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 ** As per RELEASE directions, deprecated items have been removed
15 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
16 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP
22 gc-thunk - replaced by after-gc-hook.
24 ** New SRFI modules have been added:
26 (srfi srfi-2) exports and-let*.
28 (srfi srfi-6) is a dummy module for now, since guile already provides
29 all of the srfi-6 procedures by default: open-input-string,
30 open-output-string, get-output-string.
32 (srfi srfi-8) exports receive.
34 (srfi srfi-9) exports define-record-type.
36 (srfi srfi-11) exports let-values and let*-values.
38 (srfi srfi-13) implements the SRFI String Library.
40 (srfi srfi-14) implements the SRFI Character-Set Library.
42 ** New module (ice-9 stack-catch):
44 stack-catch is like catch, but saves the current state of the stack in
45 the fluid the-last-stack. This fluid can be useful when using the
46 debugger and when re-throwing an error.
48 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
50 This has been done to prevent problems on lesser operating systems
51 that can't tolerate `*'s in file names. The exported macro continues
52 to be named `and-let*', of course.
54 On systems that support it, there is also a compatibility module named
55 (ice-9 and-let*). It will go away in the next release.
57 ** New modules (oop goops) etc.:
62 (oop goops active-slot)
63 (oop goops composite-slot)
65 The Guile Object Oriented Programming System (GOOPS) has been
66 integrated into Guile.
70 (use-modules (oop goops))
72 access GOOPS bindings.
74 We're now ready to try some basic GOOPS functionality.
78 (define-method (+ (x <string>) (y <string>))
82 (+ "abc" "de") --> "abcde"
86 (define-class <2D-vector> ()
87 (x #:init-value 0 #:accessor x-component #:init-keyword #:x)
88 (y #:init-value 0 #:accessor y-component #:init-keyword #:y))
90 (define-method write ((obj <2D-vector>) port)
91 (display (format #f "<~S, ~S>" (x-component obj) (y-component obj))
94 (define v (make <2D-vector> #:x 3 #:y 4))
97 (define-method + ((x <2D-vector>) (y <2D-vector>))
99 #:x (+ (x-component x) (x-component y))
100 #:y (+ (y-component x) (y-component y))))
104 Asking for the type of an object
106 (class-of v) --> #<<class> <2D-vector> 40241ac0>
107 <2D-vector> --> #<<class> <2D-vector> 40241ac0>
108 (class-of 1) --> #<<class> <integer> 401b2a98>
109 <integer> --> #<<class> <integer> 401b2a98>
111 (is-a? v <2D-vector>) --> #t
113 See further in the GOOPS manual and tutorial in the `doc' directory,
114 in info (goops.info) and texinfo formats.
116 ** New module (ice-9 rdelim).
118 This exports the following procedures which were previously defined
119 in the default environment:
121 read-line read-line! read-delimited read-delimited! %read-delimited!
122 %read-line write-line
124 For backwards compatibility the definitions are still imported into the
125 default environment in this version of Guile. However you should add:
127 (use-modules (ice-9 rdelim))
129 to any program which uses the definitions, since this may change in
132 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
133 can be used for similar functionality.
135 ** New module (ice-9 match)
137 This module includes Andrew K. Wright's pattern matcher:
139 (use-modules (ice-9 match))
143 (('+ x y) `(add ,x ,y))
144 (('- x y) `(sub ,x ,y))) => (add 1 2)
146 See ice-9/match.scm for brief description or
147 http://www.star-lab.com/wright/code.html for complete documentation.
149 This module requires SLIB to be installed and available from Guile.
151 ** New module (ice-9 buffered-input)
153 This module provides procedures to construct an input port from an
154 underlying source of input that reads and returns its input in chunks.
155 The underlying input source is a Scheme procedure, specified by the
156 caller, which the port invokes whenever it needs more input.
158 This is useful when building an input port whose back end is Readline
159 or a UI element such as the GtkEntry widget.
163 The reference and tutorial documentation that was previously
164 distributed separately, as `guile-doc', is now included in the core
165 Guile distribution. The documentation consists of the following
168 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
171 - The Guile Reference Manual (guile.texi) contains (or is intended to
172 contain) reference documentation on all aspects of Guile.
174 - The GOOPS Manual (goops.texi) contains both tutorial-style and
175 reference documentation for using GOOPS, Guile's Object Oriented
178 - The Revised^5 Report on the Algorithmic Language Scheme
181 See the README file in the `doc' directory for more details.
183 * Changes to the stand-alone interpreter
185 ** Evaluation of "()", the empty list, is now an error.
187 Previously, you could for example write (cons 1 ()); now you need to
188 be more explicit and write (cons 1 '()).
190 ** It's now possible to create modules with controlled environments
194 (use-modules (ice-9 safe))
195 (define m (make-safe-module))
196 ;;; m will now be a module containing only a safe subset of R5RS
197 (eval '(+ 1 2) m) --> 3
198 (eval 'load m) --> ERROR: Unbound variable: load
200 * Changes to Scheme functions and syntax
202 ** The empty combination is no longer valid syntax.
204 Previously, the expression "()" evaluated to the empty list. This has
205 been changed to signal a "missing expression" error. The correct way
206 to write the empty list as a literal constant is to use quote: "'()".
208 ** Auto-loading of compiled-code modules is deprecated.
210 Guile used to be able to automatically find and link a shared
211 libraries to satisfy requests for a module. For example, the module
212 `(foo bar)' could be implemented by placing a shared library named
213 "foo/libbar.so" (or with a different extension) in a directory on the
216 This has been found to be too tricky, and is no longer supported.
217 What you should do instead now is to write a small Scheme file that
218 explicitly calls `dynamic-link' to load the shared library and
219 `dynamic-call' to initialize it.
221 The shared libraries themselves should be installed in the usual
222 places for shared libraries, with names like "libguile-foo-bar".
224 For example, place this into a file "foo/bar.scm"
226 (define-module (foo bar))
228 (dynamic-call "foobar_init" (dynamic-link "libguile-foo-bar"))
230 The file name passed to `dynamic-link' should not contain an
231 extension. It will be provided automatically.
233 ** The module system has been made more disciplined.
235 The function `eval' will now save and restore the current module
236 around the evaluation of the specified expression. While this
237 expression is evaluated, `(current-module)' will now return the right
238 module, which is the module specified as the second argument to
241 A consequence of this change is that `eval' is not particularily
242 useful when you want allow the evaluated code to change what module is
243 designated as the current module and have this change persist from one
244 call to `eval' to the next. The read-eval-print-loop is an example
245 where `eval' is now inadequate. To compensate, there is a new
246 function `primitive-eval' that does not take a module specifier and
247 that does not save/restore the current module. You should use this
248 function together with `set-current-module', `current-module', etc
249 when you want to have more control over the state that is carried from
250 one eval to the next.
252 Additionally, it has been made sure that forms that are evaluated at
253 the top level are always evaluated with respect to the current module.
254 Previously, subforms of top-level forms such as `begin', `case',
255 etc. did not respect changes to the current module although these
256 subforms are at the top-level as well.
258 To prevent strange behaviour, the forms `define-module',
259 `use-modules', `use-syntax', and `export' have been restricted to only
260 work on the top level. The forms `define-public' and
261 `defmacro-public' only export the new binding on the top level. They
262 behave just like `define' and `defmacro', respectively, when they are
263 used in a lexical environment.
265 ** The semantics of guardians have changed.
267 The changes are for the most part compatible. An important criterion
268 was to keep the typical usage of guardians as simple as before, but to
269 make the semantics safer and (as a result) more useful.
271 *** All objects returned from guardians are now properly alive.
273 It is now guaranteed that any object referenced by an object returned
274 from a guardian is alive. It's now impossible for a guardian to
275 return a "contained" object before its "containing" object.
277 One incompatible (but probably not very important) change resulting
278 from this is that it is no longer possible to guard objects that
279 indirectly reference themselves (i.e. are parts of cycles). If you do
280 so accidentally, you'll get a warning.
282 *** There are now two types of guardians: greedy and sharing.
284 If you call (make-guardian #t) or just (make-guardian), you'll get a
285 greedy guardian, and for (make-guardian #f) a sharing guardian.
287 Greedy guardians are the default because they are more "defensive".
288 You can only greedily guard an object once. If you guard an object
289 more than once, once in a greedy guardian and the rest of times in
290 sharing guardians, then it is guaranteed that the object won't be
291 returned from sharing guardians as long as it is greedily guarded
294 Guardians returned by calls to `make-guardian' can now take one more
295 optional parameter, which says whether to throw an error in case an
296 attempt is made to greedily guard an object that is already greedily
297 guarded. The default is true, i.e. throw an error. If the parameter
298 is false, the guardian invocation returns #t if guarding was
299 successful and #f if it wasn't.
301 Also, since greedy guarding is, in effect, a side-effecting operation
302 on objects, a new function is introduced: `destroy-guardian!'.
303 Invoking this function on a guardian renders it unoperative and, if
304 the guardian is greedy, clears the "greedily guarded" property of the
305 objects that were guarded by it, thus undoing the side effect.
307 Note that all this hair is hardly very important, since guardian
308 objects are usually permanent.
310 ** Escape procedures created by call-with-current-continuation now
311 accept any number of arguments, as required by R5RS.
313 ** New function `call-with-deprecation'
315 Call a thunk, displaying a deprecation message at the first call:
318 (call-with-deprecation "`id' is deprecated. Use `identity' instead."
323 ;; `id' is deprecated. Use `identity' instead.
328 ** New function `make-object-property'
330 This function returns a new `procedure with setter' P that can be used
331 to attach a property to objects. When calling P as
335 where `obj' is any kind of object, it attaches `val' to `obj' in such
336 a way that it can be retrieved by calling P as
340 This function will replace procedure properties, symbol properties and
341 source properties eventually.
343 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
345 Instead of #&optional, #&key, etc you should now use #:optional,
346 #:key, etc. Since #:optional is a keyword, you can write it as just
347 :optional when (read-set! keywords 'prefix) is active.
349 The old reader syntax `#&' is still supported, but deprecated. It
350 will be removed in the next release.
352 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
354 `eval' is now R5RS, that is it takes two arguments.
355 The second argument is an environment specifier, i.e. either
357 (scheme-report-environment 5)
359 (interaction-environment)
365 ** New define-module option: pure
367 Tells the module system not to include any bindings from the root
372 (define-module (totally-empty-module)
375 ** New define-module option: export NAME1 ...
377 Export names NAME1 ...
379 This option is required if you want to be able to export bindings from
380 a module which doesn't import one of `define-public' or `export'.
386 :use-module (ice-9 r5rs)
389 ;;; Note that we're pure R5RS below this point!
394 ** Deprecated: scm_make_shared_substring
396 Explicit shared substrings will disappear from Guile.
398 Instead, "normal" strings will be implemented using sharing
399 internally, combined with a copy-on-write strategy.
401 ** Deprecated: scm_read_only_string_p
403 The concept of read-only strings will disappear in next release of
406 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
408 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
410 ** New function: read-string!/partial str [port_or_fdes [start [end]]]
412 Read characters from an fport or file descriptor into a string
413 STR. This procedure is scsh-compatible and can efficiently read
414 large strings. It will:
416 * attempt to fill the entire string, unless the START and/or
417 END arguments are supplied. i.e., START defaults to 0 and
418 END defaults to `(string-length str)'
420 * use the current input port if PORT_OR_FDES is not supplied.
422 * read any characters that are currently available, without
423 waiting for the rest (short reads are possible).
425 * wait for as long as it needs to for the first character to
426 become available, unless the port is in non-blocking mode
428 * return `#f' if end-of-file is encountered before reading any
429 characters, otherwise return the number of characters read.
431 * return 0 if the port is in non-blocking mode and no characters
432 are immediately available.
434 * return 0 if the request is for 0 bytes, with no end-of-file
437 ** New function: object->string OBJ
439 Return a Scheme string obtained by printing a given object.
441 ** New function: port? X
443 Returns a boolean indicating whether X is a port. Equivalent to
444 `(or (input-port? X) (output-port? X))'.
446 ** New function: file-port?
448 Determines whether a given object is a port that is related to a file.
450 ** New function: port-for-each proc
452 Apply PROC to each port in the Guile port table in turn. The
453 return value is unspecified. More specifically, PROC is applied
454 exactly once to every port that exists in the system at the time
455 PORT-FOR-EACH is invoked. Changes to the port table while
456 PORT-FOR-EACH is running have no effect as far as PORT-FOR-EACH is
459 ** New function: dup2 oldfd newfd
461 A simple wrapper for the `dup2' system call. Copies the file
462 descriptor OLDFD to descriptor number NEWFD, replacing the
463 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
464 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
465 to move away ports which are using NEWFD. The return value is
468 ** New function: close-fdes fd
470 A simple wrapper for the `close' system call. Close file
471 descriptor FD, which must be an integer. Unlike close (*note
472 close: Ports and File Descriptors.), the file descriptor will be
473 closed even if a port is using it. The return value is
476 ** New function: crypt password salt
478 Encrypts `password' using the standard unix password encryption
481 ** New function: chroot path
483 Change the root directory of the running process to `path'.
485 ** New functions: getlogin, cuserid
487 Return the login name or the user name of the current effective user
490 ** New functions: getpriority which who, setpriority which who prio
492 Get or set the priority of the running process.
494 ** New function: getpass prompt
496 Read a password from the terminal, first displaying `prompt' and
499 ** New function: flock file operation
501 Set/remove an advisory shared or exclusive lock on `file'.
503 ** New functions: sethostname name, gethostname
505 Set or get the hostname of the machine the current process is running
508 ** New function: mkstemp! tmpl
510 mkstemp creates a new unique file in the file system and returns a
511 new buffered port open for reading and writing to the file. TMPL
512 is a string specifying where the file should be created: it must
513 end with `XXXXXX' and will be changed in place to return the name
514 of the temporary file.
516 ** New function: open-input-string string
518 Return an input string port which delivers the characters from
519 `string'. This procedure, together with `open-output-string' and
520 `get-output-string' implements SRFI-6.
522 ** New function: open-output-string
524 Return an output string port which collects all data written to it.
525 The data can then be retrieved by `get-output-string'.
527 ** New function: get-output-string
529 Return the contents of an output string port.
531 ** New function: identity
535 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
536 are represented in Scheme as integers with normal host byte ordering.
538 ** New function: inet-pton family address
540 Convert a printable string network address into an integer. Note
541 that unlike the C version of this function, the result is an
542 integer with normal host byte ordering. FAMILY can be `AF_INET'
544 (inet-pton AF_INET "127.0.0.1") => 2130706433
545 (inet-pton AF_INET6 "::1") => 1
547 ** New function: inet-ntop family address
549 Convert an integer network address into a printable string. Note
550 that unlike the C version of this function, the input is an
551 integer with normal host byte ordering. FAMILY can be `AF_INET'
553 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
554 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
555 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
559 Use `identity' instead.
561 ** Deprecated: close-all-ports-except.
563 This was intended for closing ports in a child process after a fork,
564 but it has the undesirable side effect of flushing buffers.
565 port-for-each is more flexible.
567 ** The (ice-9 popen) module now attempts to set up file descriptors in
568 the child process from the current Scheme ports, instead of using the
569 current values of file descriptors 0, 1, and 2 in the parent process.
571 ** Removed function: builtin-weak-bindings
573 There is no such concept as a weak binding any more.
575 ** Removed constants: bignum-radix, scm-line-incrementors
577 ** define-method: New syntax mandatory.
579 The new method syntax is now mandatory:
581 (define-method (NAME ARG-SPEC ...) BODY ...)
582 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
584 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
585 REST-ARG ::= ARG-NAME
587 If you have old code using the old syntax, import
588 (oop goops old-define-method) before (oop goops) as in:
590 (use-modules (oop goops old-define-method) (oop goops))
592 * Changes to the gh_ interface
594 * Changes to the scm_ interface
596 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
598 Used by an application to read arbitrary number of bytes from a port.
599 Same semantics as libc read, except that scm_c_read only returns less
600 than SIZE bytes if at end-of-file.
602 Warning: Doesn't update port line and column counts!
604 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
606 Used by an application to write arbitrary number of bytes to an SCM
607 port. Similar semantics as libc write. However, unlike libc
608 write, scm_c_write writes the requested number of bytes and has no
611 Warning: Doesn't update port line and column counts!
613 ** New function: scm_init_guile ()
615 In contrast to scm_boot_guile, scm_init_guile will return normally
616 after initializing Guile. It is not available on all systems, tho.
618 ** New functions: scm_str2symbol, scm_mem2symbol
620 The function scm_str2symbol takes a const char* pointing to a zero-terminated
621 field of characters and creates a scheme symbol object from that C string.
622 The function scm_mem2symbol takes a const char* and a number of characters and
623 creates a symbol from the characters in that memory area.
625 ** New functions: scm_primitive_make_property
626 scm_primitive_property_ref
627 scm_primitive_property_set_x
628 scm_primitive_property_del_x
630 These functions implement a new way to deal with object properties.
631 See libguile/properties.c for their documentation.
633 ** New function: scm_done_free (long size)
635 This function is the inverse of scm_done_malloc. Use it to report the
636 amount of smob memory you free. The previous method, which involved
637 calling scm_done_malloc with negative argument, was somewhat
638 unintuitive (and is still available, of course).
640 ** New function: scm_c_memq (SCM obj, SCM list)
642 This function provides a fast C level alternative for scm_memq for the case
643 that the list parameter is known to be a proper list. The function is a
644 replacement for scm_sloppy_memq, but is stricter in its requirements on its
645 list input parameter, since for anything else but a proper list the function's
646 behaviour is undefined - it may even crash or loop endlessly. Further, for
647 the case that the object is not found in the list, scm_c_memq returns #f which
648 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
650 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
651 scm_remember_upto_here
653 These functions replace the function scm_remember.
655 ** Deprecated function: scm_remember
657 Use one of the new functions scm_remember_upto_here_1,
658 scm_remember_upto_here_2 or scm_remember_upto_here instead.
660 ** New function: scm_allocate_string
662 This function replaces the function scm_makstr.
664 ** Deprecated function: scm_makstr
666 Use the new function scm_allocate_string instead.
668 ** New global variable scm_gc_running_p introduced.
670 Use this variable to find out if garbage collection is being executed. Up to
671 now applications have used scm_gc_heap_lock to test if garbage collection was
672 running, which also works because of the fact that up to know only the garbage
673 collector has set this variable. But, this is an implementation detail that
674 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
675 of this variable is (and has been) not fully safe anyway.
677 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
679 Use these instead of SCM_LENGTH_MAX.
681 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
682 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
683 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
685 Use these instead of SCM_LENGTH.
687 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
688 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
689 SCM_SET_BITVECTOR_LENGTH
691 Use these instead of SCM_SETLENGTH
693 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
694 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
697 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
700 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
701 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
704 Use these instead of SCM_SETCHARS.
706 ** New macro: SCM_BITVECTOR_P
708 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
710 Use instead of SCM_COERCE_SUBSTR.
712 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
714 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
716 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
717 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
718 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
719 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
720 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
721 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
722 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
723 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
724 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
725 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
726 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
727 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
728 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
729 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
730 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
732 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
733 Use scm_memory_error instead of SCM_NALLOC.
734 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
735 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
736 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
737 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
738 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
739 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
740 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
741 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
742 Use SCM_STRINGP instead of SCM_RWSTRINGP.
743 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
744 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
745 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
746 Use a type specific setter macro instead of SCM_SETLENGTH.
747 Use a type specific setter macro instead of SCM_SETCHARS.
748 Use a type specific length macro instead of SCM_LENGTH_MAX.
749 Use SCM_GCMARKP instead of SCM_GC8MARKP.
750 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
751 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
752 Use SCM_TYP16 instead of SCM_GCTYP16.
753 Use SCM_CDR instead of SCM_GCCDR.
754 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
755 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
756 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
757 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
758 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
759 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
760 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
762 ** Removed function: scm_struct_init
764 ** Removed variable: scm_symhash_dim
766 ** Renamed function: scm_make_cont has been replaced by
767 scm_make_continuation, which has a different interface.
769 ** Deprecated function: scm_call_catching_errors
771 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
773 ** Deprecated function: scm_strhash
775 Use scm_string_hash instead.
777 ** Deprecated function: scm_vector_set_length_x
779 Instead, create a fresh vector of the desired size and copy the contents.
781 ** scm_gensym has changed prototype
783 scm_gensym now only takes one argument.
785 ** New function: scm_gentemp (SCM prefix, SCM obarray)
787 The builtin `gentemp' has now become a primitive.
789 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
792 There is now only a single symbol type scm_tc7_symbol.
793 The tag scm_tc7_lvector was not used anyway.
795 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
797 Use scm_make_smob_type and scm_set_smob_XXX instead.
799 ** New function scm_set_smob_apply.
801 This can be used to set an apply function to a smob type.
803 ** Deprecated function: scm_strprint_obj
805 Use scm_object_to_string instead.
807 ** Deprecated function: scm_wta
809 Use scm_wrong_type_arg, or another appropriate error signalling function
813 Changes since Guile 1.3.4:
815 * Changes to the distribution
817 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
819 We've changed the way we handle generated files in the Guile source
820 repository. As a result, the procedure for building trees obtained
821 from the nightly FTP snapshots or via CVS has changed:
822 - You must have appropriate versions of autoconf, automake, and
823 libtool installed on your system. See README for info on how to
824 obtain these programs.
825 - Before configuring the tree, you must first run the script
826 `autogen.sh' at the top of the source tree.
828 The Guile repository used to contain not only source files, written by
829 humans, but also some generated files, like configure scripts and
830 Makefile.in files. Even though the contents of these files could be
831 derived mechanically from other files present, we thought it would
832 make the tree easier to build if we checked them into CVS.
834 However, this approach means that minor differences between
835 developer's installed tools and habits affected the whole team.
836 So we have removed the generated files from the repository, and
837 added the autogen.sh script, which will reconstruct them
841 ** configure now has experimental options to remove support for certain
844 --disable-arrays omit array and uniform array support
845 --disable-posix omit posix interfaces
846 --disable-networking omit networking interfaces
847 --disable-regex omit regular expression interfaces
849 These are likely to become separate modules some day.
851 ** New configure option --enable-debug-freelist
853 This enables a debugging version of SCM_NEWCELL(), and also registers
854 an extra primitive, the setter `gc-set-debug-check-freelist!'.
856 Configure with the --enable-debug-freelist option to enable
857 the gc-set-debug-check-freelist! primitive, and then use:
859 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
860 (gc-set-debug-check-freelist! #f) # turn off checking
862 Checking of the freelist forces a traversal of the freelist and
863 a garbage collection before each allocation of a cell. This can
864 slow down the interpreter dramatically, so the setter should be used to
865 turn on this extra processing only when necessary.
867 ** New configure option --enable-debug-malloc
869 Include code for debugging of calls to scm_must_malloc/realloc/free.
873 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
874 2. objects reallocated by scm_must_realloc has been allocated by
876 3. reallocated objects are reallocated with the same what string
878 But, most importantly, it records the number of allocated objects of
879 each kind. This is useful when searching for memory leaks.
881 A Guile compiled with this option provides the primitive
882 `malloc-stats' which returns an alist with pairs of kind and the
883 number of objects of that kind.
885 ** All includes are now referenced relative to the root directory
887 Since some users have had problems with mixups between Guile and
888 system headers, we have decided to always refer to Guile headers via
889 their parent directories. This essentially creates a "private name
890 space" for Guile headers. This means that the compiler only is given
891 -I options for the root build and root source directory.
893 ** Header files kw.h and genio.h have been removed.
895 ** The module (ice-9 getopt-gnu-style) has been removed.
897 ** New module (ice-9 documentation)
899 Implements the interface to documentation strings associated with
902 ** New module (ice-9 time)
904 Provides a macro `time', which displays execution time of a given form.
906 ** New module (ice-9 history)
908 Loading this module enables value history in the repl.
910 * Changes to the stand-alone interpreter
912 ** New command line option --debug
914 Start Guile with debugging evaluator and backtraces enabled.
916 This is useful when debugging your .guile init file or scripts.
920 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
921 (help REGEXP) ditto for objects with names matching REGEXP (a string)
922 (help ,EXPR) gives documentation for object returned by EXPR
923 (help) gives this text
925 `help' searches among bindings exported from loaded modules, while
926 `apropos' searches among bindings visible from the "current" module.
928 Examples: (help help)
930 (help "output-string")
932 ** `help' and `apropos' now prints full module names
934 ** Dynamic linking now uses libltdl from the libtool package.
936 The old system dependent code for doing dynamic linking has been
937 replaced with calls to the libltdl functions which do all the hairy
940 The major improvement is that you can now directly pass libtool
941 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
942 will be able to do the best shared library job you can get, via
945 The way dynamic libraries are found has changed and is not really
946 portable across platforms, probably. It is therefore recommended to
947 use absolute filenames when possible.
949 If you pass a filename without an extension to `dynamic-link', it will
950 try a few appropriate ones. Thus, the most platform ignorant way is
951 to specify a name like "libfoo", without any directories and
954 ** Guile COOP threads are now compatible with LinuxThreads
956 Previously, COOP threading wasn't possible in applications linked with
957 Linux POSIX threads due to their use of the stack pointer to find the
958 thread context. This has now been fixed with a workaround which uses
959 the pthreads to allocate the stack.
961 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
963 ** Positions of erring expression in scripts
965 With version 1.3.4, the location of the erring expression in Guile
966 scipts is no longer automatically reported. (This should have been
967 documented before the 1.3.4 release.)
969 You can get this information by enabling recording of positions of
970 source expressions and running the debugging evaluator. Put this at
971 the top of your script (or in your "site" file):
973 (read-enable 'positions)
974 (debug-enable 'debug)
976 ** Backtraces in scripts
978 It is now possible to get backtraces in scripts.
982 (debug-enable 'debug 'backtrace)
984 at the top of the script.
986 (The first options enables the debugging evaluator.
987 The second enables backtraces.)
989 ** Part of module system symbol lookup now implemented in C
991 The eval closure of most modules is now implemented in C. Since this
992 was one of the bottlenecks for loading speed, Guile now loads code
993 substantially faster than before.
995 ** Attempting to get the value of an unbound variable now produces
996 an exception with a key of 'unbound-variable instead of 'misc-error.
998 ** The initial default output port is now unbuffered if it's using a
999 tty device. Previously in this situation it was line-buffered.
1001 ** New hook: after-gc-hook
1003 after-gc-hook takes over the role of gc-thunk. This hook is run at
1004 the first SCM_TICK after a GC. (Thus, the code is run at the same
1005 point during evaluation as signal handlers.)
1007 Note that this hook should be used only for diagnostic and debugging
1008 purposes. It is not certain that it will continue to be well-defined
1009 when this hook is run in the future.
1011 C programmers: Note the new C level hooks scm_before_gc_c_hook,
1012 scm_before_sweep_c_hook, scm_after_gc_c_hook.
1014 ** Improvements to garbage collector
1016 Guile 1.4 has a new policy for triggering heap allocation and
1017 determining the sizes of heap segments. It fixes a number of problems
1020 1. The new policy can handle two separate pools of cells
1021 (2-word/4-word) better. (The old policy would run wild, allocating
1022 more and more memory for certain programs.)
1024 2. The old code would sometimes allocate far too much heap so that the
1025 Guile process became gigantic. The new code avoids this.
1027 3. The old code would sometimes allocate too little so that few cells
1028 were freed at GC so that, in turn, too much time was spent in GC.
1030 4. The old code would often trigger heap allocation several times in a
1031 row. (The new scheme predicts how large the segments needs to be
1032 in order not to need further allocation.)
1034 All in all, the new GC policy will make larger applications more
1037 The new GC scheme also is prepared for POSIX threading. Threads can
1038 allocate private pools of cells ("clusters") with just a single
1039 function call. Allocation of single cells from such a cluster can
1040 then proceed without any need of inter-thread synchronization.
1042 ** New environment variables controlling GC parameters
1044 GUILE_MAX_SEGMENT_SIZE Maximal segment size
1047 Allocation of 2-word cell heaps:
1049 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
1052 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
1053 GC in percent of total heap size
1056 Allocation of 4-word cell heaps
1057 (used for real numbers and misc other objects):
1059 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
1061 (See entry "Way for application to customize GC parameters" under
1062 section "Changes to the scm_ interface" below.)
1064 ** Guile now implements reals using 4-word cells
1066 This speeds up computation with reals. (They were earlier allocated
1067 with `malloc'.) There is still some room for optimizations, however.
1069 ** Some further steps toward POSIX thread support have been taken
1071 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
1072 don't have much effect any longer, and many of them will be removed in
1076 are only handled at the top of the evaluator loop, immediately after
1077 I/O, and in scm_equalp.
1079 *** The GC can allocate thread private pools of pairs.
1081 * Changes to Scheme functions and syntax
1083 ** close-input-port and close-output-port are now R5RS
1085 These procedures have been turned into primitives and have R5RS behaviour.
1087 ** New procedure: simple-format PORT MESSAGE ARG1 ...
1089 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
1090 extended by the more sophisticated version in (ice-9 format)
1092 (simple-format port message . args)
1093 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
1094 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
1095 the escapes are replaced with corresponding members of ARGS:
1096 ~A formats using `display' and ~S formats using `write'.
1097 If DESTINATION is #t, then use the `current-output-port',
1098 if DESTINATION is #f, then return a string containing the formatted text.
1099 Does not add a trailing newline."
1101 ** string-ref: the second argument is no longer optional.
1103 ** string, list->string: no longer accept strings in their arguments,
1104 only characters, for compatibility with R5RS.
1106 ** New procedure: port-closed? PORT
1107 Returns #t if PORT is closed or #f if it is open.
1109 ** Deprecated: list*
1111 The list* functionality is now provided by cons* (SRFI-1 compliant)
1113 ** New procedure: cons* ARG1 ARG2 ... ARGn
1115 Like `list', but the last arg provides the tail of the constructed list,
1116 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
1118 Requires at least one argument. If given one argument, that argument
1119 is returned as result.
1121 This function is called `list*' in some other Schemes and in Common LISP.
1123 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
1125 ** New procedure: object-documentation OBJECT
1127 Returns the documentation string associated with OBJECT. The
1128 procedure uses a caching mechanism so that subsequent lookups are
1131 Exported by (ice-9 documentation).
1133 ** module-name now returns full names of modules
1135 Previously, only the last part of the name was returned (`session' for
1136 `(ice-9 session)'). Ex: `(ice-9 session)'.
1138 * Changes to the gh_ interface
1140 ** Deprecated: gh_int2scmb
1142 Use gh_bool2scm instead.
1144 * Changes to the scm_ interface
1146 ** Guile primitives now carry docstrings!
1148 Thanks to Greg Badros!
1150 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1152 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1153 macros and must contain a docstring that is extracted into foo.doc using a new
1154 guile-doc-snarf script (that uses guile-doc-snarf.awk).
1156 However, a major overhaul of these macros is scheduled for the next release of
1159 ** Guile primitives use a new technique for validation of arguments
1161 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
1162 the readability of argument checking.
1164 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
1166 ** New macros: SCM_PACK, SCM_UNPACK
1168 Compose/decompose an SCM value.
1170 The SCM type is now treated as an abstract data type and may be defined as a
1171 long, a void* or as a struct, depending on the architecture and compile time
1172 options. This makes it easier to find several types of bugs, for example when
1173 SCM values are treated as integers without conversion. Values of the SCM type
1174 should be treated as "atomic" values. These macros are used when
1175 composing/decomposing an SCM value, either because you want to access
1176 individual bits, or because you want to treat it as an integer value.
1178 E.g., in order to set bit 7 in an SCM value x, use the expression
1180 SCM_PACK (SCM_UNPACK (x) | 0x80)
1182 ** The name property of hooks is deprecated.
1183 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
1185 You can emulate this feature by using object properties.
1187 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
1188 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
1191 These macros will be removed in a future release of Guile.
1193 ** The following types, functions and macros from numbers.h are deprecated:
1194 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
1195 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
1197 Further, it is recommended not to rely on implementation details for guile's
1198 current implementation of bignums. It is planned to replace this
1199 implementation with gmp in the future.
1201 ** Port internals: the rw_random variable in the scm_port structure
1202 must be set to non-zero in any random access port. In recent Guile
1203 releases it was only set for bidirectional random-access ports.
1205 ** Port internals: the seek ptob procedure is now responsible for
1206 resetting the buffers if required. The change was made so that in the
1207 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
1208 the fport and strport ptobs can avoid resetting the buffers,
1209 in particular to avoid discarding unread chars. An existing port
1210 type can be fixed by adding something like the following to the
1211 beginning of the ptob seek procedure:
1213 if (pt->rw_active == SCM_PORT_READ)
1214 scm_end_input (object);
1215 else if (pt->rw_active == SCM_PORT_WRITE)
1216 ptob->flush (object);
1218 although to actually avoid resetting the buffers and discard unread
1219 chars requires further hacking that depends on the characteristics
1222 ** Deprecated functions: scm_fseek, scm_tag
1224 These functions are no longer used and will be removed in a future version.
1226 ** The scm_sysmissing procedure is no longer used in libguile.
1227 Unless it turns out to be unexpectedly useful to somebody, it will be
1228 removed in a future version.
1230 ** The format of error message strings has changed
1232 The two C procedures: scm_display_error and scm_error, as well as the
1233 primitive `scm-error', now use scm_simple_format to do their work.
1234 This means that the message strings of all code must be updated to use
1235 ~A where %s was used before, and ~S where %S was used before.
1237 During the period when there still are a lot of old Guiles out there,
1238 you might want to support both old and new versions of Guile.
1240 There are basically two methods to achieve this. Both methods use
1243 AC_CHECK_FUNCS(scm_simple_format)
1245 in your configure.in.
1247 Method 1: Use the string concatenation features of ANSI C's
1252 #ifdef HAVE_SCM_SIMPLE_FORMAT
1258 Then represent each of your error messages using a preprocessor macro:
1260 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
1264 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
1265 (define make-message string-append)
1267 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
1269 Method 2: Use the oldfmt function found in doc/oldfmt.c.
1273 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
1278 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
1282 ** Deprecated: coop_mutex_init, coop_condition_variable_init
1284 Don't use the functions coop_mutex_init and
1285 coop_condition_variable_init. They will change.
1287 Use scm_mutex_init and scm_cond_init instead.
1289 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
1290 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
1291 COND, as `scm_cond_wait' does, but it also bounds the duration
1292 of the wait. If COND has not been signaled before time ABSTIME,
1293 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
1294 returns the error code `ETIMEDOUT'.
1296 The ABSTIME parameter specifies an absolute time, with the same
1297 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
1298 to 00:00:00 GMT, January 1, 1970.
1300 ** New function: scm_cond_broadcast (scm_cond_t *COND)
1301 `scm_cond_broadcast' restarts all the threads that are waiting
1302 on the condition variable COND. Nothing happens if no threads are
1305 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
1306 `scm_key_create' allocates a new TSD key. The key is stored in
1307 the location pointed to by KEY. There is no limit on the number
1308 of keys allocated at a given time. The value initially associated
1309 with the returned key is `NULL' in all currently executing threads.
1311 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
1312 function associated with the key. When a thread terminates,
1313 DESTR_FUNCTION is called on the value associated with the key in
1314 that thread. The DESTR_FUNCTION is not called if a key is deleted
1315 with `scm_key_delete' or a value is changed with
1316 `scm_setspecific'. The order in which destructor functions are
1317 called at thread termination time is unspecified.
1319 Destructors are not yet implemented.
1321 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
1322 `scm_setspecific' changes the value associated with KEY in the
1323 calling thread, storing the given POINTER instead.
1325 ** New function: scm_getspecific (scm_key_t KEY)
1326 `scm_getspecific' returns the value currently associated with
1327 KEY in the calling thread.
1329 ** New function: scm_key_delete (scm_key_t KEY)
1330 `scm_key_delete' deallocates a TSD key. It does not check
1331 whether non-`NULL' values are associated with that key in the
1332 currently executing threads, nor call the destructor function
1333 associated with the key.
1335 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
1337 Initialize a C level hook HOOK with associated HOOK_DATA and type
1338 TYPE. (See scm_c_hook_run ().)
1340 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
1342 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
1343 is true, add it last, otherwise first. The same FUNC can be added
1344 multiple times if FUNC_DATA differ and vice versa.
1346 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
1348 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
1349 function is only removed if both FUNC and FUNC_DATA matches.
1351 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
1353 Run hook HOOK passing DATA to the hook functions.
1355 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
1356 returned is undefined.
1358 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
1359 returns a non-NULL value. This value is returned as the result of
1360 scm_c_hook_run. If all functions return NULL, NULL is returned.
1362 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
1363 returns a NULL value, and NULL is returned. If all functions returns
1364 a non-NULL value, the last value is returned.
1366 ** New C level GC hooks
1368 Five new C level hooks has been added to the garbage collector.
1370 scm_before_gc_c_hook
1373 are run before locking and after unlocking the heap. The system is
1374 thus in a mode where evaluation can take place. (Except that
1375 scm_before_gc_c_hook must not allocate new cells.)
1377 scm_before_mark_c_hook
1378 scm_before_sweep_c_hook
1379 scm_after_sweep_c_hook
1381 are run when the heap is locked. These are intended for extension of
1382 the GC in a modular fashion. Examples are the weaks and guardians
1385 ** Way for application to customize GC parameters
1387 The application can set up other default values for the GC heap
1388 allocation parameters
1390 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
1391 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
1392 GUILE_MAX_SEGMENT_SIZE,
1396 scm_default_init_heap_size_1, scm_default_min_yield_1,
1397 scm_default_init_heap_size_2, scm_default_min_yield_2,
1398 scm_default_max_segment_size
1400 respectively before callong scm_boot_guile.
1402 (See entry "New environment variables ..." in section
1403 "Changes to the stand-alone interpreter" above.)
1405 ** scm_protect_object/scm_unprotect_object now nest
1407 This means that you can call scm_protect_object multiple times on an
1408 object and count on the object being protected until
1409 scm_unprotect_object has been call the same number of times.
1411 The functions also have better time complexity.
1413 Still, it is usually possible to structure the application in a way
1414 that you don't need to use these functions. For example, if you use a
1415 protected standard Guile list to keep track of live objects rather
1416 than some custom data type, objects will die a natural death when they
1417 are no longer needed.
1419 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
1421 Guile does not provide the float representation for inexact real numbers any
1422 more. Now, only doubles are used to represent inexact real numbers. Further,
1423 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
1424 and scm_tc16_complex, respectively.
1426 ** Removed deprecated type scm_smobfuns
1428 ** Removed deprecated function scm_newsmob
1430 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
1432 There is an ongoing discussion among the developers whether to
1433 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
1434 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
1435 until this issue has been settled.
1437 ** Removed deprecated type tag scm_tc16_kw
1439 ** Added type tag scm_tc16_keyword
1441 (This was introduced already in release 1.3.4 but was not documented
1444 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
1446 * Changes to system call interfaces:
1448 ** The "select" procedure now tests port buffers for the ability to
1449 provide input or accept output. Previously only the underlying file
1450 descriptors were checked.
1452 ** New variable PIPE_BUF: the maximum number of bytes that can be
1453 atomically written to a pipe.
1455 ** If a facility is not available on the system when Guile is
1456 compiled, the corresponding primitive procedure will not be defined.
1457 Previously it would have been defined but would throw a system-error
1458 exception if called. Exception handlers which catch this case may
1459 need minor modification: an error will be thrown with key
1460 'unbound-variable instead of 'system-error. Alternatively it's
1461 now possible to use `defined?' to check whether the facility is
1464 ** Procedures which depend on the timezone should now give the correct
1465 result on systems which cache the TZ environment variable, even if TZ
1466 is changed without calling tzset.
1468 * Changes to the networking interfaces:
1470 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
1471 long integers between network and host format. For now, it's not
1472 particularly convenient to do this kind of thing, but consider:
1474 (define write-network-long
1475 (lambda (value port)
1476 (let ((v (make-uniform-vector 1 1 0)))
1477 (uniform-vector-set! v 0 (htonl value))
1478 (uniform-vector-write v port))))
1480 (define read-network-long
1482 (let ((v (make-uniform-vector 1 1 0)))
1483 (uniform-vector-read! v port)
1484 (ntohl (uniform-vector-ref v 0)))))
1486 ** If inet-aton fails, it now throws an error with key 'misc-error
1487 instead of 'system-error, since errno is not relevant.
1489 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
1490 specific keys instead of 'system-error. The latter is inappropriate
1491 since errno will not have been set. The keys are:
1492 'host-not-found, 'try-again, 'no-recovery and 'no-data.
1494 ** sethostent, setnetent, setprotoent, setservent: now take an
1495 optional argument STAYOPEN, which specifies whether the database
1496 remains open after a database entry is accessed randomly (e.g., using
1497 gethostbyname for the hosts database.) The default is #f. Previously
1501 Changes since Guile 1.3.2:
1503 * Changes to the stand-alone interpreter
1507 An initial version of the Guile debugger written by Chris Hanson has
1508 been added. The debugger is still under development but is included
1509 in the distribution anyway since it is already quite useful.
1515 after an error to enter the debugger. Type `help' inside the debugger
1516 for a description of available commands.
1518 If you prefer to have stack frames numbered and printed in
1519 anti-chronological order and prefer up in the stack to be down on the
1520 screen as is the case in gdb, you can put
1522 (debug-enable 'backwards)
1524 in your .guile startup file. (However, this means that Guile can't
1525 use indentation to indicate stack level.)
1527 The debugger is autoloaded into Guile at the first use.
1529 ** Further enhancements to backtraces
1531 There is a new debug option `width' which controls the maximum width
1532 on the screen of printed stack frames. Fancy printing parameters
1533 ("level" and "length" as in Common LISP) are adaptively adjusted for
1534 each stack frame to give maximum information while still fitting
1535 within the bounds. If the stack frame can't be made to fit by
1536 adjusting parameters, it is simply cut off at the end. This is marked
1539 ** Some modules are now only loaded when the repl is started
1541 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
1542 regex) are now loaded into (guile-user) only if the repl has been
1543 started. The effect is that the startup time for scripts has been
1544 reduced to 30% of what it was previously.
1546 Correctly written scripts load the modules they require at the top of
1547 the file and should not be affected by this change.
1549 ** Hooks are now represented as smobs
1551 * Changes to Scheme functions and syntax
1553 ** Readline support has changed again.
1555 The old (readline-activator) module is gone. Use (ice-9 readline)
1556 instead, which now contains all readline functionality. So the code
1557 to activate readline is now
1559 (use-modules (ice-9 readline))
1562 This should work at any time, including from the guile prompt.
1564 To avoid confusion about the terms of Guile's license, please only
1565 enable readline for your personal use; please don't make it the
1566 default for others. Here is why we make this rather odd-sounding
1569 Guile is normally licensed under a weakened form of the GNU General
1570 Public License, which allows you to link code with Guile without
1571 placing that code under the GPL. This exception is important to some
1574 However, since readline is distributed under the GNU General Public
1575 License, when you link Guile with readline, either statically or
1576 dynamically, you effectively change Guile's license to the strict GPL.
1577 Whenever you link any strictly GPL'd code into Guile, uses of Guile
1578 which are normally permitted become forbidden. This is a rather
1579 non-obvious consequence of the licensing terms.
1581 So, to make sure things remain clear, please let people choose for
1582 themselves whether to link GPL'd libraries like readline with Guile.
1584 ** regexp-substitute/global has changed slightly, but incompatibly.
1586 If you include a function in the item list, the string of the match
1587 object it receives is the same string passed to
1588 regexp-substitute/global, not some suffix of that string.
1589 Correspondingly, the match's positions are relative to the entire
1590 string, not the suffix.
1592 If the regexp can match the empty string, the way matches are chosen
1593 from the string has changed. regexp-substitute/global recognizes the
1594 same set of matches that list-matches does; see below.
1596 ** New function: list-matches REGEXP STRING [FLAGS]
1598 Return a list of match objects, one for every non-overlapping, maximal
1599 match of REGEXP in STRING. The matches appear in left-to-right order.
1600 list-matches only reports matches of the empty string if there are no
1601 other matches which begin on, end at, or include the empty match's
1604 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1606 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
1608 For each match of REGEXP in STRING, apply PROC to the match object,
1609 and the last value PROC returned, or INIT for the first call. Return
1610 the last value returned by PROC. We apply PROC to the matches as they
1611 appear from left to right.
1613 This function recognizes matches according to the same criteria as
1616 Thus, you could define list-matches like this:
1618 (define (list-matches regexp string . flags)
1619 (reverse! (apply fold-matches regexp string '() cons flags)))
1621 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1625 *** New function: hook? OBJ
1627 Return #t if OBJ is a hook, otherwise #f.
1629 *** New function: make-hook-with-name NAME [ARITY]
1631 Return a hook with name NAME and arity ARITY. The default value for
1632 ARITY is 0. The only effect of NAME is that it will appear when the
1633 hook object is printed to ease debugging.
1635 *** New function: hook-empty? HOOK
1637 Return #t if HOOK doesn't contain any procedures, otherwise #f.
1639 *** New function: hook->list HOOK
1641 Return a list of the procedures that are called when run-hook is
1644 ** `map' signals an error if its argument lists are not all the same length.
1646 This is the behavior required by R5RS, so this change is really a bug
1647 fix. But it seems to affect a lot of people's code, so we're
1648 mentioning it here anyway.
1650 ** Print-state handling has been made more transparent
1652 Under certain circumstances, ports are represented as a port with an
1653 associated print state. Earlier, this pair was represented as a pair
1654 (see "Some magic has been added to the printer" below). It is now
1655 indistinguishable (almost; see `get-print-state') from a port on the
1658 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
1660 Return a new port with the associated print state PRINT-STATE.
1662 *** New function: get-print-state OUTPUT-PORT
1664 Return the print state associated with this port if it exists,
1665 otherwise return #f.
1667 *** New function: directory-stream? OBJECT
1669 Returns true iff OBJECT is a directory stream --- the sort of object
1670 returned by `opendir'.
1672 ** New function: using-readline?
1674 Return #t if readline is in use in the current repl.
1676 ** structs will be removed in 1.4
1678 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
1679 and use GOOPS objects as the fundamental record type.
1681 * Changes to the scm_ interface
1683 ** structs will be removed in 1.4
1685 The entire current struct interface (struct.c, struct.h) will be
1686 replaced in Guile 1.4. We will merge GOOPS into libguile and use
1687 GOOPS objects as the fundamental record type.
1689 ** The internal representation of subr's has changed
1691 Instead of giving a hint to the subr name, the CAR field of the subr
1692 now contains an index to a subr entry in scm_subr_table.
1694 *** New variable: scm_subr_table
1696 An array of subr entries. A subr entry contains the name, properties
1697 and documentation associated with the subr. The properties and
1698 documentation slots are not yet used.
1700 ** A new scheme for "forwarding" calls to a builtin to a generic function
1702 It is now possible to extend the functionality of some Guile
1703 primitives by letting them defer a call to a GOOPS generic function on
1704 argument mismatch. This means that there is no loss of efficiency in
1709 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
1710 (define-method + ((x <string>) (y <string>))
1711 (string-append x y))
1713 + will still be as efficient as usual in numerical calculations, but
1714 can also be used for concatenating strings.
1716 Who will be the first one to extend Guile's numerical tower to
1717 rationals? :) [OK, there a few other things to fix before this can
1718 be made in a clean way.]
1720 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
1722 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
1724 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
1726 These do the same job as SCM_PROC and SCM_PROC1, but they also define
1727 a variable GENERIC which can be used by the dispatch macros below.
1729 [This is experimental code which may change soon.]
1731 *** New macros for forwarding control to a generic on arg type error
1733 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
1735 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
1737 These correspond to the scm_wta function call, and have the same
1738 behaviour until the user has called the GOOPS primitive
1739 `enable-primitive-generic!'. After that, these macros will apply the
1740 generic function GENERIC to the argument(s) instead of calling
1743 [This is experimental code which may change soon.]
1745 *** New macros for argument testing with generic dispatch
1747 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
1749 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
1751 These correspond to the SCM_ASSERT macro, but will defer control to
1752 GENERIC on error after `enable-primitive-generic!' has been called.
1754 [This is experimental code which may change soon.]
1756 ** New function: SCM scm_eval_body (SCM body, SCM env)
1758 Evaluates the body of a special form.
1760 ** The internal representation of struct's has changed
1762 Previously, four slots were allocated for the procedure(s) of entities
1763 and operators. The motivation for this representation had to do with
1764 the structure of the evaluator, the wish to support tail-recursive
1765 generic functions, and efficiency. Since the generic function
1766 dispatch mechanism has changed, there is no longer a need for such an
1767 expensive representation, and the representation has been simplified.
1769 This should not make any difference for most users.
1771 ** GOOPS support has been cleaned up.
1773 Some code has been moved from eval.c to objects.c and code in both of
1774 these compilation units has been cleaned up and better structured.
1776 *** New functions for applying generic functions
1778 New function: SCM scm_apply_generic (GENERIC, ARGS)
1779 New function: SCM scm_call_generic_0 (GENERIC)
1780 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
1781 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
1782 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
1784 ** Deprecated function: scm_make_named_hook
1786 It is now replaced by:
1788 ** New function: SCM scm_create_hook (const char *name, int arity)
1790 Creates a hook in the same way as make-hook above but also
1791 binds a variable named NAME to it.
1793 This is the typical way of creating a hook from C code.
1795 Currently, the variable is created in the "current" module.
1796 This might change when we get the new module system.
1798 [The behaviour is identical to scm_make_named_hook.]
1802 Changes since Guile 1.3:
1804 * Changes to mailing lists
1806 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
1808 See the README file to find current addresses for all the Guile
1811 * Changes to the distribution
1813 ** Readline support is no longer included with Guile by default.
1815 Based on the different license terms of Guile and Readline, we
1816 concluded that Guile should not *by default* cause the linking of
1817 Readline into an application program. Readline support is now offered
1818 as a separate module, which is linked into an application only when
1819 you explicitly specify it.
1821 Although Guile is GNU software, its distribution terms add a special
1822 exception to the usual GNU General Public License (GPL). Guile's
1823 license includes a clause that allows you to link Guile with non-free
1824 programs. We add this exception so as not to put Guile at a
1825 disadvantage vis-a-vis other extensibility packages that support other
1828 In contrast, the GNU Readline library is distributed under the GNU
1829 General Public License pure and simple. This means that you may not
1830 link Readline, even dynamically, into an application unless it is
1831 distributed under a free software license that is compatible the GPL.
1833 Because of this difference in distribution terms, an application that
1834 can use Guile may not be able to use Readline. Now users will be
1835 explicitly offered two independent decisions about the use of these
1838 You can activate the readline support by issuing
1840 (use-modules (readline-activator))
1843 from your ".guile" file, for example.
1845 * Changes to the stand-alone interpreter
1847 ** All builtins now print as primitives.
1848 Previously builtin procedures not belonging to the fundamental subr
1849 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
1850 Now, they print as #<primitive-procedure NAME>.
1852 ** Backtraces slightly more intelligible.
1853 gsubr-apply and macro transformer application frames no longer appear
1856 * Changes to Scheme functions and syntax
1858 ** Guile now correctly handles internal defines by rewriting them into
1859 their equivalent letrec. Previously, internal defines would
1860 incrementally add to the innermost environment, without checking
1861 whether the restrictions specified in RnRS were met. This lead to the
1862 correct behaviour when these restriction actually were met, but didn't
1863 catch all illegal uses. Such an illegal use could lead to crashes of
1864 the Guile interpreter or or other unwanted results. An example of
1865 incorrect internal defines that made Guile behave erratically:
1877 The problem with this example is that the definition of `c' uses the
1878 value of `b' directly. This confuses the meoization machine of Guile
1879 so that the second call of `b' (this time in a larger environment that
1880 also contains bindings for `c' and `d') refers to the binding of `c'
1881 instead of `a'. You could also make Guile crash with a variation on
1886 (define (b flag) (if flag a 1))
1887 (define c (1+ (b flag)))
1895 From now on, Guile will issue an `Unbound variable: b' error message
1900 A hook contains a list of functions which should be called on
1901 particular occasions in an existing program. Hooks are used for
1904 A window manager might have a hook before-window-map-hook. The window
1905 manager uses the function run-hooks to call all functions stored in
1906 before-window-map-hook each time a window is mapped. The user can
1907 store functions in the hook using add-hook!.
1909 In Guile, hooks are first class objects.
1911 *** New function: make-hook [N_ARGS]
1913 Return a hook for hook functions which can take N_ARGS arguments.
1914 The default value for N_ARGS is 0.
1916 (See also scm_make_named_hook below.)
1918 *** New function: add-hook! HOOK PROC [APPEND_P]
1920 Put PROC at the beginning of the list of functions stored in HOOK.
1921 If APPEND_P is supplied, and non-false, put PROC at the end instead.
1923 PROC must be able to take the number of arguments specified when the
1926 If PROC already exists in HOOK, then remove it first.
1928 *** New function: remove-hook! HOOK PROC
1930 Remove PROC from the list of functions in HOOK.
1932 *** New function: reset-hook! HOOK
1934 Clear the list of hook functions stored in HOOK.
1936 *** New function: run-hook HOOK ARG1 ...
1938 Run all hook functions stored in HOOK with arguments ARG1 ... .
1939 The number of arguments supplied must correspond to the number given
1940 when the hook was created.
1942 ** The function `dynamic-link' now takes optional keyword arguments.
1943 The only keyword argument that is currently defined is `:global
1944 BOOL'. With it, you can control whether the shared library will be
1945 linked in global mode or not. In global mode, the symbols from the
1946 linked library can be used to resolve references from other
1947 dynamically linked libraries. In non-global mode, the linked
1948 library is essentially invisible and can only be accessed via
1949 `dynamic-func', etc. The default is now to link in global mode.
1950 Previously, the default has been non-global mode.
1952 The `#:global' keyword is only effective on platforms that support
1953 the dlopen family of functions.
1955 ** New function `provided?'
1957 - Function: provided? FEATURE
1958 Return true iff FEATURE is supported by this installation of
1959 Guile. FEATURE must be a symbol naming a feature; the global
1960 variable `*features*' is a list of available features.
1962 ** Changes to the module (ice-9 expect):
1964 *** The expect-strings macro now matches `$' in a regular expression
1965 only at a line-break or end-of-file by default. Previously it would
1966 match the end of the string accumulated so far. The old behaviour
1967 can be obtained by setting the variable `expect-strings-exec-flags'
1970 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
1971 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
1972 in a regular expression will still match before a line-break or
1973 end-of-file. The default is `regexp/noteol'.
1975 *** The expect-strings macro now uses a variable
1976 `expect-strings-compile-flags' for the flags to be supplied to
1977 `make-regexp'. The default is `regexp/newline', which was previously
1980 *** The expect macro now supplies two arguments to a match procedure:
1981 the current accumulated string and a flag to indicate whether
1982 end-of-file has been reached. Previously only the string was supplied.
1983 If end-of-file is reached, the match procedure will be called an
1984 additional time with the same accumulated string as the previous call
1985 but with the flag set.
1987 ** New module (ice-9 format), implementing the Common Lisp `format' function.
1989 This code, and the documentation for it that appears here, was
1990 borrowed from SLIB, with minor adaptations for Guile.
1992 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
1993 An almost complete implementation of Common LISP format description
1994 according to the CL reference book `Common LISP' from Guy L.
1995 Steele, Digital Press. Backward compatible to most of the
1996 available Scheme format implementations.
1998 Returns `#t', `#f' or a string; has side effect of printing
1999 according to FORMAT-STRING. If DESTINATION is `#t', the output is
2000 to the current output port and `#t' is returned. If DESTINATION
2001 is `#f', a formatted string is returned as the result of the call.
2002 NEW: If DESTINATION is a string, DESTINATION is regarded as the
2003 format string; FORMAT-STRING is then the first argument and the
2004 output is returned as a string. If DESTINATION is a number, the
2005 output is to the current error port if available by the
2006 implementation. Otherwise DESTINATION must be an output port and
2009 FORMAT-STRING must be a string. In case of a formatting error
2010 format returns `#f' and prints a message on the current output or
2011 error port. Characters are output as if the string were output by
2012 the `display' function with the exception of those prefixed by a
2013 tilde (~). For a detailed description of the FORMAT-STRING syntax
2014 please consult a Common LISP format reference manual. For a test
2015 suite to verify this format implementation load `formatst.scm'.
2016 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
2018 Note: `format' is not reentrant, i.e. only one `format'-call may
2019 be executed at a time.
2022 *** Format Specification (Format version 3.0)
2024 Please consult a Common LISP format reference manual for a detailed
2025 description of the format string syntax. For a demonstration of the
2026 implemented directives see `formatst.scm'.
2028 This implementation supports directive parameters and modifiers (`:'
2029 and `@' characters). Multiple parameters must be separated by a comma
2030 (`,'). Parameters can be numerical parameters (positive or negative),
2031 character parameters (prefixed by a quote character (`''), variable
2032 parameters (`v'), number of rest arguments parameter (`#'), empty and
2033 default parameters. Directive characters are case independent. The
2034 general form of a directive is:
2036 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
2038 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
2040 *** Implemented CL Format Control Directives
2042 Documentation syntax: Uppercase characters represent the
2043 corresponding control directive characters. Lowercase characters
2044 represent control directive parameter descriptions.
2047 Any (print as `display' does).
2051 `~MINCOL,COLINC,MINPAD,PADCHARA'
2055 S-expression (print as `write' does).
2059 `~MINCOL,COLINC,MINPAD,PADCHARS'
2065 print number sign always.
2068 print comma separated.
2070 `~MINCOL,PADCHAR,COMMACHARD'
2076 print number sign always.
2079 print comma separated.
2081 `~MINCOL,PADCHAR,COMMACHARX'
2087 print number sign always.
2090 print comma separated.
2092 `~MINCOL,PADCHAR,COMMACHARO'
2098 print number sign always.
2101 print comma separated.
2103 `~MINCOL,PADCHAR,COMMACHARB'
2108 `~N,MINCOL,PADCHAR,COMMACHARR'
2112 print a number as a Roman numeral.
2115 print a number as an "old fashioned" Roman numeral.
2118 print a number as an ordinal English number.
2121 print a number as a cardinal English number.
2126 prints `y' and `ies'.
2129 as `~P but jumps 1 argument backward.'
2132 as `~@P but jumps 1 argument backward.'
2137 prints a character as the reader can understand it (i.e. `#\'
2141 prints a character as emacs does (eg. `^C' for ASCII 03).
2144 Fixed-format floating-point (prints a flonum like MMM.NNN).
2145 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
2147 If the number is positive a plus sign is printed.
2150 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
2151 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
2153 If the number is positive a plus sign is printed.
2156 General floating-point (prints a flonum either fixed or
2158 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
2160 If the number is positive a plus sign is printed.
2163 Dollars floating-point (prints a flonum in fixed with signs
2165 `~DIGITS,SCALE,WIDTH,PADCHAR$'
2167 If the number is positive a plus sign is printed.
2170 A sign is always printed and appears before the padding.
2173 The sign appears before the padding.
2181 print newline if not at the beginning of the output line.
2183 prints `~&' and then N-1 newlines.
2188 print N page separators.
2198 newline is ignored, white space left.
2201 newline is left, white space ignored.
2206 relative tabulation.
2212 Indirection (expects indirect arguments as a list).
2214 extracts indirect arguments from format arguments.
2217 Case conversion (converts by `string-downcase').
2219 converts by `string-capitalize'.
2222 converts by `string-capitalize-first'.
2225 converts by `string-upcase'.
2228 Argument Jumping (jumps 1 argument forward).
2230 jumps N arguments forward.
2233 jumps 1 argument backward.
2236 jumps N arguments backward.
2239 jumps to the 0th argument.
2242 jumps to the Nth argument (beginning from 0)
2244 `~[STR0~;STR1~;...~;STRN~]'
2245 Conditional Expression (numerical clause conditional).
2247 take argument from N.
2250 true test conditional.
2253 if-else-then conditional.
2259 default clause follows.
2262 Iteration (args come from the next argument (a list)).
2264 at most N iterations.
2267 args from next arg (a list of lists).
2270 args from the rest of arguments.
2273 args from the rest args (lists).
2284 aborts if N <= M <= K
2286 *** Not Implemented CL Format Control Directives
2289 print `#f' as an empty list (see below).
2292 print `#f' as an empty list (see below).
2298 (sorry I don't understand its semantics completely)
2300 *** Extended, Replaced and Additional Control Directives
2302 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
2303 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
2304 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
2305 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
2306 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
2307 COMMAWIDTH is the number of characters between two comma
2311 print a R4RS complex number as `~F~@Fi' with passed parameters for
2315 Pretty print formatting of an argument for scheme code lists.
2321 Flushes the output if format DESTINATION is a port.
2324 Print a `#\space' character
2326 print N `#\space' characters.
2329 Print a `#\tab' character
2331 print N `#\tab' characters.
2334 Takes N as an integer representation for a character. No arguments
2335 are consumed. N is converted to a character by `integer->char'. N
2336 must be a positive decimal number.
2339 Print out readproof. Prints out internal objects represented as
2340 `#<...>' as strings `"#<...>"' so that the format output can always
2341 be processed by `read'.
2344 Print out readproof. Prints out internal objects represented as
2345 `#<...>' as strings `"#<...>"' so that the format output can always
2346 be processed by `read'.
2349 Prints information and a copyright notice on the format
2352 prints format version.
2355 may also print number strings, i.e. passing a number as a string
2356 and format it accordingly.
2358 *** Configuration Variables
2360 The format module exports some configuration variables to suit the
2361 systems and users needs. There should be no modification necessary for
2362 the configuration that comes with Guile. Format detects automatically
2363 if the running scheme system implements floating point numbers and
2366 format:symbol-case-conv
2367 Symbols are converted by `symbol->string' so the case type of the
2368 printed symbols is implementation dependent.
2369 `format:symbol-case-conv' is a one arg closure which is either
2370 `#f' (no conversion), `string-upcase', `string-downcase' or
2371 `string-capitalize'. (default `#f')
2373 format:iobj-case-conv
2374 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
2375 implementation internal objects. (default `#f')
2378 The character prefixing the exponent value in `~E' printing.
2381 *** Compatibility With Other Format Implementations
2387 Downward compatible except for padding support and `~A', `~S',
2388 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
2389 `printf' padding support which is completely replaced by the CL
2390 `format' padding style.
2393 Downward compatible except for `~', which is not documented
2394 (ignores all characters inside the format string up to a newline
2395 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
2396 numerical and variable parameters and `:/@' modifiers in the CL
2400 Downward compatible except for `~A' and `~S' which print in
2401 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
2402 directive parameters or modifiers)).
2405 Downward compatible except for an optional destination parameter:
2406 S2C accepts a format call without a destination which returns a
2407 formatted string. This is equivalent to a #f destination in S2C.
2408 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
2409 parameters or modifiers)).
2412 ** Changes to string-handling functions.
2414 These functions were added to support the (ice-9 format) module, above.
2416 *** New function: string-upcase STRING
2417 *** New function: string-downcase STRING
2419 These are non-destructive versions of the existing string-upcase! and
2420 string-downcase! functions.
2422 *** New function: string-capitalize! STRING
2423 *** New function: string-capitalize STRING
2425 These functions convert the first letter of each word in the string to
2428 (string-capitalize "howdy there")
2431 As with the other functions, string-capitalize! modifies the string in
2432 place, while string-capitalize returns a modified copy of its argument.
2434 *** New function: string-ci->symbol STRING
2436 Return a symbol whose name is STRING, but having the same case as if
2437 the symbol had be read by `read'.
2439 Guile can be configured to be sensitive or insensitive to case
2440 differences in Scheme identifiers. If Guile is case-insensitive, all
2441 symbols are converted to lower case on input. The `string-ci->symbol'
2442 function returns a symbol whose name in STRING, transformed as Guile
2443 would if STRING were input.
2445 *** New function: substring-move! STRING1 START END STRING2 START
2447 Copy the substring of STRING1 from START (inclusive) to END
2448 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
2449 string, and the source and destination areas may overlap; in all
2450 cases, the function behaves as if all the characters were copied
2453 *** Extended functions: substring-move-left! substring-move-right!
2455 These functions now correctly copy arbitrarily overlapping substrings;
2456 they are both synonyms for substring-move!.
2459 ** New module (ice-9 getopt-long), with the function `getopt-long'.
2461 getopt-long is a function for parsing command-line arguments in a
2462 manner consistent with other GNU programs.
2464 (getopt-long ARGS GRAMMAR)
2465 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
2467 ARGS should be a list of strings. Its first element should be the
2468 name of the program; subsequent elements should be the arguments
2469 that were passed to the program on the command line. The
2470 `program-arguments' procedure returns a list of this form.
2472 GRAMMAR is a list of the form:
2473 ((OPTION (PROPERTY VALUE) ...) ...)
2475 Each OPTION should be a symbol. `getopt-long' will accept a
2476 command-line option named `--OPTION'.
2477 Each option can have the following (PROPERTY VALUE) pairs:
2479 (single-char CHAR) --- Accept `-CHAR' as a single-character
2480 equivalent to `--OPTION'. This is how to specify traditional
2482 (required? BOOL) --- If BOOL is true, the option is required.
2483 getopt-long will raise an error if it is not found in ARGS.
2484 (value BOOL) --- If BOOL is #t, the option accepts a value; if
2485 it is #f, it does not; and if it is the symbol
2486 `optional', the option may appear in ARGS with or
2488 (predicate FUNC) --- If the option accepts a value (i.e. you
2489 specified `(value #t)' for this option), then getopt
2490 will apply FUNC to the value, and throw an exception
2491 if it returns #f. FUNC should be a procedure which
2492 accepts a string and returns a boolean value; you may
2493 need to use quasiquotes to get it into GRAMMAR.
2495 The (PROPERTY VALUE) pairs may occur in any order, but each
2496 property may occur only once. By default, options do not have
2497 single-character equivalents, are not required, and do not take
2500 In ARGS, single-character options may be combined, in the usual
2501 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
2502 accepts values, then it must be the last option in the
2503 combination; the value is the next argument. So, for example, using
2504 the following grammar:
2505 ((apples (single-char #\a))
2506 (blimps (single-char #\b) (value #t))
2507 (catalexis (single-char #\c) (value #t)))
2508 the following argument lists would be acceptable:
2509 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
2510 for "blimps" and "catalexis")
2511 ("-ab" "bang" "-c" "couth") (same)
2512 ("-ac" "couth" "-b" "bang") (same)
2513 ("-abc" "couth" "bang") (an error, since `-b' is not the
2514 last option in its combination)
2516 If an option's value is optional, then `getopt-long' decides
2517 whether it has a value by looking at what follows it in ARGS. If
2518 the next element is a string, and it does not appear to be an
2519 option itself, then that string is the option's value.
2521 The value of a long option can appear as the next element in ARGS,
2522 or it can follow the option name, separated by an `=' character.
2523 Thus, using the same grammar as above, the following argument lists
2525 ("--apples" "Braeburn" "--blimps" "Goodyear")
2526 ("--apples=Braeburn" "--blimps" "Goodyear")
2527 ("--blimps" "Goodyear" "--apples=Braeburn")
2529 If the option "--" appears in ARGS, argument parsing stops there;
2530 subsequent arguments are returned as ordinary arguments, even if
2531 they resemble options. So, in the argument list:
2532 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
2533 `getopt-long' will recognize the `apples' option as having the
2534 value "Granny Smith", but it will not recognize the `blimp'
2535 option; it will return the strings "--blimp" and "Goodyear" as
2536 ordinary argument strings.
2538 The `getopt-long' function returns the parsed argument list as an
2539 assocation list, mapping option names --- the symbols from GRAMMAR
2540 --- onto their values, or #t if the option does not accept a value.
2541 Unused options do not appear in the alist.
2543 All arguments that are not the value of any option are returned
2544 as a list, associated with the empty list.
2546 `getopt-long' throws an exception if:
2547 - it finds an unrecognized option in ARGS
2548 - a required option is omitted
2549 - an option that requires an argument doesn't get one
2550 - an option that doesn't accept an argument does get one (this can
2551 only happen using the long option `--opt=value' syntax)
2552 - an option predicate fails
2557 `((lockfile-dir (required? #t)
2560 (predicate ,file-is-directory?))
2561 (verbose (required? #f)
2564 (x-includes (single-char #\x))
2565 (rnet-server (single-char #\y)
2566 (predicate ,string?))))
2568 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
2569 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
2571 => ((() "foo1" "-fred" "foo2" "foo3")
2572 (rnet-server . "lamprod")
2573 (x-includes . "/usr/include")
2574 (lockfile-dir . "/tmp")
2577 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
2579 It will be removed in a few releases.
2581 ** New syntax: lambda*
2582 ** New syntax: define*
2583 ** New syntax: define*-public
2584 ** New syntax: defmacro*
2585 ** New syntax: defmacro*-public
2586 Guile now supports optional arguments.
2588 `lambda*', `define*', `define*-public', `defmacro*' and
2589 `defmacro*-public' are identical to the non-* versions except that
2590 they use an extended type of parameter list that has the following BNF
2591 syntax (parentheses are literal, square brackets indicate grouping,
2592 and `*', `+' and `?' have the usual meaning):
2594 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
2595 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
2596 [[#&rest identifier]|[. identifier]]? ) | [identifier]
2598 ext-var-decl ::= identifier | ( identifier expression )
2600 The semantics are best illustrated with the following documentation
2601 and examples for `lambda*':
2604 lambda extended for optional and keyword arguments
2606 lambda* creates a procedure that takes optional arguments. These
2607 are specified by putting them inside brackets at the end of the
2608 paramater list, but before any dotted rest argument. For example,
2609 (lambda* (a b #&optional c d . e) '())
2610 creates a procedure with fixed arguments a and b, optional arguments c
2611 and d, and rest argument e. If the optional arguments are omitted
2612 in a call, the variables for them are unbound in the procedure. This
2613 can be checked with the bound? macro.
2615 lambda* can also take keyword arguments. For example, a procedure
2617 (lambda* (#&key xyzzy larch) '())
2618 can be called with any of the argument lists (#:xyzzy 11)
2619 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
2620 are given as keywords are bound to values.
2622 Optional and keyword arguments can also be given default values
2623 which they take on when they are not present in a call, by giving a
2624 two-item list in place of an optional argument, for example in:
2625 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
2626 foo is a fixed argument, bar is an optional argument with default
2627 value 42, and baz is a keyword argument with default value 73.
2628 Default value expressions are not evaluated unless they are needed
2629 and until the procedure is called.
2631 lambda* now supports two more special parameter list keywords.
2633 lambda*-defined procedures now throw an error by default if a
2634 keyword other than one of those specified is found in the actual
2635 passed arguments. However, specifying #&allow-other-keys
2636 immediately after the kyword argument declarations restores the
2637 previous behavior of ignoring unknown keywords. lambda* also now
2638 guarantees that if the same keyword is passed more than once, the
2639 last one passed is the one that takes effect. For example,
2640 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
2641 #:heads 37 #:tails 42 #:heads 99)
2642 would result in (99 47) being displayed.
2644 #&rest is also now provided as a synonym for the dotted syntax rest
2645 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
2646 all respects to lambda*. This is provided for more similarity to DSSSL,
2647 MIT-Scheme and Kawa among others, as well as for refugees from other
2650 Further documentation may be found in the optargs.scm file itself.
2652 The optional argument module also exports the macros `let-optional',
2653 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
2654 are not documented here because they may be removed in the future, but
2655 full documentation is still available in optargs.scm.
2657 ** New syntax: and-let*
2658 Guile now supports the `and-let*' form, described in the draft SRFI-2.
2660 Syntax: (land* (<clause> ...) <body> ...)
2661 Each <clause> should have one of the following forms:
2662 (<variable> <expression>)
2665 Each <variable> or <bound-variable> should be an identifier. Each
2666 <expression> should be a valid expression. The <body> should be a
2667 possibly empty sequence of expressions, like the <body> of a
2670 Semantics: A LAND* expression is evaluated by evaluating the
2671 <expression> or <bound-variable> of each of the <clause>s from
2672 left to right. The value of the first <expression> or
2673 <bound-variable> that evaluates to a false value is returned; the
2674 remaining <expression>s and <bound-variable>s are not evaluated.
2675 The <body> forms are evaluated iff all the <expression>s and
2676 <bound-variable>s evaluate to true values.
2678 The <expression>s and the <body> are evaluated in an environment
2679 binding each <variable> of the preceding (<variable> <expression>)
2680 clauses to the value of the <expression>. Later bindings
2681 shadow earlier bindings.
2683 Guile's and-let* macro was contributed by Michael Livshin.
2685 ** New sorting functions
2687 *** New function: sorted? SEQUENCE LESS?
2688 Returns `#t' when the sequence argument is in non-decreasing order
2689 according to LESS? (that is, there is no adjacent pair `... x y
2690 ...' for which `(less? y x)').
2692 Returns `#f' when the sequence contains at least one out-of-order
2693 pair. It is an error if the sequence is neither a list nor a
2696 *** New function: merge LIST1 LIST2 LESS?
2697 LIST1 and LIST2 are sorted lists.
2698 Returns the sorted list of all elements in LIST1 and LIST2.
2700 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
2701 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
2702 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
2703 (Here "<" should read "comes before".)
2705 *** New procedure: merge! LIST1 LIST2 LESS?
2706 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
2707 the result. If the code is compiled, and LESS? constructs no new
2708 pairs, no pairs at all will be allocated. The first pair of the
2709 result will be either the first pair of LIST1 or the first pair of
2712 *** New function: sort SEQUENCE LESS?
2713 Accepts either a list or a vector, and returns a new sequence
2714 which is sorted. The new sequence is the same type as the input.
2715 Always `(sorted? (sort sequence less?) less?)'. The original
2716 sequence is not altered in any way. The new sequence shares its
2717 elements with the old one; no elements are copied.
2719 *** New procedure: sort! SEQUENCE LESS
2720 Returns its sorted result in the original boxes. No new storage is
2721 allocated at all. Proper usage: (set! slist (sort! slist <))
2723 *** New function: stable-sort SEQUENCE LESS?
2724 Similar to `sort' but stable. That is, if "equal" elements are
2725 ordered a < b in the original sequence, they will have the same order
2728 *** New function: stable-sort! SEQUENCE LESS?
2729 Similar to `sort!' but stable.
2730 Uses temporary storage when sorting vectors.
2732 *** New functions: sort-list, sort-list!
2733 Added for compatibility with scsh.
2735 ** New built-in random number support
2737 *** New function: random N [STATE]
2738 Accepts a positive integer or real N and returns a number of the
2739 same type between zero (inclusive) and N (exclusive). The values
2740 returned have a uniform distribution.
2742 The optional argument STATE must be of the type produced by
2743 `copy-random-state' or `seed->random-state'. It defaults to the value
2744 of the variable `*random-state*'. This object is used to maintain the
2745 state of the pseudo-random-number generator and is altered as a side
2746 effect of the `random' operation.
2748 *** New variable: *random-state*
2749 Holds a data structure that encodes the internal state of the
2750 random-number generator that `random' uses by default. The nature
2751 of this data structure is implementation-dependent. It may be
2752 printed out and successfully read back in, but may or may not
2753 function correctly as a random-number state object in another
2756 *** New function: copy-random-state [STATE]
2757 Returns a new object of type suitable for use as the value of the
2758 variable `*random-state*' and as a second argument to `random'.
2759 If argument STATE is given, a copy of it is returned. Otherwise a
2760 copy of `*random-state*' is returned.
2762 *** New function: seed->random-state SEED
2763 Returns a new object of type suitable for use as the value of the
2764 variable `*random-state*' and as a second argument to `random'.
2765 SEED is a string or a number. A new state is generated and
2766 initialized using SEED.
2768 *** New function: random:uniform [STATE]
2769 Returns an uniformly distributed inexact real random number in the
2770 range between 0 and 1.
2772 *** New procedure: random:solid-sphere! VECT [STATE]
2773 Fills VECT with inexact real random numbers the sum of whose
2774 squares is less than 1.0. Thinking of VECT as coordinates in
2775 space of dimension N = `(vector-length VECT)', the coordinates are
2776 uniformly distributed within the unit N-shere. The sum of the
2777 squares of the numbers is returned. VECT can be either a vector
2778 or a uniform vector of doubles.
2780 *** New procedure: random:hollow-sphere! VECT [STATE]
2781 Fills VECT with inexact real random numbers the sum of whose squares
2782 is equal to 1.0. Thinking of VECT as coordinates in space of
2783 dimension n = `(vector-length VECT)', the coordinates are uniformly
2784 distributed over the surface of the unit n-shere. VECT can be either
2785 a vector or a uniform vector of doubles.
2787 *** New function: random:normal [STATE]
2788 Returns an inexact real in a normal distribution with mean 0 and
2789 standard deviation 1. For a normal distribution with mean M and
2790 standard deviation D use `(+ M (* D (random:normal)))'.
2792 *** New procedure: random:normal-vector! VECT [STATE]
2793 Fills VECT with inexact real random numbers which are independent and
2794 standard normally distributed (i.e., with mean 0 and variance 1).
2795 VECT can be either a vector or a uniform vector of doubles.
2797 *** New function: random:exp STATE
2798 Returns an inexact real in an exponential distribution with mean 1.
2799 For an exponential distribution with mean U use (* U (random:exp)).
2801 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
2803 These functions now operate on numbers in the range of a C unsigned
2806 These functions used to operate on numbers in the range of a C signed
2807 long; however, this seems inappropriate, because Guile integers don't
2810 ** New function: make-guardian
2811 This is an implementation of guardians as described in
2812 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
2813 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
2814 Programming Language Design and Implementation, June 1993
2815 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
2817 ** New functions: delq1!, delv1!, delete1!
2818 These procedures behave similar to delq! and friends but delete only
2819 one object if at all.
2821 ** New function: unread-string STRING PORT
2822 Unread STRING to PORT, that is, push it back onto the port so that
2823 next read operation will work on the pushed back characters.
2825 ** unread-char can now be called multiple times
2826 If unread-char is called multiple times, the unread characters will be
2827 read again in last-in first-out order.
2829 ** the procedures uniform-array-read! and uniform-array-write! now
2830 work on any kind of port, not just ports which are open on a file.
2832 ** Now 'l' in a port mode requests line buffering.
2834 ** The procedure truncate-file now works on string ports as well
2835 as file ports. If the size argument is omitted, the current
2836 file position is used.
2838 ** new procedure: seek PORT/FDES OFFSET WHENCE
2839 The arguments are the same as for the old fseek procedure, but it
2840 works on string ports as well as random-access file ports.
2842 ** the fseek procedure now works on string ports, since it has been
2843 redefined using seek.
2845 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
2846 size is not supplied.
2848 ** the newline procedure no longer flushes the port if it's not
2849 line-buffered: previously it did if it was the current output port.
2851 ** open-pipe and close-pipe are no longer primitive procedures, but
2852 an emulation can be obtained using `(use-modules (ice-9 popen))'.
2854 ** the freopen procedure has been removed.
2856 ** new procedure: drain-input PORT
2857 Drains PORT's read buffers (including any pushed-back characters)
2858 and returns the contents as a single string.
2860 ** New function: map-in-order PROC LIST1 LIST2 ...
2861 Version of `map' which guarantees that the procedure is applied to the
2862 lists in serial order.
2864 ** Renamed `serial-array-copy!' and `serial-array-map!' to
2865 `array-copy-in-order!' and `array-map-in-order!'. The old names are
2866 now obsolete and will go away in release 1.5.
2868 ** New syntax: collect BODY1 ...
2869 Version of `begin' which returns a list of the results of the body
2870 forms instead of the result of the last body form. In contrast to
2871 `begin', `collect' allows an empty body.
2873 ** New functions: read-history FILENAME, write-history FILENAME
2874 Read/write command line history from/to file. Returns #t on success
2875 and #f if an error occured.
2877 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
2879 These procedures return a list of definitions available in the specified
2880 argument, a relative module reference. In the case of no argument,
2881 `(current-module)' is now consulted for definitions to return, instead
2882 of simply returning #f, the former behavior.
2884 ** The #/ syntax for lists is no longer supported.
2886 Earlier versions of Scheme accepted this syntax, but printed a
2889 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
2891 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
2894 * Changes to the gh_ interface
2898 Now takes a second argument which is the result array. If this
2899 pointer is NULL, a new array is malloced (the old behaviour).
2901 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
2902 gh_scm2shorts, gh_scm2longs, gh_scm2floats
2906 * Changes to the scm_ interface
2908 ** Function: scm_make_named_hook (char* name, int n_args)
2910 Creates a hook in the same way as make-hook above but also
2911 binds a variable named NAME to it.
2913 This is the typical way of creating a hook from C code.
2915 Currently, the variable is created in the "current" module. This
2916 might change when we get the new module system.
2918 ** The smob interface
2920 The interface for creating smobs has changed. For documentation, see
2921 data-rep.info (made from guile-core/doc/data-rep.texi).
2923 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
2925 >>> This function will be removed in 1.3.4. <<<
2929 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
2930 This function adds a new smob type, named NAME, with instance size
2931 SIZE to the system. The return value is a tag that is used in
2932 creating instances of the type. If SIZE is 0, then no memory will
2933 be allocated when instances of the smob are created, and nothing
2934 will be freed by the default free function.
2936 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
2937 This function sets the smob marking procedure for the smob type
2938 specified by the tag TC. TC is the tag returned by
2939 `scm_make_smob_type'.
2941 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
2942 This function sets the smob freeing procedure for the smob type
2943 specified by the tag TC. TC is the tag returned by
2944 `scm_make_smob_type'.
2946 *** Function: void scm_set_smob_print (tc, print)
2948 - Function: void scm_set_smob_print (long tc,
2949 scm_sizet (*print) (SCM,
2953 This function sets the smob printing procedure for the smob type
2954 specified by the tag TC. TC is the tag returned by
2955 `scm_make_smob_type'.
2957 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
2958 This function sets the smob equality-testing predicate for the
2959 smob type specified by the tag TC. TC is the tag returned by
2960 `scm_make_smob_type'.
2962 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
2963 Make VALUE contain a smob instance of the type with type code TC and
2964 smob data DATA. VALUE must be previously declared as C type `SCM'.
2966 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
2967 This macro expands to a block of code that creates a smob instance
2968 of the type with type code TC and smob data DATA, and returns that
2969 `SCM' value. It should be the last piece of code in a block.
2971 ** The interfaces for using I/O ports and implementing port types
2972 (ptobs) have changed significantly. The new interface is based on
2973 shared access to buffers and a new set of ptob procedures.
2975 *** scm_newptob has been removed
2979 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
2981 - Function: SCM scm_make_port_type (char *type_name,
2982 int (*fill_buffer) (SCM port),
2983 void (*write_flush) (SCM port));
2985 Similarly to the new smob interface, there is a set of function
2986 setters by which the user can customize the behaviour of his port
2987 type. See ports.h (scm_set_port_XXX).
2989 ** scm_strport_to_string: New function: creates a new string from
2990 a string port's buffer.
2992 ** Plug in interface for random number generators
2993 The variable `scm_the_rng' in random.c contains a value and three
2994 function pointers which together define the current random number
2995 generator being used by the Scheme level interface and the random
2996 number library functions.
2998 The user is free to replace the default generator with the generator
3001 *** Variable: size_t scm_the_rng.rstate_size
3002 The size of the random state type used by the current RNG
3005 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
3006 Given the random STATE, return 32 random bits.
3008 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
3009 Seed random state STATE using string S of length N.
3011 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
3012 Given random state STATE, return a malloced copy.
3015 The default RNG is the MWC (Multiply With Carry) random number
3016 generator described by George Marsaglia at the Department of
3017 Statistics and Supercomputer Computations Research Institute, The
3018 Florida State University (http://stat.fsu.edu/~geo).
3020 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
3021 passes all tests in the DIEHARD test suite
3022 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
3023 costs one multiply and one add on platforms which either supports long
3024 longs (gcc does this on most systems) or have 64 bit longs. The cost
3025 is four multiply on other systems but this can be optimized by writing
3026 scm_i_uniform32 in assembler.
3028 These functions are provided through the scm_the_rng interface for use
3029 by libguile and the application.
3031 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
3032 Given the random STATE, return 32 random bits.
3033 Don't use this function directly. Instead go through the plugin
3034 interface (see "Plug in interface" above).
3036 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
3037 Initialize STATE using SEED of length N.
3039 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
3040 Return a malloc:ed copy of STATE. This function can easily be re-used
3041 in the interfaces to other RNGs.
3043 ** Random number library functions
3044 These functions use the current RNG through the scm_the_rng interface.
3045 It might be a good idea to use these functions from your C code so
3046 that only one random generator is used by all code in your program.
3048 The default random state is stored in:
3050 *** Variable: SCM scm_var_random_state
3051 Contains the vcell of the Scheme variable "*random-state*" which is
3052 used as default state by all random number functions in the Scheme
3057 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
3059 *** Function: scm_rstate *scm_c_default_rstate (void)
3060 This is a convenience function which returns the value of
3061 scm_var_random_state. An error message is generated if this value
3062 isn't a random state.
3064 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
3065 Make a new random state from the string SEED of length LENGTH.
3067 It is generally not a good idea to use multiple random states in a
3068 program. While subsequent random numbers generated from one random
3069 state are guaranteed to be reasonably independent, there is no such
3070 guarantee for numbers generated from different random states.
3072 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
3073 Return 32 random bits.
3075 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
3076 Return a sample from the uniform(0,1) distribution.
3078 *** Function: double scm_c_normal01 (scm_rstate *STATE)
3079 Return a sample from the normal(0,1) distribution.
3081 *** Function: double scm_c_exp1 (scm_rstate *STATE)
3082 Return a sample from the exp(1) distribution.
3084 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
3085 Return a sample from the discrete uniform(0,M) distribution.
3087 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
3088 Return a sample from the discrete uniform(0,M) distribution.
3089 M must be a bignum object. The returned value may be an INUM.
3093 Changes in Guile 1.3 (released Monday, October 19, 1998):
3095 * Changes to the distribution
3097 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
3098 To avoid conflicts, programs should name environment variables after
3099 themselves, except when there's a common practice establishing some
3102 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
3103 giving the former precedence, and printing a warning message if the
3104 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
3106 ** The header files related to multi-byte characters have been removed.
3107 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
3108 which referred to these explicitly will probably need to be rewritten,
3109 since the support for the variant string types has been removed; see
3112 ** The header files append.h and sequences.h have been removed. These
3113 files implemented non-R4RS operations which would encourage
3114 non-portable programming style and less easy-to-read code.
3116 * Changes to the stand-alone interpreter
3118 ** New procedures have been added to implement a "batch mode":
3120 *** Function: batch-mode?
3122 Returns a boolean indicating whether the interpreter is in batch
3125 *** Function: set-batch-mode?! ARG
3127 If ARG is true, switches the interpreter to batch mode. The `#f'
3128 case has not been implemented.
3130 ** Guile now provides full command-line editing, when run interactively.
3131 To use this feature, you must have the readline library installed.
3132 The Guile build process will notice it, and automatically include
3135 The readline library is available via anonymous FTP from any GNU
3136 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
3138 ** the-last-stack is now a fluid.
3140 * Changes to the procedure for linking libguile with your programs
3142 ** You can now use the `guile-config' utility to build programs that use Guile.
3144 Guile now includes a command-line utility called `guile-config', which
3145 can provide information about how to compile and link programs that
3148 *** `guile-config compile' prints any C compiler flags needed to use Guile.
3149 You should include this command's output on the command line you use
3150 to compile C or C++ code that #includes the Guile header files. It's
3151 usually just a `-I' flag to help the compiler find the Guile headers.
3154 *** `guile-config link' prints any linker flags necessary to link with Guile.
3156 This command writes to its standard output a list of flags which you
3157 must pass to the linker to link your code against the Guile library.
3158 The flags include '-lguile' itself, any other libraries the Guile
3159 library depends upon, and any `-L' flags needed to help the linker
3160 find those libraries.
3162 For example, here is a Makefile rule that builds a program named 'foo'
3163 from the object files ${FOO_OBJECTS}, and links them against Guile:
3166 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
3168 Previous Guile releases recommended that you use autoconf to detect
3169 which of a predefined set of libraries were present on your system.
3170 It is more robust to use `guile-config', since it records exactly which
3171 libraries the installed Guile library requires.
3173 This was originally called `build-guile', but was renamed to
3174 `guile-config' before Guile 1.3 was released, to be consistent with
3175 the analogous script for the GTK+ GUI toolkit, which is called
3179 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
3181 If you are using the GNU autoconf package to configure your program,
3182 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
3183 (described above) and gather the necessary values for use in your
3186 The GUILE_FLAGS macro expands to configure script code which runs the
3187 `guile-config' script, to find out where Guile's header files and
3188 libraries are installed. It sets two variables, marked for
3189 substitution, as by AC_SUBST.
3191 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
3192 code that uses Guile header files. This is almost always just a
3195 GUILE_LDFLAGS --- flags to pass to the linker to link a
3196 program against Guile. This includes `-lguile' for the Guile
3197 library itself, any libraries that Guile itself requires (like
3198 -lqthreads), and so on. It may also include a -L flag to tell the
3199 compiler where to find the libraries.
3201 GUILE_FLAGS is defined in the file guile.m4, in the top-level
3202 directory of the Guile distribution. You can copy it into your
3203 package's aclocal.m4 file, and then use it in your configure.in file.
3205 If you are using the `aclocal' program, distributed with GNU automake,
3206 to maintain your aclocal.m4 file, the Guile installation process
3207 installs guile.m4 where aclocal will find it. All you need to do is
3208 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
3209 this will copy the definition of GUILE_FLAGS into your aclocal.m4
3213 * Changes to Scheme functions and syntax
3215 ** Multi-byte strings have been removed, as have multi-byte and wide
3216 ports. We felt that these were the wrong approach to
3217 internationalization support.
3219 ** New function: readline [PROMPT]
3220 Read a line from the terminal, and allow the user to edit it,
3221 prompting with PROMPT. READLINE provides a large set of Emacs-like
3222 editing commands, lets the user recall previously typed lines, and
3223 works on almost every kind of terminal, including dumb terminals.
3225 READLINE assumes that the cursor is at the beginning of the line when
3226 it is invoked. Thus, you can't print a prompt yourself, and then call
3227 READLINE; you need to package up your prompt as a string, pass it to
3228 the function, and let READLINE print the prompt itself. This is
3229 because READLINE needs to know the prompt's screen width.
3231 For Guile to provide this function, you must have the readline
3232 library, version 2.1 or later, installed on your system. Readline is
3233 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
3234 any GNU mirror site.
3236 See also ADD-HISTORY function.
3238 ** New function: add-history STRING
3239 Add STRING as the most recent line in the history used by the READLINE
3240 command. READLINE does not add lines to the history itself; you must
3241 call ADD-HISTORY to make previous input available to the user.
3243 ** The behavior of the read-line function has changed.
3245 This function now uses standard C library functions to read the line,
3246 for speed. This means that it doesn not respect the value of
3247 scm-line-incrementors; it assumes that lines are delimited with
3250 (Note that this is read-line, the function that reads a line of text
3251 from a port, not readline, the function that reads a line from a
3252 terminal, providing full editing capabilities.)
3254 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
3256 This module provides some simple argument parsing. It exports one
3259 Function: getopt-gnu-style ARG-LS
3260 Parse a list of program arguments into an alist of option
3263 Each item in the list of program arguments is examined to see if
3264 it meets the syntax of a GNU long-named option. An argument like
3265 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
3266 returned alist, where MUMBLE is a keyword object with the same
3267 name as the argument. An argument like `--MUMBLE=FROB' produces
3268 an element of the form (MUMBLE . FROB), where FROB is a string.
3270 As a special case, the returned alist also contains a pair whose
3271 car is the symbol `rest'. The cdr of this pair is a list
3272 containing all the items in the argument list that are not options
3273 of the form mentioned above.
3275 The argument `--' is treated specially: all items in the argument
3276 list appearing after such an argument are not examined, and are
3277 returned in the special `rest' list.
3279 This function does not parse normal single-character switches.
3280 You will need to parse them out of the `rest' list yourself.
3282 ** The read syntax for byte vectors and short vectors has changed.
3284 Instead of #bytes(...), write #y(...).
3286 Instead of #short(...), write #h(...).
3288 This may seem nutty, but, like the other uniform vectors, byte vectors
3289 and short vectors want to have the same print and read syntax (and,
3290 more basic, want to have read syntax!). Changing the read syntax to
3291 use multiple characters after the hash sign breaks with the
3292 conventions used in R5RS and the conventions used for the other
3293 uniform vectors. It also introduces complexity in the current reader,
3294 both on the C and Scheme levels. (The Right solution is probably to
3295 change the syntax and prototypes for uniform vectors entirely.)
3298 ** The new module (ice-9 session) provides useful interactive functions.
3300 *** New procedure: (apropos REGEXP OPTION ...)
3302 Display a list of top-level variables whose names match REGEXP, and
3303 the modules they are imported from. Each OPTION should be one of the
3306 value --- Show the value of each matching variable.
3307 shadow --- Show bindings shadowed by subsequently imported modules.
3308 full --- Same as both `shadow' and `value'.
3312 guile> (apropos "trace" 'full)
3313 debug: trace #<procedure trace args>
3314 debug: untrace #<procedure untrace args>
3315 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
3316 the-scm-module: before-backtrace-hook ()
3317 the-scm-module: backtrace #<primitive-procedure backtrace>
3318 the-scm-module: after-backtrace-hook ()
3319 the-scm-module: has-shown-backtrace-hint? #f
3322 ** There are new functions and syntax for working with macros.
3324 Guile implements macros as a special object type. Any variable whose
3325 top-level binding is a macro object acts as a macro. The macro object
3326 specifies how the expression should be transformed before evaluation.
3328 *** Macro objects now print in a reasonable way, resembling procedures.
3330 *** New function: (macro? OBJ)
3331 True iff OBJ is a macro object.
3333 *** New function: (primitive-macro? OBJ)
3334 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
3335 macro transformers, implemented in eval.c rather than Scheme code.
3337 Why do we have this function?
3338 - For symmetry with procedure? and primitive-procedure?,
3339 - to allow custom print procedures to tell whether a macro is
3340 primitive, and display it differently, and
3341 - to allow compilers and user-written evaluators to distinguish
3342 builtin special forms from user-defined ones, which could be
3345 *** New function: (macro-type OBJ)
3346 Return a value indicating what kind of macro OBJ is. Possible return
3349 The symbol `syntax' --- a macro created by procedure->syntax.
3350 The symbol `macro' --- a macro created by procedure->macro.
3351 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
3352 The boolean #f --- if OBJ is not a macro object.
3354 *** New function: (macro-name MACRO)
3355 Return the name of the macro object MACRO's procedure, as returned by
3358 *** New function: (macro-transformer MACRO)
3359 Return the transformer procedure for MACRO.
3361 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
3363 Specify a new macro expander to use in the current module. Each
3364 MODULE is a module name, with the same meaning as in the `use-modules'
3365 form; each named module's exported bindings are added to the current
3366 top-level environment. TRANSFORMER is an expression evaluated in the
3367 resulting environment which must yield a procedure to use as the
3368 module's eval transformer: every expression evaluated in this module
3369 is passed to this function, and the result passed to the Guile
3372 *** macro-eval! is removed. Use local-eval instead.
3374 ** Some magic has been added to the printer to better handle user
3375 written printing routines (like record printers, closure printers).
3377 The problem is that these user written routines must have access to
3378 the current `print-state' to be able to handle fancy things like
3379 detection of circular references. These print-states have to be
3380 passed to the builtin printing routines (display, write, etc) to
3381 properly continue the print chain.
3383 We didn't want to change all existing print code so that it
3384 explicitly passes thru a print state in addition to a port. Instead,
3385 we extented the possible values that the builtin printing routines
3386 accept as a `port'. In addition to a normal port, they now also take
3387 a pair of a normal port and a print-state. Printing will go to the
3388 port and the print-state will be used to control the detection of
3389 circular references, etc. If the builtin function does not care for a
3390 print-state, it is simply ignored.
3392 User written callbacks are now called with such a pair as their
3393 `port', but because every function now accepts this pair as a PORT
3394 argument, you don't have to worry about that. In fact, it is probably
3395 safest to not check for these pairs.
3397 However, it is sometimes necessary to continue a print chain on a
3398 different port, for example to get a intermediate string
3399 representation of the printed value, mangle that string somehow, and
3400 then to finally print the mangled string. Use the new function
3402 inherit-print-state OLD-PORT NEW-PORT
3404 for this. It constructs a new `port' that prints to NEW-PORT but
3405 inherits the print-state of OLD-PORT.
3407 ** struct-vtable-offset renamed to vtable-offset-user
3409 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
3411 ** There is now a third optional argument to make-vtable-vtable
3412 (and fourth to make-struct) when constructing new types (vtables).
3413 This argument initializes field vtable-index-printer of the vtable.
3415 ** The detection of circular references has been extended to structs.
3416 That is, a structure that -- in the process of being printed -- prints
3417 itself does not lead to infinite recursion.
3419 ** There is now some basic support for fluids. Please read
3420 "libguile/fluid.h" to find out more. It is accessible from Scheme with
3421 the following functions and macros:
3423 Function: make-fluid
3425 Create a new fluid object. Fluids are not special variables or
3426 some other extension to the semantics of Scheme, but rather
3427 ordinary Scheme objects. You can store them into variables (that
3428 are still lexically scoped, of course) or into any other place you
3429 like. Every fluid has a initial value of `#f'.
3431 Function: fluid? OBJ
3433 Test whether OBJ is a fluid.
3435 Function: fluid-ref FLUID
3436 Function: fluid-set! FLUID VAL
3438 Access/modify the fluid FLUID. Modifications are only visible
3439 within the current dynamic root (that includes threads).
3441 Function: with-fluids* FLUIDS VALUES THUNK
3443 FLUIDS is a list of fluids and VALUES a corresponding list of
3444 values for these fluids. Before THUNK gets called the values are
3445 installed in the fluids and the old values of the fluids are
3446 saved in the VALUES list. When the flow of control leaves THUNK
3447 or reenters it, the values get swapped again. You might think of
3448 this as a `safe-fluid-excursion'. Note that the VALUES list is
3449 modified by `with-fluids*'.
3451 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
3453 The same as `with-fluids*' but with a different syntax. It looks
3454 just like `let', but both FLUID and VALUE are evaluated. Remember,
3455 fluids are not special variables but ordinary objects. FLUID
3456 should evaluate to a fluid.
3458 ** Changes to system call interfaces:
3460 *** close-port, close-input-port and close-output-port now return a
3461 boolean instead of an `unspecified' object. #t means that the port
3462 was successfully closed, while #f means it was already closed. It is
3463 also now possible for these procedures to raise an exception if an
3464 error occurs (some errors from write can be delayed until close.)
3466 *** the first argument to chmod, fcntl, ftell and fseek can now be a
3469 *** the third argument to fcntl is now optional.
3471 *** the first argument to chown can now be a file descriptor or a port.
3473 *** the argument to stat can now be a port.
3475 *** The following new procedures have been added (most use scsh
3478 *** procedure: close PORT/FD
3479 Similar to close-port (*note close-port: Closing Ports.), but also
3480 works on file descriptors. A side effect of closing a file
3481 descriptor is that any ports using that file descriptor are moved
3482 to a different file descriptor and have their revealed counts set
3485 *** procedure: port->fdes PORT
3486 Returns the integer file descriptor underlying PORT. As a side
3487 effect the revealed count of PORT is incremented.
3489 *** procedure: fdes->ports FDES
3490 Returns a list of existing ports which have FDES as an underlying
3491 file descriptor, without changing their revealed counts.
3493 *** procedure: fdes->inport FDES
3494 Returns an existing input port which has FDES as its underlying
3495 file descriptor, if one exists, and increments its revealed count.
3496 Otherwise, returns a new input port with a revealed count of 1.
3498 *** procedure: fdes->outport FDES
3499 Returns an existing output port which has FDES as its underlying
3500 file descriptor, if one exists, and increments its revealed count.
3501 Otherwise, returns a new output port with a revealed count of 1.
3503 The next group of procedures perform a `dup2' system call, if NEWFD
3504 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
3505 duplicated can be supplied as an integer or contained in a port. The
3506 type of value returned varies depending on which procedure is used.
3508 All procedures also have the side effect when performing `dup2' that
3509 any ports using NEWFD are moved to a different file descriptor and have
3510 their revealed counts set to zero.
3512 *** procedure: dup->fdes PORT/FD [NEWFD]
3513 Returns an integer file descriptor.
3515 *** procedure: dup->inport PORT/FD [NEWFD]
3516 Returns a new input port using the new file descriptor.
3518 *** procedure: dup->outport PORT/FD [NEWFD]
3519 Returns a new output port using the new file descriptor.
3521 *** procedure: dup PORT/FD [NEWFD]
3522 Returns a new port if PORT/FD is a port, with the same mode as the
3523 supplied port, otherwise returns an integer file descriptor.
3525 *** procedure: dup->port PORT/FD MODE [NEWFD]
3526 Returns a new port using the new file descriptor. MODE supplies a
3527 mode string for the port (*note open-file: File Ports.).
3529 *** procedure: setenv NAME VALUE
3530 Modifies the environment of the current process, which is also the
3531 default environment inherited by child processes.
3533 If VALUE is `#f', then NAME is removed from the environment.
3534 Otherwise, the string NAME=VALUE is added to the environment,
3535 replacing any existing string with name matching NAME.
3537 The return value is unspecified.
3539 *** procedure: truncate-file OBJ SIZE
3540 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
3541 can be a string containing a file name or an integer file
3542 descriptor or port open for output on the file. The underlying
3543 system calls are `truncate' and `ftruncate'.
3545 The return value is unspecified.
3547 *** procedure: setvbuf PORT MODE [SIZE]
3548 Set the buffering mode for PORT. MODE can be:
3556 block buffered, using a newly allocated buffer of SIZE bytes.
3557 However if SIZE is zero or unspecified, the port will be made
3560 This procedure should not be used after I/O has been performed with
3563 Ports are usually block buffered by default, with a default buffer
3564 size. Procedures e.g., *Note open-file: File Ports, which accept a
3565 mode string allow `0' to be added to request an unbuffered port.
3567 *** procedure: fsync PORT/FD
3568 Copies any unwritten data for the specified output file descriptor
3569 to disk. If PORT/FD is a port, its buffer is flushed before the
3570 underlying file descriptor is fsync'd. The return value is
3573 *** procedure: open-fdes PATH FLAGS [MODES]
3574 Similar to `open' but returns a file descriptor instead of a port.
3576 *** procedure: execle PATH ENV [ARG] ...
3577 Similar to `execl', but the environment of the new process is
3578 specified by ENV, which must be a list of strings as returned by
3579 the `environ' procedure.
3581 This procedure is currently implemented using the `execve' system
3582 call, but we call it `execle' because of its Scheme calling
3585 *** procedure: strerror ERRNO
3586 Returns the Unix error message corresponding to ERRNO, an integer.
3588 *** procedure: primitive-exit [STATUS]
3589 Terminate the current process without unwinding the Scheme stack.
3590 This is would typically be useful after a fork. The exit status
3591 is STATUS if supplied, otherwise zero.
3593 *** procedure: times
3594 Returns an object with information about real and processor time.
3595 The following procedures accept such an object as an argument and
3596 return a selected component:
3599 The current real time, expressed as time units relative to an
3603 The CPU time units used by the calling process.
3606 The CPU time units used by the system on behalf of the
3610 The CPU time units used by terminated child processes of the
3611 calling process, whose status has been collected (e.g., using
3615 Similarly, the CPU times units used by the system on behalf of
3616 terminated child processes.
3618 ** Removed: list-length
3619 ** Removed: list-append, list-append!
3620 ** Removed: list-reverse, list-reverse!
3622 ** array-map renamed to array-map!
3624 ** serial-array-map renamed to serial-array-map!
3626 ** catch doesn't take #f as first argument any longer
3628 Previously, it was possible to pass #f instead of a key to `catch'.
3629 That would cause `catch' to pass a jump buffer object to the procedure
3630 passed as second argument. The procedure could then use this jump
3631 buffer objekt as an argument to throw.
3633 This mechanism has been removed since its utility doesn't motivate the
3634 extra complexity it introduces.
3636 ** The `#/' notation for lists now provokes a warning message from Guile.
3637 This syntax will be removed from Guile in the near future.
3639 To disable the warning message, set the GUILE_HUSH environment
3640 variable to any non-empty value.
3642 ** The newline character now prints as `#\newline', following the
3643 normal Scheme notation, not `#\nl'.
3645 * Changes to the gh_ interface
3647 ** The gh_enter function now takes care of loading the Guile startup files.
3648 gh_enter works by calling scm_boot_guile; see the remarks below.
3650 ** Function: void gh_write (SCM x)
3652 Write the printed representation of the scheme object x to the current
3653 output port. Corresponds to the scheme level `write'.
3655 ** gh_list_length renamed to gh_length.
3657 ** vector handling routines
3659 Several major changes. In particular, gh_vector() now resembles
3660 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
3661 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
3662 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
3663 vector-related gh_ functions have been implemented.
3665 ** pair and list routines
3667 Implemented several of the R4RS pair and list functions that were
3670 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
3672 New function. Converts double arrays back and forth between Scheme
3675 * Changes to the scm_ interface
3677 ** The function scm_boot_guile now takes care of loading the startup files.
3679 Guile's primary initialization function, scm_boot_guile, now takes
3680 care of loading `boot-9.scm', in the `ice-9' module, to initialize
3681 Guile, define the module system, and put together some standard
3682 bindings. It also loads `init.scm', which is intended to hold
3683 site-specific initialization code.
3685 Since Guile cannot operate properly until boot-9.scm is loaded, there
3686 is no reason to separate loading boot-9.scm from Guile's other
3687 initialization processes.
3689 This job used to be done by scm_compile_shell_switches, which didn't
3690 make much sense; in particular, it meant that people using Guile for
3691 non-shell-like applications had to jump through hoops to get Guile
3692 initialized properly.
3694 ** The function scm_compile_shell_switches no longer loads the startup files.
3695 Now, Guile always loads the startup files, whenever it is initialized;
3696 see the notes above for scm_boot_guile and scm_load_startup_files.
3698 ** Function: scm_load_startup_files
3699 This new function takes care of loading Guile's initialization file
3700 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
3701 this is always called by the Guile initialization process, it's
3702 probably not too useful to call this yourself, but it's there anyway.
3704 ** The semantics of smob marking have changed slightly.
3706 The smob marking function (the `mark' member of the scm_smobfuns
3707 structure) is no longer responsible for setting the mark bit on the
3708 smob. The generic smob handling code in the garbage collector will
3709 set this bit. The mark function need only ensure that any other
3710 objects the smob refers to get marked.
3712 Note that this change means that the smob's GC8MARK bit is typically
3713 already set upon entry to the mark function. Thus, marking functions
3714 which look like this:
3717 if (SCM_GC8MARKP (ptr))
3719 SCM_SETGC8MARK (ptr);
3720 ... mark objects to which the smob refers ...
3723 are now incorrect, since they will return early, and fail to mark any
3724 other objects the smob refers to. Some code in the Guile library used
3727 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
3729 If you have implemented your own I/O port type, by writing the
3730 functions required by the scm_ptobfuns and then calling scm_newptob,
3731 you will need to change your functions slightly.
3733 The functions in a scm_ptobfuns structure now expect the port itself
3734 as their argument; they used to expect the `stream' member of the
3735 port's scm_port_table structure. This allows functions in an
3736 scm_ptobfuns structure to easily access the port's cell (and any flags
3737 it its CAR), and the port's scm_port_table structure.
3739 Guile now passes the I/O port itself as the `port' argument in the
3740 following scm_ptobfuns functions:
3742 int (*free) (SCM port);
3743 int (*fputc) (int, SCM port);
3744 int (*fputs) (char *, SCM port);
3745 scm_sizet (*fwrite) SCM_P ((char *ptr,
3749 int (*fflush) (SCM port);
3750 int (*fgetc) (SCM port);
3751 int (*fclose) (SCM port);
3753 The interfaces to the `mark', `print', `equalp', and `fgets' methods
3756 If you have existing code which defines its own port types, it is easy
3757 to convert your code to the new interface; simply apply SCM_STREAM to
3758 the port argument to yield the value you code used to expect.
3760 Note that since both the port and the stream have the same type in the
3761 C code --- they are both SCM values --- the C compiler will not remind
3762 you if you forget to update your scm_ptobfuns functions.
3765 ** Function: int scm_internal_select (int fds,
3769 struct timeval *timeout);
3771 This is a replacement for the `select' function provided by the OS.
3772 It enables I/O blocking and sleeping to happen for one cooperative
3773 thread without blocking other threads. It also avoids busy-loops in
3774 these situations. It is intended that all I/O blocking and sleeping
3775 will finally go through this function. Currently, this function is
3776 only available on systems providing `gettimeofday' and `select'.
3778 ** Function: SCM scm_internal_stack_catch (SCM tag,
3779 scm_catch_body_t body,
3781 scm_catch_handler_t handler,
3784 A new sibling to the other two C level `catch' functions
3785 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
3786 the stack to be saved automatically into the variable `the-last-stack'
3787 (scm_the_last_stack_var) on error. This is necessary if you want to
3788 use advanced error reporting, such as calling scm_display_error and
3789 scm_display_backtrace. (They both take a stack object as argument.)
3791 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
3793 scm_catch_handler_t handler,
3796 Spawns a new thread. It does a job similar to
3797 scm_call_with_new_thread but takes arguments more suitable when
3798 spawning threads from application C code.
3800 ** The hook scm_error_callback has been removed. It was originally
3801 intended as a way for the user to install his own error handler. But
3802 that method works badly since it intervenes between throw and catch,
3803 thereby changing the semantics of expressions like (catch #t ...).
3804 The correct way to do it is to use one of the C level catch functions
3805 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
3807 ** Removed functions:
3809 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
3810 scm_list_reverse, scm_list_reverse_x
3812 ** New macros: SCM_LISTn where n is one of the integers 0-9.
3814 These can be used for pretty list creation from C. The idea is taken
3815 from Erick Gallesio's STk.
3817 ** scm_array_map renamed to scm_array_map_x
3819 ** mbstrings are now removed
3821 This means that the type codes scm_tc7_mb_string and
3822 scm_tc7_mb_substring has been removed.
3824 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
3826 Since we no longer support multi-byte strings, these I/O functions
3827 have been simplified, and renamed. Here are their old names, and
3828 their new names and arguments:
3830 scm_gen_putc -> void scm_putc (int c, SCM port);
3831 scm_gen_puts -> void scm_puts (char *s, SCM port);
3832 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
3833 scm_gen_getc -> void scm_getc (SCM port);
3836 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
3838 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
3840 SCM_TYP7S now masks away the bit which distinguishes substrings from
3843 ** scm_catch_body_t: Backward incompatible change!
3845 Body functions to scm_internal_catch and friends do not any longer
3846 take a second argument. This is because it is no longer possible to
3847 pass a #f arg to catch.
3849 ** Calls to scm_protect_object and scm_unprotect now nest properly.
3851 The function scm_protect_object protects its argument from being freed
3852 by the garbage collector. scm_unprotect_object removes that
3855 These functions now nest properly. That is, for every object O, there
3856 is a counter which scm_protect_object(O) increments and
3857 scm_unprotect_object(O) decrements, if the counter is greater than
3858 zero. Every object's counter is zero when it is first created. If an
3859 object's counter is greater than zero, the garbage collector will not
3860 reclaim its storage.
3862 This allows you to use scm_protect_object in your code without
3863 worrying that some other function you call will call
3864 scm_unprotect_object, and allow it to be freed. Assuming that the
3865 functions you call are well-behaved, and unprotect only those objects
3866 they protect, you can follow the same rule and have confidence that
3867 objects will be freed only at appropriate times.
3870 Changes in Guile 1.2 (released Tuesday, June 24 1997):
3872 * Changes to the distribution
3874 ** Nightly snapshots are now available from ftp.red-bean.com.
3875 The old server, ftp.cyclic.com, has been relinquished to its rightful
3878 Nightly snapshots of the Guile development sources are now available via
3879 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
3881 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
3882 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
3884 ** To run Guile without installing it, the procedure has changed a bit.
3886 If you used a separate build directory to compile Guile, you'll need
3887 to include the build directory in SCHEME_LOAD_PATH, as well as the
3888 source directory. See the `INSTALL' file for examples.
3890 * Changes to the procedure for linking libguile with your programs
3892 ** The standard Guile load path for Scheme code now includes
3893 $(datadir)/guile (usually /usr/local/share/guile). This means that
3894 you can install your own Scheme files there, and Guile will find them.
3895 (Previous versions of Guile only checked a directory whose name
3896 contained the Guile version number, so you had to re-install or move
3897 your Scheme sources each time you installed a fresh version of Guile.)
3899 The load path also includes $(datadir)/guile/site; we recommend
3900 putting individual Scheme files there. If you want to install a
3901 package with multiple source files, create a directory for them under
3904 ** Guile 1.2 will now use the Rx regular expression library, if it is
3905 installed on your system. When you are linking libguile into your own
3906 programs, this means you will have to link against -lguile, -lqt (if
3907 you configured Guile with thread support), and -lrx.
3909 If you are using autoconf to generate configuration scripts for your
3910 application, the following lines should suffice to add the appropriate
3911 libraries to your link command:
3913 ### Find Rx, quickthreads and libguile.
3914 AC_CHECK_LIB(rx, main)
3915 AC_CHECK_LIB(qt, main)
3916 AC_CHECK_LIB(guile, scm_shell)
3918 The Guile 1.2 distribution does not contain sources for the Rx
3919 library, as Guile 1.0 did. If you want to use Rx, you'll need to
3920 retrieve it from a GNU FTP site and install it separately.
3922 * Changes to Scheme functions and syntax
3924 ** The dynamic linking features of Guile are now enabled by default.
3925 You can disable them by giving the `--disable-dynamic-linking' option
3928 (dynamic-link FILENAME)
3930 Find the object file denoted by FILENAME (a string) and link it
3931 into the running Guile application. When everything works out,
3932 return a Scheme object suitable for representing the linked object
3933 file. Otherwise an error is thrown. How object files are
3934 searched is system dependent.
3936 (dynamic-object? VAL)
3938 Determine whether VAL represents a dynamically linked object file.
3940 (dynamic-unlink DYNOBJ)
3942 Unlink the indicated object file from the application. DYNOBJ
3943 should be one of the values returned by `dynamic-link'.
3945 (dynamic-func FUNCTION DYNOBJ)
3947 Search the C function indicated by FUNCTION (a string or symbol)
3948 in DYNOBJ and return some Scheme object that can later be used
3949 with `dynamic-call' to actually call this function. Right now,
3950 these Scheme objects are formed by casting the address of the
3951 function to `long' and converting this number to its Scheme
3954 (dynamic-call FUNCTION DYNOBJ)
3956 Call the C function indicated by FUNCTION and DYNOBJ. The
3957 function is passed no arguments and its return value is ignored.
3958 When FUNCTION is something returned by `dynamic-func', call that
3959 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
3960 etc.), look it up in DYNOBJ; this is equivalent to
3962 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
3964 Interrupts are deferred while the C function is executing (with
3965 SCM_DEFER_INTS/SCM_ALLOW_INTS).
3967 (dynamic-args-call FUNCTION DYNOBJ ARGS)
3969 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
3970 some arguments and return its return value. The C function is
3971 expected to take two arguments and return an `int', just like
3974 int c_func (int argc, char **argv);
3976 ARGS must be a list of strings and is converted into an array of
3977 `char *'. The array is passed in ARGV and its size in ARGC. The
3978 return value is converted to a Scheme number and returned from the
3979 call to `dynamic-args-call'.
3981 When dynamic linking is disabled or not supported on your system,
3982 the above functions throw errors, but they are still available.
3984 Here is a small example that works on GNU/Linux:
3986 (define libc-obj (dynamic-link "libc.so"))
3987 (dynamic-args-call 'rand libc-obj '())
3989 See the file `libguile/DYNAMIC-LINKING' for additional comments.
3991 ** The #/ syntax for module names is depreciated, and will be removed
3992 in a future version of Guile. Instead of
4000 The latter syntax is more consistent with existing Lisp practice.
4002 ** Guile now does fancier printing of structures. Structures are the
4003 underlying implementation for records, which in turn are used to
4004 implement modules, so all of these object now print differently and in
4005 a more informative way.
4007 The Scheme printer will examine the builtin variable *struct-printer*
4008 whenever it needs to print a structure object. When this variable is
4009 not `#f' it is deemed to be a procedure and will be applied to the
4010 structure object and the output port. When *struct-printer* is `#f'
4011 or the procedure return `#f' the structure object will be printed in
4012 the boring #<struct 80458270> form.
4014 This hook is used by some routines in ice-9/boot-9.scm to implement
4015 type specific printing routines. Please read the comments there about
4018 One of the more specific uses of structs are records. The printing
4019 procedure that could be passed to MAKE-RECORD-TYPE is now actually
4020 called. It should behave like a *struct-printer* procedure (described
4023 ** Guile now supports a new R4RS-compliant syntax for keywords. A
4024 token of the form #:NAME, where NAME has the same syntax as a Scheme
4025 symbol, is the external representation of the keyword named NAME.
4026 Keyword objects print using this syntax as well, so values containing
4027 keyword objects can be read back into Guile. When used in an
4028 expression, keywords are self-quoting objects.
4030 Guile suports this read syntax, and uses this print syntax, regardless
4031 of the current setting of the `keyword' read option. The `keyword'
4032 read option only controls whether Guile recognizes the `:NAME' syntax,
4033 which is incompatible with R4RS. (R4RS says such token represent
4036 ** Guile has regular expression support again. Guile 1.0 included
4037 functions for matching regular expressions, based on the Rx library.
4038 In Guile 1.1, the Guile/Rx interface was removed to simplify the
4039 distribution, and thus Guile had no regular expression support. Guile
4040 1.2 again supports the most commonly used functions, and supports all
4041 of SCSH's regular expression functions.
4043 If your system does not include a POSIX regular expression library,
4044 and you have not linked Guile with a third-party regexp library such as
4045 Rx, these functions will not be available. You can tell whether your
4046 Guile installation includes regular expression support by checking
4047 whether the `*features*' list includes the `regex' symbol.
4049 *** regexp functions
4051 By default, Guile supports POSIX extended regular expressions. That
4052 means that the characters `(', `)', `+' and `?' are special, and must
4053 be escaped if you wish to match the literal characters.
4055 This regular expression interface was modeled after that implemented
4056 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
4057 with SCSH regular expressions.
4059 **** Function: string-match PATTERN STR [START]
4060 Compile the string PATTERN into a regular expression and compare
4061 it with STR. The optional numeric argument START specifies the
4062 position of STR at which to begin matching.
4064 `string-match' returns a "match structure" which describes what,
4065 if anything, was matched by the regular expression. *Note Match
4066 Structures::. If STR does not match PATTERN at all,
4067 `string-match' returns `#f'.
4069 Each time `string-match' is called, it must compile its PATTERN
4070 argument into a regular expression structure. This operation is
4071 expensive, which makes `string-match' inefficient if the same regular
4072 expression is used several times (for example, in a loop). For better
4073 performance, you can compile a regular expression in advance and then
4074 match strings against the compiled regexp.
4076 **** Function: make-regexp STR [FLAGS]
4077 Compile the regular expression described by STR, and return the
4078 compiled regexp structure. If STR does not describe a legal
4079 regular expression, `make-regexp' throws a
4080 `regular-expression-syntax' error.
4082 FLAGS may be the bitwise-or of one or more of the following:
4084 **** Constant: regexp/extended
4085 Use POSIX Extended Regular Expression syntax when interpreting
4086 STR. If not set, POSIX Basic Regular Expression syntax is used.
4087 If the FLAGS argument is omitted, we assume regexp/extended.
4089 **** Constant: regexp/icase
4090 Do not differentiate case. Subsequent searches using the
4091 returned regular expression will be case insensitive.
4093 **** Constant: regexp/newline
4094 Match-any-character operators don't match a newline.
4096 A non-matching list ([^...]) not containing a newline matches a
4099 Match-beginning-of-line operator (^) matches the empty string
4100 immediately after a newline, regardless of whether the FLAGS
4101 passed to regexp-exec contain regexp/notbol.
4103 Match-end-of-line operator ($) matches the empty string
4104 immediately before a newline, regardless of whether the FLAGS
4105 passed to regexp-exec contain regexp/noteol.
4107 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
4108 Match the compiled regular expression REGEXP against `str'. If
4109 the optional integer START argument is provided, begin matching
4110 from that position in the string. Return a match structure
4111 describing the results of the match, or `#f' if no match could be
4114 FLAGS may be the bitwise-or of one or more of the following:
4116 **** Constant: regexp/notbol
4117 The match-beginning-of-line operator always fails to match (but
4118 see the compilation flag regexp/newline above) This flag may be
4119 used when different portions of a string are passed to
4120 regexp-exec and the beginning of the string should not be
4121 interpreted as the beginning of the line.
4123 **** Constant: regexp/noteol
4124 The match-end-of-line operator always fails to match (but see the
4125 compilation flag regexp/newline above)
4127 **** Function: regexp? OBJ
4128 Return `#t' if OBJ is a compiled regular expression, or `#f'
4131 Regular expressions are commonly used to find patterns in one string
4132 and replace them with the contents of another string.
4134 **** Function: regexp-substitute PORT MATCH [ITEM...]
4135 Write to the output port PORT selected contents of the match
4136 structure MATCH. Each ITEM specifies what should be written, and
4137 may be one of the following arguments:
4139 * A string. String arguments are written out verbatim.
4141 * An integer. The submatch with that number is written.
4143 * The symbol `pre'. The portion of the matched string preceding
4144 the regexp match is written.
4146 * The symbol `post'. The portion of the matched string
4147 following the regexp match is written.
4149 PORT may be `#f', in which case nothing is written; instead,
4150 `regexp-substitute' constructs a string from the specified ITEMs
4153 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
4154 Similar to `regexp-substitute', but can be used to perform global
4155 substitutions on STR. Instead of taking a match structure as an
4156 argument, `regexp-substitute/global' takes two string arguments: a
4157 REGEXP string describing a regular expression, and a TARGET string
4158 which should be matched against this regular expression.
4160 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
4163 * A function may be supplied. When this function is called, it
4164 will be passed one argument: a match structure for a given
4165 regular expression match. It should return a string to be
4166 written out to PORT.
4168 * The `post' symbol causes `regexp-substitute/global' to recurse
4169 on the unmatched portion of STR. This *must* be supplied in
4170 order to perform global search-and-replace on STR; if it is
4171 not present among the ITEMs, then `regexp-substitute/global'
4172 will return after processing a single match.
4174 *** Match Structures
4176 A "match structure" is the object returned by `string-match' and
4177 `regexp-exec'. It describes which portion of a string, if any, matched
4178 the given regular expression. Match structures include: a reference to
4179 the string that was checked for matches; the starting and ending
4180 positions of the regexp match; and, if the regexp included any
4181 parenthesized subexpressions, the starting and ending positions of each
4184 In each of the regexp match functions described below, the `match'
4185 argument must be a match structure returned by a previous call to
4186 `string-match' or `regexp-exec'. Most of these functions return some
4187 information about the original target string that was matched against a
4188 regular expression; we will call that string TARGET for easy reference.
4190 **** Function: regexp-match? OBJ
4191 Return `#t' if OBJ is a match structure returned by a previous
4192 call to `regexp-exec', or `#f' otherwise.
4194 **** Function: match:substring MATCH [N]
4195 Return the portion of TARGET matched by subexpression number N.
4196 Submatch 0 (the default) represents the entire regexp match. If
4197 the regular expression as a whole matched, but the subexpression
4198 number N did not match, return `#f'.
4200 **** Function: match:start MATCH [N]
4201 Return the starting position of submatch number N.
4203 **** Function: match:end MATCH [N]
4204 Return the ending position of submatch number N.
4206 **** Function: match:prefix MATCH
4207 Return the unmatched portion of TARGET preceding the regexp match.
4209 **** Function: match:suffix MATCH
4210 Return the unmatched portion of TARGET following the regexp match.
4212 **** Function: match:count MATCH
4213 Return the number of parenthesized subexpressions from MATCH.
4214 Note that the entire regular expression match itself counts as a
4215 subexpression, and failed submatches are included in the count.
4217 **** Function: match:string MATCH
4218 Return the original TARGET string.
4220 *** Backslash Escapes
4222 Sometimes you will want a regexp to match characters like `*' or `$'
4223 exactly. For example, to check whether a particular string represents
4224 a menu entry from an Info node, it would be useful to match it against
4225 a regexp like `^* [^:]*::'. However, this won't work; because the
4226 asterisk is a metacharacter, it won't match the `*' at the beginning of
4227 the string. In this case, we want to make the first asterisk un-magic.
4229 You can do this by preceding the metacharacter with a backslash
4230 character `\'. (This is also called "quoting" the metacharacter, and
4231 is known as a "backslash escape".) When Guile sees a backslash in a
4232 regular expression, it considers the following glyph to be an ordinary
4233 character, no matter what special meaning it would ordinarily have.
4234 Therefore, we can make the above example work by changing the regexp to
4235 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
4236 to match only a single asterisk in the target string.
4238 Since the backslash is itself a metacharacter, you may force a
4239 regexp to match a backslash in the target string by preceding the
4240 backslash with itself. For example, to find variable references in a
4241 TeX program, you might want to find occurrences of the string `\let\'
4242 followed by any number of alphabetic characters. The regular expression
4243 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
4244 each match a single backslash in the target string.
4246 **** Function: regexp-quote STR
4247 Quote each special character found in STR with a backslash, and
4248 return the resulting string.
4250 *Very important:* Using backslash escapes in Guile source code (as
4251 in Emacs Lisp or C) can be tricky, because the backslash character has
4252 special meaning for the Guile reader. For example, if Guile encounters
4253 the character sequence `\n' in the middle of a string while processing
4254 Scheme code, it replaces those characters with a newline character.
4255 Similarly, the character sequence `\t' is replaced by a horizontal tab.
4256 Several of these "escape sequences" are processed by the Guile reader
4257 before your code is executed. Unrecognized escape sequences are
4258 ignored: if the characters `\*' appear in a string, they will be
4259 translated to the single character `*'.
4261 This translation is obviously undesirable for regular expressions,
4262 since we want to be able to include backslashes in a string in order to
4263 escape regexp metacharacters. Therefore, to make sure that a backslash
4264 is preserved in a string in your Guile program, you must use *two*
4265 consecutive backslashes:
4267 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
4269 The string in this example is preprocessed by the Guile reader before
4270 any code is executed. The resulting argument to `make-regexp' is the
4271 string `^\* [^:]*', which is what we really want.
4273 This also means that in order to write a regular expression that
4274 matches a single backslash character, the regular expression string in
4275 the source code must include *four* backslashes. Each consecutive pair
4276 of backslashes gets translated by the Guile reader to a single
4277 backslash, and the resulting double-backslash is interpreted by the
4278 regexp engine as matching a single backslash character. Hence:
4280 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
4282 The reason for the unwieldiness of this syntax is historical. Both
4283 regular expression pattern matchers and Unix string processing systems
4284 have traditionally used backslashes with the special meanings described
4285 above. The POSIX regular expression specification and ANSI C standard
4286 both require these semantics. Attempting to abandon either convention
4287 would cause other kinds of compatibility problems, possibly more severe
4288 ones. Therefore, without extending the Scheme reader to support
4289 strings with different quoting conventions (an ungainly and confusing
4290 extension when implemented in other languages), we must adhere to this
4291 cumbersome escape syntax.
4293 * Changes to the gh_ interface
4295 * Changes to the scm_ interface
4297 * Changes to system call interfaces:
4299 ** The value returned by `raise' is now unspecified. It throws an exception
4302 *** A new procedure `sigaction' can be used to install signal handlers
4304 (sigaction signum [action] [flags])
4306 signum is the signal number, which can be specified using the value
4309 If action is omitted, sigaction returns a pair: the CAR is the current
4310 signal hander, which will be either an integer with the value SIG_DFL
4311 (default action) or SIG_IGN (ignore), or the Scheme procedure which
4312 handles the signal, or #f if a non-Scheme procedure handles the
4313 signal. The CDR contains the current sigaction flags for the handler.
4315 If action is provided, it is installed as the new handler for signum.
4316 action can be a Scheme procedure taking one argument, or the value of
4317 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
4318 whatever signal handler was installed before sigaction was first used.
4319 Flags can optionally be specified for the new handler (SA_RESTART is
4320 always used if the system provides it, so need not be specified.) The
4321 return value is a pair with information about the old handler as
4324 This interface does not provide access to the "signal blocking"
4325 facility. Maybe this is not needed, since the thread support may
4326 provide solutions to the problem of consistent access to data
4329 *** A new procedure `flush-all-ports' is equivalent to running
4330 `force-output' on every port open for output.
4332 ** Guile now provides information on how it was built, via the new
4333 global variable, %guile-build-info. This variable records the values
4334 of the standard GNU makefile directory variables as an assocation
4335 list, mapping variable names (symbols) onto directory paths (strings).
4336 For example, to find out where the Guile link libraries were
4337 installed, you can say:
4339 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
4342 * Changes to the scm_ interface
4344 ** The new function scm_handle_by_message_noexit is just like the
4345 existing scm_handle_by_message function, except that it doesn't call
4346 exit to terminate the process. Instead, it prints a message and just
4347 returns #f. This might be a more appropriate catch-all handler for
4348 new dynamic roots and threads.
4351 Changes in Guile 1.1 (released Friday, May 16 1997):
4353 * Changes to the distribution.
4355 The Guile 1.0 distribution has been split up into several smaller
4357 guile-core --- the Guile interpreter itself.
4358 guile-tcltk --- the interface between the Guile interpreter and
4359 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
4360 is a toolkit for building graphical user interfaces.
4361 guile-rgx-ctax --- the interface between Guile and the Rx regular
4362 expression matcher, and the translator for the Ctax
4363 programming language. These are packaged together because the
4364 Ctax translator uses Rx to parse Ctax source code.
4366 This NEWS file describes the changes made to guile-core since the 1.0
4369 We no longer distribute the documentation, since it was either out of
4370 date, or incomplete. As soon as we have current documentation, we
4375 * Changes to the stand-alone interpreter
4377 ** guile now accepts command-line arguments compatible with SCSH, Olin
4378 Shivers' Scheme Shell.
4380 In general, arguments are evaluated from left to right, but there are
4381 exceptions. The following switches stop argument processing, and
4382 stash all remaining command-line arguments as the value returned by
4383 the (command-line) function.
4384 -s SCRIPT load Scheme source code from FILE, and exit
4385 -c EXPR evalute Scheme expression EXPR, and exit
4386 -- stop scanning arguments; run interactively
4388 The switches below are processed as they are encountered.
4389 -l FILE load Scheme source code from FILE
4390 -e FUNCTION after reading script, apply FUNCTION to
4391 command line arguments
4392 -ds do -s script at this point
4393 --emacs enable Emacs protocol (experimental)
4394 -h, --help display this help and exit
4395 -v, --version display version information and exit
4396 \ read arguments from following script lines
4398 So, for example, here is a Guile script named `ekko' (thanks, Olin)
4399 which re-implements the traditional "echo" command:
4401 #!/usr/local/bin/guile -s
4404 (map (lambda (arg) (display arg) (display " "))
4408 (main (command-line))
4410 Suppose we invoke this script as follows:
4412 ekko a speckled gecko
4414 Through the magic of Unix script processing (triggered by the `#!'
4415 token at the top of the file), /usr/local/bin/guile receives the
4416 following list of command-line arguments:
4418 ("-s" "./ekko" "a" "speckled" "gecko")
4420 Unix inserts the name of the script after the argument specified on
4421 the first line of the file (in this case, "-s"), and then follows that
4422 with the arguments given to the script. Guile loads the script, which
4423 defines the `main' function, and then applies it to the list of
4424 remaining command-line arguments, ("a" "speckled" "gecko").
4426 In Unix, the first line of a script file must take the following form:
4428 #!INTERPRETER ARGUMENT
4430 where INTERPRETER is the absolute filename of the interpreter
4431 executable, and ARGUMENT is a single command-line argument to pass to
4434 You may only pass one argument to the interpreter, and its length is
4435 limited. These restrictions can be annoying to work around, so Guile
4436 provides a general mechanism (borrowed from, and compatible with,
4437 SCSH) for circumventing them.
4439 If the ARGUMENT in a Guile script is a single backslash character,
4440 `\', Guile will open the script file, parse arguments from its second
4441 and subsequent lines, and replace the `\' with them. So, for example,
4442 here is another implementation of the `ekko' script:
4444 #!/usr/local/bin/guile \
4448 (for-each (lambda (arg) (display arg) (display " "))
4452 If the user invokes this script as follows:
4454 ekko a speckled gecko
4456 Unix expands this into
4458 /usr/local/bin/guile \ ekko a speckled gecko
4460 When Guile sees the `\' argument, it replaces it with the arguments
4461 read from the second line of the script, producing:
4463 /usr/local/bin/guile -e main -s ekko a speckled gecko
4465 This tells Guile to load the `ekko' script, and apply the function
4466 `main' to the argument list ("a" "speckled" "gecko").
4468 Here is how Guile parses the command-line arguments:
4469 - Each space character terminates an argument. This means that two
4470 spaces in a row introduce an empty-string argument.
4471 - The tab character is not permitted (unless you quote it with the
4472 backslash character, as described below), to avoid confusion.
4473 - The newline character terminates the sequence of arguments, and will
4474 also terminate a final non-empty argument. (However, a newline
4475 following a space will not introduce a final empty-string argument;
4476 it only terminates the argument list.)
4477 - The backslash character is the escape character. It escapes
4478 backslash, space, tab, and newline. The ANSI C escape sequences
4479 like \n and \t are also supported. These produce argument
4480 constituents; the two-character combination \n doesn't act like a
4481 terminating newline. The escape sequence \NNN for exactly three
4482 octal digits reads as the character whose ASCII code is NNN. As
4483 above, characters produced this way are argument constituents.
4484 Backslash followed by other characters is not allowed.
4486 * Changes to the procedure for linking libguile with your programs
4488 ** Guile now builds and installs a shared guile library, if your
4489 system support shared libraries. (It still builds a static library on
4490 all systems.) Guile automatically detects whether your system
4491 supports shared libraries. To prevent Guile from buildisg shared
4492 libraries, pass the `--disable-shared' flag to the configure script.
4494 Guile takes longer to compile when it builds shared libraries, because
4495 it must compile every file twice --- once to produce position-
4496 independent object code, and once to produce normal object code.
4498 ** The libthreads library has been merged into libguile.
4500 To link a program against Guile, you now need only link against
4501 -lguile and -lqt; -lthreads is no longer needed. If you are using
4502 autoconf to generate configuration scripts for your application, the
4503 following lines should suffice to add the appropriate libraries to
4506 ### Find quickthreads and libguile.
4507 AC_CHECK_LIB(qt, main)
4508 AC_CHECK_LIB(guile, scm_shell)
4510 * Changes to Scheme functions
4512 ** Guile Scheme's special syntax for keyword objects is now optional,
4513 and disabled by default.
4515 The syntax variation from R4RS made it difficult to port some
4516 interesting packages to Guile. The routines which accepted keyword
4517 arguments (mostly in the module system) have been modified to also
4518 accept symbols whose names begin with `:'.
4520 To change the keyword syntax, you must first import the (ice-9 debug)
4522 (use-modules (ice-9 debug))
4524 Then you can enable the keyword syntax as follows:
4525 (read-set! keywords 'prefix)
4527 To disable keyword syntax, do this:
4528 (read-set! keywords #f)
4530 ** Many more primitive functions accept shared substrings as
4531 arguments. In the past, these functions required normal, mutable
4532 strings as arguments, although they never made use of this
4535 ** The uniform array functions now operate on byte vectors. These
4536 functions are `array-fill!', `serial-array-copy!', `array-copy!',
4537 `serial-array-map', `array-map', `array-for-each', and
4540 ** The new functions `trace' and `untrace' implement simple debugging
4541 support for Scheme functions.
4543 The `trace' function accepts any number of procedures as arguments,
4544 and tells the Guile interpreter to display each procedure's name and
4545 arguments each time the procedure is invoked. When invoked with no
4546 arguments, `trace' returns the list of procedures currently being
4549 The `untrace' function accepts any number of procedures as arguments,
4550 and tells the Guile interpreter not to trace them any more. When
4551 invoked with no arguments, `untrace' untraces all curretly traced
4554 The tracing in Guile has an advantage over most other systems: we
4555 don't create new procedure objects, but mark the procedure objects
4556 themselves. This means that anonymous and internal procedures can be
4559 ** The function `assert-repl-prompt' has been renamed to
4560 `set-repl-prompt!'. It takes one argument, PROMPT.
4561 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
4562 - If PROMPT is a string, we use it as a prompt.
4563 - If PROMPT is a procedure accepting no arguments, we call it, and
4564 display the result as a prompt.
4565 - Otherwise, we display "> ".
4567 ** The new function `eval-string' reads Scheme expressions from a
4568 string and evaluates them, returning the value of the last expression
4569 in the string. If the string contains no expressions, it returns an
4572 ** The new function `thunk?' returns true iff its argument is a
4573 procedure of zero arguments.
4575 ** `defined?' is now a builtin function, instead of syntax. This
4576 means that its argument should be quoted. It returns #t iff its
4577 argument is bound in the current module.
4579 ** The new syntax `use-modules' allows you to add new modules to your
4580 environment without re-typing a complete `define-module' form. It
4581 accepts any number of module names as arguments, and imports their
4582 public bindings into the current module.
4584 ** The new function (module-defined? NAME MODULE) returns true iff
4585 NAME, a symbol, is defined in MODULE, a module object.
4587 ** The new function `builtin-bindings' creates and returns a hash
4588 table containing copies of all the root module's bindings.
4590 ** The new function `builtin-weak-bindings' does the same as
4591 `builtin-bindings', but creates a doubly-weak hash table.
4593 ** The `equal?' function now considers variable objects to be
4594 equivalent if they have the same name and the same value.
4596 ** The new function `command-line' returns the command-line arguments
4597 given to Guile, as a list of strings.
4599 When using guile as a script interpreter, `command-line' returns the
4600 script's arguments; those processed by the interpreter (like `-s' or
4601 `-c') are omitted. (In other words, you get the normal, expected
4602 behavior.) Any application that uses scm_shell to process its
4603 command-line arguments gets this behavior as well.
4605 ** The new function `load-user-init' looks for a file called `.guile'
4606 in the user's home directory, and loads it if it exists. This is
4607 mostly for use by the code generated by scm_compile_shell_switches,
4608 but we thought it might also be useful in other circumstances.
4610 ** The new function `log10' returns the base-10 logarithm of its
4613 ** Changes to I/O functions
4615 *** The functions `read', `primitive-load', `read-and-eval!', and
4616 `primitive-load-path' no longer take optional arguments controlling
4617 case insensitivity and a `#' parser.
4619 Case sensitivity is now controlled by a read option called
4620 `case-insensitive'. The user can add new `#' syntaxes with the
4621 `read-hash-extend' function (see below).
4623 *** The new function `read-hash-extend' allows the user to change the
4624 syntax of Guile Scheme in a somewhat controlled way.
4626 (read-hash-extend CHAR PROC)
4627 When parsing S-expressions, if we read a `#' character followed by
4628 the character CHAR, use PROC to parse an object from the stream.
4629 If PROC is #f, remove any parsing procedure registered for CHAR.
4631 The reader applies PROC to two arguments: CHAR and an input port.
4633 *** The new functions read-delimited and read-delimited! provide a
4634 general mechanism for doing delimited input on streams.
4636 (read-delimited DELIMS [PORT HANDLE-DELIM])
4637 Read until we encounter one of the characters in DELIMS (a string),
4638 or end-of-file. PORT is the input port to read from; it defaults to
4639 the current input port. The HANDLE-DELIM parameter determines how
4640 the terminating character is handled; it should be one of the
4643 'trim omit delimiter from result
4644 'peek leave delimiter character in input stream
4645 'concat append delimiter character to returned value
4646 'split return a pair: (RESULT . TERMINATOR)
4648 HANDLE-DELIM defaults to 'peek.
4650 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
4651 A side-effecting variant of `read-delimited'.
4653 The data is written into the string BUF at the indices in the
4654 half-open interval [START, END); the default interval is the whole
4655 string: START = 0 and END = (string-length BUF). The values of
4656 START and END must specify a well-defined interval in BUF, i.e.
4657 0 <= START <= END <= (string-length BUF).
4659 It returns NBYTES, the number of bytes read. If the buffer filled
4660 up without a delimiter character being found, it returns #f. If the
4661 port is at EOF when the read starts, it returns the EOF object.
4663 If an integer is returned (i.e., the read is successfully terminated
4664 by reading a delimiter character), then the HANDLE-DELIM parameter
4665 determines how to handle the terminating character. It is described
4666 above, and defaults to 'peek.
4668 (The descriptions of these functions were borrowed from the SCSH
4669 manual, by Olin Shivers and Brian Carlstrom.)
4671 *** The `%read-delimited!' function is the primitive used to implement
4672 `read-delimited' and `read-delimited!'.
4674 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
4676 This returns a pair of values: (TERMINATOR . NUM-READ).
4677 - TERMINATOR describes why the read was terminated. If it is a
4678 character or the eof object, then that is the value that terminated
4679 the read. If it is #f, the function filled the buffer without finding
4680 a delimiting character.
4681 - NUM-READ is the number of characters read into BUF.
4683 If the read is successfully terminated by reading a delimiter
4684 character, then the gobble? parameter determines what to do with the
4685 terminating character. If true, the character is removed from the
4686 input stream; if false, the character is left in the input stream
4687 where a subsequent read operation will retrieve it. In either case,
4688 the character is also the first value returned by the procedure call.
4690 (The descriptions of this function was borrowed from the SCSH manual,
4691 by Olin Shivers and Brian Carlstrom.)
4693 *** The `read-line' and `read-line!' functions have changed; they now
4694 trim the terminator by default; previously they appended it to the
4695 returned string. For the old behavior, use (read-line PORT 'concat).
4697 *** The functions `uniform-array-read!' and `uniform-array-write!' now
4698 take new optional START and END arguments, specifying the region of
4699 the array to read and write.
4701 *** The `ungetc-char-ready?' function has been removed. We feel it's
4702 inappropriate for an interface to expose implementation details this
4705 ** Changes to the Unix library and system call interface
4707 *** The new fcntl function provides access to the Unix `fcntl' system
4710 (fcntl PORT COMMAND VALUE)
4711 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
4712 Values for COMMAND are:
4714 F_DUPFD duplicate a file descriptor
4715 F_GETFD read the descriptor's close-on-exec flag
4716 F_SETFD set the descriptor's close-on-exec flag to VALUE
4717 F_GETFL read the descriptor's flags, as set on open
4718 F_SETFL set the descriptor's flags, as set on open to VALUE
4719 F_GETOWN return the process ID of a socket's owner, for SIGIO
4720 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
4721 FD_CLOEXEC not sure what this is
4723 For details, see the documentation for the fcntl system call.
4725 *** The arguments to `select' have changed, for compatibility with
4726 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
4727 expected behavior. The MILLISECONDS parameter has been changed to
4728 MICROSECONDS, to more closely resemble the underlying system call.
4729 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
4730 corresponding return set will be the same.
4732 *** The arguments to the `mknod' system call have changed. They are
4735 (mknod PATH TYPE PERMS DEV)
4736 Create a new file (`node') in the file system. PATH is the name of
4737 the file to create. TYPE is the kind of file to create; it should
4738 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
4739 permission bits to give the newly created file. If TYPE is
4740 'block-special or 'char-special, DEV specifies which device the
4741 special file refers to; its interpretation depends on the kind of
4742 special file being created.
4744 *** The `fork' function has been renamed to `primitive-fork', to avoid
4745 clashing with various SCSH forks.
4747 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
4748 and `recvfrom!'. They no longer accept a size for a second argument;
4749 you must pass a string to hold the received value. They no longer
4750 return the buffer. Instead, `recv' returns the length of the message
4751 received, and `recvfrom' returns a pair containing the packet's length
4752 and originating address.
4754 *** The file descriptor datatype has been removed, as have the
4755 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
4756 We plan to replace these functions with a SCSH-compatible interface.
4758 *** The `create' function has been removed; it's just a special case
4761 *** There are new functions to break down process termination status
4762 values. In the descriptions below, STATUS is a value returned by
4765 (status:exit-val STATUS)
4766 If the child process exited normally, this function returns the exit
4767 code for the child process (i.e., the value passed to exit, or
4768 returned from main). If the child process did not exit normally,
4769 this function returns #f.
4771 (status:stop-sig STATUS)
4772 If the child process was suspended by a signal, this function
4773 returns the signal that suspended the child. Otherwise, it returns
4776 (status:term-sig STATUS)
4777 If the child process terminated abnormally, this function returns
4778 the signal that terminated the child. Otherwise, this function
4781 POSIX promises that exactly one of these functions will return true on
4782 a valid STATUS value.
4784 These functions are compatible with SCSH.
4786 *** There are new accessors and setters for the broken-out time vectors
4787 returned by `localtime', `gmtime', and that ilk. They are:
4789 Component Accessor Setter
4790 ========================= ============ ============
4791 seconds tm:sec set-tm:sec
4792 minutes tm:min set-tm:min
4793 hours tm:hour set-tm:hour
4794 day of the month tm:mday set-tm:mday
4795 month tm:mon set-tm:mon
4796 year tm:year set-tm:year
4797 day of the week tm:wday set-tm:wday
4798 day in the year tm:yday set-tm:yday
4799 daylight saving time tm:isdst set-tm:isdst
4800 GMT offset, seconds tm:gmtoff set-tm:gmtoff
4801 name of time zone tm:zone set-tm:zone
4803 *** There are new accessors for the vectors returned by `uname',
4804 describing the host system:
4807 ============================================== ================
4808 name of the operating system implementation utsname:sysname
4809 network name of this machine utsname:nodename
4810 release level of the operating system utsname:release
4811 version level of the operating system utsname:version
4812 machine hardware platform utsname:machine
4814 *** There are new accessors for the vectors returned by `getpw',
4815 `getpwnam', `getpwuid', and `getpwent', describing entries from the
4816 system's user database:
4819 ====================== =================
4820 user name passwd:name
4821 user password passwd:passwd
4824 real name passwd:gecos
4825 home directory passwd:dir
4826 shell program passwd:shell
4828 *** There are new accessors for the vectors returned by `getgr',
4829 `getgrnam', `getgrgid', and `getgrent', describing entries from the
4830 system's group database:
4833 ======================= ============
4834 group name group:name
4835 group password group:passwd
4837 group members group:mem
4839 *** There are new accessors for the vectors returned by `gethost',
4840 `gethostbyaddr', `gethostbyname', and `gethostent', describing
4844 ========================= ===============
4845 official name of host hostent:name
4846 alias list hostent:aliases
4847 host address type hostent:addrtype
4848 length of address hostent:length
4849 list of addresses hostent:addr-list
4851 *** There are new accessors for the vectors returned by `getnet',
4852 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
4856 ========================= ===============
4857 official name of net netent:name
4858 alias list netent:aliases
4859 net number type netent:addrtype
4860 net number netent:net
4862 *** There are new accessors for the vectors returned by `getproto',
4863 `getprotobyname', `getprotobynumber', and `getprotoent', describing
4867 ========================= ===============
4868 official protocol name protoent:name
4869 alias list protoent:aliases
4870 protocol number protoent:proto
4872 *** There are new accessors for the vectors returned by `getserv',
4873 `getservbyname', `getservbyport', and `getservent', describing
4877 ========================= ===============
4878 official service name servent:name
4879 alias list servent:aliases
4880 port number servent:port
4881 protocol to use servent:proto
4883 *** There are new accessors for the sockaddr structures returned by
4884 `accept', `getsockname', `getpeername', `recvfrom!':
4887 ======================================== ===============
4888 address format (`family') sockaddr:fam
4889 path, for file domain addresses sockaddr:path
4890 address, for internet domain addresses sockaddr:addr
4891 TCP or UDP port, for internet sockaddr:port
4893 *** The `getpwent', `getgrent', `gethostent', `getnetent',
4894 `getprotoent', and `getservent' functions now return #f at the end of
4895 the user database. (They used to throw an exception.)
4897 Note that calling MUMBLEent function is equivalent to calling the
4898 corresponding MUMBLE function with no arguments.
4900 *** The `setpwent', `setgrent', `sethostent', `setnetent',
4901 `setprotoent', and `setservent' routines now take no arguments.
4903 *** The `gethost', `getproto', `getnet', and `getserv' functions now
4904 provide more useful information when they throw an exception.
4906 *** The `lnaof' function has been renamed to `inet-lnaof'.
4908 *** Guile now claims to have the `current-time' feature.
4910 *** The `mktime' function now takes an optional second argument ZONE,
4911 giving the time zone to use for the conversion. ZONE should be a
4912 string, in the same format as expected for the "TZ" environment variable.
4914 *** The `strptime' function now returns a pair (TIME . COUNT), where
4915 TIME is the parsed time as a vector, and COUNT is the number of
4916 characters from the string left unparsed. This function used to
4917 return the remaining characters as a string.
4919 *** The `gettimeofday' function has replaced the old `time+ticks' function.
4920 The return value is now (SECONDS . MICROSECONDS); the fractional
4921 component is no longer expressed in "ticks".
4923 *** The `ticks/sec' constant has been removed, in light of the above change.
4925 * Changes to the gh_ interface
4927 ** gh_eval_str() now returns an SCM object which is the result of the
4930 ** gh_scm2str() now copies the Scheme data to a caller-provided C
4933 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
4934 and returns the array
4936 ** gh_scm2str0() is gone: there is no need to distinguish
4937 null-terminated from non-null-terminated, since gh_scm2newstr() allows
4938 the user to interpret the data both ways.
4940 * Changes to the scm_ interface
4942 ** The new function scm_symbol_value0 provides an easy way to get a
4943 symbol's value from C code:
4945 SCM scm_symbol_value0 (char *NAME)
4946 Return the value of the symbol named by the null-terminated string
4947 NAME in the current module. If the symbol named NAME is unbound in
4948 the current module, return SCM_UNDEFINED.
4950 ** The new function scm_sysintern0 creates new top-level variables,
4951 without assigning them a value.
4953 SCM scm_sysintern0 (char *NAME)
4954 Create a new Scheme top-level variable named NAME. NAME is a
4955 null-terminated string. Return the variable's value cell.
4957 ** The function scm_internal_catch is the guts of catch. It handles
4958 all the mechanics of setting up a catch target, invoking the catch
4959 body, and perhaps invoking the handler if the body does a throw.
4961 The function is designed to be usable from C code, but is general
4962 enough to implement all the semantics Guile Scheme expects from throw.
4964 TAG is the catch tag. Typically, this is a symbol, but this function
4965 doesn't actually care about that.
4967 BODY is a pointer to a C function which runs the body of the catch;
4968 this is the code you can throw from. We call it like this:
4969 BODY (BODY_DATA, JMPBUF)
4971 BODY_DATA is just the BODY_DATA argument we received; we pass it
4972 through to BODY as its first argument. The caller can make
4973 BODY_DATA point to anything useful that BODY might need.
4974 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
4975 which we have just created and initialized.
4977 HANDLER is a pointer to a C function to deal with a throw to TAG,
4978 should one occur. We call it like this:
4979 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
4981 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
4982 same idea as BODY_DATA above.
4983 THROWN_TAG is the tag that the user threw to; usually this is
4984 TAG, but it could be something else if TAG was #t (i.e., a
4985 catch-all), or the user threw to a jmpbuf.
4986 THROW_ARGS is the list of arguments the user passed to the THROW
4989 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
4990 is just a pointer we pass through to HANDLER. We don't actually
4991 use either of those pointers otherwise ourselves. The idea is
4992 that, if our caller wants to communicate something to BODY or
4993 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
4994 HANDLER can then use. Think of it as a way to make BODY and
4995 HANDLER closures, not just functions; MUMBLE_DATA points to the
4998 Of course, it's up to the caller to make sure that any data a
4999 MUMBLE_DATA needs is protected from GC. A common way to do this is
5000 to make MUMBLE_DATA a pointer to data stored in an automatic
5001 structure variable; since the collector must scan the stack for
5002 references anyway, this assures that any references in MUMBLE_DATA
5005 ** The new function scm_internal_lazy_catch is exactly like
5006 scm_internal_catch, except:
5008 - It does not unwind the stack (this is the major difference).
5009 - If handler returns, its value is returned from the throw.
5010 - BODY always receives #f as its JMPBUF argument (since there's no
5011 jmpbuf associated with a lazy catch, because we don't unwind the
5014 ** scm_body_thunk is a new body function you can pass to
5015 scm_internal_catch if you want the body to be like Scheme's `catch'
5016 --- a thunk, or a function of one argument if the tag is #f.
5018 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
5019 contains the Scheme procedure to invoke as the body, and the tag
5020 we're catching. If the tag is #f, then we pass JMPBUF (created by
5021 scm_internal_catch) to the body procedure; otherwise, the body gets
5024 ** scm_handle_by_proc is a new handler function you can pass to
5025 scm_internal_catch if you want the handler to act like Scheme's catch
5026 --- call a procedure with the tag and the throw arguments.
5028 If the user does a throw to this catch, this function runs a handler
5029 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
5030 variable holding the Scheme procedure object to invoke. It ought to
5031 be a pointer to an automatic variable (i.e., one living on the stack),
5032 or the procedure object should be otherwise protected from GC.
5034 ** scm_handle_by_message is a new handler function to use with
5035 `scm_internal_catch' if you want Guile to print a message and die.
5036 It's useful for dealing with throws to uncaught keys at the top level.
5038 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
5039 message header to print; if zero, we use "guile" instead. That
5040 text is followed by a colon, then the message described by ARGS.
5042 ** The return type of scm_boot_guile is now void; the function does
5043 not return a value, and indeed, never returns at all.
5045 ** The new function scm_shell makes it easy for user applications to
5046 process command-line arguments in a way that is compatible with the
5047 stand-alone guile interpreter (which is in turn compatible with SCSH,
5050 To use the scm_shell function, first initialize any guile modules
5051 linked into your application, and then call scm_shell with the values
5052 of ARGC and ARGV your `main' function received. scm_shell will add
5053 any SCSH-style meta-arguments from the top of the script file to the
5054 argument vector, and then process the command-line arguments. This
5055 generally means loading a script file or starting up an interactive
5056 command interpreter. For details, see "Changes to the stand-alone
5059 ** The new functions scm_get_meta_args and scm_count_argv help you
5060 implement the SCSH-style meta-argument, `\'.
5062 char **scm_get_meta_args (int ARGC, char **ARGV)
5063 If the second element of ARGV is a string consisting of a single
5064 backslash character (i.e. "\\" in Scheme notation), open the file
5065 named by the following argument, parse arguments from it, and return
5066 the spliced command line. The returned array is terminated by a
5069 For details of argument parsing, see above, under "guile now accepts
5070 command-line arguments compatible with SCSH..."
5072 int scm_count_argv (char **ARGV)
5073 Count the arguments in ARGV, assuming it is terminated by a null
5076 For an example of how these functions might be used, see the source
5077 code for the function scm_shell in libguile/script.c.
5079 You will usually want to use scm_shell instead of calling this
5082 ** The new function scm_compile_shell_switches turns an array of
5083 command-line arguments into Scheme code to carry out the actions they
5084 describe. Given ARGC and ARGV, it returns a Scheme expression to
5085 evaluate, and calls scm_set_program_arguments to make any remaining
5086 command-line arguments available to the Scheme code. For example,
5087 given the following arguments:
5089 -e main -s ekko a speckled gecko
5091 scm_set_program_arguments will return the following expression:
5093 (begin (load "ekko") (main (command-line)) (quit))
5095 You will usually want to use scm_shell instead of calling this
5098 ** The function scm_shell_usage prints a usage message appropriate for
5099 an interpreter that uses scm_compile_shell_switches to handle its
5100 command-line arguments.
5102 void scm_shell_usage (int FATAL, char *MESSAGE)
5103 Print a usage message to the standard error output. If MESSAGE is
5104 non-zero, write it before the usage message, followed by a newline.
5105 If FATAL is non-zero, exit the process, using FATAL as the
5106 termination status. (If you want to be compatible with Guile,
5107 always use 1 as the exit status when terminating due to command-line
5110 You will usually want to use scm_shell instead of calling this
5113 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
5114 expressions. It used to return SCM_EOL. Earth-shattering.
5116 ** The macros for declaring scheme objects in C code have been
5117 rearranged slightly. They are now:
5119 SCM_SYMBOL (C_NAME, SCHEME_NAME)
5120 Declare a static SCM variable named C_NAME, and initialize it to
5121 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
5122 be a C identifier, and SCHEME_NAME should be a C string.
5124 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
5125 Just like SCM_SYMBOL, but make C_NAME globally visible.
5127 SCM_VCELL (C_NAME, SCHEME_NAME)
5128 Create a global variable at the Scheme level named SCHEME_NAME.
5129 Declare a static SCM variable named C_NAME, and initialize it to
5130 point to the Scheme variable's value cell.
5132 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
5133 Just like SCM_VCELL, but make C_NAME globally visible.
5135 The `guile-snarf' script writes initialization code for these macros
5136 to its standard output, given C source code as input.
5138 The SCM_GLOBAL macro is gone.
5140 ** The scm_read_line and scm_read_line_x functions have been replaced
5141 by Scheme code based on the %read-delimited! procedure (known to C
5142 code as scm_read_delimited_x). See its description above for more
5145 ** The function scm_sys_open has been renamed to scm_open. It now
5146 returns a port instead of an FD object.
5148 * The dynamic linking support has changed. For more information, see
5149 libguile/DYNAMIC-LINKING.
5154 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
5157 * Changes to the 'guile' program:
5159 ** Guile now loads some new files when it starts up. Guile first
5160 searches the load path for init.scm, and loads it if found. Then, if
5161 Guile is not being used to execute a script, and the user's home
5162 directory contains a file named `.guile', Guile loads that.
5164 ** You can now use Guile as a shell script interpreter.
5166 To paraphrase the SCSH manual:
5168 When Unix tries to execute an executable file whose first two
5169 characters are the `#!', it treats the file not as machine code to
5170 be directly executed by the native processor, but as source code
5171 to be executed by some interpreter. The interpreter to use is
5172 specified immediately after the #! sequence on the first line of
5173 the source file. The kernel reads in the name of the interpreter,
5174 and executes that instead. It passes the interpreter the source
5175 filename as its first argument, with the original arguments
5176 following. Consult the Unix man page for the `exec' system call
5177 for more information.
5179 Now you can use Guile as an interpreter, using a mechanism which is a
5180 compatible subset of that provided by SCSH.
5182 Guile now recognizes a '-s' command line switch, whose argument is the
5183 name of a file of Scheme code to load. It also treats the two
5184 characters `#!' as the start of a comment, terminated by `!#'. Thus,
5185 to make a file of Scheme code directly executable by Unix, insert the
5186 following two lines at the top of the file:
5188 #!/usr/local/bin/guile -s
5191 Guile treats the argument of the `-s' command-line switch as the name
5192 of a file of Scheme code to load, and treats the sequence `#!' as the
5193 start of a block comment, terminated by `!#'.
5195 For example, here's a version of 'echo' written in Scheme:
5197 #!/usr/local/bin/guile -s
5199 (let loop ((args (cdr (program-arguments))))
5202 (display (car args))
5203 (if (pair? (cdr args))
5205 (loop (cdr args)))))
5208 Why does `#!' start a block comment terminated by `!#', instead of the
5209 end of the line? That is the notation SCSH uses, and although we
5210 don't yet support the other SCSH features that motivate that choice,
5211 we would like to be backward-compatible with any existing Guile
5212 scripts once we do. Furthermore, if the path to Guile on your system
5213 is too long for your kernel, you can start the script with this
5217 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
5220 Note that some very old Unix systems don't support the `#!' syntax.
5223 ** You can now run Guile without installing it.
5225 Previous versions of the interactive Guile interpreter (`guile')
5226 couldn't start up unless Guile's Scheme library had been installed;
5227 they used the value of the environment variable `SCHEME_LOAD_PATH'
5228 later on in the startup process, but not to find the startup code
5229 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
5232 To run Guile without installing it, build it in the normal way, and
5233 then set the environment variable `SCHEME_LOAD_PATH' to a
5234 colon-separated list of directories, including the top-level directory
5235 of the Guile sources. For example, if you unpacked Guile so that the
5236 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
5239 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
5242 ** Guile's read-eval-print loop no longer prints #<unspecified>
5243 results. If the user wants to see this, she can evaluate the
5244 expression (assert-repl-print-unspecified #t), perhaps in her startup
5247 ** Guile no longer shows backtraces by default when an error occurs;
5248 however, it does display a message saying how to get one, and how to
5249 request that they be displayed by default. After an error, evaluate
5251 to see a backtrace, and
5252 (debug-enable 'backtrace)
5253 to see them by default.
5257 * Changes to Guile Scheme:
5259 ** Guile now distinguishes between #f and the empty list.
5261 This is for compatibility with the IEEE standard, the (possibly)
5262 upcoming Revised^5 Report on Scheme, and many extant Scheme
5265 Guile used to have #f and '() denote the same object, to make Scheme's
5266 type system more compatible with Emacs Lisp's. However, the change
5267 caused too much trouble for Scheme programmers, and we found another
5268 way to reconcile Emacs Lisp with Scheme that didn't require this.
5271 ** Guile's delq, delv, delete functions, and their destructive
5272 counterparts, delq!, delv!, and delete!, now remove all matching
5273 elements from the list, not just the first. This matches the behavior
5274 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
5275 functions which inspired them.
5277 I recognize that this change may break code in subtle ways, but it
5278 seems best to make the change before the FSF's first Guile release,
5282 ** The compiled-library-path function has been deleted from libguile.
5284 ** The facilities for loading Scheme source files have changed.
5286 *** The variable %load-path now tells Guile which directories to search
5287 for Scheme code. Its value is a list of strings, each of which names
5290 *** The variable %load-extensions now tells Guile which extensions to
5291 try appending to a filename when searching the load path. Its value
5292 is a list of strings. Its default value is ("" ".scm").
5294 *** (%search-load-path FILENAME) searches the directories listed in the
5295 value of the %load-path variable for a Scheme file named FILENAME,
5296 with all the extensions listed in %load-extensions. If it finds a
5297 match, then it returns its full filename. If FILENAME is absolute, it
5298 returns it unchanged. Otherwise, it returns #f.
5300 %search-load-path will not return matches that refer to directories.
5302 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
5303 uses %seach-load-path to find a file named FILENAME, and loads it if
5304 it finds it. If it can't read FILENAME for any reason, it throws an
5307 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
5310 *** load uses the same searching semantics as primitive-load.
5312 *** The functions %try-load, try-load-with-path, %load, load-with-path,
5313 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
5314 path, and load-module-with-path have been deleted. The functions
5315 above should serve their purposes.
5317 *** If the value of the variable %load-hook is a procedure,
5318 `primitive-load' applies its value to the name of the file being
5319 loaded (without the load path directory name prepended). If its value
5320 is #f, it is ignored. Otherwise, an error occurs.
5322 This is mostly useful for printing load notification messages.
5325 ** The function `eval!' is no longer accessible from the scheme level.
5326 We can't allow operations which introduce glocs into the scheme level,
5327 because Guile's type system can't handle these as data. Use `eval' or
5328 `read-and-eval!' (see below) as replacement.
5330 ** The new function read-and-eval! reads an expression from PORT,
5331 evaluates it, and returns the result. This is more efficient than
5332 simply calling `read' and `eval', since it is not necessary to make a
5333 copy of the expression for the evaluator to munge.
5335 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
5336 for the `read' function.
5339 ** The function `int?' has been removed; its definition was identical
5340 to that of `integer?'.
5342 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
5343 use the R4RS names for these functions.
5345 ** The function object-properties no longer returns the hash handle;
5346 it simply returns the object's property list.
5348 ** Many functions have been changed to throw errors, instead of
5349 returning #f on failure. The point of providing exception handling in
5350 the language is to simplify the logic of user code, but this is less
5351 useful if Guile's primitives don't throw exceptions.
5353 ** The function `fileno' has been renamed from `%fileno'.
5355 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
5358 * Changes to Guile's C interface:
5360 ** The library's initialization procedure has been simplified.
5361 scm_boot_guile now has the prototype:
5363 void scm_boot_guile (int ARGC,
5365 void (*main_func) (),
5368 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
5369 MAIN_FUNC should do all the work of the program (initializing other
5370 packages, reading user input, etc.) before returning. When MAIN_FUNC
5371 returns, call exit (0); this function never returns. If you want some
5372 other exit value, MAIN_FUNC may call exit itself.
5374 scm_boot_guile arranges for program-arguments to return the strings
5375 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
5376 scm_set_program_arguments with the final list, so Scheme code will
5377 know which arguments have been processed.
5379 scm_boot_guile establishes a catch-all catch handler which prints an
5380 error message and exits the process. This means that Guile exits in a
5381 coherent way when system errors occur and the user isn't prepared to
5382 handle it. If the user doesn't like this behavior, they can establish
5383 their own universal catcher in MAIN_FUNC to shadow this one.
5385 Why must the caller do all the real work from MAIN_FUNC? The garbage
5386 collector assumes that all local variables of type SCM will be above
5387 scm_boot_guile's stack frame on the stack. If you try to manipulate
5388 SCM values after this function returns, it's the luck of the draw
5389 whether the GC will be able to find the objects you allocate. So,
5390 scm_boot_guile function exits, rather than returning, to discourage
5391 people from making that mistake.
5393 The IN, OUT, and ERR arguments were removed; there are other
5394 convenient ways to override these when desired.
5396 The RESULT argument was deleted; this function should never return.
5398 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
5402 ** Guile's header files should no longer conflict with your system's
5405 In order to compile code which #included <libguile.h>, previous
5406 versions of Guile required you to add a directory containing all the
5407 Guile header files to your #include path. This was a problem, since
5408 Guile's header files have names which conflict with many systems'
5411 Now only <libguile.h> need appear in your #include path; you must
5412 refer to all Guile's other header files as <libguile/mumble.h>.
5413 Guile's installation procedure puts libguile.h in $(includedir), and
5414 the rest in $(includedir)/libguile.
5417 ** Two new C functions, scm_protect_object and scm_unprotect_object,
5418 have been added to the Guile library.
5420 scm_protect_object (OBJ) protects OBJ from the garbage collector.
5421 OBJ will not be freed, even if all other references are dropped,
5422 until someone does scm_unprotect_object (OBJ). Both functions
5425 Note that calls to scm_protect_object do not nest. You can call
5426 scm_protect_object any number of times on a given object, and the
5427 next call to scm_unprotect_object will unprotect it completely.
5429 Basically, scm_protect_object and scm_unprotect_object just
5430 maintain a list of references to things. Since the GC knows about
5431 this list, all objects it mentions stay alive. scm_protect_object
5432 adds its argument to the list; scm_unprotect_object remove its
5433 argument from the list.
5436 ** scm_eval_0str now returns the value of the last expression
5439 ** The new function scm_read_0str reads an s-expression from a
5440 null-terminated string, and returns it.
5442 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
5443 to a Scheme port object.
5445 ** The new function `scm_set_program_arguments' allows C code to set
5446 the value returned by the Scheme `program-arguments' function.
5451 * Guile no longer includes sophisticated Tcl/Tk support.
5453 The old Tcl/Tk support was unsatisfying to us, because it required the
5454 user to link against the Tcl library, as well as Tk and Guile. The
5455 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
5456 referring to widgets by names, rather than exporting widgets to Scheme
5457 code as a special datatype.
5459 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
5460 maintainers described some very interesting changes in progress to the
5461 Tcl/Tk internals, which would facilitate clean interfaces between lone
5462 Tk and other interpreters --- even for garbage-collected languages
5463 like Scheme. They expected the new Tk to be publicly available in the
5466 Since it seems that Guile might soon have a new, cleaner interface to
5467 lone Tk, and that the old Guile/Tk glue code would probably need to be
5468 completely rewritten, we (Jim Blandy and Richard Stallman) have
5469 decided not to support the old code. We'll spend the time instead on
5470 a good interface to the newer Tk, as soon as it is available.
5472 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
5475 Copyright information:
5477 Copyright (C) 1996,1997 Free Software Foundation, Inc.
5479 Permission is granted to anyone to make or distribute verbatim copies
5480 of this document as received, in any medium, provided that the
5481 copyright notice and this permission notice are preserved,
5482 thus giving the recipient permission to redistribute in turn.
5484 Permission is granted to distribute modified versions
5485 of this document, or of portions of it,
5486 under the above conditions, provided also that they
5487 carry prominent notices stating who last changed them.
5492 paragraph-separate: "[
\f]*$"