the appropriate libraries to
4689 ### Find quickthreads and libguile.
4690 AC_CHECK_LIB(qt, main)
4691 AC_CHECK_LIB(guile, scm_shell)
4693 * Changes to Scheme functions
4695 ** Guile Scheme's special syntax for keyword objects is now optional,
4696 and disabled by default.
4698 The syntax variation from R4RS made it difficult to port some
4699 interesting packages to Guile. The routines which accepted keyword
4700 arguments (mostly in the module system) have been modified to also
4701 accept symbols whose names begin with `:'.
4703 To change the keyword syntax, you must first import the (ice-9 debug)
4705 (use-modules (ice-9 debug))
4707 Then you can enable the keyword syntax as follows:
4708 (read-set! keywords 'prefix)
4710 To disable keyword syntax, do this:
4711 (read-set! keywords #f)
4713 ** Many more primitive functions accept shared substrings as
4714 arguments. In the past, these functions required normal, mutable
4715 strings as arguments, although they never made use of this
4718 ** The uniform array functions now operate on byte vectors. These
4719 functions are `array-fill!', `serial-array-copy!', `array-copy!',
4720 `serial-array-map', `array-map', `array-for-each', and
4723 ** The new functions `trace' and `untrace' implement simple debugging
4724 support for Scheme functions.
4726 The `trace' function accepts any number of procedures as arguments,
4727 and tells the Guile interpreter to display each procedure's name and
4728 arguments each time the procedure is invoked. When invoked with no
4729 arguments, `trace' returns the list of procedures currently being
4732 The `untrace' function accepts any number of procedures as arguments,
4733 and tells the Guile interpreter not to trace them any more. When
4734 invoked with no arguments, `untrace' untraces all curretly traced
4737 The tracing in Guile has an advantage over most other systems: we
4738 don't create new procedure objects, but mark the procedure objects
4739 themselves. This means that anonymous and internal procedures can be
4742 ** The function `assert-repl-prompt' has been renamed to
4743 `set-repl-prompt!'. It takes one argument, PROMPT.
4744 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
4745 - If PROMPT is a string, we use it as a prompt.
4746 - If PROMPT is a procedure accepting no arguments, we call it, and
4747 display the result as a prompt.
4748 - Otherwise, we display "> ".
4750 ** The new function `eval-string' reads Scheme expressions from a
4751 string and evaluates them, returning the value of the last expression
4752 in the string. If the string contains no expressions, it returns an
4755 ** The new function `thunk?' returns true iff its argument is a
4756 procedure of zero arguments.
4758 ** `defined?' is now a builtin function, instead of syntax. This
4759 means that its argument should be quoted. It returns #t iff its
4760 argument is bound in the current module.
4762 ** The new syntax `use-modules' allows you to add new modules to your
4763 environment without re-typing a complete `define-module' form. It
4764 accepts any number of module names as arguments, and imports their
4765 public bindings into the current module.
4767 ** The new function (module-defined? NAME MODULE) returns true iff
4768 NAME, a symbol, is defined in MODULE, a module object.
4770 ** The new function `builtin-bindings' creates and returns a hash
4771 table containing copies of all the root module's bindings.
4773 ** The new function `builtin-weak-bindings' does the same as
4774 `builtin-bindings', but creates a doubly-weak hash table.
4776 ** The `equal?' function now considers variable objects to be
4777 equivalent if they have the same name and the same value.
4779 ** The new function `command-line' returns the command-line arguments
4780 given to Guile, as a list of strings.
4782 When using guile as a script interpreter, `command-line' returns the
4783 script's arguments; those processed by the interpreter (like `-s' or
4784 `-c') are omitted. (In other words, you get the normal, expected
4785 behavior.) Any application that uses scm_shell to process its
4786 command-line arguments gets this behavior as well.
4788 ** The new function `load-user-init' looks for a file called `.guile'
4789 in the user's home directory, and loads it if it exists. This is
4790 mostly for use by the code generated by scm_compile_shell_switches,
4791 but we thought it might also be useful in other circumstances.
4793 ** The new function `log10' returns the base-10 logarithm of its
4796 ** Changes to I/O functions
4798 *** The functions `read', `primitive-load', `read-and-eval!', and
4799 `primitive-load-path' no longer take optional arguments controlling
4800 case insensitivity and a `#' parser.
4802 Case sensitivity is now controlled by a read option called
4803 `case-insensitive'. The user can add new `#' syntaxes with the
4804 `read-hash-extend' function (see below).
4806 *** The new function `read-hash-extend' allows the user to change the
4807 syntax of Guile Scheme in a somewhat controlled way.
4809 (read-hash-extend CHAR PROC)
4810 When parsing S-expressions, if we read a `#' character followed by
4811 the character CHAR, use PROC to parse an object from the stream.
4812 If PROC is #f, remove any parsing procedure registered for CHAR.
4814 The reader applies PROC to two arguments: CHAR and an input port.
4816 *** The new functions read-delimited and read-delimited! provide a
4817 general mechanism for doing delimited input on streams.
4819 (read-delimited DELIMS [PORT HANDLE-DELIM])
4820 Read until we encounter one of the characters in DELIMS (a string),
4821 or end-of-file. PORT is the input port to read from; it defaults to
4822 the current input port. The HANDLE-DELIM parameter determines how
4823 the terminating character is handled; it should be one of the
4826 'trim omit delimiter from result
4827 'peek leave delimiter character in input stream
4828 'concat append delimiter character to returned value
4829 'split return a pair: (RESULT . TERMINATOR)
4831 HANDLE-DELIM defaults to 'peek.
4833 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
4834 A side-effecting variant of `read-delimited'.
4836 The data is written into the string BUF at the indices in the
4837 half-open interval [START, END); the default interval is the whole
4838 string: START = 0 and END = (string-length BUF). The values of
4839 START and END must specify a well-defined interval in BUF, i.e.
4840 0 <= START <= END <= (string-length BUF).
4842 It returns NBYTES, the number of bytes read. If the buffer filled
4843 up without a delimiter character being found, it returns #f. If the
4844 port is at EOF when the read starts, it returns the EOF object.
4846 If an integer is returned (i.e., the read is successfully terminated
4847 by reading a delimiter character), then the HANDLE-DELIM parameter
4848 determines how to handle the terminating character. It is described
4849 above, and defaults to 'peek.
4851 (The descriptions of these functions were borrowed from the SCSH
4852 manual, by Olin Shivers and Brian Carlstrom.)
4854 *** The `%read-delimited!' function is the primitive used to implement
4855 `read-delimited' and `read-delimited!'.
4857 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
4859 This returns a pair of values: (TERMINATOR . NUM-READ).
4860 - TERMINATOR describes why the read was terminated. If it is a
4861 character or the eof object, then that is the value that terminated
4862 the read. If it is #f, the function filled the buffer without finding
4863 a delimiting character.
4864 - NUM-READ is the number of characters read into BUF.
4866 If the read is successfully terminated by reading a delimiter
4867 character, then the gobble? parameter determines what to do with the
4868 terminating character. If true, the character is removed from the
4869 input stream; if false, the character is left in the input stream
4870 where a subsequent read operation will retrieve it. In either case,
4871 the character is also the first value returned by the procedure call.
4873 (The descriptions of this function was borrowed from the SCSH manual,
4874 by Olin Shivers and Brian Carlstrom.)
4876 *** The `read-line' and `read-line!' functions have changed; they now
4877 trim the terminator by default; previously they appended it to the
4878 returned string. For the old behavior, use (read-line PORT 'concat).
4880 *** The functions `uniform-array-read!' and `uniform-array-write!' now
4881 take new optional START and END arguments, specifying the region of
4882 the array to read and write.
4884 *** The `ungetc-char-ready?' function has been removed. We feel it's
4885 inappropriate for an interface to expose implementation details this
4888 ** Changes to the Unix library and system call interface
4890 *** The new fcntl function provides access to the Unix `fcntl' system
4893 (fcntl PORT COMMAND VALUE)
4894 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
4895 Values for COMMAND are:
4897 F_DUPFD duplicate a file descriptor
4898 F_GETFD read the descriptor's close-on-exec flag
4899 F_SETFD set the descriptor's close-on-exec flag to VALUE
4900 F_GETFL read the descriptor's flags, as set on open
4901 F_SETFL set the descriptor's flags, as set on open to VALUE
4902 F_GETOWN return the process ID of a socket's owner, for SIGIO
4903 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
4904 FD_CLOEXEC not sure what this is
4906 For details, see the documentation for the fcntl system call.
4908 *** The arguments to `select' have changed, for compatibility with
4909 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
4910 expected behavior. The MILLISECONDS parameter has been changed to
4911 MICROSECONDS, to more closely resemble the underlying system call.
4912 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
4913 corresponding return set will be the same.
4915 *** The arguments to the `mknod' system call have changed. They are
4918 (mknod PATH TYPE PERMS DEV)
4919 Create a new file (`node') in the file system. PATH is the name of
4920 the file to create. TYPE is the kind of file to create; it should
4921 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
4922 permission bits to give the newly created file. If TYPE is
4923 'block-special or 'char-special, DEV specifies which device the
4924 special file refers to; its interpretation depends on the kind of
4925 special file being created.
4927 *** The `fork' function has been renamed to `primitive-fork', to avoid
4928 clashing with various SCSH forks.
4930 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
4931 and `recvfrom!'. They no longer accept a size for a second argument;
4932 you must pass a string to hold the received value. They no longer
4933 return the buffer. Instead, `recv' returns the length of the message
4934 received, and `recvfrom' returns a pair containing the packet's length
4935 and originating address.
4937 *** The file descriptor datatype has been removed, as have the
4938 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
4939 We plan to replace these functions with a SCSH-compatible interface.
4941 *** The `create' function has been removed; it's just a special case
4944 *** There are new functions to break down process termination status
4945 values. In the descriptions below, STATUS is a value returned by
4948 (status:exit-val STATUS)
4949 If the child process exited normally, this function returns the exit
4950 code for the child process (i.e., the value passed to exit, or
4951 returned from main). If the child process did not exit normally,
4952 this function returns #f.
4954 (status:stop-sig STATUS)
4955 If the child process was suspended by a signal, this function
4956 returns the signal that suspended the child. Otherwise, it returns
4959 (status:term-sig STATUS)
4960 If the child process terminated abnormally, this function returns
4961 the signal that terminated the child. Otherwise, this function
4964 POSIX promises that exactly one of these functions will return true on
4965 a valid STATUS value.
4967 These functions are compatible with SCSH.
4969 *** There are new accessors and setters for the broken-out time vectors
4970 returned by `localtime', `gmtime', and that ilk. They are:
4972 Component Accessor Setter
4973 ========================= ============ ============
4974 seconds tm:sec set-tm:sec
4975 minutes tm:min set-tm:min
4976 hours tm:hour set-tm:hour
4977 day of the month tm:mday set-tm:mday
4978 month tm:mon set-tm:mon
4979 year tm:year set-tm:year
4980 day of the week tm:wday set-tm:wday
4981 day in the year tm:yday set-tm:yday
4982 daylight saving time tm:isdst set-tm:isdst
4983 GMT offset, seconds tm:gmtoff set-tm:gmtoff
4984 name of time zone tm:zone set-tm:zone
4986 *** There are new accessors for the vectors returned by `uname',
4987 describing the host system:
4990 ============================================== ================
4991 name of the operating system implementation utsname:sysname
4992 network name of this machine utsname:nodename
4993 release level of the operating system utsname:release
4994 version level of the operating system utsname:version
4995 machine hardware platform utsname:machine
4997 *** There are new accessors for the vectors returned by `getpw',
4998 `getpwnam', `getpwuid', and `getpwent', describing entries from the
4999 system's user database:
5002 ====================== =================
5003 user name passwd:name
5004 user password passwd:passwd
5007 real name passwd:gecos
5008 home directory passwd:dir
5009 shell program passwd:shell
5011 *** There are new accessors for the vectors returned by `getgr',
5012 `getgrnam', `getgrgid', and `getgrent', describing entries from the
5013 system's group database:
5016 ======================= ============
5017 group name group:name
5018 group password group:passwd
5020 group members group:mem
5022 *** There are new accessors for the vectors returned by `gethost',
5023 `gethostbyaddr', `gethostbyname', and `gethostent', describing
5027 ========================= ===============
5028 official name of host hostent:name
5029 alias list hostent:aliases
5030 host address type hostent:addrtype
5031 length of address hostent:length
5032 list of addresses hostent:addr-list
5034 *** There are new accessors for the vectors returned by `getnet',
5035 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
5039 ========================= ===============
5040 official name of net netent:name
5041 alias list netent:aliases
5042 net number type netent:addrtype
5043 net number netent:net
5045 *** There are new accessors for the vectors returned by `getproto',
5046 `getprotobyname', `getprotobynumber', and `getprotoent', describing
5050 ========================= ===============
5051 official protocol name protoent:name
5052 alias list protoent:aliases
5053 protocol number protoent:proto
5055 *** There are new accessors for the vectors returned by `getserv',
5056 `getservbyname', `getservbyport', and `getservent', describing
5060 ========================= ===============
5061 official service name servent:name
5062 alias list servent:aliases
5063 port number servent:port
5064 protocol to use servent:proto
5066 *** There are new accessors for the sockaddr structures returned by
5067 `accept', `getsockname', `getpeername', `recvfrom!':
5070 ======================================== ===============
5071 address format (`family') sockaddr:fam
5072 path, for file domain addresses sockaddr:path
5073 address, for internet domain addresses sockaddr:addr
5074 TCP or UDP port, for internet sockaddr:port
5076 *** The `getpwent', `getgrent', `gethostent', `getnetent',
5077 `getprotoent', and `getservent' functions now return #f at the end of
5078 the user database. (They used to throw an exception.)
5080 Note that calling MUMBLEent function is equivalent to calling the
5081 corresponding MUMBLE function with no arguments.
5083 *** The `setpwent', `setgrent', `sethostent', `setnetent',
5084 `setprotoent', and `setservent' routines now take no arguments.
5086 *** The `gethost', `getproto', `getnet', and `getserv' functions now
5087 provide more useful information when they throw an exception.
5089 *** The `lnaof' function has been renamed to `inet-lnaof'.
5091 *** Guile now claims to have the `current-time' feature.
5093 *** The `mktime' function now takes an optional second argument ZONE,
5094 giving the time zone to use for the conversion. ZONE should be a
5095 string, in the same format as expected for the "TZ" environment variable.
5097 *** The `strptime' function now returns a pair (TIME . COUNT), where
5098 TIME is the parsed time as a vector, and COUNT is the number of
5099 characters from the string left unparsed. This function used to
5100 return the remaining characters as a string.
5102 *** The `gettimeofday' function has replaced the old `time+ticks' function.
5103 The return value is now (SECONDS . MICROSECONDS); the fractional
5104 component is no longer expressed in "ticks".
5106 *** The `ticks/sec' constant has been removed, in light of the above change.
5108 * Changes to the gh_ interface
5110 ** gh_eval_str() now returns an SCM object which is the result of the
5113 ** gh_scm2str() now copies the Scheme data to a caller-provided C
5116 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
5117 and returns the array
5119 ** gh_scm2str0() is gone: there is no need to distinguish
5120 null-terminated from non-null-terminated, since gh_scm2newstr() allows
5121 the user to interpret the data both ways.
5123 * Changes to the scm_ interface
5125 ** The new function scm_symbol_value0 provides an easy way to get a
5126 symbol's value from C code:
5128 SCM scm_symbol_value0 (char *NAME)
5129 Return the value of the symbol named by the null-terminated string
5130 NAME in the current module. If the symbol named NAME is unbound in
5131 the current module, return SCM_UNDEFINED.
5133 ** The new function scm_sysintern0 creates new top-level variables,
5134 without assigning them a value.
5136 SCM scm_sysintern0 (char *NAME)
5137 Create a new Scheme top-level variable named NAME. NAME is a
5138 null-terminated string. Return the variable's value cell.
5140 ** The function scm_internal_catch is the guts of catch. It handles
5141 all the mechanics of setting up a catch target, invoking the catch
5142 body, and perhaps invoking the handler if the body does a throw.
5144 The function is designed to be usable from C code, but is general
5145 enough to implement all the semantics Guile Scheme expects from throw.
5147 TAG is the catch tag. Typically, this is a symbol, but this function
5148 doesn't actually care about that.
5150 BODY is a pointer to a C function which runs the body of the catch;
5151 this is the code you can throw from. We call it like this:
5152 BODY (BODY_DATA, JMPBUF)
5154 BODY_DATA is just the BODY_DATA argument we received; we pass it
5155 through to BODY as its first argument. The caller can make
5156 BODY_DATA point to anything useful that BODY might need.
5157 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
5158 which we have just created and initialized.
5160 HANDLER is a pointer to a C function to deal with a throw to TAG,
5161 should one occur. We call it like this:
5162 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
5164 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
5165 same idea as BODY_DATA above.
5166 THROWN_TAG is the tag that the user threw to; usually this is
5167 TAG, but it could be something else if TAG was #t (i.e., a
5168 catch-all), or the user threw to a jmpbuf.
5169 THROW_ARGS is the list of arguments the user passed to the THROW
5172 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
5173 is just a pointer we pass through to HANDLER. We don't actually
5174 use either of those pointers otherwise ourselves. The idea is
5175 that, if our caller wants to communicate something to BODY or
5176 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
5177 HANDLER can then use. Think of it as a way to make BODY and
5178 HANDLER closures, not just functions; MUMBLE_DATA points to the
5181 Of course, it's up to the caller to make sure that any data a
5182 MUMBLE_DATA needs is protected from GC. A common way to do this is
5183 to make MUMBLE_DATA a pointer to data stored in an automatic
5184 structure variable; since the collector must scan the stack for
5185 references anyway, this assures that any references in MUMBLE_DATA
5188 ** The new function scm_internal_lazy_catch is exactly like
5189 scm_internal_catch, except:
5191 - It does not unwind the stack (this is the major difference).
5192 - If handler returns, its value is returned from the throw.
5193 - BODY always receives #f as its JMPBUF argument (since there's no
5194 jmpbuf associated with a lazy catch, because we don't unwind the
5197 ** scm_body_thunk is a new body function you can pass to
5198 scm_internal_catch if you want the body to be like Scheme's `catch'
5199 --- a thunk, or a function of one argument if the tag is #f.
5201 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
5202 contains the Scheme procedure to invoke as the body, and the tag
5203 we're catching. If the tag is #f, then we pass JMPBUF (created by
5204 scm_internal_catch) to the body procedure; otherwise, the body gets
5207 ** scm_handle_by_proc is a new handler function you can pass to
5208 scm_internal_catch if you want the handler to act like Scheme's catch
5209 --- call a procedure with the tag and the throw arguments.
5211 If the user does a throw to this catch, this function runs a handler
5212 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
5213 variable holding the Scheme procedure object to invoke. It ought to
5214 be a pointer to an automatic variable (i.e., one living on the stack),
5215 or the procedure object should be otherwise protected from GC.
5217 ** scm_handle_by_message is a new handler function to use with
5218 `scm_internal_catch' if you want Guile to print a message and die.
5219 It's useful for dealing with throws to uncaught keys at the top level.
5221 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
5222 message header to print; if zero, we use "guile" instead. That
5223 text is followed by a colon, then the message described by ARGS.
5225 ** The return type of scm_boot_guile is now void; the function does
5226 not return a value, and indeed, never returns at all.
5228 ** The new function scm_shell makes it easy for user applications to
5229 process command-line arguments in a way that is compatible with the
5230 stand-alone guile interpreter (which is in turn compatible with SCSH,
5233 To use the scm_shell function, first initialize any guile modules
5234 linked into your application, and then call scm_shell with the values
5235 of ARGC and ARGV your `main' function received. scm_shell will add
5236 any SCSH-style meta-arguments from the top of the script file to the
5237 argument vector, and then process the command-line arguments. This
5238 generally means loading a script file or starting up an interactive
5239 command interpreter. For details, see "Changes to the stand-alone
5242 ** The new functions scm_get_meta_args and scm_count_argv help you
5243 implement the SCSH-style meta-argument, `\'.
5245 char **scm_get_meta_args (int ARGC, char **ARGV)
5246 If the second element of ARGV is a string consisting of a single
5247 backslash character (i.e. "\\" in Scheme notation), open the file
5248 named by the following argument, parse arguments from it, and return
5249 the spliced command line. The returned array is terminated by a
5252 For details of argument parsing, see above, under "guile now accepts
5253 command-line arguments compatible with SCSH..."
5255 int scm_count_argv (char **ARGV)
5256 Count the arguments in ARGV, assuming it is terminated by a null
5259 For an example of how these functions might be used, see the source
5260 code for the function scm_shell in libguile/script.c.
5262 You will usually want to use scm_shell instead of calling this
5265 ** The new function scm_compile_shell_switches turns an array of
5266 command-line arguments into Scheme code to carry out the actions they
5267 describe. Given ARGC and ARGV, it returns a Scheme expression to
5268 evaluate, and calls scm_set_program_arguments to make any remaining
5269 command-line arguments available to the Scheme code. For example,
5270 given the following arguments:
5272 -e main -s ekko a speckled gecko
5274 scm_set_program_arguments will return the following expression:
5276 (begin (load "ekko") (main (command-line)) (quit))
5278 You will usually want to use scm_shell instead of calling this
5281 ** The function scm_shell_usage prints a usage message appropriate for
5282 an interpreter that uses scm_compile_shell_switches to handle its
5283 command-line arguments.
5285 void scm_shell_usage (int FATAL, char *MESSAGE)
5286 Print a usage message to the standard error output. If MESSAGE is
5287 non-zero, write it before the usage message, followed by a newline.
5288 If FATAL is non-zero, exit the process, using FATAL as the
5289 termination status. (If you want to be compatible with Guile,
5290 always use 1 as the exit status when terminating due to command-line
5293 You will usually want to use scm_shell instead of calling this
5296 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
5297 expressions. It used to return SCM_EOL. Earth-shattering.
5299 ** The macros for declaring scheme objects in C code have been
5300 rearranged slightly. They are now:
5302 SCM_SYMBOL (C_NAME, SCHEME_NAME)
5303 Declare a static SCM variable named C_NAME, and initialize it to
5304 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
5305 be a C identifier, and SCHEME_NAME should be a C string.
5307 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
5308 Just like SCM_SYMBOL, but make C_NAME globally visible.
5310 SCM_VCELL (C_NAME, SCHEME_NAME)
5311 Create a global variable at the Scheme level named SCHEME_NAME.
5312 Declare a static SCM variable named C_NAME, and initialize it to
5313 point to the Scheme variable's value cell.
5315 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
5316 Just like SCM_VCELL, but make C_NAME globally visible.
5318 The `guile-snarf' script writes initialization code for these macros
5319 to its standard output, given C source code as input.
5321 The SCM_GLOBAL macro is gone.
5323 ** The scm_read_line and scm_read_line_x functions have been replaced
5324 by Scheme code based on the %read-delimited! procedure (known to C
5325 code as scm_read_delimited_x). See its description above for more
5328 ** The function scm_sys_open has been renamed to scm_open. It now
5329 returns a port instead of an FD object.
5331 * The dynamic linking support has changed. For more information, see
5332 libguile/DYNAMIC-LINKING.
5337 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
5340 * Changes to the 'guile' program:
5342 ** Guile now loads some new files when it starts up. Guile first
5343 searches the load path for init.scm, and loads it if found. Then, if
5344 Guile is not being used to execute a script, and the user's home
5345 directory contains a file named `.guile', Guile loads that.
5347 ** You can now use Guile as a shell script interpreter.
5349 To paraphrase the SCSH manual:
5351 When Unix tries to execute an executable file whose first two
5352 characters are the `#!', it treats the file not as machine code to
5353 be directly executed by the native processor, but as source code
5354 to be executed by some interpreter. The interpreter to use is
5355 specified immediately after the #! sequence on the first line of
5356 the source file. The kernel reads in the name of the interpreter,
5357 and executes that instead. It passes the interpreter the source
5358 filename as its first argument, with the original arguments
5359 following. Consult the Unix man page for the `exec' system call
5360 for more information.
5362 Now you can use Guile as an interpreter, using a mechanism which is a
5363 compatible subset of that provided by SCSH.
5365 Guile now recognizes a '-s' command line switch, whose argument is the
5366 name of a file of Scheme code to load. It also treats the two
5367 characters `#!' as the start of a comment, terminated by `!#'. Thus,
5368 to make a file of Scheme code directly executable by Unix, insert the
5369 following two lines at the top of the file:
5371 #!/usr/local/bin/guile -s
5374 Guile treats the argument of the `-s' command-line switch as the name
5375 of a file of Scheme code to load, and treats the sequence `#!' as the
5376 start of a block comment, terminated by `!#'.
5378 For example, here's a version of 'echo' written in Scheme:
5380 #!/usr/local/bin/guile -s
5382 (let loop ((args (cdr (program-arguments))))
5385 (display (car args))
5386 (if (pair? (cdr args))
5388 (loop (cdr args)))))
5391 Why does `#!' start a block comment terminated by `!#', instead of the
5392 end of the line? That is the notation SCSH uses, and although we
5393 don't yet support the other SCSH features that motivate that choice,
5394 we would like to be backward-compatible with any existing Guile
5395 scripts once we do. Furthermore, if the path to Guile on your system
5396 is too long for your kernel, you can start the script with this
5400 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
5403 Note that some very old Unix systems don't support the `#!' syntax.
5406 ** You can now run Guile without installing it.
5408 Previous versions of the interactive Guile interpreter (`guile')
5409 couldn't start up unless Guile's Scheme library had been installed;
5410 they used the value of the environment variable `SCHEME_LOAD_PATH'
5411 later on in the startup process, but not to find the startup code
5412 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
5415 To run Guile without installing it, build it in the normal way, and
5416 then set the environment variable `SCHEME_LOAD_PATH' to a
5417 colon-separated list of directories, including the top-level directory
5418 of the Guile sources. For example, if you unpacked Guile so that the
5419 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
5422 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
5425 ** Guile's read-eval-print loop no longer prints #<unspecified>
5426 results. If the user wants to see this, she can evaluate the
5427 expression (assert-repl-print-unspecified #t), perhaps in her startup
5430 ** Guile no longer shows backtraces by default when an error occurs;
5431 however, it does display a message saying how to get one, and how to
5432 request that they be displayed by default. After an error, evaluate
5434 to see a backtrace, and
5435 (debug-enable 'backtrace)
5436 to see them by default.
5440 * Changes to Guile Scheme:
5442 ** Guile now distinguishes between #f and the empty list.
5444 This is for compatibility with the IEEE standard, the (possibly)
5445 upcoming Revised^5 Report on Scheme, and many extant Scheme
5448 Guile used to have #f and '() denote the same object, to make Scheme's
5449 type system more compatible with Emacs Lisp's. However, the change
5450 caused too much trouble for Scheme programmers, and we found another
5451 way to reconcile Emacs Lisp with Scheme that didn't require this.
5454 ** Guile's delq, delv, delete functions, and their destructive
5455 counterparts, delq!, delv!, and delete!, now remove all matching
5456 elements from the list, not just the first. This matches the behavior
5457 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
5458 functions which inspired them.
5460 I recognize that this change may break code in subtle ways, but it
5461 seems best to make the change before the FSF's first Guile release,
5465 ** The compiled-library-path function has been deleted from libguile.
5467 ** The facilities for loading Scheme source files have changed.
5469 *** The variable %load-path now tells Guile which directories to search
5470 for Scheme code. Its value is a list of strings, each of which names
5473 *** The variable %load-extensions now tells Guile which extensions to
5474 try appending to a filename when searching the load path. Its value
5475 is a list of strings. Its default value is ("" ".scm").
5477 *** (%search-load-path FILENAME) searches the directories listed in the
5478 value of the %load-path variable for a Scheme file named FILENAME,
5479 with all the extensions listed in %load-extensions. If it finds a
5480 match, then it returns its full filename. If FILENAME is absolute, it
5481 returns it unchanged. Otherwise, it returns #f.
5483 %search-load-path will not return matches that refer to directories.
5485 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
5486 uses %seach-load-path to find a file named FILENAME, and loads it if
5487 it finds it. If it can't read FILENAME for any reason, it throws an
5490 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
5493 *** load uses the same searching semantics as primitive-load.
5495 *** The functions %try-load, try-load-with-path, %load, load-with-path,
5496 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
5497 path, and load-module-with-path have been deleted. The functions
5498 above should serve their purposes.
5500 *** If the value of the variable %load-hook is a procedure,
5501 `primitive-load' applies its value to the name of the file being
5502 loaded (without the load path directory name prepended). If its value
5503 is #f, it is ignored. Otherwise, an error occurs.
5505 This is mostly useful for printing load notification messages.
5508 ** The function `eval!' is no longer accessible from the scheme level.
5509 We can't allow operations which introduce glocs into the scheme level,
5510 because Guile's type system can't handle these as data. Use `eval' or
5511 `read-and-eval!' (see below) as replacement.
5513 ** The new function read-and-eval! reads an expression from PORT,
5514 evaluates it, and returns the result. This is more efficient than
5515 simply calling `read' and `eval', since it is not necessary to make a
5516 copy of the expression for the evaluator to munge.
5518 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
5519 for the `read' function.
5522 ** The function `int?' has been removed; its definition was identical
5523 to that of `integer?'.
5525 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
5526 use the R4RS names for these functions.
5528 ** The function object-properties no longer returns the hash handle;
5529 it simply returns the object's property list.
5531 ** Many functions have been changed to throw errors, instead of
5532 returning #f on failure. The point of providing exception handling in
5533 the language is to simplify the logic of user code, but this is less
5534 useful if Guile's primitives don't throw exceptions.
5536 ** The function `fileno' has been renamed from `%fileno'.
5538 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
5541 * Changes to Guile's C interface:
5543 ** The library's initialization procedure has been simplified.
5544 scm_boot_guile now has the prototype:
5546 void scm_boot_guile (int ARGC,
5548 void (*main_func) (),
5551 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
5552 MAIN_FUNC should do all the work of the program (initializing other
5553 packages, reading user input, etc.) before returning. When MAIN_FUNC
5554 returns, call exit (0); this function never returns. If you want some
5555 other exit value, MAIN_FUNC may call exit itself.
5557 scm_boot_guile arranges for program-arguments to return the strings
5558 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
5559 scm_set_program_arguments with the final list, so Scheme code will
5560 know which arguments have been processed.
5562 scm_boot_guile establishes a catch-all catch handler which prints an
5563 error message and exits the process. This means that Guile exits in a
5564 coherent way when system errors occur and the user isn't prepared to
5565 handle it. If the user doesn't like this behavior, they can establish
5566 their own universal catcher in MAIN_FUNC to shadow this one.
5568 Why must the caller do all the real work from MAIN_FUNC? The garbage
5569 collector assumes that all local variables of type SCM will be above
5570 scm_boot_guile's stack frame on the stack. If you try to manipulate
5571 SCM values after this function returns, it's the luck of the draw
5572 whether the GC will be able to find the objects you allocate. So,
5573 scm_boot_guile function exits, rather than returning, to discourage
5574 people from making that mistake.
5576 The IN, OUT, and ERR arguments were removed; there are other
5577 convenient ways to override these when desired.
5579 The RESULT argument was deleted; this function should never return.
5581 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
5585 ** Guile's header files should no longer conflict with your system's
5588 In order to compile code which #included <libguile.h>, previous
5589 versions of Guile required you to add a directory containing all the
5590 Guile header files to your #include path. This was a problem, since
5591 Guile's header files have names which conflict with many systems'
5594 Now only <libguile.h> need appear in your #include path; you must
5595 refer to all Guile's other header files as <libguile/mumble.h>.
5596 Guile's installation procedure puts libguile.h in $(includedir), and
5597 the rest in $(includedir)/libguile.
5600 ** Two new C functions, scm_protect_object and scm_unprotect_object,
5601 have been added to the Guile library.
5603 scm_protect_object (OBJ) protects OBJ from the garbage collector.
5604 OBJ will not be freed, even if all other references are dropped,
5605 until someone does scm_unprotect_object (OBJ). Both functions
5608 Note that calls to scm_protect_object do not nest. You can call
5609 scm_protect_object any number of times on a given object, and the
5610 next call to scm_unprotect_object will unprotect it completely.
5612 Basically, scm_protect_object and scm_unprotect_object just
5613 maintain a list of references to things. Since the GC knows about
5614 this list, all objects it mentions stay alive. scm_protect_object
5615 adds its argument to the list; scm_unprotect_object remove its
5616 argument from the list.
5619 ** scm_eval_0str now returns the value of the last expression
5622 ** The new function scm_read_0str reads an s-expression from a
5623 null-terminated string, and returns it.
5625 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
5626 to a Scheme port object.
5628 ** The new function `scm_set_program_arguments' allows C code to set
5629 the value returned by the Scheme `program-arguments' function.
5634 * Guile no longer includes sophisticated Tcl/Tk support.
5636 The old Tcl/Tk support was unsatisfying to us, because it required the
5637 user to link against the Tcl library, as well as Tk and Guile. The
5638 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
5639 referring to widgets by names, rather than exporting widgets to Scheme
5640 code as a special datatype.
5642 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
5643 maintainers described some very interesting changes in progress to the
5644 Tcl/Tk internals, which would facilitate clean interfaces between lone
5645 Tk and other interpreters --- even for garbage-collected languages
5646 like Scheme. They expected the new Tk to be publicly available in the
5649 Since it seems that Guile might soon have a new, cleaner interface to
5650 lone Tk, and that the old Guile/Tk glue code would probably need to be
5651 completely rewritten, we (Jim Blandy and Richard Stallman) have
5652 decided not to support the old code. We'll spend the time instead on
5653 a good interface to the newer Tk, as soon as it is available.
5655 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
5658 Copyright information:
5660 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
5662 Permission is granted to anyone to make or distribute verbatim copies
5663 of this document as received, in any medium, provided that the
5664 copyright notice and this permission notice are preserved,
5665 thus giving the recipient permission to redistribute in turn.
5667 Permission is granted to distribute modified versions
5668 of this document, or of portions of it,
5669 under the above conditions, provided also that they
5670 carry prominent notices stating who last changed them.
5675 paragraph-separate: "[
\f]*$"