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 ** New module (ice-9 stack-catch):
13 stack-catch is like catch, but saves the current state of the stack in
14 the the-last-stack fluid for the debugger to inspect or in able to
17 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
19 This has been done to prevent problems on lesser operating systems
20 that can't tolerate `*'s in file names. The exported macro continues
21 to be named `and-let*', of course.
23 On systems that support it, there is also a compatibility module named
24 (ice-9 and-let*). It will go away in the next release.
26 ** New modules (oop goops) etc.:
31 (oop goops active-slot)
32 (oop goops composite-slot)
34 The Guile Object Oriented Programming System (GOOPS) has been
35 integrated into Guile.
39 (use-modules (oop goops))
41 access GOOPS bindings.
43 We're now ready to try some basic GOOPS functionality.
47 (define-method (+ (x <string>) (y <string>))
51 (+ "abc" "de") --> "abcde"
55 (define-class <2D-vector> ()
56 (x #:init-value 0 #:accessor x-component #:init-keyword #:x)
57 (y #:init-value 0 #:accessor y-component #:init-keyword #:y))
59 (define-method write ((obj <2D-vector>) port)
60 (display (format #f "<~S, ~S>" (x-component obj) (y-component obj))
63 (define v (make <2D-vector> #:x 3 #:y 4))
66 (define-method + ((x <2D-vector>) (y <2D-vector>))
68 #:x (+ (x-component x) (x-component y))
69 #:y (+ (y-component x) (y-component y))))
73 Asking for the type of an object
75 (class-of v) --> #<<class> <2D-vector> 40241ac0>
76 <2D-vector> --> #<<class> <2D-vector> 40241ac0>
77 (class-of 1) --> #<<class> <integer> 401b2a98>
78 <integer> --> #<<class> <integer> 401b2a98>
80 (is-a? v <2D-vector>) --> #t
82 See further in the GOOPS manual and tutorial in the `doc' directory,
83 in info (goops.info) and texinfo formats.
85 ** New module (ice-9 rdelim).
87 This exports the following procedures which were previously defined
88 in the default environment:
90 read-line read-line! read-delimited read-delimited! %read-delimited!
93 For backwards compatibility the definitions are still imported into the
94 default environment in this version of Guile. However you should add:
96 (use-modules (ice-9 rdelim))
98 to any program which uses the definitions, since this may change in
101 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
102 can be used for similar functionality.
104 ** New module (ice-9 match)
106 This module includes Andrew K. Wright's pattern matcher:
108 (use-modules (ice-9 match))
112 (('+ x y) `(add ,x ,y))
113 (('- x y) `(sub ,x ,y))) => (add 1 2)
115 See ice-9/match.scm for brief description or
116 http://www.star-lab.com/wright/code.html for complete documentation.
118 This module requires SLIB to be installed and available from Guile.
120 ** New module (ice-9 buffered-input)
122 This module provides procedures to construct an input port from an
123 underlying source of input that reads and returns its input in chunks.
124 The underlying input source is a Scheme procedure, specified by the
125 caller, which the port invokes whenever it needs more input.
127 This is useful when building an input port whose back end is Readline
128 or a UI element such as the GtkEntry widget.
132 The reference and tutorial documentation that was previously
133 distributed separately, as `guile-doc', is now included in the core
134 Guile distribution. The documentation consists of the following
137 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
140 - The Guile Reference Manual (guile.texi) contains (or is intended to
141 contain) reference documentation on all aspects of Guile.
143 - The GOOPS Manual (goops.texi) contains both tutorial-style and
144 reference documentation for using GOOPS, Guile's Object Oriented
147 - The Revised^4 and Revised^5 Reports on the Algorithmic Language
148 Scheme (r4rs.texi and r5rs.texi).
150 See the README file in the `doc' directory for more details.
152 * Changes to the stand-alone interpreter
154 ** Evaluation of "()", the empty list, is now an error.
156 Previously, you could for example write (cons 1 ()); now you need to
157 be more explicit and write (cons 1 '()).
159 ** It's now possible to create modules with controlled environments
163 (use-modules (ice-9 safe))
164 (define m (make-safe-module))
165 ;;; m will now be a module containing only a safe subset of R5RS
166 (eval '(+ 1 2) m) --> 3
167 (eval 'load m) --> ERROR: Unbound variable: load
169 * Changes to Scheme functions and syntax
171 ** The empty combination is no longer valid syntax.
173 Previously, the expression "()" evaluated to the empty list. This has
174 been changed to signal a "missing expression" error. The correct way
175 to write the empty list as a literal constant is to use quote: "'()".
177 ** Auto-loading of compiled-code modules is deprecated.
179 Guile used to be able to automatically find and link a shared
180 libraries to satisfy requests for a module. For example, the module
181 `(foo bar)' could be implemented by placing a shared library named
182 "foo/libbar.so" (or with a different extension) in a directory on the
185 This has been found to be too tricky, and is no longer supported.
186 What you should do instead now is to write a small Scheme file that
187 explicitly calls `dynamic-link' to load the shared library and
188 `dynamic-call' to initialize it.
190 The shared libraries themselves should be installed in the usual
191 places for shared libraries, with names like "libguile-foo-bar".
193 For example, place this into a file "foo/bar.scm"
195 (define-module (foo bar))
197 (dynamic-call "foobar_init" (dynamic-link "libguile-foo-bar"))
199 The file name passed to `dynamic-link' should not contain an
200 extension. It will be provided automatically.
202 ** The module system has been made more disciplined.
204 The function `eval' will now save and restore the current module
205 around the evaluation of the specified expression. While this
206 expression is evaluated, `(current-module)' will now return the right
207 module, which is the module specified as the second argument to
210 A consequence of this change is that `eval' is not particularily
211 useful when you want allow the evaluated code to change what module is
212 designated as the current module and have this change persist from one
213 call to `eval' to the next. The read-eval-print-loop is an example
214 where `eval' is now inadequate. To compensate, there is a new
215 function `primitive-eval' that does not take a module specifier and
216 that does not save/restore the current module. You should use this
217 function together with `set-current-module', `current-module', etc
218 when you want to have more control over the state that is carried from
219 one eval to the next.
221 Additionally, it has been made sure that forms that are evaluated at
222 the top level are always evaluated with respect to the current module.
223 Previously, subforms of top-level forms such as `begin', `case',
224 etc. did not respect changes to the current module although these
225 subforms are at the top-level as well.
227 To prevent strange behaviour, the forms `define-module',
228 `use-modules', `use-syntax', and `export' have been restricted to only
229 work on the top level. The forms `define-public' and
230 `defmacro-public' only export the new binding on the top level. They
231 behave just like `define' and `defmacro', respectively, when they are
232 used in a lexical environment.
234 ** `port-for-each' makes an additional guarantee.
236 From the docstring: @var{proc} is applied exactly once to every port
237 that exists in the system at the time @var{port-for-each} is invoked.
238 Changes to the port table while @var{port-for-each} is running have no
239 effect as far as @var{port-for-each} is concerned.
241 This guarantee is important to make (ice-9 popen) work reliable.
243 ** The semantics of guardians have changed.
245 The changes are for the most part compatible. An important criterion
246 was to keep the typical usage of guardians as simple as before, but to
247 make the semantics safer and (as a result) more useful.
249 *** All objects returned from guardians are now properly alive.
251 It is now guaranteed that any object referenced by an object returned
252 from a guardian is alive. It's now impossible for a guardian to
253 return a "contained" object before its "containing" object.
255 One incompatible (but probably not very important) change resulting
256 from this is that it is no longer possible to guard objects that
257 indirectly reference themselves (i.e. are parts of cycles). If you do
258 so accidentally, you'll get a warning.
260 *** There are now two types of guardians: greedy and sharing.
262 If you call (make-guardian #t) or just (make-guardian), you'll get a
263 greedy guardian, and for (make-guardian #f) a sharing guardian.
265 Greedy guardians are the default because they are more "defensive".
266 You can only greedily guard an object once. If you guard an object
267 more than once, once in a greedy guardian and the rest of times in
268 sharing guardians, then it is guaranteed that the object won't be
269 returned from sharing guardians as long as it is greedily guarded
272 Guardians returned by calls to `make-guardian' can now take one more
273 optional parameter, which says whether to throw an error in case an
274 attempt is made to greedily guard an object that is already greedily
275 guarded. The default is true, i.e. throw an error. If the parameter
276 is false, the guardian invocation returns #t if guarding was
277 successful and #f if it wasn't.
279 Also, since greedy guarding is, in effect, a side-effecting operation
280 on objects, a new function is introduced: `destroy-guardian!'.
281 Invoking this function on a guardian renders it unoperative and, if
282 the guardian is greedy, clears the "greedily guarded" property of the
283 objects that were guarded by it, thus undoing the side effect.
285 Note that all this hair is hardly very important, since guardian
286 objects are usually permanent.
288 ** Escape procedures created by call-with-current-continuation now
289 accept any number of arguments, as required by R5RS.
291 ** New function `call-with-deprecation'
293 Call a thunk, displaying a deprecation message at the first call:
296 (call-with-deprecation "`id' is deprecated. Use `identity' instead."
301 ;; `id' is deprecated. Use `identity' instead.
306 ** New function `make-object-property'
308 This function returns a new `procedure with setter' P that can be used
309 to attach a property to objects. When calling P as
313 where `obj' is any kind of object, it attaches `val' to `obj' in such
314 a way that it can be retrieved by calling P as
318 This function will replace procedure properties, symbol properties and
319 source properties eventually.
321 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
323 Instead of #&optional, #&key, etc you should now use #:optional,
324 #:key, etc. Since #:optional is a keyword, you can write it as just
325 :optional when (read-set! keywords 'prefix) is active.
327 The old reader syntax `#&' is still supported, but deprecated. It
328 will be removed in the next release.
330 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
332 `eval' is now R5RS, that is it takes two arguments.
333 The second argument is an environment specifier, i.e. either
335 (scheme-report-environment 5)
337 (interaction-environment)
343 ** New define-module option: pure
345 Tells the module system not to include any bindings from the root
350 (define-module (totally-empty-module)
353 ** New define-module option: export NAME1 ...
355 Export names NAME1 ...
357 This option is required if you want to be able to export bindings from
358 a module which doesn't import one of `define-public' or `export'.
364 :use-module (ice-9 r5rs)
367 ;;; Note that we're pure R5RS below this point!
372 ** Deprecated: scm_make_shared_substring
374 Explicit shared substrings will disappear from Guile.
376 Instead, "normal" strings will be implemented using sharing
377 internally, combined with a copy-on-write strategy.
379 ** Deprecated: scm_read_only_string_p
381 The concept of read-only strings will disappear in next release of
384 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
386 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
388 ** New function: read-string!/partial str [port_or_fdes [start [end]]]
390 Read characters from an fport or file descriptor into a string
391 STR. This procedure is scsh-compatible and can efficiently read
392 large strings. It will:
394 * attempt to fill the entire string, unless the START and/or
395 END arguments are supplied. i.e., START defaults to 0 and
396 END defaults to `(string-length str)'
398 * use the current input port if PORT_OR_FDES is not supplied.
400 * read any characters that are currently available, without
401 waiting for the rest (short reads are possible).
403 * wait for as long as it needs to for the first character to
404 become available, unless the port is in non-blocking mode
406 * return `#f' if end-of-file is encountered before reading any
407 characters, otherwise return the number of characters read.
409 * return 0 if the port is in non-blocking mode and no characters
410 are immediately available.
412 * return 0 if the request is for 0 bytes, with no end-of-file
415 ** New function: object->string OBJ
417 Return a Scheme string obtained by printing a given object.
419 ** New function: port? X
421 Returns a boolean indicating whether X is a port. Equivalent to
422 `(or (input-port? X) (output-port? X))'.
424 ** New function: file-port?
426 Determines whether a given object is a port that is related to a file.
428 ** New function: port-for-each proc
430 Apply PROC to each port in the Guile port table in turn. The
431 return value is unspecified.
433 ** New function: dup2 oldfd newfd
435 A simple wrapper for the `dup2' system call. Copies the file
436 descriptor OLDFD to descriptor number NEWFD, replacing the
437 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
438 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
439 to move away ports which are using NEWFD. The return value is
442 ** New function: close-fdes fd
444 A simple wrapper for the `close' system call. Close file
445 descriptor FD, which must be an integer. Unlike close (*note
446 close: Ports and File Descriptors.), the file descriptor will be
447 closed even if a port is using it. The return value is
450 ** New function: crypt password salt
452 Encrypts `password' using the standard unix password encryption
455 ** New function: chroot path
457 Change the root directory of the running process to `path'.
459 ** New functions: getlogin, cuserid
461 Return the login name or the user name of the current effective user
464 ** New functions: getpriority which who, setpriority which who prio
466 Get or set the priority of the running process.
468 ** New function: getpass prompt
470 Read a password from the terminal, first displaying `prompt' and
473 ** New function: flock file operation
475 Set/remove an advisory shared or exclusive lock on `file'.
477 ** New functions: sethostname name, gethostname
479 Set or get the hostname of the machine the current process is running
482 ** New function: mkstemp! tmpl
484 mkstemp creates a new unique file in the file system and returns a
485 new buffered port open for reading and writing to the file. TMPL
486 is a string specifying where the file should be created: it must
487 end with `XXXXXX' and will be changed in place to return the name
488 of the temporary file.
490 ** New function: open-input-string string
492 Return an input string port which delivers the characters from
493 `string'. This procedure, together with `open-output-string' and
494 `get-output-string' implements SRFI-6.
496 ** New function: open-output-string
498 Return an output string port which collects all data written to it.
499 The data can then be retrieved by `get-output-string'.
501 ** New function: get-output-string
503 Return the contents of an output string port.
505 ** New function: identity
511 Use `identity' instead.
513 ** Deprecated: close-all-ports-except.
515 This was intended for closing ports in a child process after a fork,
516 but it has the undesirable side effect of flushing buffers.
517 port-for-each is more flexible.
519 ** The (ice-9 popen) module now attempts to set up file descriptors in
520 the child process from the current Scheme ports, instead of using the
521 current values of file descriptors 0, 1, and 2 in the parent process.
523 ** Removed function: builtin-weak-bindings
525 There is no such concept as a weak binding any more.
527 ** Removed constants: bignum-radix, scm-line-incrementors
529 ** define-method: New syntax mandatory.
531 The new method syntax is now mandatory:
533 (define-method (NAME ARG-SPEC ...) BODY ...)
534 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
536 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
537 REST-ARG ::= ARG-NAME
539 If you have old code using the old syntax, import
540 (oop goops old-define-method) before (oop goops) as in:
542 (use-modules (oop goops old-define-method) (oop goops))
544 * Changes to the gh_ interface
546 * Changes to the scm_ interface
548 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
550 Used by an application to read arbitrary number of bytes from a port.
551 Same semantics as libc read, except that scm_c_read only returns less
552 than SIZE bytes if at end-of-file.
554 Warning: Doesn't update port line and column counts!
556 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
558 Used by an application to write arbitrary number of bytes to an SCM
559 port. Similar semantics as libc write. However, unlike libc
560 write, scm_c_write writes the requested number of bytes and has no
563 Warning: Doesn't update port line and column counts!
565 ** New function: scm_init_guile ()
567 In contrast to scm_boot_guile, scm_init_guile will return normally
568 after initializing Guile. It is not available on all systems, tho.
570 ** New functions: scm_str2symbol, scm_mem2symbol
572 The function scm_str2symbol takes a const char* pointing to a zero-terminated
573 field of characters and creates a scheme symbol object from that C string.
574 The function scm_mem2symbol takes a const char* and a number of characters and
575 creates a symbol from the characters in that memory area.
577 ** New functions: scm_primitive_make_property
578 scm_primitive_property_ref
579 scm_primitive_property_set_x
580 scm_primitive_property_del_x
582 These functions implement a new way to deal with object properties.
583 See libguile/properties.c for their documentation.
585 ** New function: scm_done_free (long size)
587 This function is the inverse of scm_done_malloc. Use it to report the
588 amount of smob memory you free. The previous method, which involved
589 calling scm_done_malloc with negative argument, was somewhat
590 unintuitive (and is still available, of course).
592 ** New function: scm_c_memq (SCM obj, SCM list)
594 This function provides a fast C level alternative for scm_memq for the case
595 that the list parameter is known to be a proper list. The function is a
596 replacement for scm_sloppy_memq, but is stricter in its requirements on its
597 list input parameter, since for anything else but a proper list the function's
598 behaviour is undefined - it may even crash or loop endlessly. Further, for
599 the case that the object is not found in the list, scm_c_memq returns #f which
600 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
602 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
603 scm_remember_upto_here
605 These functions replace the function scm_remember.
607 ** Deprecated function: scm_remember
609 Use one of the new functions scm_remember_upto_here_1,
610 scm_remember_upto_here_2 or scm_remember_upto_here instead.
612 ** New function: scm_allocate_string
614 This function replaces the function scm_makstr.
616 ** Deprecated function: scm_makstr
618 Use the new function scm_allocate_string instead.
620 ** New global variable scm_gc_running_p introduced.
622 Use this variable to find out if garbage collection is being executed. Up to
623 now applications have used scm_gc_heap_lock to test if garbage collection was
624 running, which also works because of the fact that up to know only the garbage
625 collector has set this variable. But, this is an implementation detail that
626 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
627 of this variable is (and has been) not fully safe anyway.
629 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
631 Use these instead of SCM_LENGTH_MAX.
633 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
634 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
635 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
637 Use these instead of SCM_LENGTH.
639 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
640 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
641 SCM_SET_BITVECTOR_LENGTH
643 Use these instead of SCM_SETLENGTH
645 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
646 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
649 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
652 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
653 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
656 Use these instead of SCM_SETCHARS.
658 ** New macro: SCM_BITVECTOR_P
660 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
662 Use instead of SCM_COERCE_SUBSTR.
664 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
666 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
668 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
669 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
670 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
671 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
672 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
673 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
674 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
675 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
676 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
677 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
678 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
679 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
680 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
681 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
682 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
684 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
685 Use scm_memory_error instead of SCM_NALLOC.
686 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
687 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
688 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
689 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
690 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
691 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
692 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
693 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
694 Use SCM_STRINGP instead of SCM_RWSTRINGP.
695 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
696 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
697 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
698 Use a type specific setter macro instead of SCM_SETLENGTH.
699 Use a type specific setter macro instead of SCM_SETCHARS.
700 Use a type specific length macro instead of SCM_LENGTH_MAX.
701 Use SCM_GCMARKP instead of SCM_GC8MARKP.
702 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
703 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
704 Use SCM_TYP16 instead of SCM_GCTYP16.
705 Use SCM_CDR instead of SCM_GCCDR.
706 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
707 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
708 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
709 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
710 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
711 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
712 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
714 ** Removed function: scm_struct_init
716 ** Removed variable: scm_symhash_dim
718 ** Renamed function: scm_make_cont has been replaced by
719 scm_make_continuation, which has a different interface.
721 ** Deprecated function: scm_call_catching_errors
723 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
725 ** Deprecated function: scm_strhash
727 Use scm_string_hash instead.
729 ** Deprecated function: scm_vector_set_length_x
731 Instead, create a fresh vector of the desired size and copy the contents.
733 ** scm_gensym has changed prototype
735 scm_gensym now only takes one argument.
737 ** New function: scm_gentemp (SCM prefix, SCM obarray)
739 The builtin `gentemp' has now become a primitive.
741 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
744 There is now only a single symbol type scm_tc7_symbol.
745 The tag scm_tc7_lvector was not used anyway.
747 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
749 Use scm_make_smob_type and scm_set_smob_XXX instead.
751 ** New function scm_set_smob_apply.
753 This can be used to set an apply function to a smob type.
755 ** Deprecated function: scm_strprint_obj
757 Use scm_object_to_string instead.
759 ** Deprecated function: scm_wta
761 Use scm_wrong_type_arg, or another appropriate error signalling function
765 Changes since Guile 1.3.4:
767 * Changes to the distribution
769 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
771 We've changed the way we handle generated files in the Guile source
772 repository. As a result, the procedure for building trees obtained
773 from the nightly FTP snapshots or via CVS has changed:
774 - You must have appropriate versions of autoconf, automake, and
775 libtool installed on your system. See README for info on how to
776 obtain these programs.
777 - Before configuring the tree, you must first run the script
778 `autogen.sh' at the top of the source tree.
780 The Guile repository used to contain not only source files, written by
781 humans, but also some generated files, like configure scripts and
782 Makefile.in files. Even though the contents of these files could be
783 derived mechanically from other files present, we thought it would
784 make the tree easier to build if we checked them into CVS.
786 However, this approach means that minor differences between
787 developer's installed tools and habits affected the whole team.
788 So we have removed the generated files from the repository, and
789 added the autogen.sh script, which will reconstruct them
793 ** configure now has experimental options to remove support for certain
796 --disable-arrays omit array and uniform array support
797 --disable-posix omit posix interfaces
798 --disable-networking omit networking interfaces
799 --disable-regex omit regular expression interfaces
801 These are likely to become separate modules some day.
803 ** New configure option --enable-debug-freelist
805 This enables a debugging version of SCM_NEWCELL(), and also registers
806 an extra primitive, the setter `gc-set-debug-check-freelist!'.
808 Configure with the --enable-debug-freelist option to enable
809 the gc-set-debug-check-freelist! primitive, and then use:
811 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
812 (gc-set-debug-check-freelist! #f) # turn off checking
814 Checking of the freelist forces a traversal of the freelist and
815 a garbage collection before each allocation of a cell. This can
816 slow down the interpreter dramatically, so the setter should be used to
817 turn on this extra processing only when necessary.
819 ** New configure option --enable-debug-malloc
821 Include code for debugging of calls to scm_must_malloc/realloc/free.
825 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
826 2. objects reallocated by scm_must_realloc has been allocated by
828 3. reallocated objects are reallocated with the same what string
830 But, most importantly, it records the number of allocated objects of
831 each kind. This is useful when searching for memory leaks.
833 A Guile compiled with this option provides the primitive
834 `malloc-stats' which returns an alist with pairs of kind and the
835 number of objects of that kind.
837 ** All includes are now referenced relative to the root directory
839 Since some users have had problems with mixups between Guile and
840 system headers, we have decided to always refer to Guile headers via
841 their parent directories. This essentially creates a "private name
842 space" for Guile headers. This means that the compiler only is given
843 -I options for the root build and root source directory.
845 ** Header files kw.h and genio.h have been removed.
847 ** The module (ice-9 getopt-gnu-style) has been removed.
849 ** New module (ice-9 documentation)
851 Implements the interface to documentation strings associated with
854 ** New module (ice-9 time)
856 Provides a macro `time', which displays execution time of a given form.
858 ** New module (ice-9 history)
860 Loading this module enables value history in the repl.
862 * Changes to the stand-alone interpreter
864 ** New command line option --debug
866 Start Guile with debugging evaluator and backtraces enabled.
868 This is useful when debugging your .guile init file or scripts.
872 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
873 (help REGEXP) ditto for objects with names matching REGEXP (a string)
874 (help ,EXPR) gives documentation for object returned by EXPR
875 (help) gives this text
877 `help' searches among bindings exported from loaded modules, while
878 `apropos' searches among bindings visible from the "current" module.
880 Examples: (help help)
882 (help "output-string")
884 ** `help' and `apropos' now prints full module names
886 ** Dynamic linking now uses libltdl from the libtool package.
888 The old system dependent code for doing dynamic linking has been
889 replaced with calls to the libltdl functions which do all the hairy
892 The major improvement is that you can now directly pass libtool
893 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
894 will be able to do the best shared library job you can get, via
897 The way dynamic libraries are found has changed and is not really
898 portable across platforms, probably. It is therefore recommended to
899 use absolute filenames when possible.
901 If you pass a filename without an extension to `dynamic-link', it will
902 try a few appropriate ones. Thus, the most platform ignorant way is
903 to specify a name like "libfoo", without any directories and
906 ** Guile COOP threads are now compatible with LinuxThreads
908 Previously, COOP threading wasn't possible in applications linked with
909 Linux POSIX threads due to their use of the stack pointer to find the
910 thread context. This has now been fixed with a workaround which uses
911 the pthreads to allocate the stack.
913 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
915 ** Positions of erring expression in scripts
917 With version 1.3.4, the location of the erring expression in Guile
918 scipts is no longer automatically reported. (This should have been
919 documented before the 1.3.4 release.)
921 You can get this information by enabling recording of positions of
922 source expressions and running the debugging evaluator. Put this at
923 the top of your script (or in your "site" file):
925 (read-enable 'positions)
926 (debug-enable 'debug)
928 ** Backtraces in scripts
930 It is now possible to get backtraces in scripts.
934 (debug-enable 'debug 'backtrace)
936 at the top of the script.
938 (The first options enables the debugging evaluator.
939 The second enables backtraces.)
941 ** Part of module system symbol lookup now implemented in C
943 The eval closure of most modules is now implemented in C. Since this
944 was one of the bottlenecks for loading speed, Guile now loads code
945 substantially faster than before.
947 ** Attempting to get the value of an unbound variable now produces
948 an exception with a key of 'unbound-variable instead of 'misc-error.
950 ** The initial default output port is now unbuffered if it's using a
951 tty device. Previously in this situation it was line-buffered.
953 ** gc-thunk is deprecated
955 gc-thunk will be removed in next release of Guile. It has been
956 replaced by after-gc-hook.
958 ** New hook: after-gc-hook
960 after-gc-hook takes over the role of gc-thunk. This hook is run at
961 the first SCM_TICK after a GC. (Thus, the code is run at the same
962 point during evaluation as signal handlers.)
964 Note that this hook should be used only for diagnostic and debugging
965 purposes. It is not certain that it will continue to be well-defined
966 when this hook is run in the future.
968 C programmers: Note the new C level hooks scm_before_gc_c_hook,
969 scm_before_sweep_c_hook, scm_after_gc_c_hook.
971 ** Improvements to garbage collector
973 Guile 1.4 has a new policy for triggering heap allocation and
974 determining the sizes of heap segments. It fixes a number of problems
977 1. The new policy can handle two separate pools of cells
978 (2-word/4-word) better. (The old policy would run wild, allocating
979 more and more memory for certain programs.)
981 2. The old code would sometimes allocate far too much heap so that the
982 Guile process became gigantic. The new code avoids this.
984 3. The old code would sometimes allocate too little so that few cells
985 were freed at GC so that, in turn, too much time was spent in GC.
987 4. The old code would often trigger heap allocation several times in a
988 row. (The new scheme predicts how large the segments needs to be
989 in order not to need further allocation.)
991 All in all, the new GC policy will make larger applications more
994 The new GC scheme also is prepared for POSIX threading. Threads can
995 allocate private pools of cells ("clusters") with just a single
996 function call. Allocation of single cells from such a cluster can
997 then proceed without any need of inter-thread synchronization.
999 ** New environment variables controlling GC parameters
1001 GUILE_MAX_SEGMENT_SIZE Maximal segment size
1004 Allocation of 2-word cell heaps:
1006 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
1009 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
1010 GC in percent of total heap size
1013 Allocation of 4-word cell heaps
1014 (used for real numbers and misc other objects):
1016 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
1018 (See entry "Way for application to customize GC parameters" under
1019 section "Changes to the scm_ interface" below.)
1021 ** Guile now implements reals using 4-word cells
1023 This speeds up computation with reals. (They were earlier allocated
1024 with `malloc'.) There is still some room for optimizations, however.
1026 ** Some further steps toward POSIX thread support have been taken
1028 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
1029 don't have much effect any longer, and many of them will be removed in
1033 are only handled at the top of the evaluator loop, immediately after
1034 I/O, and in scm_equalp.
1036 *** The GC can allocate thread private pools of pairs.
1038 * Changes to Scheme functions and syntax
1040 ** close-input-port and close-output-port are now R5RS
1042 These procedures have been turned into primitives and have R5RS behaviour.
1044 ** New procedure: simple-format PORT MESSAGE ARG1 ...
1046 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
1047 extended by the more sophisticated version in (ice-9 format)
1049 (simple-format port message . args)
1050 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
1051 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
1052 the escapes are replaced with corresponding members of ARGS:
1053 ~A formats using `display' and ~S formats using `write'.
1054 If DESTINATION is #t, then use the `current-output-port',
1055 if DESTINATION is #f, then return a string containing the formatted text.
1056 Does not add a trailing newline."
1058 ** string-ref: the second argument is no longer optional.
1060 ** string, list->string: no longer accept strings in their arguments,
1061 only characters, for compatibility with R5RS.
1063 ** New procedure: port-closed? PORT
1064 Returns #t if PORT is closed or #f if it is open.
1066 ** Deprecated: list*
1068 The list* functionality is now provided by cons* (SRFI-1 compliant)
1070 ** New procedure: cons* ARG1 ARG2 ... ARGn
1072 Like `list', but the last arg provides the tail of the constructed list,
1073 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
1075 Requires at least one argument. If given one argument, that argument
1076 is returned as result.
1078 This function is called `list*' in some other Schemes and in Common LISP.
1080 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
1082 ** New procedure: object-documentation OBJECT
1084 Returns the documentation string associated with OBJECT. The
1085 procedure uses a caching mechanism so that subsequent lookups are
1088 Exported by (ice-9 documentation).
1090 ** module-name now returns full names of modules
1092 Previously, only the last part of the name was returned (`session' for
1093 `(ice-9 session)'). Ex: `(ice-9 session)'.
1095 * Changes to the gh_ interface
1097 ** Deprecated: gh_int2scmb
1099 Use gh_bool2scm instead.
1101 * Changes to the scm_ interface
1103 ** Guile primitives now carry docstrings!
1105 Thanks to Greg Badros!
1107 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1109 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
1110 macros and must contain a docstring that is extracted into foo.doc using a new
1111 guile-doc-snarf script (that uses guile-doc-snarf.awk).
1113 However, a major overhaul of these macros is scheduled for the next release of
1116 ** Guile primitives use a new technique for validation of arguments
1118 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
1119 the readability of argument checking.
1121 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
1123 ** New macros: SCM_PACK, SCM_UNPACK
1125 Compose/decompose an SCM value.
1127 The SCM type is now treated as an abstract data type and may be defined as a
1128 long, a void* or as a struct, depending on the architecture and compile time
1129 options. This makes it easier to find several types of bugs, for example when
1130 SCM values are treated as integers without conversion. Values of the SCM type
1131 should be treated as "atomic" values. These macros are used when
1132 composing/decomposing an SCM value, either because you want to access
1133 individual bits, or because you want to treat it as an integer value.
1135 E.g., in order to set bit 7 in an SCM value x, use the expression
1137 SCM_PACK (SCM_UNPACK (x) | 0x80)
1139 ** The name property of hooks is deprecated.
1140 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
1142 You can emulate this feature by using object properties.
1144 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
1145 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
1148 These macros will be removed in a future release of Guile.
1150 ** The following types, functions and macros from numbers.h are deprecated:
1151 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
1152 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
1154 Further, it is recommended not to rely on implementation details for guile's
1155 current implementation of bignums. It is planned to replace this
1156 implementation with gmp in the future.
1158 ** Port internals: the rw_random variable in the scm_port structure
1159 must be set to non-zero in any random access port. In recent Guile
1160 releases it was only set for bidirectional random-access ports.
1162 ** Port internals: the seek ptob procedure is now responsible for
1163 resetting the buffers if required. The change was made so that in the
1164 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
1165 the fport and strport ptobs can avoid resetting the buffers,
1166 in particular to avoid discarding unread chars. An existing port
1167 type can be fixed by adding something like the following to the
1168 beginning of the ptob seek procedure:
1170 if (pt->rw_active == SCM_PORT_READ)
1171 scm_end_input (object);
1172 else if (pt->rw_active == SCM_PORT_WRITE)
1173 ptob->flush (object);
1175 although to actually avoid resetting the buffers and discard unread
1176 chars requires further hacking that depends on the characteristics
1179 ** Deprecated functions: scm_fseek, scm_tag
1181 These functions are no longer used and will be removed in a future version.
1183 ** The scm_sysmissing procedure is no longer used in libguile.
1184 Unless it turns out to be unexpectedly useful to somebody, it will be
1185 removed in a future version.
1187 ** The format of error message strings has changed
1189 The two C procedures: scm_display_error and scm_error, as well as the
1190 primitive `scm-error', now use scm_simple_format to do their work.
1191 This means that the message strings of all code must be updated to use
1192 ~A where %s was used before, and ~S where %S was used before.
1194 During the period when there still are a lot of old Guiles out there,
1195 you might want to support both old and new versions of Guile.
1197 There are basically two methods to achieve this. Both methods use
1200 AC_CHECK_FUNCS(scm_simple_format)
1202 in your configure.in.
1204 Method 1: Use the string concatenation features of ANSI C's
1209 #ifdef HAVE_SCM_SIMPLE_FORMAT
1215 Then represent each of your error messages using a preprocessor macro:
1217 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
1221 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
1222 (define make-message string-append)
1224 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
1226 Method 2: Use the oldfmt function found in doc/oldfmt.c.
1230 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
1235 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
1239 ** Deprecated: coop_mutex_init, coop_condition_variable_init
1241 Don't use the functions coop_mutex_init and
1242 coop_condition_variable_init. They will change.
1244 Use scm_mutex_init and scm_cond_init instead.
1246 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
1247 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
1248 COND, as `scm_cond_wait' does, but it also bounds the duration
1249 of the wait. If COND has not been signaled before time ABSTIME,
1250 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
1251 returns the error code `ETIMEDOUT'.
1253 The ABSTIME parameter specifies an absolute time, with the same
1254 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
1255 to 00:00:00 GMT, January 1, 1970.
1257 ** New function: scm_cond_broadcast (scm_cond_t *COND)
1258 `scm_cond_broadcast' restarts all the threads that are waiting
1259 on the condition variable COND. Nothing happens if no threads are
1262 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
1263 `scm_key_create' allocates a new TSD key. The key is stored in
1264 the location pointed to by KEY. There is no limit on the number
1265 of keys allocated at a given time. The value initially associated
1266 with the returned key is `NULL' in all currently executing threads.
1268 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
1269 function associated with the key. When a thread terminates,
1270 DESTR_FUNCTION is called on the value associated with the key in
1271 that thread. The DESTR_FUNCTION is not called if a key is deleted
1272 with `scm_key_delete' or a value is changed with
1273 `scm_setspecific'. The order in which destructor functions are
1274 called at thread termination time is unspecified.
1276 Destructors are not yet implemented.
1278 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
1279 `scm_setspecific' changes the value associated with KEY in the
1280 calling thread, storing the given POINTER instead.
1282 ** New function: scm_getspecific (scm_key_t KEY)
1283 `scm_getspecific' returns the value currently associated with
1284 KEY in the calling thread.
1286 ** New function: scm_key_delete (scm_key_t KEY)
1287 `scm_key_delete' deallocates a TSD key. It does not check
1288 whether non-`NULL' values are associated with that key in the
1289 currently executing threads, nor call the destructor function
1290 associated with the key.
1292 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
1294 Initialize a C level hook HOOK with associated HOOK_DATA and type
1295 TYPE. (See scm_c_hook_run ().)
1297 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
1299 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
1300 is true, add it last, otherwise first. The same FUNC can be added
1301 multiple times if FUNC_DATA differ and vice versa.
1303 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
1305 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
1306 function is only removed if both FUNC and FUNC_DATA matches.
1308 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
1310 Run hook HOOK passing DATA to the hook functions.
1312 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
1313 returned is undefined.
1315 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
1316 returns a non-NULL value. This value is returned as the result of
1317 scm_c_hook_run. If all functions return NULL, NULL is returned.
1319 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
1320 returns a NULL value, and NULL is returned. If all functions returns
1321 a non-NULL value, the last value is returned.
1323 ** New C level GC hooks
1325 Five new C level hooks has been added to the garbage collector.
1327 scm_before_gc_c_hook
1330 are run before locking and after unlocking the heap. The system is
1331 thus in a mode where evaluation can take place. (Except that
1332 scm_before_gc_c_hook must not allocate new cells.)
1334 scm_before_mark_c_hook
1335 scm_before_sweep_c_hook
1336 scm_after_sweep_c_hook
1338 are run when the heap is locked. These are intended for extension of
1339 the GC in a modular fashion. Examples are the weaks and guardians
1342 ** Way for application to customize GC parameters
1344 The application can set up other default values for the GC heap
1345 allocation parameters
1347 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
1348 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
1349 GUILE_MAX_SEGMENT_SIZE,
1353 scm_default_init_heap_size_1, scm_default_min_yield_1,
1354 scm_default_init_heap_size_2, scm_default_min_yield_2,
1355 scm_default_max_segment_size
1357 respectively before callong scm_boot_guile.
1359 (See entry "New environment variables ..." in section
1360 "Changes to the stand-alone interpreter" above.)
1362 ** scm_protect_object/scm_unprotect_object now nest
1364 This means that you can call scm_protect_object multiple times on an
1365 object and count on the object being protected until
1366 scm_unprotect_object has been call the same number of times.
1368 The functions also have better time complexity.
1370 Still, it is usually possible to structure the application in a way
1371 that you don't need to use these functions. For example, if you use a
1372 protected standard Guile list to keep track of live objects rather
1373 than some custom data type, objects will die a natural death when they
1374 are no longer needed.
1376 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
1378 Guile does not provide the float representation for inexact real numbers any
1379 more. Now, only doubles are used to represent inexact real numbers. Further,
1380 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
1381 and scm_tc16_complex, respectively.
1383 ** Removed deprecated type scm_smobfuns
1385 ** Removed deprecated function scm_newsmob
1387 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
1389 There is an ongoing discussion among the developers whether to
1390 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
1391 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
1392 until this issue has been settled.
1394 ** Removed deprecated type tag scm_tc16_kw
1396 ** Added type tag scm_tc16_keyword
1398 (This was introduced already in release 1.3.4 but was not documented
1401 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
1403 * Changes to system call interfaces:
1405 ** The "select" procedure now tests port buffers for the ability to
1406 provide input or accept output. Previously only the underlying file
1407 descriptors were checked.
1409 ** New variable PIPE_BUF: the maximum number of bytes that can be
1410 atomically written to a pipe.
1412 ** If a facility is not available on the system when Guile is
1413 compiled, the corresponding primitive procedure will not be defined.
1414 Previously it would have been defined but would throw a system-error
1415 exception if called. Exception handlers which catch this case may
1416 need minor modification: an error will be thrown with key
1417 'unbound-variable instead of 'system-error. Alternatively it's
1418 now possible to use `defined?' to check whether the facility is
1421 ** Procedures which depend on the timezone should now give the correct
1422 result on systems which cache the TZ environment variable, even if TZ
1423 is changed without calling tzset.
1425 * Changes to the networking interfaces:
1427 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
1428 long integers between network and host format. For now, it's not
1429 particularly convenient to do this kind of thing, but consider:
1431 (define write-network-long
1432 (lambda (value port)
1433 (let ((v (make-uniform-vector 1 1 0)))
1434 (uniform-vector-set! v 0 (htonl value))
1435 (uniform-vector-write v port))))
1437 (define read-network-long
1439 (let ((v (make-uniform-vector 1 1 0)))
1440 (uniform-vector-read! v port)
1441 (ntohl (uniform-vector-ref v 0)))))
1443 ** If inet-aton fails, it now throws an error with key 'misc-error
1444 instead of 'system-error, since errno is not relevant.
1446 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
1447 specific keys instead of 'system-error. The latter is inappropriate
1448 since errno will not have been set. The keys are:
1449 'host-not-found, 'try-again, 'no-recovery and 'no-data.
1451 ** sethostent, setnetent, setprotoent, setservent: now take an
1452 optional argument STAYOPEN, which specifies whether the database
1453 remains open after a database entry is accessed randomly (e.g., using
1454 gethostbyname for the hosts database.) The default is #f. Previously
1458 Changes since Guile 1.3.2:
1460 * Changes to the stand-alone interpreter
1464 An initial version of the Guile debugger written by Chris Hanson has
1465 been added. The debugger is still under development but is included
1466 in the distribution anyway since it is already quite useful.
1472 after an error to enter the debugger. Type `help' inside the debugger
1473 for a description of available commands.
1475 If you prefer to have stack frames numbered and printed in
1476 anti-chronological order and prefer up in the stack to be down on the
1477 screen as is the case in gdb, you can put
1479 (debug-enable 'backwards)
1481 in your .guile startup file. (However, this means that Guile can't
1482 use indentation to indicate stack level.)
1484 The debugger is autoloaded into Guile at the first use.
1486 ** Further enhancements to backtraces
1488 There is a new debug option `width' which controls the maximum width
1489 on the screen of printed stack frames. Fancy printing parameters
1490 ("level" and "length" as in Common LISP) are adaptively adjusted for
1491 each stack frame to give maximum information while still fitting
1492 within the bounds. If the stack frame can't be made to fit by
1493 adjusting parameters, it is simply cut off at the end. This is marked
1496 ** Some modules are now only loaded when the repl is started
1498 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
1499 regex) are now loaded into (guile-user) only if the repl has been
1500 started. The effect is that the startup time for scripts has been
1501 reduced to 30% of what it was previously.
1503 Correctly written scripts load the modules they require at the top of
1504 the file and should not be affected by this change.
1506 ** Hooks are now represented as smobs
1508 * Changes to Scheme functions and syntax
1510 ** Readline support has changed again.
1512 The old (readline-activator) module is gone. Use (ice-9 readline)
1513 instead, which now contains all readline functionality. So the code
1514 to activate readline is now
1516 (use-modules (ice-9 readline))
1519 This should work at any time, including from the guile prompt.
1521 To avoid confusion about the terms of Guile's license, please only
1522 enable readline for your personal use; please don't make it the
1523 default for others. Here is why we make this rather odd-sounding
1526 Guile is normally licensed under a weakened form of the GNU General
1527 Public License, which allows you to link code with Guile without
1528 placing that code under the GPL. This exception is important to some
1531 However, since readline is distributed under the GNU General Public
1532 License, when you link Guile with readline, either statically or
1533 dynamically, you effectively change Guile's license to the strict GPL.
1534 Whenever you link any strictly GPL'd code into Guile, uses of Guile
1535 which are normally permitted become forbidden. This is a rather
1536 non-obvious consequence of the licensing terms.
1538 So, to make sure things remain clear, please let people choose for
1539 themselves whether to link GPL'd libraries like readline with Guile.
1541 ** regexp-substitute/global has changed slightly, but incompatibly.
1543 If you include a function in the item list, the string of the match
1544 object it receives is the same string passed to
1545 regexp-substitute/global, not some suffix of that string.
1546 Correspondingly, the match's positions are relative to the entire
1547 string, not the suffix.
1549 If the regexp can match the empty string, the way matches are chosen
1550 from the string has changed. regexp-substitute/global recognizes the
1551 same set of matches that list-matches does; see below.
1553 ** New function: list-matches REGEXP STRING [FLAGS]
1555 Return a list of match objects, one for every non-overlapping, maximal
1556 match of REGEXP in STRING. The matches appear in left-to-right order.
1557 list-matches only reports matches of the empty string if there are no
1558 other matches which begin on, end at, or include the empty match's
1561 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1563 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
1565 For each match of REGEXP in STRING, apply PROC to the match object,
1566 and the last value PROC returned, or INIT for the first call. Return
1567 the last value returned by PROC. We apply PROC to the matches as they
1568 appear from left to right.
1570 This function recognizes matches according to the same criteria as
1573 Thus, you could define list-matches like this:
1575 (define (list-matches regexp string . flags)
1576 (reverse! (apply fold-matches regexp string '() cons flags)))
1578 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1582 *** New function: hook? OBJ
1584 Return #t if OBJ is a hook, otherwise #f.
1586 *** New function: make-hook-with-name NAME [ARITY]
1588 Return a hook with name NAME and arity ARITY. The default value for
1589 ARITY is 0. The only effect of NAME is that it will appear when the
1590 hook object is printed to ease debugging.
1592 *** New function: hook-empty? HOOK
1594 Return #t if HOOK doesn't contain any procedures, otherwise #f.
1596 *** New function: hook->list HOOK
1598 Return a list of the procedures that are called when run-hook is
1601 ** `map' signals an error if its argument lists are not all the same length.
1603 This is the behavior required by R5RS, so this change is really a bug
1604 fix. But it seems to affect a lot of people's code, so we're
1605 mentioning it here anyway.
1607 ** Print-state handling has been made more transparent
1609 Under certain circumstances, ports are represented as a port with an
1610 associated print state. Earlier, this pair was represented as a pair
1611 (see "Some magic has been added to the printer" below). It is now
1612 indistinguishable (almost; see `get-print-state') from a port on the
1615 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
1617 Return a new port with the associated print state PRINT-STATE.
1619 *** New function: get-print-state OUTPUT-PORT
1621 Return the print state associated with this port if it exists,
1622 otherwise return #f.
1624 *** New function: directory-stream? OBJECT
1626 Returns true iff OBJECT is a directory stream --- the sort of object
1627 returned by `opendir'.
1629 ** New function: using-readline?
1631 Return #t if readline is in use in the current repl.
1633 ** structs will be removed in 1.4
1635 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
1636 and use GOOPS objects as the fundamental record type.
1638 * Changes to the scm_ interface
1640 ** structs will be removed in 1.4
1642 The entire current struct interface (struct.c, struct.h) will be
1643 replaced in Guile 1.4. We will merge GOOPS into libguile and use
1644 GOOPS objects as the fundamental record type.
1646 ** The internal representation of subr's has changed
1648 Instead of giving a hint to the subr name, the CAR field of the subr
1649 now contains an index to a subr entry in scm_subr_table.
1651 *** New variable: scm_subr_table
1653 An array of subr entries. A subr entry contains the name, properties
1654 and documentation associated with the subr. The properties and
1655 documentation slots are not yet used.
1657 ** A new scheme for "forwarding" calls to a builtin to a generic function
1659 It is now possible to extend the functionality of some Guile
1660 primitives by letting them defer a call to a GOOPS generic function on
1661 argument mismatch. This means that there is no loss of efficiency in
1666 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
1667 (define-method + ((x <string>) (y <string>))
1668 (string-append x y))
1670 + will still be as efficient as usual in numerical calculations, but
1671 can also be used for concatenating strings.
1673 Who will be the first one to extend Guile's numerical tower to
1674 rationals? :) [OK, there a few other things to fix before this can
1675 be made in a clean way.]
1677 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
1679 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
1681 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
1683 These do the same job as SCM_PROC and SCM_PROC1, but they also define
1684 a variable GENERIC which can be used by the dispatch macros below.
1686 [This is experimental code which may change soon.]
1688 *** New macros for forwarding control to a generic on arg type error
1690 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
1692 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
1694 These correspond to the scm_wta function call, and have the same
1695 behaviour until the user has called the GOOPS primitive
1696 `enable-primitive-generic!'. After that, these macros will apply the
1697 generic function GENERIC to the argument(s) instead of calling
1700 [This is experimental code which may change soon.]
1702 *** New macros for argument testing with generic dispatch
1704 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
1706 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
1708 These correspond to the SCM_ASSERT macro, but will defer control to
1709 GENERIC on error after `enable-primitive-generic!' has been called.
1711 [This is experimental code which may change soon.]
1713 ** New function: SCM scm_eval_body (SCM body, SCM env)
1715 Evaluates the body of a special form.
1717 ** The internal representation of struct's has changed
1719 Previously, four slots were allocated for the procedure(s) of entities
1720 and operators. The motivation for this representation had to do with
1721 the structure of the evaluator, the wish to support tail-recursive
1722 generic functions, and efficiency. Since the generic function
1723 dispatch mechanism has changed, there is no longer a need for such an
1724 expensive representation, and the representation has been simplified.
1726 This should not make any difference for most users.
1728 ** GOOPS support has been cleaned up.
1730 Some code has been moved from eval.c to objects.c and code in both of
1731 these compilation units has been cleaned up and better structured.
1733 *** New functions for applying generic functions
1735 New function: SCM scm_apply_generic (GENERIC, ARGS)
1736 New function: SCM scm_call_generic_0 (GENERIC)
1737 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
1738 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
1739 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
1741 ** Deprecated function: scm_make_named_hook
1743 It is now replaced by:
1745 ** New function: SCM scm_create_hook (const char *name, int arity)
1747 Creates a hook in the same way as make-hook above but also
1748 binds a variable named NAME to it.
1750 This is the typical way of creating a hook from C code.
1752 Currently, the variable is created in the "current" module.
1753 This might change when we get the new module system.
1755 [The behaviour is identical to scm_make_named_hook.]
1759 Changes since Guile 1.3:
1761 * Changes to mailing lists
1763 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
1765 See the README file to find current addresses for all the Guile
1768 * Changes to the distribution
1770 ** Readline support is no longer included with Guile by default.
1772 Based on the different license terms of Guile and Readline, we
1773 concluded that Guile should not *by default* cause the linking of
1774 Readline into an application program. Readline support is now offered
1775 as a separate module, which is linked into an application only when
1776 you explicitly specify it.
1778 Although Guile is GNU software, its distribution terms add a special
1779 exception to the usual GNU General Public License (GPL). Guile's
1780 license includes a clause that allows you to link Guile with non-free
1781 programs. We add this exception so as not to put Guile at a
1782 disadvantage vis-a-vis other extensibility packages that support other
1785 In contrast, the GNU Readline library is distributed under the GNU
1786 General Public License pure and simple. This means that you may not
1787 link Readline, even dynamically, into an application unless it is
1788 distributed under a free software license that is compatible the GPL.
1790 Because of this difference in distribution terms, an application that
1791 can use Guile may not be able to use Readline. Now users will be
1792 explicitly offered two independent decisions about the use of these
1795 You can activate the readline support by issuing
1797 (use-modules (readline-activator))
1800 from your ".guile" file, for example.
1802 * Changes to the stand-alone interpreter
1804 ** All builtins now print as primitives.
1805 Previously builtin procedures not belonging to the fundamental subr
1806 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
1807 Now, they print as #<primitive-procedure NAME>.
1809 ** Backtraces slightly more intelligible.
1810 gsubr-apply and macro transformer application frames no longer appear
1813 * Changes to Scheme functions and syntax
1815 ** Guile now correctly handles internal defines by rewriting them into
1816 their equivalent letrec. Previously, internal defines would
1817 incrementally add to the innermost environment, without checking
1818 whether the restrictions specified in RnRS were met. This lead to the
1819 correct behaviour when these restriction actually were met, but didn't
1820 catch all illegal uses. Such an illegal use could lead to crashes of
1821 the Guile interpreter or or other unwanted results. An example of
1822 incorrect internal defines that made Guile behave erratically:
1834 The problem with this example is that the definition of `c' uses the
1835 value of `b' directly. This confuses the meoization machine of Guile
1836 so that the second call of `b' (this time in a larger environment that
1837 also contains bindings for `c' and `d') refers to the binding of `c'
1838 instead of `a'. You could also make Guile crash with a variation on
1843 (define (b flag) (if flag a 1))
1844 (define c (1+ (b flag)))
1852 From now on, Guile will issue an `Unbound variable: b' error message
1857 A hook contains a list of functions which should be called on
1858 particular occasions in an existing program. Hooks are used for
1861 A window manager might have a hook before-window-map-hook. The window
1862 manager uses the function run-hooks to call all functions stored in
1863 before-window-map-hook each time a window is mapped. The user can
1864 store functions in the hook using add-hook!.
1866 In Guile, hooks are first class objects.
1868 *** New function: make-hook [N_ARGS]
1870 Return a hook for hook functions which can take N_ARGS arguments.
1871 The default value for N_ARGS is 0.
1873 (See also scm_make_named_hook below.)
1875 *** New function: add-hook! HOOK PROC [APPEND_P]
1877 Put PROC at the beginning of the list of functions stored in HOOK.
1878 If APPEND_P is supplied, and non-false, put PROC at the end instead.
1880 PROC must be able to take the number of arguments specified when the
1883 If PROC already exists in HOOK, then remove it first.
1885 *** New function: remove-hook! HOOK PROC
1887 Remove PROC from the list of functions in HOOK.
1889 *** New function: reset-hook! HOOK
1891 Clear the list of hook functions stored in HOOK.
1893 *** New function: run-hook HOOK ARG1 ...
1895 Run all hook functions stored in HOOK with arguments ARG1 ... .
1896 The number of arguments supplied must correspond to the number given
1897 when the hook was created.
1899 ** The function `dynamic-link' now takes optional keyword arguments.
1900 The only keyword argument that is currently defined is `:global
1901 BOOL'. With it, you can control whether the shared library will be
1902 linked in global mode or not. In global mode, the symbols from the
1903 linked library can be used to resolve references from other
1904 dynamically linked libraries. In non-global mode, the linked
1905 library is essentially invisible and can only be accessed via
1906 `dynamic-func', etc. The default is now to link in global mode.
1907 Previously, the default has been non-global mode.
1909 The `#:global' keyword is only effective on platforms that support
1910 the dlopen family of functions.
1912 ** New function `provided?'
1914 - Function: provided? FEATURE
1915 Return true iff FEATURE is supported by this installation of
1916 Guile. FEATURE must be a symbol naming a feature; the global
1917 variable `*features*' is a list of available features.
1919 ** Changes to the module (ice-9 expect):
1921 *** The expect-strings macro now matches `$' in a regular expression
1922 only at a line-break or end-of-file by default. Previously it would
1923 match the end of the string accumulated so far. The old behaviour
1924 can be obtained by setting the variable `expect-strings-exec-flags'
1927 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
1928 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
1929 in a regular expression will still match before a line-break or
1930 end-of-file. The default is `regexp/noteol'.
1932 *** The expect-strings macro now uses a variable
1933 `expect-strings-compile-flags' for the flags to be supplied to
1934 `make-regexp'. The default is `regexp/newline', which was previously
1937 *** The expect macro now supplies two arguments to a match procedure:
1938 the current accumulated string and a flag to indicate whether
1939 end-of-file has been reached. Previously only the string was supplied.
1940 If end-of-file is reached, the match procedure will be called an
1941 additional time with the same accumulated string as the previous call
1942 but with the flag set.
1944 ** New module (ice-9 format), implementing the Common Lisp `format' function.
1946 This code, and the documentation for it that appears here, was
1947 borrowed from SLIB, with minor adaptations for Guile.
1949 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
1950 An almost complete implementation of Common LISP format description
1951 according to the CL reference book `Common LISP' from Guy L.
1952 Steele, Digital Press. Backward compatible to most of the
1953 available Scheme format implementations.
1955 Returns `#t', `#f' or a string; has side effect of printing
1956 according to FORMAT-STRING. If DESTINATION is `#t', the output is
1957 to the current output port and `#t' is returned. If DESTINATION
1958 is `#f', a formatted string is returned as the result of the call.
1959 NEW: If DESTINATION is a string, DESTINATION is regarded as the
1960 format string; FORMAT-STRING is then the first argument and the
1961 output is returned as a string. If DESTINATION is a number, the
1962 output is to the current error port if available by the
1963 implementation. Otherwise DESTINATION must be an output port and
1966 FORMAT-STRING must be a string. In case of a formatting error
1967 format returns `#f' and prints a message on the current output or
1968 error port. Characters are output as if the string were output by
1969 the `display' function with the exception of those prefixed by a
1970 tilde (~). For a detailed description of the FORMAT-STRING syntax
1971 please consult a Common LISP format reference manual. For a test
1972 suite to verify this format implementation load `formatst.scm'.
1973 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
1975 Note: `format' is not reentrant, i.e. only one `format'-call may
1976 be executed at a time.
1979 *** Format Specification (Format version 3.0)
1981 Please consult a Common LISP format reference manual for a detailed
1982 description of the format string syntax. For a demonstration of the
1983 implemented directives see `formatst.scm'.
1985 This implementation supports directive parameters and modifiers (`:'
1986 and `@' characters). Multiple parameters must be separated by a comma
1987 (`,'). Parameters can be numerical parameters (positive or negative),
1988 character parameters (prefixed by a quote character (`''), variable
1989 parameters (`v'), number of rest arguments parameter (`#'), empty and
1990 default parameters. Directive characters are case independent. The
1991 general form of a directive is:
1993 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
1995 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
1997 *** Implemented CL Format Control Directives
1999 Documentation syntax: Uppercase characters represent the
2000 corresponding control directive characters. Lowercase characters
2001 represent control directive parameter descriptions.
2004 Any (print as `display' does).
2008 `~MINCOL,COLINC,MINPAD,PADCHARA'
2012 S-expression (print as `write' does).
2016 `~MINCOL,COLINC,MINPAD,PADCHARS'
2022 print number sign always.
2025 print comma separated.
2027 `~MINCOL,PADCHAR,COMMACHARD'
2033 print number sign always.
2036 print comma separated.
2038 `~MINCOL,PADCHAR,COMMACHARX'
2044 print number sign always.
2047 print comma separated.
2049 `~MINCOL,PADCHAR,COMMACHARO'
2055 print number sign always.
2058 print comma separated.
2060 `~MINCOL,PADCHAR,COMMACHARB'
2065 `~N,MINCOL,PADCHAR,COMMACHARR'
2069 print a number as a Roman numeral.
2072 print a number as an "old fashioned" Roman numeral.
2075 print a number as an ordinal English number.
2078 print a number as a cardinal English number.
2083 prints `y' and `ies'.
2086 as `~P but jumps 1 argument backward.'
2089 as `~@P but jumps 1 argument backward.'
2094 prints a character as the reader can understand it (i.e. `#\'
2098 prints a character as emacs does (eg. `^C' for ASCII 03).
2101 Fixed-format floating-point (prints a flonum like MMM.NNN).
2102 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
2104 If the number is positive a plus sign is printed.
2107 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
2108 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
2110 If the number is positive a plus sign is printed.
2113 General floating-point (prints a flonum either fixed or
2115 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
2117 If the number is positive a plus sign is printed.
2120 Dollars floating-point (prints a flonum in fixed with signs
2122 `~DIGITS,SCALE,WIDTH,PADCHAR$'
2124 If the number is positive a plus sign is printed.
2127 A sign is always printed and appears before the padding.
2130 The sign appears before the padding.
2138 print newline if not at the beginning of the output line.
2140 prints `~&' and then N-1 newlines.
2145 print N page separators.
2155 newline is ignored, white space left.
2158 newline is left, white space ignored.
2163 relative tabulation.
2169 Indirection (expects indirect arguments as a list).
2171 extracts indirect arguments from format arguments.
2174 Case conversion (converts by `string-downcase').
2176 converts by `string-capitalize'.
2179 converts by `string-capitalize-first'.
2182 converts by `string-upcase'.
2185 Argument Jumping (jumps 1 argument forward).
2187 jumps N arguments forward.
2190 jumps 1 argument backward.
2193 jumps N arguments backward.
2196 jumps to the 0th argument.
2199 jumps to the Nth argument (beginning from 0)
2201 `~[STR0~;STR1~;...~;STRN~]'
2202 Conditional Expression (numerical clause conditional).
2204 take argument from N.
2207 true test conditional.
2210 if-else-then conditional.
2216 default clause follows.
2219 Iteration (args come from the next argument (a list)).
2221 at most N iterations.
2224 args from next arg (a list of lists).
2227 args from the rest of arguments.
2230 args from the rest args (lists).
2241 aborts if N <= M <= K
2243 *** Not Implemented CL Format Control Directives
2246 print `#f' as an empty list (see below).
2249 print `#f' as an empty list (see below).
2255 (sorry I don't understand its semantics completely)
2257 *** Extended, Replaced and Additional Control Directives
2259 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
2260 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
2261 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
2262 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
2263 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
2264 COMMAWIDTH is the number of characters between two comma
2268 print a R4RS complex number as `~F~@Fi' with passed parameters for
2272 Pretty print formatting of an argument for scheme code lists.
2278 Flushes the output if format DESTINATION is a port.
2281 Print a `#\space' character
2283 print N `#\space' characters.
2286 Print a `#\tab' character
2288 print N `#\tab' characters.
2291 Takes N as an integer representation for a character. No arguments
2292 are consumed. N is converted to a character by `integer->char'. N
2293 must be a positive decimal number.
2296 Print out readproof. Prints out internal objects represented as
2297 `#<...>' as strings `"#<...>"' so that the format output can always
2298 be processed by `read'.
2301 Print out readproof. Prints out internal objects represented as
2302 `#<...>' as strings `"#<...>"' so that the format output can always
2303 be processed by `read'.
2306 Prints information and a copyright notice on the format
2309 prints format version.
2312 may also print number strings, i.e. passing a number as a string
2313 and format it accordingly.
2315 *** Configuration Variables
2317 The format module exports some configuration variables to suit the
2318 systems and users needs. There should be no modification necessary for
2319 the configuration that comes with Guile. Format detects automatically
2320 if the running scheme system implements floating point numbers and
2323 format:symbol-case-conv
2324 Symbols are converted by `symbol->string' so the case type of the
2325 printed symbols is implementation dependent.
2326 `format:symbol-case-conv' is a one arg closure which is either
2327 `#f' (no conversion), `string-upcase', `string-downcase' or
2328 `string-capitalize'. (default `#f')
2330 format:iobj-case-conv
2331 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
2332 implementation internal objects. (default `#f')
2335 The character prefixing the exponent value in `~E' printing.
2338 *** Compatibility With Other Format Implementations
2344 Downward compatible except for padding support and `~A', `~S',
2345 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
2346 `printf' padding support which is completely replaced by the CL
2347 `format' padding style.
2350 Downward compatible except for `~', which is not documented
2351 (ignores all characters inside the format string up to a newline
2352 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
2353 numerical and variable parameters and `:/@' modifiers in the CL
2357 Downward compatible except for `~A' and `~S' which print in
2358 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
2359 directive parameters or modifiers)).
2362 Downward compatible except for an optional destination parameter:
2363 S2C accepts a format call without a destination which returns a
2364 formatted string. This is equivalent to a #f destination in S2C.
2365 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
2366 parameters or modifiers)).
2369 ** Changes to string-handling functions.
2371 These functions were added to support the (ice-9 format) module, above.
2373 *** New function: string-upcase STRING
2374 *** New function: string-downcase STRING
2376 These are non-destructive versions of the existing string-upcase! and
2377 string-downcase! functions.
2379 *** New function: string-capitalize! STRING
2380 *** New function: string-capitalize STRING
2382 These functions convert the first letter of each word in the string to
2385 (string-capitalize "howdy there")
2388 As with the other functions, string-capitalize! modifies the string in
2389 place, while string-capitalize returns a modified copy of its argument.
2391 *** New function: string-ci->symbol STRING
2393 Return a symbol whose name is STRING, but having the same case as if
2394 the symbol had be read by `read'.
2396 Guile can be configured to be sensitive or insensitive to case
2397 differences in Scheme identifiers. If Guile is case-insensitive, all
2398 symbols are converted to lower case on input. The `string-ci->symbol'
2399 function returns a symbol whose name in STRING, transformed as Guile
2400 would if STRING were input.
2402 *** New function: substring-move! STRING1 START END STRING2 START
2404 Copy the substring of STRING1 from START (inclusive) to END
2405 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
2406 string, and the source and destination areas may overlap; in all
2407 cases, the function behaves as if all the characters were copied
2410 *** Extended functions: substring-move-left! substring-move-right!
2412 These functions now correctly copy arbitrarily overlapping substrings;
2413 they are both synonyms for substring-move!.
2416 ** New module (ice-9 getopt-long), with the function `getopt-long'.
2418 getopt-long is a function for parsing command-line arguments in a
2419 manner consistent with other GNU programs.
2421 (getopt-long ARGS GRAMMAR)
2422 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
2424 ARGS should be a list of strings. Its first element should be the
2425 name of the program; subsequent elements should be the arguments
2426 that were passed to the program on the command line. The
2427 `program-arguments' procedure returns a list of this form.
2429 GRAMMAR is a list of the form:
2430 ((OPTION (PROPERTY VALUE) ...) ...)
2432 Each OPTION should be a symbol. `getopt-long' will accept a
2433 command-line option named `--OPTION'.
2434 Each option can have the following (PROPERTY VALUE) pairs:
2436 (single-char CHAR) --- Accept `-CHAR' as a single-character
2437 equivalent to `--OPTION'. This is how to specify traditional
2439 (required? BOOL) --- If BOOL is true, the option is required.
2440 getopt-long will raise an error if it is not found in ARGS.
2441 (value BOOL) --- If BOOL is #t, the option accepts a value; if
2442 it is #f, it does not; and if it is the symbol
2443 `optional', the option may appear in ARGS with or
2445 (predicate FUNC) --- If the option accepts a value (i.e. you
2446 specified `(value #t)' for this option), then getopt
2447 will apply FUNC to the value, and throw an exception
2448 if it returns #f. FUNC should be a procedure which
2449 accepts a string and returns a boolean value; you may
2450 need to use quasiquotes to get it into GRAMMAR.
2452 The (PROPERTY VALUE) pairs may occur in any order, but each
2453 property may occur only once. By default, options do not have
2454 single-character equivalents, are not required, and do not take
2457 In ARGS, single-character options may be combined, in the usual
2458 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
2459 accepts values, then it must be the last option in the
2460 combination; the value is the next argument. So, for example, using
2461 the following grammar:
2462 ((apples (single-char #\a))
2463 (blimps (single-char #\b) (value #t))
2464 (catalexis (single-char #\c) (value #t)))
2465 the following argument lists would be acceptable:
2466 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
2467 for "blimps" and "catalexis")
2468 ("-ab" "bang" "-c" "couth") (same)
2469 ("-ac" "couth" "-b" "bang") (same)
2470 ("-abc" "couth" "bang") (an error, since `-b' is not the
2471 last option in its combination)
2473 If an option's value is optional, then `getopt-long' decides
2474 whether it has a value by looking at what follows it in ARGS. If
2475 the next element is a string, and it does not appear to be an
2476 option itself, then that string is the option's value.
2478 The value of a long option can appear as the next element in ARGS,
2479 or it can follow the option name, separated by an `=' character.
2480 Thus, using the same grammar as above, the following argument lists
2482 ("--apples" "Braeburn" "--blimps" "Goodyear")
2483 ("--apples=Braeburn" "--blimps" "Goodyear")
2484 ("--blimps" "Goodyear" "--apples=Braeburn")
2486 If the option "--" appears in ARGS, argument parsing stops there;
2487 subsequent arguments are returned as ordinary arguments, even if
2488 they resemble options. So, in the argument list:
2489 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
2490 `getopt-long' will recognize the `apples' option as having the
2491 value "Granny Smith", but it will not recognize the `blimp'
2492 option; it will return the strings "--blimp" and "Goodyear" as
2493 ordinary argument strings.
2495 The `getopt-long' function returns the parsed argument list as an
2496 assocation list, mapping option names --- the symbols from GRAMMAR
2497 --- onto their values, or #t if the option does not accept a value.
2498 Unused options do not appear in the alist.
2500 All arguments that are not the value of any option are returned
2501 as a list, associated with the empty list.
2503 `getopt-long' throws an exception if:
2504 - it finds an unrecognized option in ARGS
2505 - a required option is omitted
2506 - an option that requires an argument doesn't get one
2507 - an option that doesn't accept an argument does get one (this can
2508 only happen using the long option `--opt=value' syntax)
2509 - an option predicate fails
2514 `((lockfile-dir (required? #t)
2517 (predicate ,file-is-directory?))
2518 (verbose (required? #f)
2521 (x-includes (single-char #\x))
2522 (rnet-server (single-char #\y)
2523 (predicate ,string?))))
2525 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
2526 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
2528 => ((() "foo1" "-fred" "foo2" "foo3")
2529 (rnet-server . "lamprod")
2530 (x-includes . "/usr/include")
2531 (lockfile-dir . "/tmp")
2534 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
2536 It will be removed in a few releases.
2538 ** New syntax: lambda*
2539 ** New syntax: define*
2540 ** New syntax: define*-public
2541 ** New syntax: defmacro*
2542 ** New syntax: defmacro*-public
2543 Guile now supports optional arguments.
2545 `lambda*', `define*', `define*-public', `defmacro*' and
2546 `defmacro*-public' are identical to the non-* versions except that
2547 they use an extended type of parameter list that has the following BNF
2548 syntax (parentheses are literal, square brackets indicate grouping,
2549 and `*', `+' and `?' have the usual meaning):
2551 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
2552 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
2553 [[#&rest identifier]|[. identifier]]? ) | [identifier]
2555 ext-var-decl ::= identifier | ( identifier expression )
2557 The semantics are best illustrated with the following documentation
2558 and examples for `lambda*':
2561 lambda extended for optional and keyword arguments
2563 lambda* creates a procedure that takes optional arguments. These
2564 are specified by putting them inside brackets at the end of the
2565 paramater list, but before any dotted rest argument. For example,
2566 (lambda* (a b #&optional c d . e) '())
2567 creates a procedure with fixed arguments a and b, optional arguments c
2568 and d, and rest argument e. If the optional arguments are omitted
2569 in a call, the variables for them are unbound in the procedure. This
2570 can be checked with the bound? macro.
2572 lambda* can also take keyword arguments. For example, a procedure
2574 (lambda* (#&key xyzzy larch) '())
2575 can be called with any of the argument lists (#:xyzzy 11)
2576 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
2577 are given as keywords are bound to values.
2579 Optional and keyword arguments can also be given default values
2580 which they take on when they are not present in a call, by giving a
2581 two-item list in place of an optional argument, for example in:
2582 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
2583 foo is a fixed argument, bar is an optional argument with default
2584 value 42, and baz is a keyword argument with default value 73.
2585 Default value expressions are not evaluated unless they are needed
2586 and until the procedure is called.
2588 lambda* now supports two more special parameter list keywords.
2590 lambda*-defined procedures now throw an error by default if a
2591 keyword other than one of those specified is found in the actual
2592 passed arguments. However, specifying #&allow-other-keys
2593 immediately after the kyword argument declarations restores the
2594 previous behavior of ignoring unknown keywords. lambda* also now
2595 guarantees that if the same keyword is passed more than once, the
2596 last one passed is the one that takes effect. For example,
2597 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
2598 #:heads 37 #:tails 42 #:heads 99)
2599 would result in (99 47) being displayed.
2601 #&rest is also now provided as a synonym for the dotted syntax rest
2602 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
2603 all respects to lambda*. This is provided for more similarity to DSSSL,
2604 MIT-Scheme and Kawa among others, as well as for refugees from other
2607 Further documentation may be found in the optargs.scm file itself.
2609 The optional argument module also exports the macros `let-optional',
2610 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
2611 are not documented here because they may be removed in the future, but
2612 full documentation is still available in optargs.scm.
2614 ** New syntax: and-let*
2615 Guile now supports the `and-let*' form, described in the draft SRFI-2.
2617 Syntax: (land* (<clause> ...) <body> ...)
2618 Each <clause> should have one of the following forms:
2619 (<variable> <expression>)
2622 Each <variable> or <bound-variable> should be an identifier. Each
2623 <expression> should be a valid expression. The <body> should be a
2624 possibly empty sequence of expressions, like the <body> of a
2627 Semantics: A LAND* expression is evaluated by evaluating the
2628 <expression> or <bound-variable> of each of the <clause>s from
2629 left to right. The value of the first <expression> or
2630 <bound-variable> that evaluates to a false value is returned; the
2631 remaining <expression>s and <bound-variable>s are not evaluated.
2632 The <body> forms are evaluated iff all the <expression>s and
2633 <bound-variable>s evaluate to true values.
2635 The <expression>s and the <body> are evaluated in an environment
2636 binding each <variable> of the preceding (<variable> <expression>)
2637 clauses to the value of the <expression>. Later bindings
2638 shadow earlier bindings.
2640 Guile's and-let* macro was contributed by Michael Livshin.
2642 ** New sorting functions
2644 *** New function: sorted? SEQUENCE LESS?
2645 Returns `#t' when the sequence argument is in non-decreasing order
2646 according to LESS? (that is, there is no adjacent pair `... x y
2647 ...' for which `(less? y x)').
2649 Returns `#f' when the sequence contains at least one out-of-order
2650 pair. It is an error if the sequence is neither a list nor a
2653 *** New function: merge LIST1 LIST2 LESS?
2654 LIST1 and LIST2 are sorted lists.
2655 Returns the sorted list of all elements in LIST1 and LIST2.
2657 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
2658 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
2659 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
2660 (Here "<" should read "comes before".)
2662 *** New procedure: merge! LIST1 LIST2 LESS?
2663 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
2664 the result. If the code is compiled, and LESS? constructs no new
2665 pairs, no pairs at all will be allocated. The first pair of the
2666 result will be either the first pair of LIST1 or the first pair of
2669 *** New function: sort SEQUENCE LESS?
2670 Accepts either a list or a vector, and returns a new sequence
2671 which is sorted. The new sequence is the same type as the input.
2672 Always `(sorted? (sort sequence less?) less?)'. The original
2673 sequence is not altered in any way. The new sequence shares its
2674 elements with the old one; no elements are copied.
2676 *** New procedure: sort! SEQUENCE LESS
2677 Returns its sorted result in the original boxes. No new storage is
2678 allocated at all. Proper usage: (set! slist (sort! slist <))
2680 *** New function: stable-sort SEQUENCE LESS?
2681 Similar to `sort' but stable. That is, if "equal" elements are
2682 ordered a < b in the original sequence, they will have the same order
2685 *** New function: stable-sort! SEQUENCE LESS?
2686 Similar to `sort!' but stable.
2687 Uses temporary storage when sorting vectors.
2689 *** New functions: sort-list, sort-list!
2690 Added for compatibility with scsh.
2692 ** New built-in random number support
2694 *** New function: random N [STATE]
2695 Accepts a positive integer or real N and returns a number of the
2696 same type between zero (inclusive) and N (exclusive). The values
2697 returned have a uniform distribution.
2699 The optional argument STATE must be of the type produced by
2700 `copy-random-state' or `seed->random-state'. It defaults to the value
2701 of the variable `*random-state*'. This object is used to maintain the
2702 state of the pseudo-random-number generator and is altered as a side
2703 effect of the `random' operation.
2705 *** New variable: *random-state*
2706 Holds a data structure that encodes the internal state of the
2707 random-number generator that `random' uses by default. The nature
2708 of this data structure is implementation-dependent. It may be
2709 printed out and successfully read back in, but may or may not
2710 function correctly as a random-number state object in another
2713 *** New function: copy-random-state [STATE]
2714 Returns a new object of type suitable for use as the value of the
2715 variable `*random-state*' and as a second argument to `random'.
2716 If argument STATE is given, a copy of it is returned. Otherwise a
2717 copy of `*random-state*' is returned.
2719 *** New function: seed->random-state SEED
2720 Returns a new object of type suitable for use as the value of the
2721 variable `*random-state*' and as a second argument to `random'.
2722 SEED is a string or a number. A new state is generated and
2723 initialized using SEED.
2725 *** New function: random:uniform [STATE]
2726 Returns an uniformly distributed inexact real random number in the
2727 range between 0 and 1.
2729 *** New procedure: random:solid-sphere! VECT [STATE]
2730 Fills VECT with inexact real random numbers the sum of whose
2731 squares is less than 1.0. Thinking of VECT as coordinates in
2732 space of dimension N = `(vector-length VECT)', the coordinates are
2733 uniformly distributed within the unit N-shere. The sum of the
2734 squares of the numbers is returned. VECT can be either a vector
2735 or a uniform vector of doubles.
2737 *** New procedure: random:hollow-sphere! VECT [STATE]
2738 Fills VECT with inexact real random numbers the sum of whose squares
2739 is equal to 1.0. Thinking of VECT as coordinates in space of
2740 dimension n = `(vector-length VECT)', the coordinates are uniformly
2741 distributed over the surface of the unit n-shere. VECT can be either
2742 a vector or a uniform vector of doubles.
2744 *** New function: random:normal [STATE]
2745 Returns an inexact real in a normal distribution with mean 0 and
2746 standard deviation 1. For a normal distribution with mean M and
2747 standard deviation D use `(+ M (* D (random:normal)))'.
2749 *** New procedure: random:normal-vector! VECT [STATE]
2750 Fills VECT with inexact real random numbers which are independent and
2751 standard normally distributed (i.e., with mean 0 and variance 1).
2752 VECT can be either a vector or a uniform vector of doubles.
2754 *** New function: random:exp STATE
2755 Returns an inexact real in an exponential distribution with mean 1.
2756 For an exponential distribution with mean U use (* U (random:exp)).
2758 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
2760 These functions now operate on numbers in the range of a C unsigned
2763 These functions used to operate on numbers in the range of a C signed
2764 long; however, this seems inappropriate, because Guile integers don't
2767 ** New function: make-guardian
2768 This is an implementation of guardians as described in
2769 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
2770 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
2771 Programming Language Design and Implementation, June 1993
2772 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
2774 ** New functions: delq1!, delv1!, delete1!
2775 These procedures behave similar to delq! and friends but delete only
2776 one object if at all.
2778 ** New function: unread-string STRING PORT
2779 Unread STRING to PORT, that is, push it back onto the port so that
2780 next read operation will work on the pushed back characters.
2782 ** unread-char can now be called multiple times
2783 If unread-char is called multiple times, the unread characters will be
2784 read again in last-in first-out order.
2786 ** the procedures uniform-array-read! and uniform-array-write! now
2787 work on any kind of port, not just ports which are open on a file.
2789 ** Now 'l' in a port mode requests line buffering.
2791 ** The procedure truncate-file now works on string ports as well
2792 as file ports. If the size argument is omitted, the current
2793 file position is used.
2795 ** new procedure: seek PORT/FDES OFFSET WHENCE
2796 The arguments are the same as for the old fseek procedure, but it
2797 works on string ports as well as random-access file ports.
2799 ** the fseek procedure now works on string ports, since it has been
2800 redefined using seek.
2802 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
2803 size is not supplied.
2805 ** the newline procedure no longer flushes the port if it's not
2806 line-buffered: previously it did if it was the current output port.
2808 ** open-pipe and close-pipe are no longer primitive procedures, but
2809 an emulation can be obtained using `(use-modules (ice-9 popen))'.
2811 ** the freopen procedure has been removed.
2813 ** new procedure: drain-input PORT
2814 Drains PORT's read buffers (including any pushed-back characters)
2815 and returns the contents as a single string.
2817 ** New function: map-in-order PROC LIST1 LIST2 ...
2818 Version of `map' which guarantees that the procedure is applied to the
2819 lists in serial order.
2821 ** Renamed `serial-array-copy!' and `serial-array-map!' to
2822 `array-copy-in-order!' and `array-map-in-order!'. The old names are
2823 now obsolete and will go away in release 1.5.
2825 ** New syntax: collect BODY1 ...
2826 Version of `begin' which returns a list of the results of the body
2827 forms instead of the result of the last body form. In contrast to
2828 `begin', `collect' allows an empty body.
2830 ** New functions: read-history FILENAME, write-history FILENAME
2831 Read/write command line history from/to file. Returns #t on success
2832 and #f if an error occured.
2834 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
2836 These procedures return a list of definitions available in the specified
2837 argument, a relative module reference. In the case of no argument,
2838 `(current-module)' is now consulted for definitions to return, instead
2839 of simply returning #f, the former behavior.
2841 ** The #/ syntax for lists is no longer supported.
2843 Earlier versions of Scheme accepted this syntax, but printed a
2846 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
2848 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
2851 * Changes to the gh_ interface
2855 Now takes a second argument which is the result array. If this
2856 pointer is NULL, a new array is malloced (the old behaviour).
2858 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
2859 gh_scm2shorts, gh_scm2longs, gh_scm2floats
2863 * Changes to the scm_ interface
2865 ** Function: scm_make_named_hook (char* name, int n_args)
2867 Creates a hook in the same way as make-hook above but also
2868 binds a variable named NAME to it.
2870 This is the typical way of creating a hook from C code.
2872 Currently, the variable is created in the "current" module. This
2873 might change when we get the new module system.
2875 ** The smob interface
2877 The interface for creating smobs has changed. For documentation, see
2878 data-rep.info (made from guile-core/doc/data-rep.texi).
2880 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
2882 >>> This function will be removed in 1.3.4. <<<
2886 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
2887 This function adds a new smob type, named NAME, with instance size
2888 SIZE to the system. The return value is a tag that is used in
2889 creating instances of the type. If SIZE is 0, then no memory will
2890 be allocated when instances of the smob are created, and nothing
2891 will be freed by the default free function.
2893 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
2894 This function sets the smob marking procedure for the smob type
2895 specified by the tag TC. TC is the tag returned by
2896 `scm_make_smob_type'.
2898 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
2899 This function sets the smob freeing procedure for the smob type
2900 specified by the tag TC. TC is the tag returned by
2901 `scm_make_smob_type'.
2903 *** Function: void scm_set_smob_print (tc, print)
2905 - Function: void scm_set_smob_print (long tc,
2906 scm_sizet (*print) (SCM,
2910 This function sets the smob printing procedure for the smob type
2911 specified by the tag TC. TC is the tag returned by
2912 `scm_make_smob_type'.
2914 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
2915 This function sets the smob equality-testing predicate for the
2916 smob type specified by the tag TC. TC is the tag returned by
2917 `scm_make_smob_type'.
2919 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
2920 Make VALUE contain a smob instance of the type with type code TC and
2921 smob data DATA. VALUE must be previously declared as C type `SCM'.
2923 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
2924 This macro expands to a block of code that creates a smob instance
2925 of the type with type code TC and smob data DATA, and returns that
2926 `SCM' value. It should be the last piece of code in a block.
2928 ** The interfaces for using I/O ports and implementing port types
2929 (ptobs) have changed significantly. The new interface is based on
2930 shared access to buffers and a new set of ptob procedures.
2932 *** scm_newptob has been removed
2936 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
2938 - Function: SCM scm_make_port_type (char *type_name,
2939 int (*fill_buffer) (SCM port),
2940 void (*write_flush) (SCM port));
2942 Similarly to the new smob interface, there is a set of function
2943 setters by which the user can customize the behaviour of his port
2944 type. See ports.h (scm_set_port_XXX).
2946 ** scm_strport_to_string: New function: creates a new string from
2947 a string port's buffer.
2949 ** Plug in interface for random number generators
2950 The variable `scm_the_rng' in random.c contains a value and three
2951 function pointers which together define the current random number
2952 generator being used by the Scheme level interface and the random
2953 number library functions.
2955 The user is free to replace the default generator with the generator
2958 *** Variable: size_t scm_the_rng.rstate_size
2959 The size of the random state type used by the current RNG
2962 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
2963 Given the random STATE, return 32 random bits.
2965 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
2966 Seed random state STATE using string S of length N.
2968 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
2969 Given random state STATE, return a malloced copy.
2972 The default RNG is the MWC (Multiply With Carry) random number
2973 generator described by George Marsaglia at the Department of
2974 Statistics and Supercomputer Computations Research Institute, The
2975 Florida State University (http://stat.fsu.edu/~geo).
2977 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
2978 passes all tests in the DIEHARD test suite
2979 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
2980 costs one multiply and one add on platforms which either supports long
2981 longs (gcc does this on most systems) or have 64 bit longs. The cost
2982 is four multiply on other systems but this can be optimized by writing
2983 scm_i_uniform32 in assembler.
2985 These functions are provided through the scm_the_rng interface for use
2986 by libguile and the application.
2988 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
2989 Given the random STATE, return 32 random bits.
2990 Don't use this function directly. Instead go through the plugin
2991 interface (see "Plug in interface" above).
2993 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
2994 Initialize STATE using SEED of length N.
2996 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
2997 Return a malloc:ed copy of STATE. This function can easily be re-used
2998 in the interfaces to other RNGs.
3000 ** Random number library functions
3001 These functions use the current RNG through the scm_the_rng interface.
3002 It might be a good idea to use these functions from your C code so
3003 that only one random generator is used by all code in your program.
3005 The default random state is stored in:
3007 *** Variable: SCM scm_var_random_state
3008 Contains the vcell of the Scheme variable "*random-state*" which is
3009 used as default state by all random number functions in the Scheme
3014 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
3016 *** Function: scm_rstate *scm_c_default_rstate (void)
3017 This is a convenience function which returns the value of
3018 scm_var_random_state. An error message is generated if this value
3019 isn't a random state.
3021 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
3022 Make a new random state from the string SEED of length LENGTH.
3024 It is generally not a good idea to use multiple random states in a
3025 program. While subsequent random numbers generated from one random
3026 state are guaranteed to be reasonably independent, there is no such
3027 guarantee for numbers generated from different random states.
3029 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
3030 Return 32 random bits.
3032 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
3033 Return a sample from the uniform(0,1) distribution.
3035 *** Function: double scm_c_normal01 (scm_rstate *STATE)
3036 Return a sample from the normal(0,1) distribution.
3038 *** Function: double scm_c_exp1 (scm_rstate *STATE)
3039 Return a sample from the exp(1) distribution.
3041 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
3042 Return a sample from the discrete uniform(0,M) distribution.
3044 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
3045 Return a sample from the discrete uniform(0,M) distribution.
3046 M must be a bignum object. The returned value may be an INUM.
3050 Changes in Guile 1.3 (released Monday, October 19, 1998):
3052 * Changes to the distribution
3054 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
3055 To avoid conflicts, programs should name environment variables after
3056 themselves, except when there's a common practice establishing some
3059 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
3060 giving the former precedence, and printing a warning message if the
3061 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
3063 ** The header files related to multi-byte characters have been removed.
3064 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
3065 which referred to these explicitly will probably need to be rewritten,
3066 since the support for the variant string types has been removed; see
3069 ** The header files append.h and sequences.h have been removed. These
3070 files implemented non-R4RS operations which would encourage
3071 non-portable programming style and less easy-to-read code.
3073 * Changes to the stand-alone interpreter
3075 ** New procedures have been added to implement a "batch mode":
3077 *** Function: batch-mode?
3079 Returns a boolean indicating whether the interpreter is in batch
3082 *** Function: set-batch-mode?! ARG
3084 If ARG is true, switches the interpreter to batch mode. The `#f'
3085 case has not been implemented.
3087 ** Guile now provides full command-line editing, when run interactively.
3088 To use this feature, you must have the readline library installed.
3089 The Guile build process will notice it, and automatically include
3092 The readline library is available via anonymous FTP from any GNU
3093 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
3095 ** the-last-stack is now a fluid.
3097 * Changes to the procedure for linking libguile with your programs
3099 ** You can now use the `guile-config' utility to build programs that use Guile.
3101 Guile now includes a command-line utility called `guile-config', which
3102 can provide information about how to compile and link programs that
3105 *** `guile-config compile' prints any C compiler flags needed to use Guile.
3106 You should include this command's output on the command line you use
3107 to compile C or C++ code that #includes the Guile header files. It's
3108 usually just a `-I' flag to help the compiler find the Guile headers.
3111 *** `guile-config link' prints any linker flags necessary to link with Guile.
3113 This command writes to its standard output a list of flags which you
3114 must pass to the linker to link your code against the Guile library.
3115 The flags include '-lguile' itself, any other libraries the Guile
3116 library depends upon, and any `-L' flags needed to help the linker
3117 find those libraries.
3119 For example, here is a Makefile rule that builds a program named 'foo'
3120 from the object files ${FOO_OBJECTS}, and links them against Guile:
3123 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
3125 Previous Guile releases recommended that you use autoconf to detect
3126 which of a predefined set of libraries were present on your system.
3127 It is more robust to use `guile-config', since it records exactly which
3128 libraries the installed Guile library requires.
3130 This was originally called `build-guile', but was renamed to
3131 `guile-config' before Guile 1.3 was released, to be consistent with
3132 the analogous script for the GTK+ GUI toolkit, which is called
3136 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
3138 If you are using the GNU autoconf package to configure your program,
3139 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
3140 (described above) and gather the necessary values for use in your
3143 The GUILE_FLAGS macro expands to configure script code which runs the
3144 `guile-config' script, to find out where Guile's header files and
3145 libraries are installed. It sets two variables, marked for
3146 substitution, as by AC_SUBST.
3148 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
3149 code that uses Guile header files. This is almost always just a
3152 GUILE_LDFLAGS --- flags to pass to the linker to link a
3153 program against Guile. This includes `-lguile' for the Guile
3154 library itself, any libraries that Guile itself requires (like
3155 -lqthreads), and so on. It may also include a -L flag to tell the
3156 compiler where to find the libraries.
3158 GUILE_FLAGS is defined in the file guile.m4, in the top-level
3159 directory of the Guile distribution. You can copy it into your
3160 package's aclocal.m4 file, and then use it in your configure.in file.
3162 If you are using the `aclocal' program, distributed with GNU automake,
3163 to maintain your aclocal.m4 file, the Guile installation process
3164 installs guile.m4 where aclocal will find it. All you need to do is
3165 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
3166 this will copy the definition of GUILE_FLAGS into your aclocal.m4
3170 * Changes to Scheme functions and syntax
3172 ** Multi-byte strings have been removed, as have multi-byte and wide
3173 ports. We felt that these were the wrong approach to
3174 internationalization support.
3176 ** New function: readline [PROMPT]
3177 Read a line from the terminal, and allow the user to edit it,
3178 prompting with PROMPT. READLINE provides a large set of Emacs-like
3179 editing commands, lets the user recall previously typed lines, and
3180 works on almost every kind of terminal, including dumb terminals.
3182 READLINE assumes that the cursor is at the beginning of the line when
3183 it is invoked. Thus, you can't print a prompt yourself, and then call
3184 READLINE; you need to package up your prompt as a string, pass it to
3185 the function, and let READLINE print the prompt itself. This is
3186 because READLINE needs to know the prompt's screen width.
3188 For Guile to provide this function, you must have the readline
3189 library, version 2.1 or later, installed on your system. Readline is
3190 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
3191 any GNU mirror site.
3193 See also ADD-HISTORY function.
3195 ** New function: add-history STRING
3196 Add STRING as the most recent line in the history used by the READLINE
3197 command. READLINE does not add lines to the history itself; you must
3198 call ADD-HISTORY to make previous input available to the user.
3200 ** The behavior of the read-line function has changed.
3202 This function now uses standard C library functions to read the line,
3203 for speed. This means that it doesn not respect the value of
3204 scm-line-incrementors; it assumes that lines are delimited with
3207 (Note that this is read-line, the function that reads a line of text
3208 from a port, not readline, the function that reads a line from a
3209 terminal, providing full editing capabilities.)
3211 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
3213 This module provides some simple argument parsing. It exports one
3216 Function: getopt-gnu-style ARG-LS
3217 Parse a list of program arguments into an alist of option
3220 Each item in the list of program arguments is examined to see if
3221 it meets the syntax of a GNU long-named option. An argument like
3222 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
3223 returned alist, where MUMBLE is a keyword object with the same
3224 name as the argument. An argument like `--MUMBLE=FROB' produces
3225 an element of the form (MUMBLE . FROB), where FROB is a string.
3227 As a special case, the returned alist also contains a pair whose
3228 car is the symbol `rest'. The cdr of this pair is a list
3229 containing all the items in the argument list that are not options
3230 of the form mentioned above.
3232 The argument `--' is treated specially: all items in the argument
3233 list appearing after such an argument are not examined, and are
3234 returned in the special `rest' list.
3236 This function does not parse normal single-character switches.
3237 You will need to parse them out of the `rest' list yourself.
3239 ** The read syntax for byte vectors and short vectors has changed.
3241 Instead of #bytes(...), write #y(...).
3243 Instead of #short(...), write #h(...).
3245 This may seem nutty, but, like the other uniform vectors, byte vectors
3246 and short vectors want to have the same print and read syntax (and,
3247 more basic, want to have read syntax!). Changing the read syntax to
3248 use multiple characters after the hash sign breaks with the
3249 conventions used in R5RS and the conventions used for the other
3250 uniform vectors. It also introduces complexity in the current reader,
3251 both on the C and Scheme levels. (The Right solution is probably to
3252 change the syntax and prototypes for uniform vectors entirely.)
3255 ** The new module (ice-9 session) provides useful interactive functions.
3257 *** New procedure: (apropos REGEXP OPTION ...)
3259 Display a list of top-level variables whose names match REGEXP, and
3260 the modules they are imported from. Each OPTION should be one of the
3263 value --- Show the value of each matching variable.
3264 shadow --- Show bindings shadowed by subsequently imported modules.
3265 full --- Same as both `shadow' and `value'.
3269 guile> (apropos "trace" 'full)
3270 debug: trace #<procedure trace args>
3271 debug: untrace #<procedure untrace args>
3272 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
3273 the-scm-module: before-backtrace-hook ()
3274 the-scm-module: backtrace #<primitive-procedure backtrace>
3275 the-scm-module: after-backtrace-hook ()
3276 the-scm-module: has-shown-backtrace-hint? #f
3279 ** There are new functions and syntax for working with macros.
3281 Guile implements macros as a special object type. Any variable whose
3282 top-level binding is a macro object acts as a macro. The macro object
3283 specifies how the expression should be transformed before evaluation.
3285 *** Macro objects now print in a reasonable way, resembling procedures.
3287 *** New function: (macro? OBJ)
3288 True iff OBJ is a macro object.
3290 *** New function: (primitive-macro? OBJ)
3291 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
3292 macro transformers, implemented in eval.c rather than Scheme code.
3294 Why do we have this function?
3295 - For symmetry with procedure? and primitive-procedure?,
3296 - to allow custom print procedures to tell whether a macro is
3297 primitive, and display it differently, and
3298 - to allow compilers and user-written evaluators to distinguish
3299 builtin special forms from user-defined ones, which could be
3302 *** New function: (macro-type OBJ)
3303 Return a value indicating what kind of macro OBJ is. Possible return
3306 The symbol `syntax' --- a macro created by procedure->syntax.
3307 The symbol `macro' --- a macro created by procedure->macro.
3308 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
3309 The boolean #f --- if OBJ is not a macro object.
3311 *** New function: (macro-name MACRO)
3312 Return the name of the macro object MACRO's procedure, as returned by
3315 *** New function: (macro-transformer MACRO)
3316 Return the transformer procedure for MACRO.
3318 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
3320 Specify a new macro expander to use in the current module. Each
3321 MODULE is a module name, with the same meaning as in the `use-modules'
3322 form; each named module's exported bindings are added to the current
3323 top-level environment. TRANSFORMER is an expression evaluated in the
3324 resulting environment which must yield a procedure to use as the
3325 module's eval transformer: every expression evaluated in this module
3326 is passed to this function, and the result passed to the Guile
3329 *** macro-eval! is removed. Use local-eval instead.
3331 ** Some magic has been added to the printer to better handle user
3332 written printing routines (like record printers, closure printers).
3334 The problem is that these user written routines must have access to
3335 the current `print-state' to be able to handle fancy things like
3336 detection of circular references. These print-states have to be
3337 passed to the builtin printing routines (display, write, etc) to
3338 properly continue the print chain.
3340 We didn't want to change all existing print code so that it
3341 explicitly passes thru a print state in addition to a port. Instead,
3342 we extented the possible values that the builtin printing routines
3343 accept as a `port'. In addition to a normal port, they now also take
3344 a pair of a normal port and a print-state. Printing will go to the
3345 port and the print-state will be used to control the detection of
3346 circular references, etc. If the builtin function does not care for a
3347 print-state, it is simply ignored.
3349 User written callbacks are now called with such a pair as their
3350 `port', but because every function now accepts this pair as a PORT
3351 argument, you don't have to worry about that. In fact, it is probably
3352 safest to not check for these pairs.
3354 However, it is sometimes necessary to continue a print chain on a
3355 different port, for example to get a intermediate string
3356 representation of the printed value, mangle that string somehow, and
3357 then to finally print the mangled string. Use the new function
3359 inherit-print-state OLD-PORT NEW-PORT
3361 for this. It constructs a new `port' that prints to NEW-PORT but
3362 inherits the print-state of OLD-PORT.
3364 ** struct-vtable-offset renamed to vtable-offset-user
3366 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
3368 ** There is now a third optional argument to make-vtable-vtable
3369 (and fourth to make-struct) when constructing new types (vtables).
3370 This argument initializes field vtable-index-printer of the vtable.
3372 ** The detection of circular references has been extended to structs.
3373 That is, a structure that -- in the process of being printed -- prints
3374 itself does not lead to infinite recursion.
3376 ** There is now some basic support for fluids. Please read
3377 "libguile/fluid.h" to find out more. It is accessible from Scheme with
3378 the following functions and macros:
3380 Function: make-fluid
3382 Create a new fluid object. Fluids are not special variables or
3383 some other extension to the semantics of Scheme, but rather
3384 ordinary Scheme objects. You can store them into variables (that
3385 are still lexically scoped, of course) or into any other place you
3386 like. Every fluid has a initial value of `#f'.
3388 Function: fluid? OBJ
3390 Test whether OBJ is a fluid.
3392 Function: fluid-ref FLUID
3393 Function: fluid-set! FLUID VAL
3395 Access/modify the fluid FLUID. Modifications are only visible
3396 within the current dynamic root (that includes threads).
3398 Function: with-fluids* FLUIDS VALUES THUNK
3400 FLUIDS is a list of fluids and VALUES a corresponding list of
3401 values for these fluids. Before THUNK gets called the values are
3402 installed in the fluids and the old values of the fluids are
3403 saved in the VALUES list. When the flow of control leaves THUNK
3404 or reenters it, the values get swapped again. You might think of
3405 this as a `safe-fluid-excursion'. Note that the VALUES list is
3406 modified by `with-fluids*'.
3408 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
3410 The same as `with-fluids*' but with a different syntax. It looks
3411 just like `let', but both FLUID and VALUE are evaluated. Remember,
3412 fluids are not special variables but ordinary objects. FLUID
3413 should evaluate to a fluid.
3415 ** Changes to system call interfaces:
3417 *** close-port, close-input-port and close-output-port now return a
3418 boolean instead of an `unspecified' object. #t means that the port
3419 was successfully closed, while #f means it was already closed. It is
3420 also now possible for these procedures to raise an exception if an
3421 error occurs (some errors from write can be delayed until close.)
3423 *** the first argument to chmod, fcntl, ftell and fseek can now be a
3426 *** the third argument to fcntl is now optional.
3428 *** the first argument to chown can now be a file descriptor or a port.
3430 *** the argument to stat can now be a port.
3432 *** The following new procedures have been added (most use scsh
3435 *** procedure: close PORT/FD
3436 Similar to close-port (*note close-port: Closing Ports.), but also
3437 works on file descriptors. A side effect of closing a file
3438 descriptor is that any ports using that file descriptor are moved
3439 to a different file descriptor and have their revealed counts set
3442 *** procedure: port->fdes PORT
3443 Returns the integer file descriptor underlying PORT. As a side
3444 effect the revealed count of PORT is incremented.
3446 *** procedure: fdes->ports FDES
3447 Returns a list of existing ports which have FDES as an underlying
3448 file descriptor, without changing their revealed counts.
3450 *** procedure: fdes->inport FDES
3451 Returns an existing input port which has FDES as its underlying
3452 file descriptor, if one exists, and increments its revealed count.
3453 Otherwise, returns a new input port with a revealed count of 1.
3455 *** procedure: fdes->outport FDES
3456 Returns an existing output port which has FDES as its underlying
3457 file descriptor, if one exists, and increments its revealed count.
3458 Otherwise, returns a new output port with a revealed count of 1.
3460 The next group of procedures perform a `dup2' system call, if NEWFD
3461 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
3462 duplicated can be supplied as an integer or contained in a port. The
3463 type of value returned varies depending on which procedure is used.
3465 All procedures also have the side effect when performing `dup2' that
3466 any ports using NEWFD are moved to a different file descriptor and have
3467 their revealed counts set to zero.
3469 *** procedure: dup->fdes PORT/FD [NEWFD]
3470 Returns an integer file descriptor.
3472 *** procedure: dup->inport PORT/FD [NEWFD]
3473 Returns a new input port using the new file descriptor.
3475 *** procedure: dup->outport PORT/FD [NEWFD]
3476 Returns a new output port using the new file descriptor.
3478 *** procedure: dup PORT/FD [NEWFD]
3479 Returns a new port if PORT/FD is a port, with the same mode as the
3480 supplied port, otherwise returns an integer file descriptor.
3482 *** procedure: dup->port PORT/FD MODE [NEWFD]
3483 Returns a new port using the new file descriptor. MODE supplies a
3484 mode string for the port (*note open-file: File Ports.).
3486 *** procedure: setenv NAME VALUE
3487 Modifies the environment of the current process, which is also the
3488 default environment inherited by child processes.
3490 If VALUE is `#f', then NAME is removed from the environment.
3491 Otherwise, the string NAME=VALUE is added to the environment,
3492 replacing any existing string with name matching NAME.
3494 The return value is unspecified.
3496 *** procedure: truncate-file OBJ SIZE
3497 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
3498 can be a string containing a file name or an integer file
3499 descriptor or port open for output on the file. The underlying
3500 system calls are `truncate' and `ftruncate'.
3502 The return value is unspecified.
3504 *** procedure: setvbuf PORT MODE [SIZE]
3505 Set the buffering mode for PORT. MODE can be:
3513 block buffered, using a newly allocated buffer of SIZE bytes.
3514 However if SIZE is zero or unspecified, the port will be made
3517 This procedure should not be used after I/O has been performed with
3520 Ports are usually block buffered by default, with a default buffer
3521 size. Procedures e.g., *Note open-file: File Ports, which accept a
3522 mode string allow `0' to be added to request an unbuffered port.
3524 *** procedure: fsync PORT/FD
3525 Copies any unwritten data for the specified output file descriptor
3526 to disk. If PORT/FD is a port, its buffer is flushed before the
3527 underlying file descriptor is fsync'd. The return value is
3530 *** procedure: open-fdes PATH FLAGS [MODES]
3531 Similar to `open' but returns a file descriptor instead of a port.
3533 *** procedure: execle PATH ENV [ARG] ...
3534 Similar to `execl', but the environment of the new process is
3535 specified by ENV, which must be a list of strings as returned by
3536 the `environ' procedure.
3538 This procedure is currently implemented using the `execve' system
3539 call, but we call it `execle' because of its Scheme calling
3542 *** procedure: strerror ERRNO
3543 Returns the Unix error message corresponding to ERRNO, an integer.
3545 *** procedure: primitive-exit [STATUS]
3546 Terminate the current process without unwinding the Scheme stack.
3547 This is would typically be useful after a fork. The exit status
3548 is STATUS if supplied, otherwise zero.
3550 *** procedure: times
3551 Returns an object with information about real and processor time.
3552 The following procedures accept such an object as an argument and
3553 return a selected component:
3556 The current real time, expressed as time units relative to an
3560 The CPU time units used by the calling process.
3563 The CPU time units used by the system on behalf of the
3567 The CPU time units used by terminated child processes of the
3568 calling process, whose status has been collected (e.g., using
3572 Similarly, the CPU times units used by the system on behalf of
3573 terminated child processes.
3575 ** Removed: list-length
3576 ** Removed: list-append, list-append!
3577 ** Removed: list-reverse, list-reverse!
3579 ** array-map renamed to array-map!
3581 ** serial-array-map renamed to serial-array-map!
3583 ** catch doesn't take #f as first argument any longer
3585 Previously, it was possible to pass #f instead of a key to `catch'.
3586 That would cause `catch' to pass a jump buffer object to the procedure
3587 passed as second argument. The procedure could then use this jump
3588 buffer objekt as an argument to throw.
3590 This mechanism has been removed since its utility doesn't motivate the
3591 extra complexity it introduces.
3593 ** The `#/' notation for lists now provokes a warning message from Guile.
3594 This syntax will be removed from Guile in the near future.
3596 To disable the warning message, set the GUILE_HUSH environment
3597 variable to any non-empty value.
3599 ** The newline character now prints as `#\newline', following the
3600 normal Scheme notation, not `#\nl'.
3602 * Changes to the gh_ interface
3604 ** The gh_enter function now takes care of loading the Guile startup files.
3605 gh_enter works by calling scm_boot_guile; see the remarks below.
3607 ** Function: void gh_write (SCM x)
3609 Write the printed representation of the scheme object x to the current
3610 output port. Corresponds to the scheme level `write'.
3612 ** gh_list_length renamed to gh_length.
3614 ** vector handling routines
3616 Several major changes. In particular, gh_vector() now resembles
3617 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
3618 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
3619 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
3620 vector-related gh_ functions have been implemented.
3622 ** pair and list routines
3624 Implemented several of the R4RS pair and list functions that were
3627 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
3629 New function. Converts double arrays back and forth between Scheme
3632 * Changes to the scm_ interface
3634 ** The function scm_boot_guile now takes care of loading the startup files.
3636 Guile's primary initialization function, scm_boot_guile, now takes
3637 care of loading `boot-9.scm', in the `ice-9' module, to initialize
3638 Guile, define the module system, and put together some standard
3639 bindings. It also loads `init.scm', which is intended to hold
3640 site-specific initialization code.
3642 Since Guile cannot operate properly until boot-9.scm is loaded, there
3643 is no reason to separate loading boot-9.scm from Guile's other
3644 initialization processes.
3646 This job used to be done by scm_compile_shell_switches, which didn't
3647 make much sense; in particular, it meant that people using Guile for
3648 non-shell-like applications had to jump through hoops to get Guile
3649 initialized properly.
3651 ** The function scm_compile_shell_switches no longer loads the startup files.
3652 Now, Guile always loads the startup files, whenever it is initialized;
3653 see the notes above for scm_boot_guile and scm_load_startup_files.
3655 ** Function: scm_load_startup_files
3656 This new function takes care of loading Guile's initialization file
3657 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
3658 this is always called by the Guile initialization process, it's
3659 probably not too useful to call this yourself, but it's there anyway.
3661 ** The semantics of smob marking have changed slightly.
3663 The smob marking function (the `mark' member of the scm_smobfuns
3664 structure) is no longer responsible for setting the mark bit on the
3665 smob. The generic smob handling code in the garbage collector will
3666 set this bit. The mark function need only ensure that any other
3667 objects the smob refers to get marked.
3669 Note that this change means that the smob's GC8MARK bit is typically
3670 already set upon entry to the mark function. Thus, marking functions
3671 which look like this:
3674 if (SCM_GC8MARKP (ptr))
3676 SCM_SETGC8MARK (ptr);
3677 ... mark objects to which the smob refers ...
3680 are now incorrect, since they will return early, and fail to mark any
3681 other objects the smob refers to. Some code in the Guile library used
3684 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
3686 If you have implemented your own I/O port type, by writing the
3687 functions required by the scm_ptobfuns and then calling scm_newptob,
3688 you will need to change your functions slightly.
3690 The functions in a scm_ptobfuns structure now expect the port itself
3691 as their argument; they used to expect the `stream' member of the
3692 port's scm_port_table structure. This allows functions in an
3693 scm_ptobfuns structure to easily access the port's cell (and any flags
3694 it its CAR), and the port's scm_port_table structure.
3696 Guile now passes the I/O port itself as the `port' argument in the
3697 following scm_ptobfuns functions:
3699 int (*free) (SCM port);
3700 int (*fputc) (int, SCM port);
3701 int (*fputs) (char *, SCM port);
3702 scm_sizet (*fwrite) SCM_P ((char *ptr,
3706 int (*fflush) (SCM port);
3707 int (*fgetc) (SCM port);
3708 int (*fclose) (SCM port);
3710 The interfaces to the `mark', `print', `equalp', and `fgets' methods
3713 If you have existing code which defines its own port types, it is easy
3714 to convert your code to the new interface; simply apply SCM_STREAM to
3715 the port argument to yield the value you code used to expect.
3717 Note that since both the port and the stream have the same type in the
3718 C code --- they are both SCM values --- the C compiler will not remind
3719 you if you forget to update your scm_ptobfuns functions.
3722 ** Function: int scm_internal_select (int fds,
3726 struct timeval *timeout);
3728 This is a replacement for the `select' function provided by the OS.
3729 It enables I/O blocking and sleeping to happen for one cooperative
3730 thread without blocking other threads. It also avoids busy-loops in
3731 these situations. It is intended that all I/O blocking and sleeping
3732 will finally go through this function. Currently, this function is
3733 only available on systems providing `gettimeofday' and `select'.
3735 ** Function: SCM scm_internal_stack_catch (SCM tag,
3736 scm_catch_body_t body,
3738 scm_catch_handler_t handler,
3741 A new sibling to the other two C level `catch' functions
3742 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
3743 the stack to be saved automatically into the variable `the-last-stack'
3744 (scm_the_last_stack_var) on error. This is necessary if you want to
3745 use advanced error reporting, such as calling scm_display_error and
3746 scm_display_backtrace. (They both take a stack object as argument.)
3748 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
3750 scm_catch_handler_t handler,
3753 Spawns a new thread. It does a job similar to
3754 scm_call_with_new_thread but takes arguments more suitable when
3755 spawning threads from application C code.
3757 ** The hook scm_error_callback has been removed. It was originally
3758 intended as a way for the user to install his own error handler. But
3759 that method works badly since it intervenes between throw and catch,
3760 thereby changing the semantics of expressions like (catch #t ...).
3761 The correct way to do it is to use one of the C level catch functions
3762 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
3764 ** Removed functions:
3766 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
3767 scm_list_reverse, scm_list_reverse_x
3769 ** New macros: SCM_LISTn where n is one of the integers 0-9.
3771 These can be used for pretty list creation from C. The idea is taken
3772 from Erick Gallesio's STk.
3774 ** scm_array_map renamed to scm_array_map_x
3776 ** mbstrings are now removed
3778 This means that the type codes scm_tc7_mb_string and
3779 scm_tc7_mb_substring has been removed.
3781 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
3783 Since we no longer support multi-byte strings, these I/O functions
3784 have been simplified, and renamed. Here are their old names, and
3785 their new names and arguments:
3787 scm_gen_putc -> void scm_putc (int c, SCM port);
3788 scm_gen_puts -> void scm_puts (char *s, SCM port);
3789 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
3790 scm_gen_getc -> void scm_getc (SCM port);
3793 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
3795 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
3797 SCM_TYP7S now masks away the bit which distinguishes substrings from
3800 ** scm_catch_body_t: Backward incompatible change!
3802 Body functions to scm_internal_catch and friends do not any longer
3803 take a second argument. This is because it is no longer possible to
3804 pass a #f arg to catch.
3806 ** Calls to scm_protect_object and scm_unprotect now nest properly.
3808 The function scm_protect_object protects its argument from being freed
3809 by the garbage collector. scm_unprotect_object removes that
3812 These functions now nest properly. That is, for every object O, there
3813 is a counter which scm_protect_object(O) increments and
3814 scm_unprotect_object(O) decrements, if the counter is greater than
3815 zero. Every object's counter is zero when it is first created. If an
3816 object's counter is greater than zero, the garbage collector will not
3817 reclaim its storage.
3819 This allows you to use scm_protect_object in your code without
3820 worrying that some other function you call will call
3821 scm_unprotect_object, and allow it to be freed. Assuming that the
3822 functions you call are well-behaved, and unprotect only those objects
3823 they protect, you can follow the same rule and have confidence that
3824 objects will be freed only at appropriate times.
3827 Changes in Guile 1.2 (released Tuesday, June 24 1997):
3829 * Changes to the distribution
3831 ** Nightly snapshots are now available from ftp.red-bean.com.
3832 The old server, ftp.cyclic.com, has been relinquished to its rightful
3835 Nightly snapshots of the Guile development sources are now available via
3836 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
3838 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
3839 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
3841 ** To run Guile without installing it, the procedure has changed a bit.
3843 If you used a separate build directory to compile Guile, you'll need
3844 to include the build directory in SCHEME_LOAD_PATH, as well as the
3845 source directory. See the `INSTALL' file for examples.
3847 * Changes to the procedure for linking libguile with your programs
3849 ** The standard Guile load path for Scheme code now includes
3850 $(datadir)/guile (usually /usr/local/share/guile). This means that
3851 you can install your own Scheme files there, and Guile will find them.
3852 (Previous versions of Guile only checked a directory whose name
3853 contained the Guile version number, so you had to re-install or move
3854 your Scheme sources each time you installed a fresh version of Guile.)
3856 The load path also includes $(datadir)/guile/site; we recommend
3857 putting individual Scheme files there. If you want to install a
3858 package with multiple source files, create a directory for them under
3861 ** Guile 1.2 will now use the Rx regular expression library, if it is
3862 installed on your system. When you are linking libguile into your own
3863 programs, this means you will have to link against -lguile, -lqt (if
3864 you configured Guile with thread support), and -lrx.
3866 If you are using autoconf to generate configuration scripts for your
3867 application, the following lines should suffice to add the appropriate
3868 libraries to your link command:
3870 ### Find Rx, quickthreads and libguile.
3871 AC_CHECK_LIB(rx, main)
3872 AC_CHECK_LIB(qt, main)
3873 AC_CHECK_LIB(guile, scm_shell)
3875 The Guile 1.2 distribution does not contain sources for the Rx
3876 library, as Guile 1.0 did. If you want to use Rx, you'll need to
3877 retrieve it from a GNU FTP site and install it separately.
3879 * Changes to Scheme functions and syntax
3881 ** The dynamic linking features of Guile are now enabled by default.
3882 You can disable them by giving the `--disable-dynamic-linking' option
3885 (dynamic-link FILENAME)
3887 Find the object file denoted by FILENAME (a string) and link it
3888 into the running Guile application. When everything works out,
3889 return a Scheme object suitable for representing the linked object
3890 file. Otherwise an error is thrown. How object files are
3891 searched is system dependent.
3893 (dynamic-object? VAL)
3895 Determine whether VAL represents a dynamically linked object file.
3897 (dynamic-unlink DYNOBJ)
3899 Unlink the indicated object file from the application. DYNOBJ
3900 should be one of the values returned by `dynamic-link'.
3902 (dynamic-func FUNCTION DYNOBJ)
3904 Search the C function indicated by FUNCTION (a string or symbol)
3905 in DYNOBJ and return some Scheme object that can later be used
3906 with `dynamic-call' to actually call this function. Right now,
3907 these Scheme objects are formed by casting the address of the
3908 function to `long' and converting this number to its Scheme
3911 (dynamic-call FUNCTION DYNOBJ)
3913 Call the C function indicated by FUNCTION and DYNOBJ. The
3914 function is passed no arguments and its return value is ignored.
3915 When FUNCTION is something returned by `dynamic-func', call that
3916 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
3917 etc.), look it up in DYNOBJ; this is equivalent to
3919 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
3921 Interrupts are deferred while the C function is executing (with
3922 SCM_DEFER_INTS/SCM_ALLOW_INTS).
3924 (dynamic-args-call FUNCTION DYNOBJ ARGS)
3926 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
3927 some arguments and return its return value. The C function is
3928 expected to take two arguments and return an `int', just like
3931 int c_func (int argc, char **argv);
3933 ARGS must be a list of strings and is converted into an array of
3934 `char *'. The array is passed in ARGV and its size in ARGC. The
3935 return value is converted to a Scheme number and returned from the
3936 call to `dynamic-args-call'.
3938 When dynamic linking is disabled or not supported on your system,
3939 the above functions throw errors, but they are still available.
3941 Here is a small example that works on GNU/Linux:
3943 (define libc-obj (dynamic-link "libc.so"))
3944 (dynamic-args-call 'rand libc-obj '())
3946 See the file `libguile/DYNAMIC-LINKING' for additional comments.
3948 ** The #/ syntax for module names is depreciated, and will be removed
3949 in a future version of Guile. Instead of
3957 The latter syntax is more consistent with existing Lisp practice.
3959 ** Guile now does fancier printing of structures. Structures are the
3960 underlying implementation for records, which in turn are used to
3961 implement modules, so all of these object now print differently and in
3962 a more informative way.
3964 The Scheme printer will examine the builtin variable *struct-printer*
3965 whenever it needs to print a structure object. When this variable is
3966 not `#f' it is deemed to be a procedure and will be applied to the
3967 structure object and the output port. When *struct-printer* is `#f'
3968 or the procedure return `#f' the structure object will be printed in
3969 the boring #<struct 80458270> form.
3971 This hook is used by some routines in ice-9/boot-9.scm to implement
3972 type specific printing routines. Please read the comments there about
3975 One of the more specific uses of structs are records. The printing
3976 procedure that could be passed to MAKE-RECORD-TYPE is now actually
3977 called. It should behave like a *struct-printer* procedure (described
3980 ** Guile now supports a new R4RS-compliant syntax for keywords. A
3981 token of the form #:NAME, where NAME has the same syntax as a Scheme
3982 symbol, is the external representation of the keyword named NAME.
3983 Keyword objects print using this syntax as well, so values containing
3984 keyword objects can be read back into Guile. When used in an
3985 expression, keywords are self-quoting objects.
3987 Guile suports this read syntax, and uses this print syntax, regardless
3988 of the current setting of the `keyword' read option. The `keyword'
3989 read option only controls whether Guile recognizes the `:NAME' syntax,
3990 which is incompatible with R4RS. (R4RS says such token represent
3993 ** Guile has regular expression support again. Guile 1.0 included
3994 functions for matching regular expressions, based on the Rx library.
3995 In Guile 1.1, the Guile/Rx interface was removed to simplify the
3996 distribution, and thus Guile had no regular expression support. Guile
3997 1.2 again supports the most commonly used functions, and supports all
3998 of SCSH's regular expression functions.
4000 If your system does not include a POSIX regular expression library,
4001 and you have not linked Guile with a third-party regexp library such as
4002 Rx, these functions will not be available. You can tell whether your
4003 Guile installation includes regular expression support by checking
4004 whether the `*features*' list includes the `regex' symbol.
4006 *** regexp functions
4008 By default, Guile supports POSIX extended regular expressions. That
4009 means that the characters `(', `)', `+' and `?' are special, and must
4010 be escaped if you wish to match the literal characters.
4012 This regular expression interface was modeled after that implemented
4013 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
4014 with SCSH regular expressions.
4016 **** Function: string-match PATTERN STR [START]
4017 Compile the string PATTERN into a regular expression and compare
4018 it with STR. The optional numeric argument START specifies the
4019 position of STR at which to begin matching.
4021 `string-match' returns a "match structure" which describes what,
4022 if anything, was matched by the regular expression. *Note Match
4023 Structures::. If STR does not match PATTERN at all,
4024 `string-match' returns `#f'.
4026 Each time `string-match' is called, it must compile its PATTERN
4027 argument into a regular expression structure. This operation is
4028 expensive, which makes `string-match' inefficient if the same regular
4029 expression is used several times (for example, in a loop). For better
4030 performance, you can compile a regular expression in advance and then
4031 match strings against the compiled regexp.
4033 **** Function: make-regexp STR [FLAGS]
4034 Compile the regular expression described by STR, and return the
4035 compiled regexp structure. If STR does not describe a legal
4036 regular expression, `make-regexp' throws a
4037 `regular-expression-syntax' error.
4039 FLAGS may be the bitwise-or of one or more of the following:
4041 **** Constant: regexp/extended
4042 Use POSIX Extended Regular Expression syntax when interpreting
4043 STR. If not set, POSIX Basic Regular Expression syntax is used.
4044 If the FLAGS argument is omitted, we assume regexp/extended.
4046 **** Constant: regexp/icase
4047 Do not differentiate case. Subsequent searches using the
4048 returned regular expression will be case insensitive.
4050 **** Constant: regexp/newline
4051 Match-any-character operators don't match a newline.
4053 A non-matching list ([^...]) not containing a newline matches a
4056 Match-beginning-of-line operator (^) matches the empty string
4057 immediately after a newline, regardless of whether the FLAGS
4058 passed to regexp-exec contain regexp/notbol.
4060 Match-end-of-line operator ($) matches the empty string
4061 immediately before a newline, regardless of whether the FLAGS
4062 passed to regexp-exec contain regexp/noteol.
4064 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
4065 Match the compiled regular expression REGEXP against `str'. If
4066 the optional integer START argument is provided, begin matching
4067 from that position in the string. Return a match structure
4068 describing the results of the match, or `#f' if no match could be
4071 FLAGS may be the bitwise-or of one or more of the following:
4073 **** Constant: regexp/notbol
4074 The match-beginning-of-line operator always fails to match (but
4075 see the compilation flag regexp/newline above) This flag may be
4076 used when different portions of a string are passed to
4077 regexp-exec and the beginning of the string should not be
4078 interpreted as the beginning of the line.
4080 **** Constant: regexp/noteol
4081 The match-end-of-line operator always fails to match (but see the
4082 compilation flag regexp/newline above)
4084 **** Function: regexp? OBJ
4085 Return `#t' if OBJ is a compiled regular expression, or `#f'
4088 Regular expressions are commonly used to find patterns in one string
4089 and replace them with the contents of another string.
4091 **** Function: regexp-substitute PORT MATCH [ITEM...]
4092 Write to the output port PORT selected contents of the match
4093 structure MATCH. Each ITEM specifies what should be written, and
4094 may be one of the following arguments:
4096 * A string. String arguments are written out verbatim.
4098 * An integer. The submatch with that number is written.
4100 * The symbol `pre'. The portion of the matched string preceding
4101 the regexp match is written.
4103 * The symbol `post'. The portion of the matched string
4104 following the regexp match is written.
4106 PORT may be `#f', in which case nothing is written; instead,
4107 `regexp-substitute' constructs a string from the specified ITEMs
4110 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
4111 Similar to `regexp-substitute', but can be used to perform global
4112 substitutions on STR. Instead of taking a match structure as an
4113 argument, `regexp-substitute/global' takes two string arguments: a
4114 REGEXP string describing a regular expression, and a TARGET string
4115 which should be matched against this regular expression.
4117 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
4120 * A function may be supplied. When this function is called, it
4121 will be passed one argument: a match structure for a given
4122 regular expression match. It should return a string to be
4123 written out to PORT.
4125 * The `post' symbol causes `regexp-substitute/global' to recurse
4126 on the unmatched portion of STR. This *must* be supplied in
4127 order to perform global search-and-replace on STR; if it is
4128 not present among the ITEMs, then `regexp-substitute/global'
4129 will return after processing a single match.
4131 *** Match Structures
4133 A "match structure" is the object returned by `string-match' and
4134 `regexp-exec'. It describes which portion of a string, if any, matched
4135 the given regular expression. Match structures include: a reference to
4136 the string that was checked for matches; the starting and ending
4137 positions of the regexp match; and, if the regexp included any
4138 parenthesized subexpressions, the starting and ending positions of each
4141 In each of the regexp match functions described below, the `match'
4142 argument must be a match structure returned by a previous call to
4143 `string-match' or `regexp-exec'. Most of these functions return some
4144 information about the original target string that was matched against a
4145 regular expression; we will call that string TARGET for easy reference.
4147 **** Function: regexp-match? OBJ
4148 Return `#t' if OBJ is a match structure returned by a previous
4149 call to `regexp-exec', or `#f' otherwise.
4151 **** Function: match:substring MATCH [N]
4152 Return the portion of TARGET matched by subexpression number N.
4153 Submatch 0 (the default) represents the entire regexp match. If
4154 the regular expression as a whole matched, but the subexpression
4155 number N did not match, return `#f'.
4157 **** Function: match:start MATCH [N]
4158 Return the starting position of submatch number N.
4160 **** Function: match:end MATCH [N]
4161 Return the ending position of submatch number N.
4163 **** Function: match:prefix MATCH
4164 Return the unmatched portion of TARGET preceding the regexp match.
4166 **** Function: match:suffix MATCH
4167 Return the unmatched portion of TARGET following the regexp match.
4169 **** Function: match:count MATCH
4170 Return the number of parenthesized subexpressions from MATCH.
4171 Note that the entire regular expression match itself counts as a
4172 subexpression, and failed submatches are included in the count.
4174 **** Function: match:string MATCH
4175 Return the original TARGET string.
4177 *** Backslash Escapes
4179 Sometimes you will want a regexp to match characters like `*' or `$'
4180 exactly. For example, to check whether a particular string represents
4181 a menu entry from an Info node, it would be useful to match it against
4182 a regexp like `^* [^:]*::'. However, this won't work; because the
4183 asterisk is a metacharacter, it won't match the `*' at the beginning of
4184 the string. In this case, we want to make the first asterisk un-magic.
4186 You can do this by preceding the metacharacter with a backslash
4187 character `\'. (This is also called "quoting" the metacharacter, and
4188 is known as a "backslash escape".) When Guile sees a backslash in a
4189 regular expression, it considers the following glyph to be an ordinary
4190 character, no matter what special meaning it would ordinarily have.
4191 Therefore, we can make the above example work by changing the regexp to
4192 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
4193 to match only a single asterisk in the target string.
4195 Since the backslash is itself a metacharacter, you may force a
4196 regexp to match a backslash in the target string by preceding the
4197 backslash with itself. For example, to find variable references in a
4198 TeX program, you might want to find occurrences of the string `\let\'
4199 followed by any number of alphabetic characters. The regular expression
4200 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
4201 each match a single backslash in the target string.
4203 **** Function: regexp-quote STR
4204 Quote each special character found in STR with a backslash, and
4205 return the resulting string.
4207 *Very important:* Using backslash escapes in Guile source code (as
4208 in Emacs Lisp or C) can be tricky, because the backslash character has
4209 special meaning for the Guile reader. For example, if Guile encounters
4210 the character sequence `\n' in the middle of a string while processing
4211 Scheme code, it replaces those characters with a newline character.
4212 Similarly, the character sequence `\t' is replaced by a horizontal tab.
4213 Several of these "escape sequences" are processed by the Guile reader
4214 before your code is executed. Unrecognized escape sequences are
4215 ignored: if the characters `\*' appear in a string, they will be
4216 translated to the single character `*'.
4218 This translation is obviously undesirable for regular expressions,
4219 since we want to be able to include backslashes in a string in order to
4220 escape regexp metacharacters. Therefore, to make sure that a backslash
4221 is preserved in a string in your Guile program, you must use *two*
4222 consecutive backslashes:
4224 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
4226 The string in this example is preprocessed by the Guile reader before
4227 any code is executed. The resulting argument to `make-regexp' is the
4228 string `^\* [^:]*', which is what we really want.
4230 This also means that in order to write a regular expression that
4231 matches a single backslash character, the regular expression string in
4232 the source code must include *four* backslashes. Each consecutive pair
4233 of backslashes gets translated by the Guile reader to a single
4234 backslash, and the resulting double-backslash is interpreted by the
4235 regexp engine as matching a single backslash character. Hence:
4237 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
4239 The reason for the unwieldiness of this syntax is historical. Both
4240 regular expression pattern matchers and Unix string processing systems
4241 have traditionally used backslashes with the special meanings described
4242 above. The POSIX regular expression specification and ANSI C standard
4243 both require these semantics. Attempting to abandon either convention
4244 would cause other kinds of compatibility problems, possibly more severe
4245 ones. Therefore, without extending the Scheme reader to support
4246 strings with different quoting conventions (an ungainly and confusing
4247 extension when implemented in other languages), we must adhere to this
4248 cumbersome escape syntax.
4250 * Changes to the gh_ interface
4252 * Changes to the scm_ interface
4254 * Changes to system call interfaces:
4256 ** The value returned by `raise' is now unspecified. It throws an exception
4259 *** A new procedure `sigaction' can be used to install signal handlers
4261 (sigaction signum [action] [flags])
4263 signum is the signal number, which can be specified using the value
4266 If action is omitted, sigaction returns a pair: the CAR is the current
4267 signal hander, which will be either an integer with the value SIG_DFL
4268 (default action) or SIG_IGN (ignore), or the Scheme procedure which
4269 handles the signal, or #f if a non-Scheme procedure handles the
4270 signal. The CDR contains the current sigaction flags for the handler.
4272 If action is provided, it is installed as the new handler for signum.
4273 action can be a Scheme procedure taking one argument, or the value of
4274 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
4275 whatever signal handler was installed before sigaction was first used.
4276 Flags can optionally be specified for the new handler (SA_RESTART is
4277 always used if the system provides it, so need not be specified.) The
4278 return value is a pair with information about the old handler as
4281 This interface does not provide access to the "signal blocking"
4282 facility. Maybe this is not needed, since the thread support may
4283 provide solutions to the problem of consistent access to data
4286 *** A new procedure `flush-all-ports' is equivalent to running
4287 `force-output' on every port open for output.
4289 ** Guile now provides information on how it was built, via the new
4290 global variable, %guile-build-info. This variable records the values
4291 of the standard GNU makefile directory variables as an assocation
4292 list, mapping variable names (symbols) onto directory paths (strings).
4293 For example, to find out where the Guile link libraries were
4294 installed, you can say:
4296 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
4299 * Changes to the scm_ interface
4301 ** The new function scm_handle_by_message_noexit is just like the
4302 existing scm_handle_by_message function, except that it doesn't call
4303 exit to terminate the process. Instead, it prints a message and just
4304 returns #f. This might be a more appropriate catch-all handler for
4305 new dynamic roots and threads.
4308 Changes in Guile 1.1 (released Friday, May 16 1997):
4310 * Changes to the distribution.
4312 The Guile 1.0 distribution has been split up into several smaller
4314 guile-core --- the Guile interpreter itself.
4315 guile-tcltk --- the interface between the Guile interpreter and
4316 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
4317 is a toolkit for building graphical user interfaces.
4318 guile-rgx-ctax --- the interface between Guile and the Rx regular
4319 expression matcher, and the translator for the Ctax
4320 programming language. These are packaged together because the
4321 Ctax translator uses Rx to parse Ctax source code.
4323 This NEWS file describes the changes made to guile-core since the 1.0
4326 We no longer distribute the documentation, since it was either out of
4327 date, or incomplete. As soon as we have current documentation, we
4332 * Changes to the stand-alone interpreter
4334 ** guile now accepts command-line arguments compatible with SCSH, Olin
4335 Shivers' Scheme Shell.
4337 In general, arguments are evaluated from left to right, but there are
4338 exceptions. The following switches stop argument processing, and
4339 stash all remaining command-line arguments as the value returned by
4340 the (command-line) function.
4341 -s SCRIPT load Scheme source code from FILE, and exit
4342 -c EXPR evalute Scheme expression EXPR, and exit
4343 -- stop scanning arguments; run interactively
4345 The switches below are processed as they are encountered.
4346 -l FILE load Scheme source code from FILE
4347 -e FUNCTION after reading script, apply FUNCTION to
4348 command line arguments
4349 -ds do -s script at this point
4350 --emacs enable Emacs protocol (experimental)
4351 -h, --help display this help and exit
4352 -v, --version display version information and exit
4353 \ read arguments from following script lines
4355 So, for example, here is a Guile script named `ekko' (thanks, Olin)
4356 which re-implements the traditional "echo" command:
4358 #!/usr/local/bin/guile -s
4361 (map (lambda (arg) (display arg) (display " "))
4365 (main (command-line))
4367 Suppose we invoke this script as follows:
4369 ekko a speckled gecko
4371 Through the magic of Unix script processing (triggered by the `#!'
4372 token at the top of the file), /usr/local/bin/guile receives the
4373 following list of command-line arguments:
4375 ("-s" "./ekko" "a" "speckled" "gecko")
4377 Unix inserts the name of the script after the argument specified on
4378 the first line of the file (in this case, "-s"), and then follows that
4379 with the arguments given to the script. Guile loads the script, which
4380 defines the `main' function, and then applies it to the list of
4381 remaining command-line arguments, ("a" "speckled" "gecko").
4383 In Unix, the first line of a script file must take the following form:
4385 #!INTERPRETER ARGUMENT
4387 where INTERPRETER is the absolute filename of the interpreter
4388 executable, and ARGUMENT is a single command-line argument to pass to
4391 You may only pass one argument to the interpreter, and its length is
4392 limited. These restrictions can be annoying to work around, so Guile
4393 provides a general mechanism (borrowed from, and compatible with,
4394 SCSH) for circumventing them.
4396 If the ARGUMENT in a Guile script is a single backslash character,
4397 `\', Guile will open the script file, parse arguments from its second
4398 and subsequent lines, and replace the `\' with them. So, for example,
4399 here is another implementation of the `ekko' script:
4401 #!/usr/local/bin/guile \
4405 (for-each (lambda (arg) (display arg) (display " "))
4409 If the user invokes this script as follows:
4411 ekko a speckled gecko
4413 Unix expands this into
4415 /usr/local/bin/guile \ ekko a speckled gecko
4417 When Guile sees the `\' argument, it replaces it with the arguments
4418 read from the second line of the script, producing:
4420 /usr/local/bin/guile -e main -s ekko a speckled gecko
4422 This tells Guile to load the `ekko' script, and apply the function
4423 `main' to the argument list ("a" "speckled" "gecko").
4425 Here is how Guile parses the command-line arguments:
4426 - Each space character terminates an argument. This means that two
4427 spaces in a row introduce an empty-string argument.
4428 - The tab character is not permitted (unless you quote it with the
4429 backslash character, as described below), to avoid confusion.
4430 - The newline character terminates the sequence of arguments, and will
4431 also terminate a final non-empty argument. (However, a newline
4432 following a space will not introduce a final empty-string argument;
4433 it only terminates the argument list.)
4434 - The backslash character is the escape character. It escapes
4435 backslash, space, tab, and newline. The ANSI C escape sequences
4436 like \n and \t are also supported. These produce argument
4437 constituents; the two-character combination \n doesn't act like a
4438 terminating newline. The escape sequence \NNN for exactly three
4439 octal digits reads as the character whose ASCII code is NNN. As
4440 above, characters produced this way are argument constituents.
4441 Backslash followed by other characters is not allowed.
4443 * Changes to the procedure for linking libguile with your programs
4445 ** Guile now builds and installs a shared guile library, if your
4446 system support shared libraries. (It still builds a static library on
4447 all systems.) Guile automatically detects whether your system
4448 supports shared libraries. To prevent Guile from buildisg shared
4449 libraries, pass the `--disable-shared' flag to the configure script.
4451 Guile takes longer to compile when it builds shared libraries, because
4452 it must compile every file twice --- once to produce position-
4453 independent object code, and once to produce normal object code.
4455 ** The libthreads library has been merged into libguile.
4457 To link a program against Guile, you now need only link against
4458 -lguile and -lqt; -lthreads is no longer needed. If you are using
4459 autoconf to generate configuration scripts for your application, the
4460 following lines should suffice to add the appropriate libraries to
4463 ### Find quickthreads and libguile.
4464 AC_CHECK_LIB(qt, main)
4465 AC_CHECK_LIB(guile, scm_shell)
4467 * Changes to Scheme functions
4469 ** Guile Scheme's special syntax for keyword objects is now optional,
4470 and disabled by default.
4472 The syntax variation from R4RS made it difficult to port some
4473 interesting packages to Guile. The routines which accepted keyword
4474 arguments (mostly in the module system) have been modified to also
4475 accept symbols whose names begin with `:'.
4477 To change the keyword syntax, you must first import the (ice-9 debug)
4479 (use-modules (ice-9 debug))
4481 Then you can enable the keyword syntax as follows:
4482 (read-set! keywords 'prefix)
4484 To disable keyword syntax, do this:
4485 (read-set! keywords #f)
4487 ** Many more primitive functions accept shared substrings as
4488 arguments. In the past, these functions required normal, mutable
4489 strings as arguments, although they never made use of this
4492 ** The uniform array functions now operate on byte vectors. These
4493 functions are `array-fill!', `serial-array-copy!', `array-copy!',
4494 `serial-array-map', `array-map', `array-for-each', and
4497 ** The new functions `trace' and `untrace' implement simple debugging
4498 support for Scheme functions.
4500 The `trace' function accepts any number of procedures as arguments,
4501 and tells the Guile interpreter to display each procedure's name and
4502 arguments each time the procedure is invoked. When invoked with no
4503 arguments, `trace' returns the list of procedures currently being
4506 The `untrace' function accepts any number of procedures as arguments,
4507 and tells the Guile interpreter not to trace them any more. When
4508 invoked with no arguments, `untrace' untraces all curretly traced
4511 The tracing in Guile has an advantage over most other systems: we
4512 don't create new procedure objects, but mark the procedure objects
4513 themselves. This means that anonymous and internal procedures can be
4516 ** The function `assert-repl-prompt' has been renamed to
4517 `set-repl-prompt!'. It takes one argument, PROMPT.
4518 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
4519 - If PROMPT is a string, we use it as a prompt.
4520 - If PROMPT is a procedure accepting no arguments, we call it, and
4521 display the result as a prompt.
4522 - Otherwise, we display "> ".
4524 ** The new function `eval-string' reads Scheme expressions from a
4525 string and evaluates them, returning the value of the last expression
4526 in the string. If the string contains no expressions, it returns an
4529 ** The new function `thunk?' returns true iff its argument is a
4530 procedure of zero arguments.
4532 ** `defined?' is now a builtin function, instead of syntax. This
4533 means that its argument should be quoted. It returns #t iff its
4534 argument is bound in the current module.
4536 ** The new syntax `use-modules' allows you to add new modules to your
4537 environment without re-typing a complete `define-module' form. It
4538 accepts any number of module names as arguments, and imports their
4539 public bindings into the current module.
4541 ** The new function (module-defined? NAME MODULE) returns true iff
4542 NAME, a symbol, is defined in MODULE, a module object.
4544 ** The new function `builtin-bindings' creates and returns a hash
4545 table containing copies of all the root module's bindings.
4547 ** The new function `builtin-weak-bindings' does the same as
4548 `builtin-bindings', but creates a doubly-weak hash table.
4550 ** The `equal?' function now considers variable objects to be
4551 equivalent if they have the same name and the same value.
4553 ** The new function `command-line' returns the command-line arguments
4554 given to Guile, as a list of strings.
4556 When using guile as a script interpreter, `command-line' returns the
4557 script's arguments; those processed by the interpreter (like `-s' or
4558 `-c') are omitted. (In other words, you get the normal, expected
4559 behavior.) Any application that uses scm_shell to process its
4560 command-line arguments gets this behavior as well.
4562 ** The new function `load-user-init' looks for a file called `.guile'
4563 in the user's home directory, and loads it if it exists. This is
4564 mostly for use by the code generated by scm_compile_shell_switches,
4565 but we thought it might also be useful in other circumstances.
4567 ** The new function `log10' returns the base-10 logarithm of its
4570 ** Changes to I/O functions
4572 *** The functions `read', `primitive-load', `read-and-eval!', and
4573 `primitive-load-path' no longer take optional arguments controlling
4574 case insensitivity and a `#' parser.
4576 Case sensitivity is now controlled by a read option called
4577 `case-insensitive'. The user can add new `#' syntaxes with the
4578 `read-hash-extend' function (see below).
4580 *** The new function `read-hash-extend' allows the user to change the
4581 syntax of Guile Scheme in a somewhat controlled way.
4583 (read-hash-extend CHAR PROC)
4584 When parsing S-expressions, if we read a `#' character followed by
4585 the character CHAR, use PROC to parse an object from the stream.
4586 If PROC is #f, remove any parsing procedure registered for CHAR.
4588 The reader applies PROC to two arguments: CHAR and an input port.
4590 *** The new functions read-delimited and read-delimited! provide a
4591 general mechanism for doing delimited input on streams.
4593 (read-delimited DELIMS [PORT HANDLE-DELIM])
4594 Read until we encounter one of the characters in DELIMS (a string),
4595 or end-of-file. PORT is the input port to read from; it defaults to
4596 the current input port. The HANDLE-DELIM parameter determines how
4597 the terminating character is handled; it should be one of the
4600 'trim omit delimiter from result
4601 'peek leave delimiter character in input stream
4602 'concat append delimiter character to returned value
4603 'split return a pair: (RESULT . TERMINATOR)
4605 HANDLE-DELIM defaults to 'peek.
4607 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
4608 A side-effecting variant of `read-delimited'.
4610 The data is written into the string BUF at the indices in the
4611 half-open interval [START, END); the default interval is the whole
4612 string: START = 0 and END = (string-length BUF). The values of
4613 START and END must specify a well-defined interval in BUF, i.e.
4614 0 <= START <= END <= (string-length BUF).
4616 It returns NBYTES, the number of bytes read. If the buffer filled
4617 up without a delimiter character being found, it returns #f. If the
4618 port is at EOF when the read starts, it returns the EOF object.
4620 If an integer is returned (i.e., the read is successfully terminated
4621 by reading a delimiter character), then the HANDLE-DELIM parameter
4622 determines how to handle the terminating character. It is described
4623 above, and defaults to 'peek.
4625 (The descriptions of these functions were borrowed from the SCSH
4626 manual, by Olin Shivers and Brian Carlstrom.)
4628 *** The `%read-delimited!' function is the primitive used to implement
4629 `read-delimited' and `read-delimited!'.
4631 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
4633 This returns a pair of values: (TERMINATOR . NUM-READ).
4634 - TERMINATOR describes why the read was terminated. If it is a
4635 character or the eof object, then that is the value that terminated
4636 the read. If it is #f, the function filled the buffer without finding
4637 a delimiting character.
4638 - NUM-READ is the number of characters read into BUF.
4640 If the read is successfully terminated by reading a delimiter
4641 character, then the gobble? parameter determines what to do with the
4642 terminating character. If true, the character is removed from the
4643 input stream; if false, the character is left in the input stream
4644 where a subsequent read operation will retrieve it. In either case,
4645 the character is also the first value returned by the procedure call.
4647 (The descriptions of this function was borrowed from the SCSH manual,
4648 by Olin Shivers and Brian Carlstrom.)
4650 *** The `read-line' and `read-line!' functions have changed; they now
4651 trim the terminator by default; previously they appended it to the
4652 returned string. For the old behavior, use (read-line PORT 'concat).
4654 *** The functions `uniform-array-read!' and `uniform-array-write!' now
4655 take new optional START and END arguments, specifying the region of
4656 the array to read and write.
4658 *** The `ungetc-char-ready?' function has been removed. We feel it's
4659 inappropriate for an interface to expose implementation details this
4662 ** Changes to the Unix library and system call interface
4664 *** The new fcntl function provides access to the Unix `fcntl' system
4667 (fcntl PORT COMMAND VALUE)
4668 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
4669 Values for COMMAND are:
4671 F_DUPFD duplicate a file descriptor
4672 F_GETFD read the descriptor's close-on-exec flag
4673 F_SETFD set the descriptor's close-on-exec flag to VALUE
4674 F_GETFL read the descriptor's flags, as set on open
4675 F_SETFL set the descriptor's flags, as set on open to VALUE
4676 F_GETOWN return the process ID of a socket's owner, for SIGIO
4677 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
4678 FD_CLOEXEC not sure what this is
4680 For details, see the documentation for the fcntl system call.
4682 *** The arguments to `select' have changed, for compatibility with
4683 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
4684 expected behavior. The MILLISECONDS parameter has been changed to
4685 MICROSECONDS, to more closely resemble the underlying system call.
4686 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
4687 corresponding return set will be the same.
4689 *** The arguments to the `mknod' system call have changed. They are
4692 (mknod PATH TYPE PERMS DEV)
4693 Create a new file (`node') in the file system. PATH is the name of
4694 the file to create. TYPE is the kind of file to create; it should
4695 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
4696 permission bits to give the newly created file. If TYPE is
4697 'block-special or 'char-special, DEV specifies which device the
4698 special file refers to; its interpretation depends on the kind of
4699 special file being created.
4701 *** The `fork' function has been renamed to `primitive-fork', to avoid
4702 clashing with various SCSH forks.
4704 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
4705 and `recvfrom!'. They no longer accept a size for a second argument;
4706 you must pass a string to hold the received value. They no longer
4707 return the buffer. Instead, `recv' returns the length of the message
4708 received, and `recvfrom' returns a pair containing the packet's length
4709 and originating address.
4711 *** The file descriptor datatype has been removed, as have the
4712 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
4713 We plan to replace these functions with a SCSH-compatible interface.
4715 *** The `create' function has been removed; it's just a special case
4718 *** There are new functions to break down process termination status
4719 values. In the descriptions below, STATUS is a value returned by
4722 (status:exit-val STATUS)
4723 If the child process exited normally, this function returns the exit
4724 code for the child process (i.e., the value passed to exit, or
4725 returned from main). If the child process did not exit normally,
4726 this function returns #f.
4728 (status:stop-sig STATUS)
4729 If the child process was suspended by a signal, this function
4730 returns the signal that suspended the child. Otherwise, it returns
4733 (status:term-sig STATUS)
4734 If the child process terminated abnormally, this function returns
4735 the signal that terminated the child. Otherwise, this function
4738 POSIX promises that exactly one of these functions will return true on
4739 a valid STATUS value.
4741 These functions are compatible with SCSH.
4743 *** There are new accessors and setters for the broken-out time vectors
4744 returned by `localtime', `gmtime', and that ilk. They are:
4746 Component Accessor Setter
4747 ========================= ============ ============
4748 seconds tm:sec set-tm:sec
4749 minutes tm:min set-tm:min
4750 hours tm:hour set-tm:hour
4751 day of the month tm:mday set-tm:mday
4752 month tm:mon set-tm:mon
4753 year tm:year set-tm:year
4754 day of the week tm:wday set-tm:wday
4755 day in the year tm:yday set-tm:yday
4756 daylight saving time tm:isdst set-tm:isdst
4757 GMT offset, seconds tm:gmtoff set-tm:gmtoff
4758 name of time zone tm:zone set-tm:zone
4760 *** There are new accessors for the vectors returned by `uname',
4761 describing the host system:
4764 ============================================== ================
4765 name of the operating system implementation utsname:sysname
4766 network name of this machine utsname:nodename
4767 release level of the operating system utsname:release
4768 version level of the operating system utsname:version
4769 machine hardware platform utsname:machine
4771 *** There are new accessors for the vectors returned by `getpw',
4772 `getpwnam', `getpwuid', and `getpwent', describing entries from the
4773 system's user database:
4776 ====================== =================
4777 user name passwd:name
4778 user password passwd:passwd
4781 real name passwd:gecos
4782 home directory passwd:dir
4783 shell program passwd:shell
4785 *** There are new accessors for the vectors returned by `getgr',
4786 `getgrnam', `getgrgid', and `getgrent', describing entries from the
4787 system's group database:
4790 ======================= ============
4791 group name group:name
4792 group password group:passwd
4794 group members group:mem
4796 *** There are new accessors for the vectors returned by `gethost',
4797 `gethostbyaddr', `gethostbyname', and `gethostent', describing
4801 ========================= ===============
4802 official name of host hostent:name
4803 alias list hostent:aliases
4804 host address type hostent:addrtype
4805 length of address hostent:length
4806 list of addresses hostent:addr-list
4808 *** There are new accessors for the vectors returned by `getnet',
4809 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
4813 ========================= ===============
4814 official name of net netent:name
4815 alias list netent:aliases
4816 net number type netent:addrtype
4817 net number netent:net
4819 *** There are new accessors for the vectors returned by `getproto',
4820 `getprotobyname', `getprotobynumber', and `getprotoent', describing
4824 ========================= ===============
4825 official protocol name protoent:name
4826 alias list protoent:aliases
4827 protocol number protoent:proto
4829 *** There are new accessors for the vectors returned by `getserv',
4830 `getservbyname', `getservbyport', and `getservent', describing
4834 ========================= ===============
4835 official service name servent:name
4836 alias list servent:aliases
4837 port number servent:port
4838 protocol to use servent:proto
4840 *** There are new accessors for the sockaddr structures returned by
4841 `accept', `getsockname', `getpeername', `recvfrom!':
4844 ======================================== ===============
4845 address format (`family') sockaddr:fam
4846 path, for file domain addresses sockaddr:path
4847 address, for internet domain addresses sockaddr:addr
4848 TCP or UDP port, for internet sockaddr:port
4850 *** The `getpwent', `getgrent', `gethostent', `getnetent',
4851 `getprotoent', and `getservent' functions now return #f at the end of
4852 the user database. (They used to throw an exception.)
4854 Note that calling MUMBLEent function is equivalent to calling the
4855 corresponding MUMBLE function with no arguments.
4857 *** The `setpwent', `setgrent', `sethostent', `setnetent',
4858 `setprotoent', and `setservent' routines now take no arguments.
4860 *** The `gethost', `getproto', `getnet', and `getserv' functions now
4861 provide more useful information when they throw an exception.
4863 *** The `lnaof' function has been renamed to `inet-lnaof'.
4865 *** Guile now claims to have the `current-time' feature.
4867 *** The `mktime' function now takes an optional second argument ZONE,
4868 giving the time zone to use for the conversion. ZONE should be a
4869 string, in the same format as expected for the "TZ" environment variable.
4871 *** The `strptime' function now returns a pair (TIME . COUNT), where
4872 TIME is the parsed time as a vector, and COUNT is the number of
4873 characters from the string left unparsed. This function used to
4874 return the remaining characters as a string.
4876 *** The `gettimeofday' function has replaced the old `time+ticks' function.
4877 The return value is now (SECONDS . MICROSECONDS); the fractional
4878 component is no longer expressed in "ticks".
4880 *** The `ticks/sec' constant has been removed, in light of the above change.
4882 * Changes to the gh_ interface
4884 ** gh_eval_str() now returns an SCM object which is the result of the
4887 ** gh_scm2str() now copies the Scheme data to a caller-provided C
4890 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
4891 and returns the array
4893 ** gh_scm2str0() is gone: there is no need to distinguish
4894 null-terminated from non-null-terminated, since gh_scm2newstr() allows
4895 the user to interpret the data both ways.
4897 * Changes to the scm_ interface
4899 ** The new function scm_symbol_value0 provides an easy way to get a
4900 symbol's value from C code:
4902 SCM scm_symbol_value0 (char *NAME)
4903 Return the value of the symbol named by the null-terminated string
4904 NAME in the current module. If the symbol named NAME is unbound in
4905 the current module, return SCM_UNDEFINED.
4907 ** The new function scm_sysintern0 creates new top-level variables,
4908 without assigning them a value.
4910 SCM scm_sysintern0 (char *NAME)
4911 Create a new Scheme top-level variable named NAME. NAME is a
4912 null-terminated string. Return the variable's value cell.
4914 ** The function scm_internal_catch is the guts of catch. It handles
4915 all the mechanics of setting up a catch target, invoking the catch
4916 body, and perhaps invoking the handler if the body does a throw.
4918 The function is designed to be usable from C code, but is general
4919 enough to implement all the semantics Guile Scheme expects from throw.
4921 TAG is the catch tag. Typically, this is a symbol, but this function
4922 doesn't actually care about that.
4924 BODY is a pointer to a C function which runs the body of the catch;
4925 this is the code you can throw from. We call it like this:
4926 BODY (BODY_DATA, JMPBUF)
4928 BODY_DATA is just the BODY_DATA argument we received; we pass it
4929 through to BODY as its first argument. The caller can make
4930 BODY_DATA point to anything useful that BODY might need.
4931 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
4932 which we have just created and initialized.
4934 HANDLER is a pointer to a C function to deal with a throw to TAG,
4935 should one occur. We call it like this:
4936 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
4938 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
4939 same idea as BODY_DATA above.
4940 THROWN_TAG is the tag that the user threw to; usually this is
4941 TAG, but it could be something else if TAG was #t (i.e., a
4942 catch-all), or the user threw to a jmpbuf.
4943 THROW_ARGS is the list of arguments the user passed to the THROW
4946 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
4947 is just a pointer we pass through to HANDLER. We don't actually
4948 use either of those pointers otherwise ourselves. The idea is
4949 that, if our caller wants to communicate something to BODY or
4950 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
4951 HANDLER can then use. Think of it as a way to make BODY and
4952 HANDLER closures, not just functions; MUMBLE_DATA points to the
4955 Of course, it's up to the caller to make sure that any data a
4956 MUMBLE_DATA needs is protected from GC. A common way to do this is
4957 to make MUMBLE_DATA a pointer to data stored in an automatic
4958 structure variable; since the collector must scan the stack for
4959 references anyway, this assures that any references in MUMBLE_DATA
4962 ** The new function scm_internal_lazy_catch is exactly like
4963 scm_internal_catch, except:
4965 - It does not unwind the stack (this is the major difference).
4966 - If handler returns, its value is returned from the throw.
4967 - BODY always receives #f as its JMPBUF argument (since there's no
4968 jmpbuf associated with a lazy catch, because we don't unwind the
4971 ** scm_body_thunk is a new body function you can pass to
4972 scm_internal_catch if you want the body to be like Scheme's `catch'
4973 --- a thunk, or a function of one argument if the tag is #f.
4975 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
4976 contains the Scheme procedure to invoke as the body, and the tag
4977 we're catching. If the tag is #f, then we pass JMPBUF (created by
4978 scm_internal_catch) to the body procedure; otherwise, the body gets
4981 ** scm_handle_by_proc is a new handler function you can pass to
4982 scm_internal_catch if you want the handler to act like Scheme's catch
4983 --- call a procedure with the tag and the throw arguments.
4985 If the user does a throw to this catch, this function runs a handler
4986 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
4987 variable holding the Scheme procedure object to invoke. It ought to
4988 be a pointer to an automatic variable (i.e., one living on the stack),
4989 or the procedure object should be otherwise protected from GC.
4991 ** scm_handle_by_message is a new handler function to use with
4992 `scm_internal_catch' if you want Guile to print a message and die.
4993 It's useful for dealing with throws to uncaught keys at the top level.
4995 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
4996 message header to print; if zero, we use "guile" instead. That
4997 text is followed by a colon, then the message described by ARGS.
4999 ** The return type of scm_boot_guile is now void; the function does
5000 not return a value, and indeed, never returns at all.
5002 ** The new function scm_shell makes it easy for user applications to
5003 process command-line arguments in a way that is compatible with the
5004 stand-alone guile interpreter (which is in turn compatible with SCSH,
5007 To use the scm_shell function, first initialize any guile modules
5008 linked into your application, and then call scm_shell with the values
5009 of ARGC and ARGV your `main' function received. scm_shell will add
5010 any SCSH-style meta-arguments from the top of the script file to the
5011 argument vector, and then process the command-line arguments. This
5012 generally means loading a script file or starting up an interactive
5013 command interpreter. For details, see "Changes to the stand-alone
5016 ** The new functions scm_get_meta_args and scm_count_argv help you
5017 implement the SCSH-style meta-argument, `\'.
5019 char **scm_get_meta_args (int ARGC, char **ARGV)
5020 If the second element of ARGV is a string consisting of a single
5021 backslash character (i.e. "\\" in Scheme notation), open the file
5022 named by the following argument, parse arguments from it, and return
5023 the spliced command line. The returned array is terminated by a
5026 For details of argument parsing, see above, under "guile now accepts
5027 command-line arguments compatible with SCSH..."
5029 int scm_count_argv (char **ARGV)
5030 Count the arguments in ARGV, assuming it is terminated by a null
5033 For an example of how these functions might be used, see the source
5034 code for the function scm_shell in libguile/script.c.
5036 You will usually want to use scm_shell instead of calling this
5039 ** The new function scm_compile_shell_switches turns an array of
5040 command-line arguments into Scheme code to carry out the actions they
5041 describe. Given ARGC and ARGV, it returns a Scheme expression to
5042 evaluate, and calls scm_set_program_arguments to make any remaining
5043 command-line arguments available to the Scheme code. For example,
5044 given the following arguments:
5046 -e main -s ekko a speckled gecko
5048 scm_set_program_arguments will return the following expression:
5050 (begin (load "ekko") (main (command-line)) (quit))
5052 You will usually want to use scm_shell instead of calling this
5055 ** The function scm_shell_usage prints a usage message appropriate for
5056 an interpreter that uses scm_compile_shell_switches to handle its
5057 command-line arguments.
5059 void scm_shell_usage (int FATAL, char *MESSAGE)
5060 Print a usage message to the standard error output. If MESSAGE is
5061 non-zero, write it before the usage message, followed by a newline.
5062 If FATAL is non-zero, exit the process, using FATAL as the
5063 termination status. (If you want to be compatible with Guile,
5064 always use 1 as the exit status when terminating due to command-line
5067 You will usually want to use scm_shell instead of calling this
5070 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
5071 expressions. It used to return SCM_EOL. Earth-shattering.
5073 ** The macros for declaring scheme objects in C code have been
5074 rearranged slightly. They are now:
5076 SCM_SYMBOL (C_NAME, SCHEME_NAME)
5077 Declare a static SCM variable named C_NAME, and initialize it to
5078 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
5079 be a C identifier, and SCHEME_NAME should be a C string.
5081 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
5082 Just like SCM_SYMBOL, but make C_NAME globally visible.
5084 SCM_VCELL (C_NAME, SCHEME_NAME)
5085 Create a global variable at the Scheme level named SCHEME_NAME.
5086 Declare a static SCM variable named C_NAME, and initialize it to
5087 point to the Scheme variable's value cell.
5089 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
5090 Just like SCM_VCELL, but make C_NAME globally visible.
5092 The `guile-snarf' script writes initialization code for these macros
5093 to its standard output, given C source code as input.
5095 The SCM_GLOBAL macro is gone.
5097 ** The scm_read_line and scm_read_line_x functions have been replaced
5098 by Scheme code based on the %read-delimited! procedure (known to C
5099 code as scm_read_delimited_x). See its description above for more
5102 ** The function scm_sys_open has been renamed to scm_open. It now
5103 returns a port instead of an FD object.
5105 * The dynamic linking support has changed. For more information, see
5106 libguile/DYNAMIC-LINKING.
5111 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
5114 * Changes to the 'guile' program:
5116 ** Guile now loads some new files when it starts up. Guile first
5117 searches the load path for init.scm, and loads it if found. Then, if
5118 Guile is not being used to execute a script, and the user's home
5119 directory contains a file named `.guile', Guile loads that.
5121 ** You can now use Guile as a shell script interpreter.
5123 To paraphrase the SCSH manual:
5125 When Unix tries to execute an executable file whose first two
5126 characters are the `#!', it treats the file not as machine code to
5127 be directly executed by the native processor, but as source code
5128 to be executed by some interpreter. The interpreter to use is
5129 specified immediately after the #! sequence on the first line of
5130 the source file. The kernel reads in the name of the interpreter,
5131 and executes that instead. It passes the interpreter the source
5132 filename as its first argument, with the original arguments
5133 following. Consult the Unix man page for the `exec' system call
5134 for more information.
5136 Now you can use Guile as an interpreter, using a mechanism which is a
5137 compatible subset of that provided by SCSH.
5139 Guile now recognizes a '-s' command line switch, whose argument is the
5140 name of a file of Scheme code to load. It also treats the two
5141 characters `#!' as the start of a comment, terminated by `!#'. Thus,
5142 to make a file of Scheme code directly executable by Unix, insert the
5143 following two lines at the top of the file:
5145 #!/usr/local/bin/guile -s
5148 Guile treats the argument of the `-s' command-line switch as the name
5149 of a file of Scheme code to load, and treats the sequence `#!' as the
5150 start of a block comment, terminated by `!#'.
5152 For example, here's a version of 'echo' written in Scheme:
5154 #!/usr/local/bin/guile -s
5156 (let loop ((args (cdr (program-arguments))))
5159 (display (car args))
5160 (if (pair? (cdr args))
5162 (loop (cdr args)))))
5165 Why does `#!' start a block comment terminated by `!#', instead of the
5166 end of the line? That is the notation SCSH uses, and although we
5167 don't yet support the other SCSH features that motivate that choice,
5168 we would like to be backward-compatible with any existing Guile
5169 scripts once we do. Furthermore, if the path to Guile on your system
5170 is too long for your kernel, you can start the script with this
5174 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
5177 Note that some very old Unix systems don't support the `#!' syntax.
5180 ** You can now run Guile without installing it.
5182 Previous versions of the interactive Guile interpreter (`guile')
5183 couldn't start up unless Guile's Scheme library had been installed;
5184 they used the value of the environment variable `SCHEME_LOAD_PATH'
5185 later on in the startup process, but not to find the startup code
5186 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
5189 To run Guile without installing it, build it in the normal way, and
5190 then set the environment variable `SCHEME_LOAD_PATH' to a
5191 colon-separated list of directories, including the top-level directory
5192 of the Guile sources. For example, if you unpacked Guile so that the
5193 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
5196 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
5199 ** Guile's read-eval-print loop no longer prints #<unspecified>
5200 results. If the user wants to see this, she can evaluate the
5201 expression (assert-repl-print-unspecified #t), perhaps in her startup
5204 ** Guile no longer shows backtraces by default when an error occurs;
5205 however, it does display a message saying how to get one, and how to
5206 request that they be displayed by default. After an error, evaluate
5208 to see a backtrace, and
5209 (debug-enable 'backtrace)
5210 to see them by default.
5214 * Changes to Guile Scheme:
5216 ** Guile now distinguishes between #f and the empty list.
5218 This is for compatibility with the IEEE standard, the (possibly)
5219 upcoming Revised^5 Report on Scheme, and many extant Scheme
5222 Guile used to have #f and '() denote the same object, to make Scheme's
5223 type system more compatible with Emacs Lisp's. However, the change
5224 caused too much trouble for Scheme programmers, and we found another
5225 way to reconcile Emacs Lisp with Scheme that didn't require this.
5228 ** Guile's delq, delv, delete functions, and their destructive
5229 counterparts, delq!, delv!, and delete!, now remove all matching
5230 elements from the list, not just the first. This matches the behavior
5231 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
5232 functions which inspired them.
5234 I recognize that this change may break code in subtle ways, but it
5235 seems best to make the change before the FSF's first Guile release,
5239 ** The compiled-library-path function has been deleted from libguile.
5241 ** The facilities for loading Scheme source files have changed.
5243 *** The variable %load-path now tells Guile which directories to search
5244 for Scheme code. Its value is a list of strings, each of which names
5247 *** The variable %load-extensions now tells Guile which extensions to
5248 try appending to a filename when searching the load path. Its value
5249 is a list of strings. Its default value is ("" ".scm").
5251 *** (%search-load-path FILENAME) searches the directories listed in the
5252 value of the %load-path variable for a Scheme file named FILENAME,
5253 with all the extensions listed in %load-extensions. If it finds a
5254 match, then it returns its full filename. If FILENAME is absolute, it
5255 returns it unchanged. Otherwise, it returns #f.
5257 %search-load-path will not return matches that refer to directories.
5259 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
5260 uses %seach-load-path to find a file named FILENAME, and loads it if
5261 it finds it. If it can't read FILENAME for any reason, it throws an
5264 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
5267 *** load uses the same searching semantics as primitive-load.
5269 *** The functions %try-load, try-load-with-path, %load, load-with-path,
5270 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
5271 path, and load-module-with-path have been deleted. The functions
5272 above should serve their purposes.
5274 *** If the value of the variable %load-hook is a procedure,
5275 `primitive-load' applies its value to the name of the file being
5276 loaded (without the load path directory name prepended). If its value
5277 is #f, it is ignored. Otherwise, an error occurs.
5279 This is mostly useful for printing load notification messages.
5282 ** The function `eval!' is no longer accessible from the scheme level.
5283 We can't allow operations which introduce glocs into the scheme level,
5284 because Guile's type system can't handle these as data. Use `eval' or
5285 `read-and-eval!' (see below) as replacement.
5287 ** The new function read-and-eval! reads an expression from PORT,
5288 evaluates it, and returns the result. This is more efficient than
5289 simply calling `read' and `eval', since it is not necessary to make a
5290 copy of the expression for the evaluator to munge.
5292 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
5293 for the `read' function.
5296 ** The function `int?' has been removed; its definition was identical
5297 to that of `integer?'.
5299 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
5300 use the R4RS names for these functions.
5302 ** The function object-properties no longer returns the hash handle;
5303 it simply returns the object's property list.
5305 ** Many functions have been changed to throw errors, instead of
5306 returning #f on failure. The point of providing exception handling in
5307 the language is to simplify the logic of user code, but this is less
5308 useful if Guile's primitives don't throw exceptions.
5310 ** The function `fileno' has been renamed from `%fileno'.
5312 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
5315 * Changes to Guile's C interface:
5317 ** The library's initialization procedure has been simplified.
5318 scm_boot_guile now has the prototype:
5320 void scm_boot_guile (int ARGC,
5322 void (*main_func) (),
5325 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
5326 MAIN_FUNC should do all the work of the program (initializing other
5327 packages, reading user input, etc.) before returning. When MAIN_FUNC
5328 returns, call exit (0); this function never returns. If you want some
5329 other exit value, MAIN_FUNC may call exit itself.
5331 scm_boot_guile arranges for program-arguments to return the strings
5332 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
5333 scm_set_program_arguments with the final list, so Scheme code will
5334 know which arguments have been processed.
5336 scm_boot_guile establishes a catch-all catch handler which prints an
5337 error message and exits the process. This means that Guile exits in a
5338 coherent way when system errors occur and the user isn't prepared to
5339 handle it. If the user doesn't like this behavior, they can establish
5340 their own universal catcher in MAIN_FUNC to shadow this one.
5342 Why must the caller do all the real work from MAIN_FUNC? The garbage
5343 collector assumes that all local variables of type SCM will be above
5344 scm_boot_guile's stack frame on the stack. If you try to manipulate
5345 SCM values after this function returns, it's the luck of the draw
5346 whether the GC will be able to find the objects you allocate. So,
5347 scm_boot_guile function exits, rather than returning, to discourage
5348 people from making that mistake.
5350 The IN, OUT, and ERR arguments were removed; there are other
5351 convenient ways to override these when desired.
5353 The RESULT argument was deleted; this function should never return.
5355 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
5359 ** Guile's header files should no longer conflict with your system's
5362 In order to compile code which #included <libguile.h>, previous
5363 versions of Guile required you to add a directory containing all the
5364 Guile header files to your #include path. This was a problem, since
5365 Guile's header files have names which conflict with many systems'
5368 Now only <libguile.h> need appear in your #include path; you must
5369 refer to all Guile's other header files as <libguile/mumble.h>.
5370 Guile's installation procedure puts libguile.h in $(includedir), and
5371 the rest in $(includedir)/libguile.
5374 ** Two new C functions, scm_protect_object and scm_unprotect_object,
5375 have been added to the Guile library.
5377 scm_protect_object (OBJ) protects OBJ from the garbage collector.
5378 OBJ will not be freed, even if all other references are dropped,
5379 until someone does scm_unprotect_object (OBJ). Both functions
5382 Note that calls to scm_protect_object do not nest. You can call
5383 scm_protect_object any number of times on a given object, and the
5384 next call to scm_unprotect_object will unprotect it completely.
5386 Basically, scm_protect_object and scm_unprotect_object just
5387 maintain a list of references to things. Since the GC knows about
5388 this list, all objects it mentions stay alive. scm_protect_object
5389 adds its argument to the list; scm_unprotect_object remove its
5390 argument from the list.
5393 ** scm_eval_0str now returns the value of the last expression
5396 ** The new function scm_read_0str reads an s-expression from a
5397 null-terminated string, and returns it.
5399 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
5400 to a Scheme port object.
5402 ** The new function `scm_set_program_arguments' allows C code to set
5403 the value returned by the Scheme `program-arguments' function.
5408 * Guile no longer includes sophisticated Tcl/Tk support.
5410 The old Tcl/Tk support was unsatisfying to us, because it required the
5411 user to link against the Tcl library, as well as Tk and Guile. The
5412 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
5413 referring to widgets by names, rather than exporting widgets to Scheme
5414 code as a special datatype.
5416 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
5417 maintainers described some very interesting changes in progress to the
5418 Tcl/Tk internals, which would facilitate clean interfaces between lone
5419 Tk and other interpreters --- even for garbage-collected languages
5420 like Scheme. They expected the new Tk to be publicly available in the
5423 Since it seems that Guile might soon have a new, cleaner interface to
5424 lone Tk, and that the old Guile/Tk glue code would probably need to be
5425 completely rewritten, we (Jim Blandy and Richard Stallman) have
5426 decided not to support the old code. We'll spend the time instead on
5427 a good interface to the newer Tk, as soon as it is available.
5429 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
5432 Copyright information:
5434 Copyright (C) 1996,1997 Free Software Foundation, Inc.
5436 Permission is granted to anyone to make or distribute verbatim copies
5437 of this document as received, in any medium, provided that the
5438 copyright notice and this permission notice are preserved,
5439 thus giving the recipient permission to redistribute in turn.
5441 Permission is granted to distribute modified versions
5442 of this document, or of portions of it,
5443 under the above conditions, provided also that they
5444 carry prominent notices stating who last changed them.
5449 paragraph-separate: "[
\f]*$"