1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002 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 the stable branch:
9 * Changes to the distribution
11 ** Guile now provide and uses an "effective" version number.
13 Guile now provides scm_effective_version and effective-version
14 functions which return the "effective" version number. This is just
15 the normal full version string without the final micro-version number,
16 so the current effective-version is "1.6". The effective version
17 should remain unchanged during a stable series, and should be used for
18 items like the versioned share directory name
19 i.e. /usr/share/guile/1.6.
21 Providing an unchanging version number during a stable release for
22 things like the versioned share directory can be particularly
23 important for Guile "add-on" packages, since it provides a directory
24 that they can install to that won't be changed out from under them
25 with each micro release during a stable series.
27 ** There are two new thread implementation options: "null" and "coop-pthreads".
29 When you configure "--with-threads=null", you will get the usual
30 threading API (call-with-new-thread, make-mutex, etc), but you can't
31 actually create new threads. Also, "--with-threads=no" is now
32 equivalent to "--with-threads=null". This means that the thread API
33 is always present, although you might not be able to create new
36 The "coop-pthread" (or shorter: "copt") thread implementation will use
37 portable POSIX threads but will restrict them so that only one thread
38 can execute 'in Guile' at any one time. This option will give you the
39 same basic behavior as the old "coop" option, but hopefully in a more
42 The default is now "coop-pthread", unless your platform doesn't have
43 pthreads, in which case "null" threads are used.
45 ** Guile now includes its own version of libltdl.
47 We now use a modified version of libltdl that allows us to make
48 improvements to it without having to rely on libtool releases.
50 * Changes to the standalone interpreter
52 ** New command line option `--no-debug'.
54 Specifying `--no-debug' on the command line will keep the debugging
55 evaluator turned off, even for interactive sessions.
57 ** User-init file ~/.guile is now loaded with the debugging evaluator.
59 Previously, the normal evaluator would have been used. Using the
60 debugging evaluator gives better error messages.
62 * Changes to Scheme functions and syntax
64 ** New function: effective-version
66 Returns the "effective" version number. This is just the normal full
67 version string without the final micro-version number. See "Changes
68 to the distribution" above.
72 Futures is a way of providing an alternative evaluation policy, very
73 similar in principle to "promises". Like promises, futures allow the
74 main process to continue instantly, but while promises postpone
75 evaluation ("lazy" evaluation) until the value is requested, futures
76 immediately starts evaluation in a parallel thread.
78 Futures are good when you want to express that "I'll need the value of
79 this computation sometime soon" and want to allow processing to go on
80 in the background until that time arrives.
82 ** New syntax: future FORM
84 Begin evaluation of FORM in a parallel thread and return the future
85 immediately. (Akin to 'delay'.)
87 ** New procedure: future-ref FUTURE
89 Return the computed value of the future. Wait if the computation is
90 not finished. (Akin to 'force'.)
92 ** New syntax: parallel FORM ...
94 Compute the results of FORM ... in parallel (in a separate thread for
95 each form) and return them as multiple values.
97 ** New syntax: letpar ((VAR EXP) ...) BODYFORM ...
99 Like 'let' but evaluates the binding expressions EXP ... in parallel.
101 ** New functions: par-map, par-for-each PROC ARGLIST ...
103 Like 'map' and 'for-each' but evaluate the procedure PROC in a
104 separate thread for each (set of) argument(s). All applications are
105 guaranteed to be completed before the procedure returns.
107 ** New functions: n-par-map, n-par-for-each N PROC ARGLIST ...
109 Like 'par-map' and 'par-for-each' but evaluate the procedure PROC in N
110 threads. This is useful when PROC uses large amounts of resources
111 and/or the argument list(s) is/are long so that one thread per (set
112 of) argument(s) would consume too much system resources. On a
113 dual-CPU system, N = 4 would often be a good choice.
115 ** Fair mutexes and condition variables
117 Fair mutexes and condition variables have been added. The fairness
118 means that scheduling is arranged to give as equal time shares as
119 possible and that threads are awakened in a first-in-first-out
120 manner. This is not guaranteed with standard mutexes and condition
123 In addition, fair mutexes are recursive. Locking a fair mutex that
124 you have already locked will succeed. Every call to lock-mutex must
125 be matched with a call to unlock-mutex. Only the last call to
126 unlock-mutex will actually unlock the mutex.
128 A fair condition variable must be used together with a fair mutex,
129 just as a standard condition variable must be used together with a
132 ** New functions: make-fair-mutex, make-fair-condition-variable'
134 Make a new fair mutex and a new fair condition variable respectively.
136 ** New function 'try-mutex'.
138 This function will attempt to lock a mutex but will return immediately
139 instead if blocking and indicate failure.
141 ** Waiting on a condition variable can have a timeout.
143 The funtion 'wait-condition-variable' now takes a third, optional
144 argument that specifies the point in time where the waiting should be
147 ** New function 'broadcast-condition-variable'.
149 ** New functions 'all-threads' and 'current-thread'.
151 ** Signals and system asyncs work better with threads.
153 The function 'sigaction' now takes a fourth, optional, argument that
154 specifies the thread that the handler should run in. When the
155 argument is omitted, the handler will run in the thread that called
158 Likewise, 'system-async-mark' takes a second, optional, argument that
159 specifies the thread that the async should run in. When it is
160 omitted, the async will run in the thread that called
163 C code can use the new functions scm_sigaction_for_thread and
164 scm_system_async_mark_for_thread to pass the new thread argument.
166 ** The function 'system-async' is deprecated.
168 You can now pass any zero-argument procedure to 'system-async-mark'.
169 The function 'system-async' will just return its argument unchanged
172 ** New functions 'call-with-blocked-asyncs' and
173 'call-with-unblocked-asyncs'
175 The expression (call-with-blocked-asyncs PROC) will call PROC and will
176 block execution of system asyncs for the current thread by one level
177 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
178 procedure and will unblock the execution of system asyncs by one
179 level for the current thread.
181 Only system asyncs are affected by these functions.
183 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
185 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
186 instead. Those functions are easier to use correctly and can be
189 ** New function 'unsetenv'.
191 ** New macro 'define-syntax-public'.
193 It works like 'define-syntax' and also exports the defined macro (but
196 ** There is support for Infinity and NaNs.
198 Following PLT Scheme, Guile can now work with infinite numbers, and
201 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
202 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
203 "+nan.0"). These numbers are inexact and have no exact counterpart.
205 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
206 sign of the dividend. The infinities are integers, and they answer #t
207 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
208 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
219 ERROR: Numerical overflow
221 Two new predicates 'inf?' and 'nan?' can be used to test for the
224 ** Inexact zero can have a sign.
226 Guile can now distinguish between plus and minus inexact zero, if your
227 platform supports this, too. The two zeros are equal according to
228 '=', but not according to 'eqv?'. For example
239 ** We now have uninterned symbols.
241 The new function 'make-symbol' will return a uninterned symbol. This
242 is a symbol that is unique and is guaranteed to remain unique.
243 However, uninterned symbols can not yet be read back in.
245 Use the new function 'symbol-interned?' to check whether a symbol is
248 ** pretty-print has more options.
250 The function pretty-print from the (ice-9 pretty-print) module can now
251 also be invoked with keyword arguments that control things like
252 maximum output width. See its online documentation.
254 ** Variables have no longer a special behavior for `equal?'.
256 Previously, comparing two variables with `equal?' would recursivly
257 compare their values. This is no longer done. Variables are now only
258 `equal?' if they are `eq?'.
260 ** `(begin)' is now valid.
262 You can now use an empty `begin' form. It will yield #<unspecified>
263 when evaluated and simply be ignored in a definition context.
265 ** Removed: substring-move-left!, substring-move-right!
267 Use `substring-move!' instead.
269 ** Deprecated: procedure->macro
271 Change your code to use either procedure->memoizing-macro or, probably better,
272 to use r5rs macros. Also, be aware that macro expansion will not be done
273 during evaluation, but prior to evaluation.
275 ** Soft ports now allow a `char-ready?' procedure
277 The vector argument to `make-soft-port' can now have a length of
278 either 5 or 6. (Previously the length had to be 5.) The optional 6th
279 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
280 that returns the number of characters that can be read immediately
281 without the soft port blocking.
283 ** New debugging feature: breakpoints.
285 Guile now has breakpoints. For details see the `Debugging Features'
286 chapter in the reference manual.
288 ** Deprecated: undefine
290 There is no replacement for undefine.
292 * Changes to the C interface
294 ** New function: scm_effective_version
296 Returns the "effective" version number. This is just the normal full
297 version string without the final micro-version number. See "Changes
298 to the distribution" above.
300 ** The function scm_call_with_new_thread has a new prototype.
302 Instead of taking a list with the thunk and handler, these two
303 arguments are now passed directly:
305 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
307 This is an incompatible change.
309 ** The value 'scm_mask_ints' is no longer writable.
311 Previously, you could set scm_mask_ints directly. This is no longer
312 possible. Use scm_c_call_with_blocked_asyncs and
313 scm_c_call_with_unblocked_asyncs instead.
315 ** New functions scm_c_call_with_blocked_asyncs and
316 scm_c_call_with_unblocked_asyncs
318 Like scm_call_with_blocked_asyncs etc. but for C functions.
320 ** New snarfer macro SCM_DEFINE_PUBLIC.
322 This is like SCM_DEFINE, but also calls scm_c_export for the defined
323 function in the init section.
325 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
327 ** New macros SCM_VECTOR_REF and SCM_VECTOR_SET.
329 Use these in preference to SCM_VELTS.
331 ** The SCM_VELTS macros now returns a read-only vector. For writing,
332 use the new macros SCM_WRITABLE_VELTS or SCM_VECTOR_SET. The use of
333 SCM_WRITABLE_VELTS is discouraged, though.
335 ** Garbage collector rewrite.
337 The garbage collector is cleaned up a lot, and now uses lazy
338 sweeping. This is reflected in the output of (gc-stats); since cells
339 are being freed when they are allocated, the cells-allocated field
340 stays roughly constant.
342 For malloc related triggers, the behavior is changed. It uses the same
343 heuristic as the cell-triggered collections. It may be tuned with the
344 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
345 for minimum yield of malloc related triggers. The default is 40.
346 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
349 Debugging operations for the freelist have been deprecated, along with
350 the C variables that control garbage collection. The environment
351 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
352 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
354 ** The function scm_definedp has been renamed to scm_defined_p
356 The name scm_definedp is deprecated.
358 ** The struct scm_cell has been renamed to scm_t_cell
360 This is in accordance to Guile's naming scheme for types. Note that
361 the name scm_cell is now used for a function that allocates and
362 initializes a new cell (see below).
364 ** New functions for memory management
366 A new set of functions for memory management has been added since the
367 old way (scm_must_malloc, scm_must_free, etc) was error prone and
368 indeed, Guile itself contained some long standing bugs that could
369 cause aborts in long running programs.
371 The new functions are more symmetrical and do not need cooperation
372 from smob free routines, among other improvements.
374 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
375 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
376 scm_gc_free, scm_gc_register_collectable_memory, and
377 scm_gc_unregister_collectable_memory. Refer to the manual for more
378 details and for upgrading instructions.
380 The old functions for memory management have been deprecated. They
381 are: scm_must_malloc, scm_must_realloc, scm_must_free,
382 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
384 ** New function: scm_str2string
386 This function creates a scheme string from a 0-terminated C string. The input
389 ** Declarations of exported features are marked with SCM_API.
391 Every declaration of a feature that belongs to the exported Guile API
392 has been marked by adding the macro "SCM_API" to the start of the
393 declaration. This macro can expand into different things, the most
394 common of which is just "extern" for Unix platforms. On Win32, it can
395 be used to control which symbols are exported from a DLL.
397 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
398 will expand into "__declspec (dllimport) extern", which is needed for
399 linking to the Guile DLL in Windows.
401 There are also SCM_RL_IMPORT, QT_IMPORT, SCM_SRFI1314_IMPORT, and
402 SCM_SRFI4_IMPORT, for the corresponding libraries.
404 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
406 Use the new functions scm_cell and scm_double_cell instead. The old macros
407 had problems because with them allocation and initialization was separated and
408 the GC could sometimes observe half initialized cells. Only careful coding by
409 the user of SCM_NEWCELL and SCM_NEWCELL2 could make this safe and efficient.
411 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
413 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
416 ** SRCBRKP has been deprecated.
418 Use scm_c_source_property_breakpoint_p instead.
420 ** Deprecated: scm_makmacro
422 Change your code to use either scm_makmmacro or, probably better, to use r5rs
423 macros. Also, be aware that macro expansion will not be done during
424 evaluation, but prior to evaluation.
426 ** Removed from scm_root_state: def_inp, def_outp, def_errp, together
427 with corresponding macros scm_def_inp, scm_def_outp and scm_def_errp.
428 These were undocumented and unused copies of the standard ports at the
429 time that Guile was initialised. Normally the current ports should be
430 used instead, obtained from scm_current_input_port () etc. If an
431 application needs to retain earlier ports, it should save them in a
432 gc-protected location.
434 ** Removed compile time option MEMOIZE_LOCALS
436 Now, caching of local variable positions during memoization is mandatory.
437 However, the option to disable the caching has most probably not been used
440 ** Removed compile time option SCM_RECKLESS
442 Full number of arguments checking of closures is mandatory now. However, the
443 option to disable the checking has most probably not been used anyway.
445 ** Removed compile time option SCM_CAUTIOUS
447 Full number of arguments checking of closures is mandatory now. However, the
448 option to disable the checking has most probably not been used anyway.
450 ** Deprecated configure flags USE_THREADS and GUILE_ISELECT
452 Previously, when the C preprocessor macro USE_THREADS was defined,
453 libguile included a thread API. This API is now always included, even
454 when threads are not really supported. Thus, you don't need to test
457 Analogously, GUILE_ISELECT was defined when the function
458 scm_internal_select was provided by Guile. This function is now
459 always defined, and GUILE_ISELECT with it.
461 ** Removed definitions: scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify,
462 s_t_ify, scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
463 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2, scm_tc16_allocated,
464 SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY, SCM_IM_0_COND,
465 SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED, scm_debug_newcell,
466 scm_debug_newcell2, scm_substring_move_left_x, scm_substring_move_right_x,
467 long_long, ulong_long, scm_sizet, SCM_WNA, SCM_OUTOFRANGE, SCM_NALLOC,
468 SCM_HUP_SIGNAL, SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL,
469 SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL,
470 SCM_SIG_ORD, SCM_ORD_SIG, SCM_NUM_SIGS, moddata, registered_mods,
471 scm_register_module_xxx, scm_registered_modules,
472 scm_clear_registered_modules, scm_wta, *top-level-lookup-closure*,
473 scm_top_level_lookup_closure_var, scm_system_transformer, scm_eval_3,
474 scm_eval2, SCM_SETAND_CAR, SCM_SETOR_CAR, SCM_SETAND_CDR, SCM_SETOR_CDR,
475 SCM_FREEP, SCM_NFREEP, SCM_GC8MARKP, SCM_SETGC8MARK, SCM_CLRGC8MARK,
476 SCM_GCTYP16, SCM_GCCDR, scm_remember, scm_protect_object,
477 scm_unprotect_object, root_module_lookup_closure, scm_sym_app,
478 scm_sym_modules, module_prefix, make_modules_in_var,
479 beautify_user_module_x_var, try_module_autoload_var, scm_module_full_name,
480 scm_the_root_module, scm_make_module, scm_ensure_user_module,
481 scm_load_scheme_module, scm_port, scm_ptob_descriptor, scm_port_rw_active,
482 scm_close_all_ports_except, scm_rstate, scm_rng, scm_i_rstate,
483 SCM_SLOPPY_STRINGP, SCM_RWSTRINGP, SCM_STRING_UCHARS, SCM_STRING_CHARS,
484 scm_read_only_string_p, scm_makstr, scm_makfromstr,
485 scm_make_shared_substring, scm_tc7_substring, SCM_SLOPPY_CONSP,
486 SCM_SLOPPY_NCONSP, scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
487 sym_huh, scm_variable_set_name_hint, scm_builtin_variable, SCM_VARVCELL,
488 SCM_UDVARIABLEP, SCM_DEFVARIABLEP, scm_internal_with_fluids,
489 scm_make_gsubr, scm_make_gsubr_with_generic, scm_create_hook, list*,
490 SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4, SCM_LIST5,
491 SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9, scm_listify, scm_sloppy_memq,
492 scm_sloppy_memv, scm_sloppy_member, scm_end_of_file_key,
493 scm_read_and_eval_x, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
494 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
495 scm_subr_entry, SCM_SUBR_DOC, scm_make_subr_opt, scm_make_subr,
496 scm_make_subr_with_generic, setjmp_type, setjmp_type,
497 scm_call_catching_errors, scm_make_smob_type_mfpe, scm_set_smob_mfpe,
498 scm_strprint_obj, scm_read_0str, scm_eval_0str, SCM_CHARS, SCM_UCHARS,
499 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
500 SCM_LENGTH_MAX, SCM_LENGTH, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
501 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR, scm_strhash,
502 scm_sym2vcell, scm_sym2ovcell_soft, scm_sym2ovcell,
503 scm_intern_obarray_soft, scm_intern_obarray, scm_intern, scm_intern0,
504 scm_sysintern, scm_sysintern0, scm_sysintern0_no_module_lookup,
505 scm_symbol_value0, scm_string_to_obarray_symbol, scm_intern_symbol,
506 scm_unintern_symbol, scm_symbol_binding, scm_symbol_interned_p,
507 scm_symbol_bound_p, scm_symbol_set_x, scm_gentemp,
508 scm_init_symbols_deprecated, s_vector_set_length_x, scm_vector_set_length_x,
509 scm_contregs, scm_debug_info, scm_debug_frame, SCM_DSIDEVAL, SCM_OPDIRP,
510 scm_fport, scm_option, SCM_CONST_LONG, SCM_VCELL, SCM_GLOBAL_VCELL,
511 SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT, scm_srcprops, scm_srcprops_chunk,
512 scm_info_frame, scm_stack, scm_array, scm_array_dim, SCM_ARRAY_CONTIGUOUS,
513 SCM_HUGE_LENGTH, SCM_FUNC_NAME, SCM_WTA, RETURN_SCM_WTA,
514 SCM_VALIDATE_NUMBER_COPY, SCM_VALIDATE_NUMBER_DEF_COPY,
515 SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
516 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
517 SCM_VALIDATE_RWSTRING, SCM_VALIDATE_OPDIR, DIGITS, scm_small_istr2int,
518 scm_istr2int, scm_istr2flo, scm_istring2number, scm_istr2int,
519 scm_istr2flo, scm_istring2number, scm_vtable_index_vcell, scm_si_vcell,
520 SCM_ECONSP, SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
521 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable
523 Changes since Guile 1.4:
525 * Changes to the distribution
527 ** A top-level TODO file is included.
529 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
531 Guile now always uses three numbers to represent the version,
532 i.e. "1.6.5". The first number, 1, is the major version number, the
533 second number, 6, is the minor version number, and the third number,
534 5, is the micro version number. Changes in major version number
535 indicate major changes in Guile.
537 Minor version numbers that are even denote stable releases, and odd
538 minor version numbers denote development versions (which may be
539 unstable). The micro version number indicates a minor sub-revision of
540 a given MAJOR.MINOR release.
542 In keeping with the new scheme, (minor-version) and scm_minor_version
543 no longer return everything but the major version number. They now
544 just return the minor version number. Two new functions
545 (micro-version) and scm_micro_version have been added to report the
546 micro version number.
548 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
550 ** New preprocessor definitions are available for checking versions.
552 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
553 SCM_MICRO_VERSION to the appropriate integer values.
555 ** Guile now actively warns about deprecated features.
557 The new configure option `--enable-deprecated=LEVEL' and the
558 environment variable GUILE_WARN_DEPRECATED control this mechanism.
559 See INSTALL and README for more information.
561 ** Guile is much more likely to work on 64-bit architectures.
563 Guile now compiles and passes "make check" with only two UNRESOLVED GC
564 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
565 for the use of a test machine, and thanks to Stefan Jahn for ia64
568 ** New functions: setitimer and getitimer.
570 These implement a fairly direct interface to the libc functions of the
573 ** The #. reader extension is now disabled by default.
575 For safety reasons, #. evaluation is disabled by default. To
576 re-enable it, set the fluid read-eval? to #t. For example:
578 (fluid-set! read-eval? #t)
580 but make sure you realize the potential security risks involved. With
581 read-eval? enabled, reading a data file from an untrusted source can
584 ** New SRFI modules have been added:
586 SRFI-0 `cond-expand' is now supported in Guile, without requiring
589 (srfi srfi-1) is a library containing many useful pair- and list-processing
592 (srfi srfi-2) exports and-let*.
594 (srfi srfi-4) implements homogeneous numeric vector datatypes.
596 (srfi srfi-6) is a dummy module for now, since guile already provides
597 all of the srfi-6 procedures by default: open-input-string,
598 open-output-string, get-output-string.
600 (srfi srfi-8) exports receive.
602 (srfi srfi-9) exports define-record-type.
604 (srfi srfi-10) exports define-reader-ctor and implements the reader
607 (srfi srfi-11) exports let-values and let*-values.
609 (srfi srfi-13) implements the SRFI String Library.
611 (srfi srfi-14) implements the SRFI Character-Set Library.
613 (srfi srfi-17) implements setter and getter-with-setter and redefines
614 some accessor procedures as procedures with getters. (such as car,
615 cdr, vector-ref etc.)
617 (srfi srfi-19) implements the SRFI Time/Date Library.
619 ** New scripts / "executable modules"
621 Subdirectory "scripts" contains Scheme modules that are packaged to
622 also be executable as scripts. At this time, these scripts are available:
631 See README there for more info.
633 These scripts can be invoked from the shell with the new program
634 "guile-tools", which keeps track of installation directory for you.
637 $ guile-tools display-commentary srfi/*.scm
639 guile-tools is copied to the standard $bindir on "make install".
641 ** New module (ice-9 stack-catch):
643 stack-catch is like catch, but saves the current state of the stack in
644 the fluid the-last-stack. This fluid can be useful when using the
645 debugger and when re-throwing an error.
647 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
649 This has been done to prevent problems on lesser operating systems
650 that can't tolerate `*'s in file names. The exported macro continues
651 to be named `and-let*', of course.
653 On systems that support it, there is also a compatibility module named
654 (ice-9 and-let*). It will go away in the next release.
656 ** New modules (oop goops) etc.:
661 (oop goops active-slot)
662 (oop goops composite-slot)
664 The Guile Object Oriented Programming System (GOOPS) has been
665 integrated into Guile. For further information, consult the GOOPS
666 manual and tutorial in the `doc' directory.
668 ** New module (ice-9 rdelim).
670 This exports the following procedures which were previously defined
671 in the default environment:
673 read-line read-line! read-delimited read-delimited! %read-delimited!
674 %read-line write-line
676 For backwards compatibility the definitions are still imported into the
677 default environment in this version of Guile. However you should add:
679 (use-modules (ice-9 rdelim))
681 to any program which uses the definitions, since this may change in
684 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
685 can be used for similar functionality.
687 ** New module (ice-9 rw)
689 This is a subset of the (scsh rw) module from guile-scsh. Currently
690 it defines two procedures:
692 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
694 Read characters from a port or file descriptor into a string STR.
695 A port must have an underlying file descriptor -- a so-called
696 fport. This procedure is scsh-compatible and can efficiently read
699 *** New function: write-string/partial str [port_or_fdes [start [end]]]
701 Write characters from a string STR to a port or file descriptor.
702 A port must have an underlying file descriptor -- a so-called
703 fport. This procedure is mostly compatible and can efficiently
706 ** New module (ice-9 match)
708 This module includes Andrew K. Wright's pattern matcher. See
709 ice-9/match.scm for brief description or
711 http://www.star-lab.com/wright/code.html
713 for complete documentation.
715 ** New module (ice-9 buffered-input)
717 This module provides procedures to construct an input port from an
718 underlying source of input that reads and returns its input in chunks.
719 The underlying input source is a Scheme procedure, specified by the
720 caller, which the port invokes whenever it needs more input.
722 This is useful when building an input port whose back end is Readline
723 or a UI element such as the GtkEntry widget.
727 The reference and tutorial documentation that was previously
728 distributed separately, as `guile-doc', is now included in the core
729 Guile distribution. The documentation consists of the following
732 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
735 - The Guile Reference Manual (guile.texi) contains (or is intended to
736 contain) reference documentation on all aspects of Guile.
738 - The GOOPS Manual (goops.texi) contains both tutorial-style and
739 reference documentation for using GOOPS, Guile's Object Oriented
742 - The Revised^5 Report on the Algorithmic Language Scheme
745 See the README file in the `doc' directory for more details.
747 ** There are a couple of examples in the examples/ directory now.
749 * Changes to the stand-alone interpreter
751 ** New command line option `--use-srfi'
753 Using this option, SRFI modules can be loaded on startup and be
754 available right from the beginning. This makes programming portable
755 Scheme programs easier.
757 The option `--use-srfi' expects a comma-separated list of numbers,
758 each representing a SRFI number to be loaded into the interpreter
759 before starting evaluating a script file or the REPL. Additionally,
760 the feature identifier for the loaded SRFIs is recognized by
761 `cond-expand' when using this option.
764 $ guile --use-srfi=8,13
765 guile> (receive (x z) (values 1 2) (+ 1 2))
767 guile> (string-pad "bla" 20)
770 ** Guile now always starts up in the `(guile-user)' module.
772 Previously, scripts executed via the `-s' option would run in the
773 `(guile)' module and the repl would run in the `(guile-user)' module.
774 Now every user action takes place in the `(guile-user)' module by
777 * Changes to Scheme functions and syntax
779 ** Character classifiers work for non-ASCII characters.
781 The predicates `char-alphabetic?', `char-numeric?',
782 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
783 no longer check whether their arguments are ASCII characters.
784 Previously, a character would only be considered alphabetic when it
785 was also ASCII, for example.
787 ** Previously deprecated Scheme functions have been removed:
789 tag - no replacement.
790 fseek - replaced by seek.
791 list* - replaced by cons*.
793 ** It's now possible to create modules with controlled environments
797 (use-modules (ice-9 safe))
798 (define m (make-safe-module))
799 ;;; m will now be a module containing only a safe subset of R5RS
800 (eval '(+ 1 2) m) --> 3
801 (eval 'load m) --> ERROR: Unbound variable: load
803 ** Evaluation of "()", the empty list, is now an error.
805 Previously, the expression "()" evaluated to the empty list. This has
806 been changed to signal a "missing expression" error. The correct way
807 to write the empty list as a literal constant is to use quote: "'()".
809 ** New concept of `Guile Extensions'.
811 A Guile Extension is just a ordinary shared library that can be linked
812 at run-time. We found it advantageous to give this simple concept a
813 dedicated name to distinguish the issues related to shared libraries
814 from the issues related to the module system.
816 *** New function: load-extension
818 Executing (load-extension lib init) is mostly equivalent to
820 (dynamic-call init (dynamic-link lib))
822 except when scm_register_extension has been called previously.
823 Whenever appropriate, you should use `load-extension' instead of
824 dynamic-link and dynamic-call.
826 *** New C function: scm_c_register_extension
828 This function registers a initialization function for use by
829 `load-extension'. Use it when you don't want specific extensions to
830 be loaded as shared libraries (for example on platforms that don't
831 support dynamic linking).
833 ** Auto-loading of compiled-code modules is deprecated.
835 Guile used to be able to automatically find and link a shared
836 library to satisfy requests for a module. For example, the module
837 `(foo bar)' could be implemented by placing a shared library named
838 "foo/libbar.so" (or with a different extension) in a directory on the
841 This has been found to be too tricky, and is no longer supported. The
842 shared libraries are now called "extensions". You should now write a
843 small Scheme file that calls `load-extension' to load the shared
844 library and initialize it explicitely.
846 The shared libraries themselves should be installed in the usual
847 places for shared libraries, with names like "libguile-foo-bar".
849 For example, place this into a file "foo/bar.scm"
851 (define-module (foo bar))
853 (load-extension "libguile-foo-bar" "foobar_init")
855 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
857 `eval' is now R5RS, that is it takes two arguments.
858 The second argument is an environment specifier, i.e. either
860 (scheme-report-environment 5)
862 (interaction-environment)
868 ** The module system has been made more disciplined.
870 The function `eval' will save and restore the current module around
871 the evaluation of the specified expression. While this expression is
872 evaluated, `(current-module)' will now return the right module, which
873 is the module specified as the second argument to `eval'.
875 A consequence of this change is that `eval' is not particularly
876 useful when you want allow the evaluated code to change what module is
877 designated as the current module and have this change persist from one
878 call to `eval' to the next. The read-eval-print-loop is an example
879 where `eval' is now inadequate. To compensate, there is a new
880 function `primitive-eval' that does not take a module specifier and
881 that does not save/restore the current module. You should use this
882 function together with `set-current-module', `current-module', etc
883 when you want to have more control over the state that is carried from
884 one eval to the next.
886 Additionally, it has been made sure that forms that are evaluated at
887 the top level are always evaluated with respect to the current module.
888 Previously, subforms of top-level forms such as `begin', `case',
889 etc. did not respect changes to the current module although these
890 subforms are at the top-level as well.
892 To prevent strange behavior, the forms `define-module',
893 `use-modules', `use-syntax', and `export' have been restricted to only
894 work on the top level. The forms `define-public' and
895 `defmacro-public' only export the new binding on the top level. They
896 behave just like `define' and `defmacro', respectively, when they are
897 used in a lexical environment.
899 Also, `export' will no longer silently re-export bindings imported
900 from a used module. It will emit a `deprecation' warning and will
901 cease to perform any re-export in the next version. If you actually
902 want to re-export bindings, use the new `re-export' in place of
903 `export'. The new `re-export' will not make copies of variables when
904 rexporting them, as `export' did wrongly.
906 ** Module system now allows selection and renaming of imported bindings
908 Previously, when using `use-modules' or the `#:use-module' clause in
909 the `define-module' form, all the bindings (association of symbols to
910 values) for imported modules were added to the "current module" on an
911 as-is basis. This has been changed to allow finer control through two
912 new facilities: selection and renaming.
914 You can now select which of the imported module's bindings are to be
915 visible in the current module by using the `:select' clause. This
916 clause also can be used to rename individual bindings. For example:
918 ;; import all bindings no questions asked
919 (use-modules (ice-9 common-list))
921 ;; import four bindings, renaming two of them;
922 ;; the current module sees: every some zonk-y zonk-n
923 (use-modules ((ice-9 common-list)
926 (remove-if-not . zonk-n))))
928 You can also programmatically rename all selected bindings using the
929 `:renamer' clause, which specifies a proc that takes a symbol and
930 returns another symbol. Because it is common practice to use a prefix,
931 we now provide the convenience procedure `symbol-prefix-proc'. For
934 ;; import four bindings, renaming two of them specifically,
935 ;; and all four w/ prefix "CL:";
936 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
937 (use-modules ((ice-9 common-list)
940 (remove-if-not . zonk-n))
941 :renamer (symbol-prefix-proc 'CL:)))
943 ;; import four bindings, renaming two of them specifically,
944 ;; and all four by upcasing.
945 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
946 (define (upcase-symbol sym)
947 (string->symbol (string-upcase (symbol->string sym))))
949 (use-modules ((ice-9 common-list)
952 (remove-if-not . zonk-n))
953 :renamer upcase-symbol))
955 Note that programmatic renaming is done *after* individual renaming.
956 Also, the above examples show `use-modules', but the same facilities are
957 available for the `#:use-module' clause of `define-module'.
959 See manual for more info.
961 ** The semantics of guardians have changed.
963 The changes are for the most part compatible. An important criterion
964 was to keep the typical usage of guardians as simple as before, but to
965 make the semantics safer and (as a result) more useful.
967 *** All objects returned from guardians are now properly alive.
969 It is now guaranteed that any object referenced by an object returned
970 from a guardian is alive. It's now impossible for a guardian to
971 return a "contained" object before its "containing" object.
973 One incompatible (but probably not very important) change resulting
974 from this is that it is no longer possible to guard objects that
975 indirectly reference themselves (i.e. are parts of cycles). If you do
976 so accidentally, you'll get a warning.
978 *** There are now two types of guardians: greedy and sharing.
980 If you call (make-guardian #t) or just (make-guardian), you'll get a
981 greedy guardian, and for (make-guardian #f) a sharing guardian.
983 Greedy guardians are the default because they are more "defensive".
984 You can only greedily guard an object once. If you guard an object
985 more than once, once in a greedy guardian and the rest of times in
986 sharing guardians, then it is guaranteed that the object won't be
987 returned from sharing guardians as long as it is greedily guarded
990 Guardians returned by calls to `make-guardian' can now take one more
991 optional parameter, which says whether to throw an error in case an
992 attempt is made to greedily guard an object that is already greedily
993 guarded. The default is true, i.e. throw an error. If the parameter
994 is false, the guardian invocation returns #t if guarding was
995 successful and #f if it wasn't.
997 Also, since greedy guarding is, in effect, a side-effecting operation
998 on objects, a new function is introduced: `destroy-guardian!'.
999 Invoking this function on a guardian renders it unoperative and, if
1000 the guardian is greedy, clears the "greedily guarded" property of the
1001 objects that were guarded by it, thus undoing the side effect.
1003 Note that all this hair is hardly very important, since guardian
1004 objects are usually permanent.
1006 ** Continuations created by call-with-current-continuation now accept
1007 any number of arguments, as required by R5RS.
1009 ** New function `issue-deprecation-warning'
1011 This function is used to display the deprecation messages that are
1012 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1015 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1019 ;; `id' is deprecated. Use `identity' instead.
1024 ** New syntax `begin-deprecated'
1026 When deprecated features are included (as determined by the configure
1027 option --enable-deprecated), `begin-deprecated' is identical to
1028 `begin'. When deprecated features are excluded, it always evaluates
1029 to `#f', ignoring the body forms.
1031 ** New function `make-object-property'
1033 This function returns a new `procedure with setter' P that can be used
1034 to attach a property to objects. When calling P as
1038 where `obj' is any kind of object, it attaches `val' to `obj' in such
1039 a way that it can be retrieved by calling P as
1043 This function will replace procedure properties, symbol properties and
1044 source properties eventually.
1046 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1048 Instead of #&optional, #&key, etc you should now use #:optional,
1049 #:key, etc. Since #:optional is a keyword, you can write it as just
1050 :optional when (read-set! keywords 'prefix) is active.
1052 The old reader syntax `#&' is still supported, but deprecated. It
1053 will be removed in the next release.
1055 ** New define-module option: pure
1057 Tells the module system not to include any bindings from the root
1062 (define-module (totally-empty-module)
1065 ** New define-module option: export NAME1 ...
1067 Export names NAME1 ...
1069 This option is required if you want to be able to export bindings from
1070 a module which doesn't import one of `define-public' or `export'.
1074 (define-module (foo)
1076 :use-module (ice-9 r5rs)
1079 ;;; Note that we're pure R5RS below this point!
1084 ** New function: object->string OBJ
1086 Return a Scheme string obtained by printing a given object.
1088 ** New function: port? X
1090 Returns a boolean indicating whether X is a port. Equivalent to
1091 `(or (input-port? X) (output-port? X))'.
1093 ** New function: file-port?
1095 Determines whether a given object is a port that is related to a file.
1097 ** New function: port-for-each proc
1099 Apply PROC to each port in the Guile port table in turn. The return
1100 value is unspecified. More specifically, PROC is applied exactly once
1101 to every port that exists in the system at the time PORT-FOR-EACH is
1102 invoked. Changes to the port table while PORT-FOR-EACH is running
1103 have no effect as far as PORT-FOR-EACH is concerned.
1105 ** New function: dup2 oldfd newfd
1107 A simple wrapper for the `dup2' system call. Copies the file
1108 descriptor OLDFD to descriptor number NEWFD, replacing the
1109 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1110 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1111 to move away ports which are using NEWFD. The return value is
1114 ** New function: close-fdes fd
1116 A simple wrapper for the `close' system call. Close file
1117 descriptor FD, which must be an integer. Unlike close (*note
1118 close: Ports and File Descriptors.), the file descriptor will be
1119 closed even if a port is using it. The return value is
1122 ** New function: crypt password salt
1124 Encrypts `password' using the standard unix password encryption
1127 ** New function: chroot path
1129 Change the root directory of the running process to `path'.
1131 ** New functions: getlogin, cuserid
1133 Return the login name or the user name of the current effective user
1136 ** New functions: getpriority which who, setpriority which who prio
1138 Get or set the priority of the running process.
1140 ** New function: getpass prompt
1142 Read a password from the terminal, first displaying `prompt' and
1145 ** New function: flock file operation
1147 Set/remove an advisory shared or exclusive lock on `file'.
1149 ** New functions: sethostname name, gethostname
1151 Set or get the hostname of the machine the current process is running
1154 ** New function: mkstemp! tmpl
1156 mkstemp creates a new unique file in the file system and returns a
1157 new buffered port open for reading and writing to the file. TMPL
1158 is a string specifying where the file should be created: it must
1159 end with `XXXXXX' and will be changed in place to return the name
1160 of the temporary file.
1162 ** New function: open-input-string string
1164 Return an input string port which delivers the characters from
1165 `string'. This procedure, together with `open-output-string' and
1166 `get-output-string' implements SRFI-6.
1168 ** New function: open-output-string
1170 Return an output string port which collects all data written to it.
1171 The data can then be retrieved by `get-output-string'.
1173 ** New function: get-output-string
1175 Return the contents of an output string port.
1177 ** New function: identity
1179 Return the argument.
1181 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1182 are represented in Scheme as integers with normal host byte ordering.
1184 ** New function: inet-pton family address
1186 Convert a printable string network address into an integer. Note that
1187 unlike the C version of this function, the result is an integer with
1188 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1191 (inet-pton AF_INET "127.0.0.1") => 2130706433
1192 (inet-pton AF_INET6 "::1") => 1
1194 ** New function: inet-ntop family address
1196 Convert an integer network address into a printable string. Note that
1197 unlike the C version of this function, the input is an integer with
1198 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1201 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1202 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1203 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1207 Use `identity' instead.
1213 ** Deprecated: return-it
1217 ** Deprecated: string-character-length
1219 Use `string-length' instead.
1221 ** Deprecated: flags
1223 Use `logior' instead.
1225 ** Deprecated: close-all-ports-except.
1227 This was intended for closing ports in a child process after a fork,
1228 but it has the undesirable side effect of flushing buffers.
1229 port-for-each is more flexible.
1231 ** The (ice-9 popen) module now attempts to set up file descriptors in
1232 the child process from the current Scheme ports, instead of using the
1233 current values of file descriptors 0, 1, and 2 in the parent process.
1235 ** Removed function: builtin-weak-bindings
1237 There is no such concept as a weak binding any more.
1239 ** Removed constants: bignum-radix, scm-line-incrementors
1241 ** define-method: New syntax mandatory.
1243 The new method syntax is now mandatory:
1245 (define-method (NAME ARG-SPEC ...) BODY ...)
1246 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1248 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1249 REST-ARG ::= ARG-NAME
1251 If you have old code using the old syntax, import
1252 (oop goops old-define-method) before (oop goops) as in:
1254 (use-modules (oop goops old-define-method) (oop goops))
1256 ** Deprecated function: builtin-variable
1257 Removed function: builtin-bindings
1259 There is no longer a distinction between builtin or other variables.
1260 Use module system operations for all variables.
1262 ** Lazy-catch handlers are no longer allowed to return.
1264 That is, a call to `throw', `error', etc is now guaranteed to not
1267 ** Bugfixes for (ice-9 getopt-long)
1269 This module is now tested using test-suite/tests/getopt-long.test.
1270 The following bugs have been fixed:
1272 *** Parsing for options that are specified to have `optional' args now checks
1273 if the next element is an option instead of unconditionally taking it as the
1276 *** An error is now thrown for `--opt=val' when the option description
1277 does not specify `(value #t)' or `(value optional)'. This condition used to
1278 be accepted w/o error, contrary to the documentation.
1280 *** The error message for unrecognized options is now more informative.
1281 It used to be "not a record", an artifact of the implementation.
1283 *** The error message for `--opt' terminating the arg list (no value), when
1284 `(value #t)' is specified, is now more informative. It used to be "not enough
1287 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1288 The expansion used to be like so:
1290 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1292 Note that the "5d" is dropped. Now it is like so:
1294 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1296 This enables single-char options to have adjoining arguments as long as their
1297 constituent characters are not potential single-char options.
1299 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1301 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1302 property `arglist', which can be retrieved by `arity'. The result is that
1303 `arity' can give more detailed information than before:
1307 guile> (use-modules (ice-9 optargs))
1308 guile> (define* (foo #:optional a b c) a)
1310 0 or more arguments in `lambda*:G0'.
1315 3 optional arguments: `a', `b' and `c'.
1316 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
1318 2 required arguments: `a' and `b', 2 keyword arguments: `c'
1319 and `d', other keywords allowed.
1320 guile> (define* (baz a b #:optional c #:rest r) a)
1322 2 required arguments: `a' and `b', 1 optional argument: `c',
1325 * Changes to the C interface
1327 ** Types have been renamed from scm_*_t to scm_t_*.
1329 This has been done for POSIX sake. It reserves identifiers ending
1330 with "_t". What a concept.
1332 The old names are still available with status `deprecated'.
1334 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
1336 ** Deprecated features have been removed.
1340 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
1341 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
1343 *** C Functions removed
1345 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
1346 scm_fseek - replaced by scm_seek.
1347 gc-thunk - replaced by after-gc-hook.
1348 gh_int2scmb - replaced by gh_bool2scm.
1349 scm_tc_dblr - replaced by scm_tc16_real.
1350 scm_tc_dblc - replaced by scm_tc16_complex.
1351 scm_list_star - replaced by scm_cons_star.
1353 ** Deprecated: scm_makfromstr
1355 Use scm_mem2string instead.
1357 ** Deprecated: scm_make_shared_substring
1359 Explicit shared substrings will disappear from Guile.
1361 Instead, "normal" strings will be implemented using sharing
1362 internally, combined with a copy-on-write strategy.
1364 ** Deprecated: scm_read_only_string_p
1366 The concept of read-only strings will disappear in next release of
1369 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
1371 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
1373 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
1375 Call a procedure with the indicated number of arguments.
1379 scm_call_1 (proc, arg1);
1381 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
1383 Call a procedure with the indicated number of arguments and a list
1388 scm_apply_1 (proc, arg1, args);
1390 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
1392 Create a list of the given number of elements.
1394 ** Renamed function: scm_listify has been replaced by scm_list_n.
1396 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
1397 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
1399 Use functions scm_list_N instead.
1401 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
1403 Used by an application to read arbitrary number of bytes from a port.
1404 Same semantics as libc read, except that scm_c_read only returns less
1405 than SIZE bytes if at end-of-file.
1407 Warning: Doesn't update port line and column counts!
1409 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
1411 Used by an application to write arbitrary number of bytes to an SCM
1412 port. Similar semantics as libc write. However, unlike libc
1413 write, scm_c_write writes the requested number of bytes and has no
1416 Warning: Doesn't update port line and column counts!
1418 ** New function: scm_init_guile ()
1420 In contrast to scm_boot_guile, scm_init_guile will return normally
1421 after initializing Guile. It is not available on all systems, tho.
1423 ** New functions: scm_str2symbol, scm_mem2symbol
1425 The function scm_str2symbol takes a const char* pointing to a zero-terminated
1426 field of characters and creates a scheme symbol object from that C string.
1427 The function scm_mem2symbol takes a const char* and a number of characters and
1428 creates a symbol from the characters in that memory area.
1430 ** New functions: scm_primitive_make_property
1431 scm_primitive_property_ref
1432 scm_primitive_property_set_x
1433 scm_primitive_property_del_x
1435 These functions implement a new way to deal with object properties.
1436 See libguile/properties.c for their documentation.
1438 ** New function: scm_done_free (long size)
1440 This function is the inverse of scm_done_malloc. Use it to report the
1441 amount of smob memory you free. The previous method, which involved
1442 calling scm_done_malloc with negative argument, was somewhat
1443 unintuitive (and is still available, of course).
1445 ** New function: scm_c_memq (SCM obj, SCM list)
1447 This function provides a fast C level alternative for scm_memq for the case
1448 that the list parameter is known to be a proper list. The function is a
1449 replacement for scm_sloppy_memq, but is stricter in its requirements on its
1450 list input parameter, since for anything else but a proper list the function's
1451 behaviour is undefined - it may even crash or loop endlessly. Further, for
1452 the case that the object is not found in the list, scm_c_memq returns #f which
1453 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
1455 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
1456 scm_remember_upto_here
1458 These functions replace the function scm_remember.
1460 ** Deprecated function: scm_remember
1462 Use one of the new functions scm_remember_upto_here_1,
1463 scm_remember_upto_here_2 or scm_remember_upto_here instead.
1465 ** New function: scm_allocate_string
1467 This function replaces the function scm_makstr.
1469 ** Deprecated function: scm_makstr
1471 Use the new function scm_allocate_string instead.
1473 ** New global variable scm_gc_running_p introduced.
1475 Use this variable to find out if garbage collection is being executed. Up to
1476 now applications have used scm_gc_heap_lock to test if garbage collection was
1477 running, which also works because of the fact that up to know only the garbage
1478 collector has set this variable. But, this is an implementation detail that
1479 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
1480 of this variable is (and has been) not fully safe anyway.
1482 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
1484 Use these instead of SCM_LENGTH_MAX.
1486 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
1487 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
1488 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
1490 Use these instead of SCM_LENGTH.
1492 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
1493 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
1494 SCM_SET_BITVECTOR_LENGTH
1496 Use these instead of SCM_SETLENGTH
1498 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
1499 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
1502 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
1505 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
1506 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
1509 Use these instead of SCM_SETCHARS.
1511 ** New macro: SCM_BITVECTOR_P
1513 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
1515 Use instead of SCM_COERCE_SUBSTR.
1517 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
1519 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
1521 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
1522 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
1523 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
1524 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
1525 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
1526 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
1527 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
1528 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
1529 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
1530 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
1531 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
1532 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
1533 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
1534 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
1535 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
1537 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
1538 Use scm_memory_error instead of SCM_NALLOC.
1539 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
1540 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
1541 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
1542 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
1543 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
1544 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
1545 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
1546 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
1547 Use SCM_STRINGP instead of SCM_RWSTRINGP.
1548 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
1549 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
1550 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
1551 Use a type specific setter macro instead of SCM_SETLENGTH.
1552 Use a type specific setter macro instead of SCM_SETCHARS.
1553 Use a type specific length macro instead of SCM_LENGTH_MAX.
1554 Use SCM_GCMARKP instead of SCM_GC8MARKP.
1555 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
1556 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
1557 Use SCM_TYP16 instead of SCM_GCTYP16.
1558 Use SCM_CDR instead of SCM_GCCDR.
1559 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
1560 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
1561 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
1562 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
1563 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
1564 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
1565 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
1567 ** Removed function: scm_struct_init
1569 ** Removed variable: scm_symhash_dim
1571 ** Renamed function: scm_make_cont has been replaced by
1572 scm_make_continuation, which has a different interface.
1574 ** Deprecated function: scm_call_catching_errors
1576 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
1578 ** Deprecated function: scm_strhash
1580 Use scm_string_hash instead.
1582 ** Deprecated function: scm_vector_set_length_x
1584 Instead, create a fresh vector of the desired size and copy the contents.
1586 ** scm_gensym has changed prototype
1588 scm_gensym now only takes one argument.
1590 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
1593 There is now only a single symbol type scm_tc7_symbol.
1594 The tag scm_tc7_lvector was not used anyway.
1596 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
1598 Use scm_make_smob_type and scm_set_smob_XXX instead.
1600 ** New function scm_set_smob_apply.
1602 This can be used to set an apply function to a smob type.
1604 ** Deprecated function: scm_strprint_obj
1606 Use scm_object_to_string instead.
1608 ** Deprecated function: scm_wta
1610 Use scm_wrong_type_arg, or another appropriate error signalling function
1613 ** Explicit support for obarrays has been deprecated.
1615 Use `scm_str2symbol' and the generic hashtable functions instead.
1617 ** The concept of `vcells' has been deprecated.
1619 The data type `variable' is now used exclusively. `Vcells' have been
1620 a low-level concept so you are likely not affected by this change.
1622 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
1623 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
1625 Use scm_c_define or scm_c_lookup instead, as appropriate.
1627 *** New functions: scm_c_module_lookup, scm_c_lookup,
1628 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
1629 scm_module_define, scm_define.
1631 These functions work with variables instead of with vcells.
1633 ** New functions for creating and defining `subr's and `gsubr's.
1635 The new functions more clearly distinguish between creating a subr (or
1636 gsubr) object and adding it to the current module.
1638 These new functions are available: scm_c_make_subr, scm_c_define_subr,
1639 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
1640 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
1641 scm_c_define_gsubr_with_generic.
1643 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
1644 scm_make_subr_with_generic, scm_make_gsubr,
1645 scm_make_gsubr_with_generic.
1647 Use the new ones from above instead.
1649 ** C interface to the module system has changed.
1651 While we suggest that you avoid as many explicit module system
1652 operations from C as possible for the time being, the C interface has
1653 been made more similar to the high-level Scheme module system.
1655 *** New functions: scm_c_define_module, scm_c_use_module,
1656 scm_c_export, scm_c_resolve_module.
1658 They mostly work like their Scheme namesakes. scm_c_define_module
1659 takes a function that is called a context where the new module is
1662 *** Deprecated functions: scm_the_root_module, scm_make_module,
1663 scm_ensure_user_module, scm_load_scheme_module.
1665 Use the new functions instead.
1667 ** Renamed function: scm_internal_with_fluids becomes
1670 scm_internal_with_fluids is available as a deprecated function.
1672 ** New function: scm_c_with_fluid.
1674 Just like scm_c_with_fluids, but takes one fluid and one value instead
1677 ** Deprecated typedefs: long_long, ulong_long.
1679 They are of questionable utility and they pollute the global
1682 ** Deprecated typedef: scm_sizet
1684 It is of questionable utility now that Guile requires ANSI C, and is
1687 ** Deprecated typedefs: scm_port_rw_active, scm_port,
1688 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
1689 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
1691 Made more compliant with the naming policy by adding a _t at the end.
1693 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
1694 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
1696 With the exception of the mysterious scm_2ulong2big, they are still
1697 available under new names (scm_i_mkbig etc). These functions are not
1698 intended to be used in user code. You should avoid dealing with
1699 bignums directly, and should deal with numbers in general (which can
1702 ** Change in behavior: scm_num2long, scm_num2ulong
1704 The scm_num2[u]long functions don't any longer accept an inexact
1705 argument. This change in behavior is motivated by concordance with
1706 R5RS: It is more common that a primitive doesn't want to accept an
1707 inexact for an exact.
1709 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
1710 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
1711 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
1714 These are conversion functions between the various ANSI C integral
1715 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
1716 accept an inexact argument.
1718 ** New functions: scm_float2num, scm_double2num,
1719 scm_num2float, scm_num2double.
1721 These are conversion functions between the two ANSI C float types and
1724 ** New number validation macros:
1725 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
1729 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
1731 These are just nicer-named old scm_protect_object and
1732 scm_unprotect_object.
1734 ** Deprecated functions: scm_protect_object, scm_unprotect_object
1736 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
1738 These functions can be used to register pointers to locations that
1741 ** Deprecated function: scm_create_hook.
1743 Its sins are: misleading name, non-modularity and lack of general
1747 Changes since Guile 1.3.4:
1749 * Changes to the distribution
1751 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
1753 We've changed the way we handle generated files in the Guile source
1754 repository. As a result, the procedure for building trees obtained
1755 from the nightly FTP snapshots or via CVS has changed:
1756 - You must have appropriate versions of autoconf, automake, and
1757 libtool installed on your system. See README for info on how to
1758 obtain these programs.
1759 - Before configuring the tree, you must first run the script
1760 `autogen.sh' at the top of the source tree.
1762 The Guile repository used to contain not only source files, written by
1763 humans, but also some generated files, like configure scripts and
1764 Makefile.in files. Even though the contents of these files could be
1765 derived mechanically from other files present, we thought it would
1766 make the tree easier to build if we checked them into CVS.
1768 However, this approach means that minor differences between
1769 developer's installed tools and habits affected the whole team.
1770 So we have removed the generated files from the repository, and
1771 added the autogen.sh script, which will reconstruct them
1775 ** configure now has experimental options to remove support for certain
1778 --disable-arrays omit array and uniform array support
1779 --disable-posix omit posix interfaces
1780 --disable-networking omit networking interfaces
1781 --disable-regex omit regular expression interfaces
1783 These are likely to become separate modules some day.
1785 ** New configure option --enable-debug-freelist
1787 This enables a debugging version of SCM_NEWCELL(), and also registers
1788 an extra primitive, the setter `gc-set-debug-check-freelist!'.
1790 Configure with the --enable-debug-freelist option to enable
1791 the gc-set-debug-check-freelist! primitive, and then use:
1793 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
1794 (gc-set-debug-check-freelist! #f) # turn off checking
1796 Checking of the freelist forces a traversal of the freelist and
1797 a garbage collection before each allocation of a cell. This can
1798 slow down the interpreter dramatically, so the setter should be used to
1799 turn on this extra processing only when necessary.
1801 ** New configure option --enable-debug-malloc
1803 Include code for debugging of calls to scm_must_malloc/realloc/free.
1807 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
1808 2. objects reallocated by scm_must_realloc has been allocated by
1810 3. reallocated objects are reallocated with the same what string
1812 But, most importantly, it records the number of allocated objects of
1813 each kind. This is useful when searching for memory leaks.
1815 A Guile compiled with this option provides the primitive
1816 `malloc-stats' which returns an alist with pairs of kind and the
1817 number of objects of that kind.
1819 ** All includes are now referenced relative to the root directory
1821 Since some users have had problems with mixups between Guile and
1822 system headers, we have decided to always refer to Guile headers via
1823 their parent directories. This essentially creates a "private name
1824 space" for Guile headers. This means that the compiler only is given
1825 -I options for the root build and root source directory.
1827 ** Header files kw.h and genio.h have been removed.
1829 ** The module (ice-9 getopt-gnu-style) has been removed.
1831 ** New module (ice-9 documentation)
1833 Implements the interface to documentation strings associated with
1836 ** New module (ice-9 time)
1838 Provides a macro `time', which displays execution time of a given form.
1840 ** New module (ice-9 history)
1842 Loading this module enables value history in the repl.
1844 * Changes to the stand-alone interpreter
1846 ** New command line option --debug
1848 Start Guile with debugging evaluator and backtraces enabled.
1850 This is useful when debugging your .guile init file or scripts.
1852 ** New help facility
1854 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
1855 (help REGEXP) ditto for objects with names matching REGEXP (a string)
1856 (help 'NAME) gives documentation for NAME, even if it is not an object
1857 (help ,EXPR) gives documentation for object returned by EXPR
1858 (help (my module)) gives module commentary for `(my module)'
1859 (help) gives this text
1861 `help' searches among bindings exported from loaded modules, while
1862 `apropos' searches among bindings visible from the "current" module.
1864 Examples: (help help)
1866 (help "output-string")
1868 ** `help' and `apropos' now prints full module names
1870 ** Dynamic linking now uses libltdl from the libtool package.
1872 The old system dependent code for doing dynamic linking has been
1873 replaced with calls to the libltdl functions which do all the hairy
1876 The major improvement is that you can now directly pass libtool
1877 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
1878 will be able to do the best shared library job you can get, via
1881 The way dynamic libraries are found has changed and is not really
1882 portable across platforms, probably. It is therefore recommended to
1883 use absolute filenames when possible.
1885 If you pass a filename without an extension to `dynamic-link', it will
1886 try a few appropriate ones. Thus, the most platform ignorant way is
1887 to specify a name like "libfoo", without any directories and
1890 ** Guile COOP threads are now compatible with LinuxThreads
1892 Previously, COOP threading wasn't possible in applications linked with
1893 Linux POSIX threads due to their use of the stack pointer to find the
1894 thread context. This has now been fixed with a workaround which uses
1895 the pthreads to allocate the stack.
1897 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
1899 ** Positions of erring expression in scripts
1901 With version 1.3.4, the location of the erring expression in Guile
1902 scipts is no longer automatically reported. (This should have been
1903 documented before the 1.3.4 release.)
1905 You can get this information by enabling recording of positions of
1906 source expressions and running the debugging evaluator. Put this at
1907 the top of your script (or in your "site" file):
1909 (read-enable 'positions)
1910 (debug-enable 'debug)
1912 ** Backtraces in scripts
1914 It is now possible to get backtraces in scripts.
1918 (debug-enable 'debug 'backtrace)
1920 at the top of the script.
1922 (The first options enables the debugging evaluator.
1923 The second enables backtraces.)
1925 ** Part of module system symbol lookup now implemented in C
1927 The eval closure of most modules is now implemented in C. Since this
1928 was one of the bottlenecks for loading speed, Guile now loads code
1929 substantially faster than before.
1931 ** Attempting to get the value of an unbound variable now produces
1932 an exception with a key of 'unbound-variable instead of 'misc-error.
1934 ** The initial default output port is now unbuffered if it's using a
1935 tty device. Previously in this situation it was line-buffered.
1937 ** New hook: after-gc-hook
1939 after-gc-hook takes over the role of gc-thunk. This hook is run at
1940 the first SCM_TICK after a GC. (Thus, the code is run at the same
1941 point during evaluation as signal handlers.)
1943 Note that this hook should be used only for diagnostic and debugging
1944 purposes. It is not certain that it will continue to be well-defined
1945 when this hook is run in the future.
1947 C programmers: Note the new C level hooks scm_before_gc_c_hook,
1948 scm_before_sweep_c_hook, scm_after_gc_c_hook.
1950 ** Improvements to garbage collector
1952 Guile 1.4 has a new policy for triggering heap allocation and
1953 determining the sizes of heap segments. It fixes a number of problems
1956 1. The new policy can handle two separate pools of cells
1957 (2-word/4-word) better. (The old policy would run wild, allocating
1958 more and more memory for certain programs.)
1960 2. The old code would sometimes allocate far too much heap so that the
1961 Guile process became gigantic. The new code avoids this.
1963 3. The old code would sometimes allocate too little so that few cells
1964 were freed at GC so that, in turn, too much time was spent in GC.
1966 4. The old code would often trigger heap allocation several times in a
1967 row. (The new scheme predicts how large the segments needs to be
1968 in order not to need further allocation.)
1970 All in all, the new GC policy will make larger applications more
1973 The new GC scheme also is prepared for POSIX threading. Threads can
1974 allocate private pools of cells ("clusters") with just a single
1975 function call. Allocation of single cells from such a cluster can
1976 then proceed without any need of inter-thread synchronization.
1978 ** New environment variables controlling GC parameters
1980 GUILE_MAX_SEGMENT_SIZE Maximal segment size
1983 Allocation of 2-word cell heaps:
1985 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
1988 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
1989 GC in percent of total heap size
1992 Allocation of 4-word cell heaps
1993 (used for real numbers and misc other objects):
1995 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
1997 (See entry "Way for application to customize GC parameters" under
1998 section "Changes to the scm_ interface" below.)
2000 ** Guile now implements reals using 4-word cells
2002 This speeds up computation with reals. (They were earlier allocated
2003 with `malloc'.) There is still some room for optimizations, however.
2005 ** Some further steps toward POSIX thread support have been taken
2007 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2008 don't have much effect any longer, and many of them will be removed in
2012 are only handled at the top of the evaluator loop, immediately after
2013 I/O, and in scm_equalp.
2015 *** The GC can allocate thread private pools of pairs.
2017 * Changes to Scheme functions and syntax
2019 ** close-input-port and close-output-port are now R5RS
2021 These procedures have been turned into primitives and have R5RS behaviour.
2023 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2025 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2026 extended by the more sophisticated version in (ice-9 format)
2028 (simple-format port message . args)
2029 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2030 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2031 the escapes are replaced with corresponding members of ARGS:
2032 ~A formats using `display' and ~S formats using `write'.
2033 If DESTINATION is #t, then use the `current-output-port',
2034 if DESTINATION is #f, then return a string containing the formatted text.
2035 Does not add a trailing newline."
2037 ** string-ref: the second argument is no longer optional.
2039 ** string, list->string: no longer accept strings in their arguments,
2040 only characters, for compatibility with R5RS.
2042 ** New procedure: port-closed? PORT
2043 Returns #t if PORT is closed or #f if it is open.
2045 ** Deprecated: list*
2047 The list* functionality is now provided by cons* (SRFI-1 compliant)
2049 ** New procedure: cons* ARG1 ARG2 ... ARGn
2051 Like `list', but the last arg provides the tail of the constructed list,
2052 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2054 Requires at least one argument. If given one argument, that argument
2055 is returned as result.
2057 This function is called `list*' in some other Schemes and in Common LISP.
2059 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2061 ** New procedure: object-documentation OBJECT
2063 Returns the documentation string associated with OBJECT. The
2064 procedure uses a caching mechanism so that subsequent lookups are
2067 Exported by (ice-9 documentation).
2069 ** module-name now returns full names of modules
2071 Previously, only the last part of the name was returned (`session' for
2072 `(ice-9 session)'). Ex: `(ice-9 session)'.
2074 * Changes to the gh_ interface
2076 ** Deprecated: gh_int2scmb
2078 Use gh_bool2scm instead.
2080 * Changes to the scm_ interface
2082 ** Guile primitives now carry docstrings!
2084 Thanks to Greg Badros!
2086 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2088 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2089 macros and must contain a docstring that is extracted into foo.doc using a new
2090 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2092 However, a major overhaul of these macros is scheduled for the next release of
2095 ** Guile primitives use a new technique for validation of arguments
2097 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2098 the readability of argument checking.
2100 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2102 ** New macros: SCM_PACK, SCM_UNPACK
2104 Compose/decompose an SCM value.
2106 The SCM type is now treated as an abstract data type and may be defined as a
2107 long, a void* or as a struct, depending on the architecture and compile time
2108 options. This makes it easier to find several types of bugs, for example when
2109 SCM values are treated as integers without conversion. Values of the SCM type
2110 should be treated as "atomic" values. These macros are used when
2111 composing/decomposing an SCM value, either because you want to access
2112 individual bits, or because you want to treat it as an integer value.
2114 E.g., in order to set bit 7 in an SCM value x, use the expression
2116 SCM_PACK (SCM_UNPACK (x) | 0x80)
2118 ** The name property of hooks is deprecated.
2119 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2121 You can emulate this feature by using object properties.
2123 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2124 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2127 These macros will be removed in a future release of Guile.
2129 ** The following types, functions and macros from numbers.h are deprecated:
2130 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2131 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2133 Further, it is recommended not to rely on implementation details for guile's
2134 current implementation of bignums. It is planned to replace this
2135 implementation with gmp in the future.
2137 ** Port internals: the rw_random variable in the scm_port structure
2138 must be set to non-zero in any random access port. In recent Guile
2139 releases it was only set for bidirectional random-access ports.
2141 ** Port internals: the seek ptob procedure is now responsible for
2142 resetting the buffers if required. The change was made so that in the
2143 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2144 the fport and strport ptobs can avoid resetting the buffers,
2145 in particular to avoid discarding unread chars. An existing port
2146 type can be fixed by adding something like the following to the
2147 beginning of the ptob seek procedure:
2149 if (pt->rw_active == SCM_PORT_READ)
2150 scm_end_input (object);
2151 else if (pt->rw_active == SCM_PORT_WRITE)
2152 ptob->flush (object);
2154 although to actually avoid resetting the buffers and discard unread
2155 chars requires further hacking that depends on the characteristics
2158 ** Deprecated functions: scm_fseek, scm_tag
2160 These functions are no longer used and will be removed in a future version.
2162 ** The scm_sysmissing procedure is no longer used in libguile.
2163 Unless it turns out to be unexpectedly useful to somebody, it will be
2164 removed in a future version.
2166 ** The format of error message strings has changed
2168 The two C procedures: scm_display_error and scm_error, as well as the
2169 primitive `scm-error', now use scm_simple_format to do their work.
2170 This means that the message strings of all code must be updated to use
2171 ~A where %s was used before, and ~S where %S was used before.
2173 During the period when there still are a lot of old Guiles out there,
2174 you might want to support both old and new versions of Guile.
2176 There are basically two methods to achieve this. Both methods use
2179 AC_CHECK_FUNCS(scm_simple_format)
2181 in your configure.in.
2183 Method 1: Use the string concatenation features of ANSI C's
2188 #ifdef HAVE_SCM_SIMPLE_FORMAT
2194 Then represent each of your error messages using a preprocessor macro:
2196 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2200 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2201 (define make-message string-append)
2203 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2205 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2209 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2214 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2218 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2220 Don't use the functions coop_mutex_init and
2221 coop_condition_variable_init. They will change.
2223 Use scm_mutex_init and scm_cond_init instead.
2225 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2226 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2227 COND, as `scm_cond_wait' does, but it also bounds the duration
2228 of the wait. If COND has not been signaled before time ABSTIME,
2229 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2230 returns the error code `ETIMEDOUT'.
2232 The ABSTIME parameter specifies an absolute time, with the same
2233 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2234 to 00:00:00 GMT, January 1, 1970.
2236 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2237 `scm_cond_broadcast' restarts all the threads that are waiting
2238 on the condition variable COND. Nothing happens if no threads are
2241 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2242 `scm_key_create' allocates a new TSD key. The key is stored in
2243 the location pointed to by KEY. There is no limit on the number
2244 of keys allocated at a given time. The value initially associated
2245 with the returned key is `NULL' in all currently executing threads.
2247 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2248 function associated with the key. When a thread terminates,
2249 DESTR_FUNCTION is called on the value associated with the key in
2250 that thread. The DESTR_FUNCTION is not called if a key is deleted
2251 with `scm_key_delete' or a value is changed with
2252 `scm_setspecific'. The order in which destructor functions are
2253 called at thread termination time is unspecified.
2255 Destructors are not yet implemented.
2257 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2258 `scm_setspecific' changes the value associated with KEY in the
2259 calling thread, storing the given POINTER instead.
2261 ** New function: scm_getspecific (scm_key_t KEY)
2262 `scm_getspecific' returns the value currently associated with
2263 KEY in the calling thread.
2265 ** New function: scm_key_delete (scm_key_t KEY)
2266 `scm_key_delete' deallocates a TSD key. It does not check
2267 whether non-`NULL' values are associated with that key in the
2268 currently executing threads, nor call the destructor function
2269 associated with the key.
2271 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2273 Initialize a C level hook HOOK with associated HOOK_DATA and type
2274 TYPE. (See scm_c_hook_run ().)
2276 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2278 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2279 is true, add it last, otherwise first. The same FUNC can be added
2280 multiple times if FUNC_DATA differ and vice versa.
2282 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2284 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2285 function is only removed if both FUNC and FUNC_DATA matches.
2287 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2289 Run hook HOOK passing DATA to the hook functions.
2291 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2292 returned is undefined.
2294 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2295 returns a non-NULL value. This value is returned as the result of
2296 scm_c_hook_run. If all functions return NULL, NULL is returned.
2298 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2299 returns a NULL value, and NULL is returned. If all functions returns
2300 a non-NULL value, the last value is returned.
2302 ** New C level GC hooks
2304 Five new C level hooks has been added to the garbage collector.
2306 scm_before_gc_c_hook
2309 are run before locking and after unlocking the heap. The system is
2310 thus in a mode where evaluation can take place. (Except that
2311 scm_before_gc_c_hook must not allocate new cells.)
2313 scm_before_mark_c_hook
2314 scm_before_sweep_c_hook
2315 scm_after_sweep_c_hook
2317 are run when the heap is locked. These are intended for extension of
2318 the GC in a modular fashion. Examples are the weaks and guardians
2321 ** Way for application to customize GC parameters
2323 The application can set up other default values for the GC heap
2324 allocation parameters
2326 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
2327 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
2328 GUILE_MAX_SEGMENT_SIZE,
2332 scm_default_init_heap_size_1, scm_default_min_yield_1,
2333 scm_default_init_heap_size_2, scm_default_min_yield_2,
2334 scm_default_max_segment_size
2336 respectively before callong scm_boot_guile.
2338 (See entry "New environment variables ..." in section
2339 "Changes to the stand-alone interpreter" above.)
2341 ** scm_protect_object/scm_unprotect_object now nest
2343 This means that you can call scm_protect_object multiple times on an
2344 object and count on the object being protected until
2345 scm_unprotect_object has been call the same number of times.
2347 The functions also have better time complexity.
2349 Still, it is usually possible to structure the application in a way
2350 that you don't need to use these functions. For example, if you use a
2351 protected standard Guile list to keep track of live objects rather
2352 than some custom data type, objects will die a natural death when they
2353 are no longer needed.
2355 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
2357 Guile does not provide the float representation for inexact real numbers any
2358 more. Now, only doubles are used to represent inexact real numbers. Further,
2359 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
2360 and scm_tc16_complex, respectively.
2362 ** Removed deprecated type scm_smobfuns
2364 ** Removed deprecated function scm_newsmob
2366 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
2368 There is an ongoing discussion among the developers whether to
2369 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
2370 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
2371 until this issue has been settled.
2373 ** Removed deprecated type tag scm_tc16_kw
2375 ** Added type tag scm_tc16_keyword
2377 (This was introduced already in release 1.3.4 but was not documented
2380 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
2382 * Changes to system call interfaces:
2384 ** The "select" procedure now tests port buffers for the ability to
2385 provide input or accept output. Previously only the underlying file
2386 descriptors were checked.
2388 ** New variable PIPE_BUF: the maximum number of bytes that can be
2389 atomically written to a pipe.
2391 ** If a facility is not available on the system when Guile is
2392 compiled, the corresponding primitive procedure will not be defined.
2393 Previously it would have been defined but would throw a system-error
2394 exception if called. Exception handlers which catch this case may
2395 need minor modification: an error will be thrown with key
2396 'unbound-variable instead of 'system-error. Alternatively it's
2397 now possible to use `defined?' to check whether the facility is
2400 ** Procedures which depend on the timezone should now give the correct
2401 result on systems which cache the TZ environment variable, even if TZ
2402 is changed without calling tzset.
2404 * Changes to the networking interfaces:
2406 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
2407 long integers between network and host format. For now, it's not
2408 particularly convenient to do this kind of thing, but consider:
2410 (define write-network-long
2411 (lambda (value port)
2412 (let ((v (make-uniform-vector 1 1 0)))
2413 (uniform-vector-set! v 0 (htonl value))
2414 (uniform-vector-write v port))))
2416 (define read-network-long
2418 (let ((v (make-uniform-vector 1 1 0)))
2419 (uniform-vector-read! v port)
2420 (ntohl (uniform-vector-ref v 0)))))
2422 ** If inet-aton fails, it now throws an error with key 'misc-error
2423 instead of 'system-error, since errno is not relevant.
2425 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
2426 specific keys instead of 'system-error. The latter is inappropriate
2427 since errno will not have been set. The keys are:
2428 'host-not-found, 'try-again, 'no-recovery and 'no-data.
2430 ** sethostent, setnetent, setprotoent, setservent: now take an
2431 optional argument STAYOPEN, which specifies whether the database
2432 remains open after a database entry is accessed randomly (e.g., using
2433 gethostbyname for the hosts database.) The default is #f. Previously
2437 Changes since Guile 1.3.2:
2439 * Changes to the stand-alone interpreter
2443 An initial version of the Guile debugger written by Chris Hanson has
2444 been added. The debugger is still under development but is included
2445 in the distribution anyway since it is already quite useful.
2451 after an error to enter the debugger. Type `help' inside the debugger
2452 for a description of available commands.
2454 If you prefer to have stack frames numbered and printed in
2455 anti-chronological order and prefer up in the stack to be down on the
2456 screen as is the case in gdb, you can put
2458 (debug-enable 'backwards)
2460 in your .guile startup file. (However, this means that Guile can't
2461 use indentation to indicate stack level.)
2463 The debugger is autoloaded into Guile at the first use.
2465 ** Further enhancements to backtraces
2467 There is a new debug option `width' which controls the maximum width
2468 on the screen of printed stack frames. Fancy printing parameters
2469 ("level" and "length" as in Common LISP) are adaptively adjusted for
2470 each stack frame to give maximum information while still fitting
2471 within the bounds. If the stack frame can't be made to fit by
2472 adjusting parameters, it is simply cut off at the end. This is marked
2475 ** Some modules are now only loaded when the repl is started
2477 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
2478 regex) are now loaded into (guile-user) only if the repl has been
2479 started. The effect is that the startup time for scripts has been
2480 reduced to 30% of what it was previously.
2482 Correctly written scripts load the modules they require at the top of
2483 the file and should not be affected by this change.
2485 ** Hooks are now represented as smobs
2487 * Changes to Scheme functions and syntax
2489 ** Readline support has changed again.
2491 The old (readline-activator) module is gone. Use (ice-9 readline)
2492 instead, which now contains all readline functionality. So the code
2493 to activate readline is now
2495 (use-modules (ice-9 readline))
2498 This should work at any time, including from the guile prompt.
2500 To avoid confusion about the terms of Guile's license, please only
2501 enable readline for your personal use; please don't make it the
2502 default for others. Here is why we make this rather odd-sounding
2505 Guile is normally licensed under a weakened form of the GNU General
2506 Public License, which allows you to link code with Guile without
2507 placing that code under the GPL. This exception is important to some
2510 However, since readline is distributed under the GNU General Public
2511 License, when you link Guile with readline, either statically or
2512 dynamically, you effectively change Guile's license to the strict GPL.
2513 Whenever you link any strictly GPL'd code into Guile, uses of Guile
2514 which are normally permitted become forbidden. This is a rather
2515 non-obvious consequence of the licensing terms.
2517 So, to make sure things remain clear, please let people choose for
2518 themselves whether to link GPL'd libraries like readline with Guile.
2520 ** regexp-substitute/global has changed slightly, but incompatibly.
2522 If you include a function in the item list, the string of the match
2523 object it receives is the same string passed to
2524 regexp-substitute/global, not some suffix of that string.
2525 Correspondingly, the match's positions are relative to the entire
2526 string, not the suffix.
2528 If the regexp can match the empty string, the way matches are chosen
2529 from the string has changed. regexp-substitute/global recognizes the
2530 same set of matches that list-matches does; see below.
2532 ** New function: list-matches REGEXP STRING [FLAGS]
2534 Return a list of match objects, one for every non-overlapping, maximal
2535 match of REGEXP in STRING. The matches appear in left-to-right order.
2536 list-matches only reports matches of the empty string if there are no
2537 other matches which begin on, end at, or include the empty match's
2540 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2542 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
2544 For each match of REGEXP in STRING, apply PROC to the match object,
2545 and the last value PROC returned, or INIT for the first call. Return
2546 the last value returned by PROC. We apply PROC to the matches as they
2547 appear from left to right.
2549 This function recognizes matches according to the same criteria as
2552 Thus, you could define list-matches like this:
2554 (define (list-matches regexp string . flags)
2555 (reverse! (apply fold-matches regexp string '() cons flags)))
2557 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2561 *** New function: hook? OBJ
2563 Return #t if OBJ is a hook, otherwise #f.
2565 *** New function: make-hook-with-name NAME [ARITY]
2567 Return a hook with name NAME and arity ARITY. The default value for
2568 ARITY is 0. The only effect of NAME is that it will appear when the
2569 hook object is printed to ease debugging.
2571 *** New function: hook-empty? HOOK
2573 Return #t if HOOK doesn't contain any procedures, otherwise #f.
2575 *** New function: hook->list HOOK
2577 Return a list of the procedures that are called when run-hook is
2580 ** `map' signals an error if its argument lists are not all the same length.
2582 This is the behavior required by R5RS, so this change is really a bug
2583 fix. But it seems to affect a lot of people's code, so we're
2584 mentioning it here anyway.
2586 ** Print-state handling has been made more transparent
2588 Under certain circumstances, ports are represented as a port with an
2589 associated print state. Earlier, this pair was represented as a pair
2590 (see "Some magic has been added to the printer" below). It is now
2591 indistinguishable (almost; see `get-print-state') from a port on the
2594 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
2596 Return a new port with the associated print state PRINT-STATE.
2598 *** New function: get-print-state OUTPUT-PORT
2600 Return the print state associated with this port if it exists,
2601 otherwise return #f.
2603 *** New function: directory-stream? OBJECT
2605 Returns true iff OBJECT is a directory stream --- the sort of object
2606 returned by `opendir'.
2608 ** New function: using-readline?
2610 Return #t if readline is in use in the current repl.
2612 ** structs will be removed in 1.4
2614 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
2615 and use GOOPS objects as the fundamental record type.
2617 * Changes to the scm_ interface
2619 ** structs will be removed in 1.4
2621 The entire current struct interface (struct.c, struct.h) will be
2622 replaced in Guile 1.4. We will merge GOOPS into libguile and use
2623 GOOPS objects as the fundamental record type.
2625 ** The internal representation of subr's has changed
2627 Instead of giving a hint to the subr name, the CAR field of the subr
2628 now contains an index to a subr entry in scm_subr_table.
2630 *** New variable: scm_subr_table
2632 An array of subr entries. A subr entry contains the name, properties
2633 and documentation associated with the subr. The properties and
2634 documentation slots are not yet used.
2636 ** A new scheme for "forwarding" calls to a builtin to a generic function
2638 It is now possible to extend the functionality of some Guile
2639 primitives by letting them defer a call to a GOOPS generic function on
2640 argument mismatch. This means that there is no loss of efficiency in
2645 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
2646 (define-method + ((x <string>) (y <string>))
2647 (string-append x y))
2649 + will still be as efficient as usual in numerical calculations, but
2650 can also be used for concatenating strings.
2652 Who will be the first one to extend Guile's numerical tower to
2653 rationals? :) [OK, there a few other things to fix before this can
2654 be made in a clean way.]
2656 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
2658 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
2660 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
2662 These do the same job as SCM_PROC and SCM_PROC1, but they also define
2663 a variable GENERIC which can be used by the dispatch macros below.
2665 [This is experimental code which may change soon.]
2667 *** New macros for forwarding control to a generic on arg type error
2669 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
2671 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
2673 These correspond to the scm_wta function call, and have the same
2674 behaviour until the user has called the GOOPS primitive
2675 `enable-primitive-generic!'. After that, these macros will apply the
2676 generic function GENERIC to the argument(s) instead of calling
2679 [This is experimental code which may change soon.]
2681 *** New macros for argument testing with generic dispatch
2683 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
2685 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
2687 These correspond to the SCM_ASSERT macro, but will defer control to
2688 GENERIC on error after `enable-primitive-generic!' has been called.
2690 [This is experimental code which may change soon.]
2692 ** New function: SCM scm_eval_body (SCM body, SCM env)
2694 Evaluates the body of a special form.
2696 ** The internal representation of struct's has changed
2698 Previously, four slots were allocated for the procedure(s) of entities
2699 and operators. The motivation for this representation had to do with
2700 the structure of the evaluator, the wish to support tail-recursive
2701 generic functions, and efficiency. Since the generic function
2702 dispatch mechanism has changed, there is no longer a need for such an
2703 expensive representation, and the representation has been simplified.
2705 This should not make any difference for most users.
2707 ** GOOPS support has been cleaned up.
2709 Some code has been moved from eval.c to objects.c and code in both of
2710 these compilation units has been cleaned up and better structured.
2712 *** New functions for applying generic functions
2714 New function: SCM scm_apply_generic (GENERIC, ARGS)
2715 New function: SCM scm_call_generic_0 (GENERIC)
2716 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
2717 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
2718 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
2720 ** Deprecated function: scm_make_named_hook
2722 It is now replaced by:
2724 ** New function: SCM scm_create_hook (const char *name, int arity)
2726 Creates a hook in the same way as make-hook above but also
2727 binds a variable named NAME to it.
2729 This is the typical way of creating a hook from C code.
2731 Currently, the variable is created in the "current" module.
2732 This might change when we get the new module system.
2734 [The behaviour is identical to scm_make_named_hook.]
2738 Changes since Guile 1.3:
2740 * Changes to mailing lists
2742 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
2744 See the README file to find current addresses for all the Guile
2747 * Changes to the distribution
2749 ** Readline support is no longer included with Guile by default.
2751 Based on the different license terms of Guile and Readline, we
2752 concluded that Guile should not *by default* cause the linking of
2753 Readline into an application program. Readline support is now offered
2754 as a separate module, which is linked into an application only when
2755 you explicitly specify it.
2757 Although Guile is GNU software, its distribution terms add a special
2758 exception to the usual GNU General Public License (GPL). Guile's
2759 license includes a clause that allows you to link Guile with non-free
2760 programs. We add this exception so as not to put Guile at a
2761 disadvantage vis-a-vis other extensibility packages that support other
2764 In contrast, the GNU Readline library is distributed under the GNU
2765 General Public License pure and simple. This means that you may not
2766 link Readline, even dynamically, into an application unless it is
2767 distributed under a free software license that is compatible the GPL.
2769 Because of this difference in distribution terms, an application that
2770 can use Guile may not be able to use Readline. Now users will be
2771 explicitly offered two independent decisions about the use of these
2774 You can activate the readline support by issuing
2776 (use-modules (readline-activator))
2779 from your ".guile" file, for example.
2781 * Changes to the stand-alone interpreter
2783 ** All builtins now print as primitives.
2784 Previously builtin procedures not belonging to the fundamental subr
2785 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
2786 Now, they print as #<primitive-procedure NAME>.
2788 ** Backtraces slightly more intelligible.
2789 gsubr-apply and macro transformer application frames no longer appear
2792 * Changes to Scheme functions and syntax
2794 ** Guile now correctly handles internal defines by rewriting them into
2795 their equivalent letrec. Previously, internal defines would
2796 incrementally add to the innermost environment, without checking
2797 whether the restrictions specified in RnRS were met. This lead to the
2798 correct behaviour when these restriction actually were met, but didn't
2799 catch all illegal uses. Such an illegal use could lead to crashes of
2800 the Guile interpreter or or other unwanted results. An example of
2801 incorrect internal defines that made Guile behave erratically:
2813 The problem with this example is that the definition of `c' uses the
2814 value of `b' directly. This confuses the meoization machine of Guile
2815 so that the second call of `b' (this time in a larger environment that
2816 also contains bindings for `c' and `d') refers to the binding of `c'
2817 instead of `a'. You could also make Guile crash with a variation on
2822 (define (b flag) (if flag a 1))
2823 (define c (1+ (b flag)))
2831 From now on, Guile will issue an `Unbound variable: b' error message
2836 A hook contains a list of functions which should be called on
2837 particular occasions in an existing program. Hooks are used for
2840 A window manager might have a hook before-window-map-hook. The window
2841 manager uses the function run-hooks to call all functions stored in
2842 before-window-map-hook each time a window is mapped. The user can
2843 store functions in the hook using add-hook!.
2845 In Guile, hooks are first class objects.
2847 *** New function: make-hook [N_ARGS]
2849 Return a hook for hook functions which can take N_ARGS arguments.
2850 The default value for N_ARGS is 0.
2852 (See also scm_make_named_hook below.)
2854 *** New function: add-hook! HOOK PROC [APPEND_P]
2856 Put PROC at the beginning of the list of functions stored in HOOK.
2857 If APPEND_P is supplied, and non-false, put PROC at the end instead.
2859 PROC must be able to take the number of arguments specified when the
2862 If PROC already exists in HOOK, then remove it first.
2864 *** New function: remove-hook! HOOK PROC
2866 Remove PROC from the list of functions in HOOK.
2868 *** New function: reset-hook! HOOK
2870 Clear the list of hook functions stored in HOOK.
2872 *** New function: run-hook HOOK ARG1 ...
2874 Run all hook functions stored in HOOK with arguments ARG1 ... .
2875 The number of arguments supplied must correspond to the number given
2876 when the hook was created.
2878 ** The function `dynamic-link' now takes optional keyword arguments.
2879 The only keyword argument that is currently defined is `:global
2880 BOOL'. With it, you can control whether the shared library will be
2881 linked in global mode or not. In global mode, the symbols from the
2882 linked library can be used to resolve references from other
2883 dynamically linked libraries. In non-global mode, the linked
2884 library is essentially invisible and can only be accessed via
2885 `dynamic-func', etc. The default is now to link in global mode.
2886 Previously, the default has been non-global mode.
2888 The `#:global' keyword is only effective on platforms that support
2889 the dlopen family of functions.
2891 ** New function `provided?'
2893 - Function: provided? FEATURE
2894 Return true iff FEATURE is supported by this installation of
2895 Guile. FEATURE must be a symbol naming a feature; the global
2896 variable `*features*' is a list of available features.
2898 ** Changes to the module (ice-9 expect):
2900 *** The expect-strings macro now matches `$' in a regular expression
2901 only at a line-break or end-of-file by default. Previously it would
2902 match the end of the string accumulated so far. The old behaviour
2903 can be obtained by setting the variable `expect-strings-exec-flags'
2906 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
2907 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
2908 in a regular expression will still match before a line-break or
2909 end-of-file. The default is `regexp/noteol'.
2911 *** The expect-strings macro now uses a variable
2912 `expect-strings-compile-flags' for the flags to be supplied to
2913 `make-regexp'. The default is `regexp/newline', which was previously
2916 *** The expect macro now supplies two arguments to a match procedure:
2917 the current accumulated string and a flag to indicate whether
2918 end-of-file has been reached. Previously only the string was supplied.
2919 If end-of-file is reached, the match procedure will be called an
2920 additional time with the same accumulated string as the previous call
2921 but with the flag set.
2923 ** New module (ice-9 format), implementing the Common Lisp `format' function.
2925 This code, and the documentation for it that appears here, was
2926 borrowed from SLIB, with minor adaptations for Guile.
2928 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
2929 An almost complete implementation of Common LISP format description
2930 according to the CL reference book `Common LISP' from Guy L.
2931 Steele, Digital Press. Backward compatible to most of the
2932 available Scheme format implementations.
2934 Returns `#t', `#f' or a string; has side effect of printing
2935 according to FORMAT-STRING. If DESTINATION is `#t', the output is
2936 to the current output port and `#t' is returned. If DESTINATION
2937 is `#f', a formatted string is returned as the result of the call.
2938 NEW: If DESTINATION is a string, DESTINATION is regarded as the
2939 format string; FORMAT-STRING is then the first argument and the
2940 output is returned as a string. If DESTINATION is a number, the
2941 output is to the current error port if available by the
2942 implementation. Otherwise DESTINATION must be an output port and
2945 FORMAT-STRING must be a string. In case of a formatting error
2946 format returns `#f' and prints a message on the current output or
2947 error port. Characters are output as if the string were output by
2948 the `display' function with the exception of those prefixed by a
2949 tilde (~). For a detailed description of the FORMAT-STRING syntax
2950 please consult a Common LISP format reference manual. For a test
2951 suite to verify this format implementation load `formatst.scm'.
2952 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
2954 Note: `format' is not reentrant, i.e. only one `format'-call may
2955 be executed at a time.
2958 *** Format Specification (Format version 3.0)
2960 Please consult a Common LISP format reference manual for a detailed
2961 description of the format string syntax. For a demonstration of the
2962 implemented directives see `formatst.scm'.
2964 This implementation supports directive parameters and modifiers (`:'
2965 and `@' characters). Multiple parameters must be separated by a comma
2966 (`,'). Parameters can be numerical parameters (positive or negative),
2967 character parameters (prefixed by a quote character (`''), variable
2968 parameters (`v'), number of rest arguments parameter (`#'), empty and
2969 default parameters. Directive characters are case independent. The
2970 general form of a directive is:
2972 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
2974 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
2976 *** Implemented CL Format Control Directives
2978 Documentation syntax: Uppercase characters represent the
2979 corresponding control directive characters. Lowercase characters
2980 represent control directive parameter descriptions.
2983 Any (print as `display' does).
2987 `~MINCOL,COLINC,MINPAD,PADCHARA'
2991 S-expression (print as `write' does).
2995 `~MINCOL,COLINC,MINPAD,PADCHARS'
3001 print number sign always.
3004 print comma separated.
3006 `~MINCOL,PADCHAR,COMMACHARD'
3012 print number sign always.
3015 print comma separated.
3017 `~MINCOL,PADCHAR,COMMACHARX'
3023 print number sign always.
3026 print comma separated.
3028 `~MINCOL,PADCHAR,COMMACHARO'
3034 print number sign always.
3037 print comma separated.
3039 `~MINCOL,PADCHAR,COMMACHARB'
3044 `~N,MINCOL,PADCHAR,COMMACHARR'
3048 print a number as a Roman numeral.
3051 print a number as an "old fashioned" Roman numeral.
3054 print a number as an ordinal English number.
3057 print a number as a cardinal English number.
3062 prints `y' and `ies'.
3065 as `~P but jumps 1 argument backward.'
3068 as `~@P but jumps 1 argument backward.'
3073 prints a character as the reader can understand it (i.e. `#\'
3077 prints a character as emacs does (eg. `^C' for ASCII 03).
3080 Fixed-format floating-point (prints a flonum like MMM.NNN).
3081 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3083 If the number is positive a plus sign is printed.
3086 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3087 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3089 If the number is positive a plus sign is printed.
3092 General floating-point (prints a flonum either fixed or
3094 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3096 If the number is positive a plus sign is printed.
3099 Dollars floating-point (prints a flonum in fixed with signs
3101 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3103 If the number is positive a plus sign is printed.
3106 A sign is always printed and appears before the padding.
3109 The sign appears before the padding.
3117 print newline if not at the beginning of the output line.
3119 prints `~&' and then N-1 newlines.
3124 print N page separators.
3134 newline is ignored, white space left.
3137 newline is left, white space ignored.
3142 relative tabulation.
3148 Indirection (expects indirect arguments as a list).
3150 extracts indirect arguments from format arguments.
3153 Case conversion (converts by `string-downcase').
3155 converts by `string-capitalize'.
3158 converts by `string-capitalize-first'.
3161 converts by `string-upcase'.
3164 Argument Jumping (jumps 1 argument forward).
3166 jumps N arguments forward.
3169 jumps 1 argument backward.
3172 jumps N arguments backward.
3175 jumps to the 0th argument.
3178 jumps to the Nth argument (beginning from 0)
3180 `~[STR0~;STR1~;...~;STRN~]'
3181 Conditional Expression (numerical clause conditional).
3183 take argument from N.
3186 true test conditional.
3189 if-else-then conditional.
3195 default clause follows.
3198 Iteration (args come from the next argument (a list)).
3200 at most N iterations.
3203 args from next arg (a list of lists).
3206 args from the rest of arguments.
3209 args from the rest args (lists).
3220 aborts if N <= M <= K
3222 *** Not Implemented CL Format Control Directives
3225 print `#f' as an empty list (see below).
3228 print `#f' as an empty list (see below).
3234 (sorry I don't understand its semantics completely)
3236 *** Extended, Replaced and Additional Control Directives
3238 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3239 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3240 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3241 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3242 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3243 COMMAWIDTH is the number of characters between two comma
3247 print a R4RS complex number as `~F~@Fi' with passed parameters for
3251 Pretty print formatting of an argument for scheme code lists.
3257 Flushes the output if format DESTINATION is a port.
3260 Print a `#\space' character
3262 print N `#\space' characters.
3265 Print a `#\tab' character
3267 print N `#\tab' characters.
3270 Takes N as an integer representation for a character. No arguments
3271 are consumed. N is converted to a character by `integer->char'. N
3272 must be a positive decimal number.
3275 Print out readproof. Prints out internal objects represented as
3276 `#<...>' as strings `"#<...>"' so that the format output can always
3277 be processed by `read'.
3280 Print out readproof. Prints out internal objects represented as
3281 `#<...>' as strings `"#<...>"' so that the format output can always
3282 be processed by `read'.
3285 Prints information and a copyright notice on the format
3288 prints format version.
3291 may also print number strings, i.e. passing a number as a string
3292 and format it accordingly.
3294 *** Configuration Variables
3296 The format module exports some configuration variables to suit the
3297 systems and users needs. There should be no modification necessary for
3298 the configuration that comes with Guile. Format detects automatically
3299 if the running scheme system implements floating point numbers and
3302 format:symbol-case-conv
3303 Symbols are converted by `symbol->string' so the case type of the
3304 printed symbols is implementation dependent.
3305 `format:symbol-case-conv' is a one arg closure which is either
3306 `#f' (no conversion), `string-upcase', `string-downcase' or
3307 `string-capitalize'. (default `#f')
3309 format:iobj-case-conv
3310 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3311 implementation internal objects. (default `#f')
3314 The character prefixing the exponent value in `~E' printing.
3317 *** Compatibility With Other Format Implementations
3323 Downward compatible except for padding support and `~A', `~S',
3324 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
3325 `printf' padding support which is completely replaced by the CL
3326 `format' padding style.
3329 Downward compatible except for `~', which is not documented
3330 (ignores all characters inside the format string up to a newline
3331 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
3332 numerical and variable parameters and `:/@' modifiers in the CL
3336 Downward compatible except for `~A' and `~S' which print in
3337 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
3338 directive parameters or modifiers)).
3341 Downward compatible except for an optional destination parameter:
3342 S2C accepts a format call without a destination which returns a
3343 formatted string. This is equivalent to a #f destination in S2C.
3344 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
3345 parameters or modifiers)).
3348 ** Changes to string-handling functions.
3350 These functions were added to support the (ice-9 format) module, above.
3352 *** New function: string-upcase STRING
3353 *** New function: string-downcase STRING
3355 These are non-destructive versions of the existing string-upcase! and
3356 string-downcase! functions.
3358 *** New function: string-capitalize! STRING
3359 *** New function: string-capitalize STRING
3361 These functions convert the first letter of each word in the string to
3364 (string-capitalize "howdy there")
3367 As with the other functions, string-capitalize! modifies the string in
3368 place, while string-capitalize returns a modified copy of its argument.
3370 *** New function: string-ci->symbol STRING
3372 Return a symbol whose name is STRING, but having the same case as if
3373 the symbol had be read by `read'.
3375 Guile can be configured to be sensitive or insensitive to case
3376 differences in Scheme identifiers. If Guile is case-insensitive, all
3377 symbols are converted to lower case on input. The `string-ci->symbol'
3378 function returns a symbol whose name in STRING, transformed as Guile
3379 would if STRING were input.
3381 *** New function: substring-move! STRING1 START END STRING2 START
3383 Copy the substring of STRING1 from START (inclusive) to END
3384 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
3385 string, and the source and destination areas may overlap; in all
3386 cases, the function behaves as if all the characters were copied
3389 *** Extended functions: substring-move-left! substring-move-right!
3391 These functions now correctly copy arbitrarily overlapping substrings;
3392 they are both synonyms for substring-move!.
3395 ** New module (ice-9 getopt-long), with the function `getopt-long'.
3397 getopt-long is a function for parsing command-line arguments in a
3398 manner consistent with other GNU programs.
3400 (getopt-long ARGS GRAMMAR)
3401 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
3403 ARGS should be a list of strings. Its first element should be the
3404 name of the program; subsequent elements should be the arguments
3405 that were passed to the program on the command line. The
3406 `program-arguments' procedure returns a list of this form.
3408 GRAMMAR is a list of the form:
3409 ((OPTION (PROPERTY VALUE) ...) ...)
3411 Each OPTION should be a symbol. `getopt-long' will accept a
3412 command-line option named `--OPTION'.
3413 Each option can have the following (PROPERTY VALUE) pairs:
3415 (single-char CHAR) --- Accept `-CHAR' as a single-character
3416 equivalent to `--OPTION'. This is how to specify traditional
3418 (required? BOOL) --- If BOOL is true, the option is required.
3419 getopt-long will raise an error if it is not found in ARGS.
3420 (value BOOL) --- If BOOL is #t, the option accepts a value; if
3421 it is #f, it does not; and if it is the symbol
3422 `optional', the option may appear in ARGS with or
3424 (predicate FUNC) --- If the option accepts a value (i.e. you
3425 specified `(value #t)' for this option), then getopt
3426 will apply FUNC to the value, and throw an exception
3427 if it returns #f. FUNC should be a procedure which
3428 accepts a string and returns a boolean value; you may
3429 need to use quasiquotes to get it into GRAMMAR.
3431 The (PROPERTY VALUE) pairs may occur in any order, but each
3432 property may occur only once. By default, options do not have
3433 single-character equivalents, are not required, and do not take
3436 In ARGS, single-character options may be combined, in the usual
3437 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
3438 accepts values, then it must be the last option in the
3439 combination; the value is the next argument. So, for example, using
3440 the following grammar:
3441 ((apples (single-char #\a))
3442 (blimps (single-char #\b) (value #t))
3443 (catalexis (single-char #\c) (value #t)))
3444 the following argument lists would be acceptable:
3445 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
3446 for "blimps" and "catalexis")
3447 ("-ab" "bang" "-c" "couth") (same)
3448 ("-ac" "couth" "-b" "bang") (same)
3449 ("-abc" "couth" "bang") (an error, since `-b' is not the
3450 last option in its combination)
3452 If an option's value is optional, then `getopt-long' decides
3453 whether it has a value by looking at what follows it in ARGS. If
3454 the next element is a string, and it does not appear to be an
3455 option itself, then that string is the option's value.
3457 The value of a long option can appear as the next element in ARGS,
3458 or it can follow the option name, separated by an `=' character.
3459 Thus, using the same grammar as above, the following argument lists
3461 ("--apples" "Braeburn" "--blimps" "Goodyear")
3462 ("--apples=Braeburn" "--blimps" "Goodyear")
3463 ("--blimps" "Goodyear" "--apples=Braeburn")
3465 If the option "--" appears in ARGS, argument parsing stops there;
3466 subsequent arguments are returned as ordinary arguments, even if
3467 they resemble options. So, in the argument list:
3468 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
3469 `getopt-long' will recognize the `apples' option as having the
3470 value "Granny Smith", but it will not recognize the `blimp'
3471 option; it will return the strings "--blimp" and "Goodyear" as
3472 ordinary argument strings.
3474 The `getopt-long' function returns the parsed argument list as an
3475 assocation list, mapping option names --- the symbols from GRAMMAR
3476 --- onto their values, or #t if the option does not accept a value.
3477 Unused options do not appear in the alist.
3479 All arguments that are not the value of any option are returned
3480 as a list, associated with the empty list.
3482 `getopt-long' throws an exception if:
3483 - it finds an unrecognized option in ARGS
3484 - a required option is omitted
3485 - an option that requires an argument doesn't get one
3486 - an option that doesn't accept an argument does get one (this can
3487 only happen using the long option `--opt=value' syntax)
3488 - an option predicate fails
3493 `((lockfile-dir (required? #t)
3496 (predicate ,file-is-directory?))
3497 (verbose (required? #f)
3500 (x-includes (single-char #\x))
3501 (rnet-server (single-char #\y)
3502 (predicate ,string?))))
3504 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
3505 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
3507 => ((() "foo1" "-fred" "foo2" "foo3")
3508 (rnet-server . "lamprod")
3509 (x-includes . "/usr/include")
3510 (lockfile-dir . "/tmp")
3513 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
3515 It will be removed in a few releases.
3517 ** New syntax: lambda*
3518 ** New syntax: define*
3519 ** New syntax: define*-public
3520 ** New syntax: defmacro*
3521 ** New syntax: defmacro*-public
3522 Guile now supports optional arguments.
3524 `lambda*', `define*', `define*-public', `defmacro*' and
3525 `defmacro*-public' are identical to the non-* versions except that
3526 they use an extended type of parameter list that has the following BNF
3527 syntax (parentheses are literal, square brackets indicate grouping,
3528 and `*', `+' and `?' have the usual meaning):
3530 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
3531 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
3532 [[#&rest identifier]|[. identifier]]? ) | [identifier]
3534 ext-var-decl ::= identifier | ( identifier expression )
3536 The semantics are best illustrated with the following documentation
3537 and examples for `lambda*':
3540 lambda extended for optional and keyword arguments
3542 lambda* creates a procedure that takes optional arguments. These
3543 are specified by putting them inside brackets at the end of the
3544 paramater list, but before any dotted rest argument. For example,
3545 (lambda* (a b #&optional c d . e) '())
3546 creates a procedure with fixed arguments a and b, optional arguments c
3547 and d, and rest argument e. If the optional arguments are omitted
3548 in a call, the variables for them are unbound in the procedure. This
3549 can be checked with the bound? macro.
3551 lambda* can also take keyword arguments. For example, a procedure
3553 (lambda* (#&key xyzzy larch) '())
3554 can be called with any of the argument lists (#:xyzzy 11)
3555 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
3556 are given as keywords are bound to values.
3558 Optional and keyword arguments can also be given default values
3559 which they take on when they are not present in a call, by giving a
3560 two-item list in place of an optional argument, for example in:
3561 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
3562 foo is a fixed argument, bar is an optional argument with default
3563 value 42, and baz is a keyword argument with default value 73.
3564 Default value expressions are not evaluated unless they are needed
3565 and until the procedure is called.
3567 lambda* now supports two more special parameter list keywords.
3569 lambda*-defined procedures now throw an error by default if a
3570 keyword other than one of those specified is found in the actual
3571 passed arguments. However, specifying #&allow-other-keys
3572 immediately after the kyword argument declarations restores the
3573 previous behavior of ignoring unknown keywords. lambda* also now
3574 guarantees that if the same keyword is passed more than once, the
3575 last one passed is the one that takes effect. For example,
3576 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
3577 #:heads 37 #:tails 42 #:heads 99)
3578 would result in (99 47) being displayed.
3580 #&rest is also now provided as a synonym for the dotted syntax rest
3581 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
3582 all respects to lambda*. This is provided for more similarity to DSSSL,
3583 MIT-Scheme and Kawa among others, as well as for refugees from other
3586 Further documentation may be found in the optargs.scm file itself.
3588 The optional argument module also exports the macros `let-optional',
3589 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
3590 are not documented here because they may be removed in the future, but
3591 full documentation is still available in optargs.scm.
3593 ** New syntax: and-let*
3594 Guile now supports the `and-let*' form, described in the draft SRFI-2.
3596 Syntax: (land* (<clause> ...) <body> ...)
3597 Each <clause> should have one of the following forms:
3598 (<variable> <expression>)
3601 Each <variable> or <bound-variable> should be an identifier. Each
3602 <expression> should be a valid expression. The <body> should be a
3603 possibly empty sequence of expressions, like the <body> of a
3606 Semantics: A LAND* expression is evaluated by evaluating the
3607 <expression> or <bound-variable> of each of the <clause>s from
3608 left to right. The value of the first <expression> or
3609 <bound-variable> that evaluates to a false value is returned; the
3610 remaining <expression>s and <bound-variable>s are not evaluated.
3611 The <body> forms are evaluated iff all the <expression>s and
3612 <bound-variable>s evaluate to true values.
3614 The <expression>s and the <body> are evaluated in an environment
3615 binding each <variable> of the preceding (<variable> <expression>)
3616 clauses to the value of the <expression>. Later bindings
3617 shadow earlier bindings.
3619 Guile's and-let* macro was contributed by Michael Livshin.
3621 ** New sorting functions
3623 *** New function: sorted? SEQUENCE LESS?
3624 Returns `#t' when the sequence argument is in non-decreasing order
3625 according to LESS? (that is, there is no adjacent pair `... x y
3626 ...' for which `(less? y x)').
3628 Returns `#f' when the sequence contains at least one out-of-order
3629 pair. It is an error if the sequence is neither a list nor a
3632 *** New function: merge LIST1 LIST2 LESS?
3633 LIST1 and LIST2 are sorted lists.
3634 Returns the sorted list of all elements in LIST1 and LIST2.
3636 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
3637 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
3638 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
3639 (Here "<" should read "comes before".)
3641 *** New procedure: merge! LIST1 LIST2 LESS?
3642 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
3643 the result. If the code is compiled, and LESS? constructs no new
3644 pairs, no pairs at all will be allocated. The first pair of the
3645 result will be either the first pair of LIST1 or the first pair of
3648 *** New function: sort SEQUENCE LESS?
3649 Accepts either a list or a vector, and returns a new sequence
3650 which is sorted. The new sequence is the same type as the input.
3651 Always `(sorted? (sort sequence less?) less?)'. The original
3652 sequence is not altered in any way. The new sequence shares its
3653 elements with the old one; no elements are copied.
3655 *** New procedure: sort! SEQUENCE LESS
3656 Returns its sorted result in the original boxes. No new storage is
3657 allocated at all. Proper usage: (set! slist (sort! slist <))
3659 *** New function: stable-sort SEQUENCE LESS?
3660 Similar to `sort' but stable. That is, if "equal" elements are
3661 ordered a < b in the original sequence, they will have the same order
3664 *** New function: stable-sort! SEQUENCE LESS?
3665 Similar to `sort!' but stable.
3666 Uses temporary storage when sorting vectors.
3668 *** New functions: sort-list, sort-list!
3669 Added for compatibility with scsh.
3671 ** New built-in random number support
3673 *** New function: random N [STATE]
3674 Accepts a positive integer or real N and returns a number of the
3675 same type between zero (inclusive) and N (exclusive). The values
3676 returned have a uniform distribution.
3678 The optional argument STATE must be of the type produced by
3679 `copy-random-state' or `seed->random-state'. It defaults to the value
3680 of the variable `*random-state*'. This object is used to maintain the
3681 state of the pseudo-random-number generator and is altered as a side
3682 effect of the `random' operation.
3684 *** New variable: *random-state*
3685 Holds a data structure that encodes the internal state of the
3686 random-number generator that `random' uses by default. The nature
3687 of this data structure is implementation-dependent. It may be
3688 printed out and successfully read back in, but may or may not
3689 function correctly as a random-number state object in another
3692 *** New function: copy-random-state [STATE]
3693 Returns a new object of type suitable for use as the value of the
3694 variable `*random-state*' and as a second argument to `random'.
3695 If argument STATE is given, a copy of it is returned. Otherwise a
3696 copy of `*random-state*' is returned.
3698 *** New function: seed->random-state SEED
3699 Returns a new object of type suitable for use as the value of the
3700 variable `*random-state*' and as a second argument to `random'.
3701 SEED is a string or a number. A new state is generated and
3702 initialized using SEED.
3704 *** New function: random:uniform [STATE]
3705 Returns an uniformly distributed inexact real random number in the
3706 range between 0 and 1.
3708 *** New procedure: random:solid-sphere! VECT [STATE]
3709 Fills VECT with inexact real random numbers the sum of whose
3710 squares is less than 1.0. Thinking of VECT as coordinates in
3711 space of dimension N = `(vector-length VECT)', the coordinates are
3712 uniformly distributed within the unit N-shere. The sum of the
3713 squares of the numbers is returned. VECT can be either a vector
3714 or a uniform vector of doubles.
3716 *** New procedure: random:hollow-sphere! VECT [STATE]
3717 Fills VECT with inexact real random numbers the sum of whose squares
3718 is equal to 1.0. Thinking of VECT as coordinates in space of
3719 dimension n = `(vector-length VECT)', the coordinates are uniformly
3720 distributed over the surface of the unit n-shere. VECT can be either
3721 a vector or a uniform vector of doubles.
3723 *** New function: random:normal [STATE]
3724 Returns an inexact real in a normal distribution with mean 0 and
3725 standard deviation 1. For a normal distribution with mean M and
3726 standard deviation D use `(+ M (* D (random:normal)))'.
3728 *** New procedure: random:normal-vector! VECT [STATE]
3729 Fills VECT with inexact real random numbers which are independent and
3730 standard normally distributed (i.e., with mean 0 and variance 1).
3731 VECT can be either a vector or a uniform vector of doubles.
3733 *** New function: random:exp STATE
3734 Returns an inexact real in an exponential distribution with mean 1.
3735 For an exponential distribution with mean U use (* U (random:exp)).
3737 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
3739 These functions now operate on numbers in the range of a C unsigned
3742 These functions used to operate on numbers in the range of a C signed
3743 long; however, this seems inappropriate, because Guile integers don't
3746 ** New function: make-guardian
3747 This is an implementation of guardians as described in
3748 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
3749 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
3750 Programming Language Design and Implementation, June 1993
3751 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
3753 ** New functions: delq1!, delv1!, delete1!
3754 These procedures behave similar to delq! and friends but delete only
3755 one object if at all.
3757 ** New function: unread-string STRING PORT
3758 Unread STRING to PORT, that is, push it back onto the port so that
3759 next read operation will work on the pushed back characters.
3761 ** unread-char can now be called multiple times
3762 If unread-char is called multiple times, the unread characters will be
3763 read again in last-in first-out order.
3765 ** the procedures uniform-array-read! and uniform-array-write! now
3766 work on any kind of port, not just ports which are open on a file.
3768 ** Now 'l' in a port mode requests line buffering.
3770 ** The procedure truncate-file now works on string ports as well
3771 as file ports. If the size argument is omitted, the current
3772 file position is used.
3774 ** new procedure: seek PORT/FDES OFFSET WHENCE
3775 The arguments are the same as for the old fseek procedure, but it
3776 works on string ports as well as random-access file ports.
3778 ** the fseek procedure now works on string ports, since it has been
3779 redefined using seek.
3781 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
3782 size is not supplied.
3784 ** the newline procedure no longer flushes the port if it's not
3785 line-buffered: previously it did if it was the current output port.
3787 ** open-pipe and close-pipe are no longer primitive procedures, but
3788 an emulation can be obtained using `(use-modules (ice-9 popen))'.
3790 ** the freopen procedure has been removed.
3792 ** new procedure: drain-input PORT
3793 Drains PORT's read buffers (including any pushed-back characters)
3794 and returns the contents as a single string.
3796 ** New function: map-in-order PROC LIST1 LIST2 ...
3797 Version of `map' which guarantees that the procedure is applied to the
3798 lists in serial order.
3800 ** Renamed `serial-array-copy!' and `serial-array-map!' to
3801 `array-copy-in-order!' and `array-map-in-order!'. The old names are
3802 now obsolete and will go away in release 1.5.
3804 ** New syntax: collect BODY1 ...
3805 Version of `begin' which returns a list of the results of the body
3806 forms instead of the result of the last body form. In contrast to
3807 `begin', `collect' allows an empty body.
3809 ** New functions: read-history FILENAME, write-history FILENAME
3810 Read/write command line history from/to file. Returns #t on success
3811 and #f if an error occured.
3813 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
3815 These procedures return a list of definitions available in the specified
3816 argument, a relative module reference. In the case of no argument,
3817 `(current-module)' is now consulted for definitions to return, instead
3818 of simply returning #f, the former behavior.
3820 ** The #/ syntax for lists is no longer supported.
3822 Earlier versions of Scheme accepted this syntax, but printed a
3825 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
3827 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
3830 * Changes to the gh_ interface
3834 Now takes a second argument which is the result array. If this
3835 pointer is NULL, a new array is malloced (the old behaviour).
3837 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
3838 gh_scm2shorts, gh_scm2longs, gh_scm2floats
3842 * Changes to the scm_ interface
3844 ** Function: scm_make_named_hook (char* name, int n_args)
3846 Creates a hook in the same way as make-hook above but also
3847 binds a variable named NAME to it.
3849 This is the typical way of creating a hook from C code.
3851 Currently, the variable is created in the "current" module. This
3852 might change when we get the new module system.
3854 ** The smob interface
3856 The interface for creating smobs has changed. For documentation, see
3857 data-rep.info (made from guile-core/doc/data-rep.texi).
3859 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
3861 >>> This function will be removed in 1.3.4. <<<
3865 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
3866 This function adds a new smob type, named NAME, with instance size
3867 SIZE to the system. The return value is a tag that is used in
3868 creating instances of the type. If SIZE is 0, then no memory will
3869 be allocated when instances of the smob are created, and nothing
3870 will be freed by the default free function.
3872 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
3873 This function sets the smob marking procedure for the smob type
3874 specified by the tag TC. TC is the tag returned by
3875 `scm_make_smob_type'.
3877 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
3878 This function sets the smob freeing procedure for the smob type
3879 specified by the tag TC. TC is the tag returned by
3880 `scm_make_smob_type'.
3882 *** Function: void scm_set_smob_print (tc, print)
3884 - Function: void scm_set_smob_print (long tc,
3885 scm_sizet (*print) (SCM,
3889 This function sets the smob printing procedure for the smob type
3890 specified by the tag TC. TC is the tag returned by
3891 `scm_make_smob_type'.
3893 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
3894 This function sets the smob equality-testing predicate for the
3895 smob type specified by the tag TC. TC is the tag returned by
3896 `scm_make_smob_type'.
3898 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
3899 Make VALUE contain a smob instance of the type with type code TC and
3900 smob data DATA. VALUE must be previously declared as C type `SCM'.
3902 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
3903 This macro expands to a block of code that creates a smob instance
3904 of the type with type code TC and smob data DATA, and returns that
3905 `SCM' value. It should be the last piece of code in a block.
3907 ** The interfaces for using I/O ports and implementing port types
3908 (ptobs) have changed significantly. The new interface is based on
3909 shared access to buffers and a new set of ptob procedures.
3911 *** scm_newptob has been removed
3915 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
3917 - Function: SCM scm_make_port_type (char *type_name,
3918 int (*fill_buffer) (SCM port),
3919 void (*write_flush) (SCM port));
3921 Similarly to the new smob interface, there is a set of function
3922 setters by which the user can customize the behaviour of his port
3923 type. See ports.h (scm_set_port_XXX).
3925 ** scm_strport_to_string: New function: creates a new string from
3926 a string port's buffer.
3928 ** Plug in interface for random number generators
3929 The variable `scm_the_rng' in random.c contains a value and three
3930 function pointers which together define the current random number
3931 generator being used by the Scheme level interface and the random
3932 number library functions.
3934 The user is free to replace the default generator with the generator
3937 *** Variable: size_t scm_the_rng.rstate_size
3938 The size of the random state type used by the current RNG
3941 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
3942 Given the random STATE, return 32 random bits.
3944 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
3945 Seed random state STATE using string S of length N.
3947 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
3948 Given random state STATE, return a malloced copy.
3951 The default RNG is the MWC (Multiply With Carry) random number
3952 generator described by George Marsaglia at the Department of
3953 Statistics and Supercomputer Computations Research Institute, The
3954 Florida State University (http://stat.fsu.edu/~geo).
3956 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
3957 passes all tests in the DIEHARD test suite
3958 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
3959 costs one multiply and one add on platforms which either supports long
3960 longs (gcc does this on most systems) or have 64 bit longs. The cost
3961 is four multiply on other systems but this can be optimized by writing
3962 scm_i_uniform32 in assembler.
3964 These functions are provided through the scm_the_rng interface for use
3965 by libguile and the application.
3967 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
3968 Given the random STATE, return 32 random bits.
3969 Don't use this function directly. Instead go through the plugin
3970 interface (see "Plug in interface" above).
3972 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
3973 Initialize STATE using SEED of length N.
3975 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
3976 Return a malloc:ed copy of STATE. This function can easily be re-used
3977 in the interfaces to other RNGs.
3979 ** Random number library functions
3980 These functions use the current RNG through the scm_the_rng interface.
3981 It might be a good idea to use these functions from your C code so
3982 that only one random generator is used by all code in your program.
3984 The default random state is stored in:
3986 *** Variable: SCM scm_var_random_state
3987 Contains the vcell of the Scheme variable "*random-state*" which is
3988 used as default state by all random number functions in the Scheme
3993 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
3995 *** Function: scm_rstate *scm_c_default_rstate (void)
3996 This is a convenience function which returns the value of
3997 scm_var_random_state. An error message is generated if this value
3998 isn't a random state.
4000 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4001 Make a new random state from the string SEED of length LENGTH.
4003 It is generally not a good idea to use multiple random states in a
4004 program. While subsequent random numbers generated from one random
4005 state are guaranteed to be reasonably independent, there is no such
4006 guarantee for numbers generated from different random states.
4008 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4009 Return 32 random bits.
4011 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4012 Return a sample from the uniform(0,1) distribution.
4014 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4015 Return a sample from the normal(0,1) distribution.
4017 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4018 Return a sample from the exp(1) distribution.
4020 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4021 Return a sample from the discrete uniform(0,M) distribution.
4023 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4024 Return a sample from the discrete uniform(0,M) distribution.
4025 M must be a bignum object. The returned value may be an INUM.
4029 Changes in Guile 1.3 (released Monday, October 19, 1998):
4031 * Changes to the distribution
4033 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4034 To avoid conflicts, programs should name environment variables after
4035 themselves, except when there's a common practice establishing some
4038 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4039 giving the former precedence, and printing a warning message if the
4040 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4042 ** The header files related to multi-byte characters have been removed.
4043 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4044 which referred to these explicitly will probably need to be rewritten,
4045 since the support for the variant string types has been removed; see
4048 ** The header files append.h and sequences.h have been removed. These
4049 files implemented non-R4RS operations which would encourage
4050 non-portable programming style and less easy-to-read code.
4052 * Changes to the stand-alone interpreter
4054 ** New procedures have been added to implement a "batch mode":
4056 *** Function: batch-mode?
4058 Returns a boolean indicating whether the interpreter is in batch
4061 *** Function: set-batch-mode?! ARG
4063 If ARG is true, switches the interpreter to batch mode. The `#f'
4064 case has not been implemented.
4066 ** Guile now provides full command-line editing, when run interactively.
4067 To use this feature, you must have the readline library installed.
4068 The Guile build process will notice it, and automatically include
4071 The readline library is available via anonymous FTP from any GNU
4072 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4074 ** the-last-stack is now a fluid.
4076 * Changes to the procedure for linking libguile with your programs
4078 ** You can now use the `guile-config' utility to build programs that use Guile.
4080 Guile now includes a command-line utility called `guile-config', which
4081 can provide information about how to compile and link programs that
4084 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4085 You should include this command's output on the command line you use
4086 to compile C or C++ code that #includes the Guile header files. It's
4087 usually just a `-I' flag to help the compiler find the Guile headers.
4090 *** `guile-config link' prints any linker flags necessary to link with Guile.
4092 This command writes to its standard output a list of flags which you
4093 must pass to the linker to link your code against the Guile library.
4094 The flags include '-lguile' itself, any other libraries the Guile
4095 library depends upon, and any `-L' flags needed to help the linker
4096 find those libraries.
4098 For example, here is a Makefile rule that builds a program named 'foo'
4099 from the object files ${FOO_OBJECTS}, and links them against Guile:
4102 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4104 Previous Guile releases recommended that you use autoconf to detect
4105 which of a predefined set of libraries were present on your system.
4106 It is more robust to use `guile-config', since it records exactly which
4107 libraries the installed Guile library requires.
4109 This was originally called `build-guile', but was renamed to
4110 `guile-config' before Guile 1.3 was released, to be consistent with
4111 the analogous script for the GTK+ GUI toolkit, which is called
4115 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4117 If you are using the GNU autoconf package to configure your program,
4118 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4119 (described above) and gather the necessary values for use in your
4122 The GUILE_FLAGS macro expands to configure script code which runs the
4123 `guile-config' script, to find out where Guile's header files and
4124 libraries are installed. It sets two variables, marked for
4125 substitution, as by AC_SUBST.
4127 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4128 code that uses Guile header files. This is almost always just a
4131 GUILE_LDFLAGS --- flags to pass to the linker to link a
4132 program against Guile. This includes `-lguile' for the Guile
4133 library itself, any libraries that Guile itself requires (like
4134 -lqthreads), and so on. It may also include a -L flag to tell the
4135 compiler where to find the libraries.
4137 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4138 directory of the Guile distribution. You can copy it into your
4139 package's aclocal.m4 file, and then use it in your configure.in file.
4141 If you are using the `aclocal' program, distributed with GNU automake,
4142 to maintain your aclocal.m4 file, the Guile installation process
4143 installs guile.m4 where aclocal will find it. All you need to do is
4144 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4145 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4149 * Changes to Scheme functions and syntax
4151 ** Multi-byte strings have been removed, as have multi-byte and wide
4152 ports. We felt that these were the wrong approach to
4153 internationalization support.
4155 ** New function: readline [PROMPT]
4156 Read a line from the terminal, and allow the user to edit it,
4157 prompting with PROMPT. READLINE provides a large set of Emacs-like
4158 editing commands, lets the user recall previously typed lines, and
4159 works on almost every kind of terminal, including dumb terminals.
4161 READLINE assumes that the cursor is at the beginning of the line when
4162 it is invoked. Thus, you can't print a prompt yourself, and then call
4163 READLINE; you need to package up your prompt as a string, pass it to
4164 the function, and let READLINE print the prompt itself. This is
4165 because READLINE needs to know the prompt's screen width.
4167 For Guile to provide this function, you must have the readline
4168 library, version 2.1 or later, installed on your system. Readline is
4169 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4170 any GNU mirror site.
4172 See also ADD-HISTORY function.
4174 ** New function: add-history STRING
4175 Add STRING as the most recent line in the history used by the READLINE
4176 command. READLINE does not add lines to the history itself; you must
4177 call ADD-HISTORY to make previous input available to the user.
4179 ** The behavior of the read-line function has changed.
4181 This function now uses standard C library functions to read the line,
4182 for speed. This means that it doesn not respect the value of
4183 scm-line-incrementors; it assumes that lines are delimited with
4186 (Note that this is read-line, the function that reads a line of text
4187 from a port, not readline, the function that reads a line from a
4188 terminal, providing full editing capabilities.)
4190 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4192 This module provides some simple argument parsing. It exports one
4195 Function: getopt-gnu-style ARG-LS
4196 Parse a list of program arguments into an alist of option
4199 Each item in the list of program arguments is examined to see if
4200 it meets the syntax of a GNU long-named option. An argument like
4201 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4202 returned alist, where MUMBLE is a keyword object with the same
4203 name as the argument. An argument like `--MUMBLE=FROB' produces
4204 an element of the form (MUMBLE . FROB), where FROB is a string.
4206 As a special case, the returned alist also contains a pair whose
4207 car is the symbol `rest'. The cdr of this pair is a list
4208 containing all the items in the argument list that are not options
4209 of the form mentioned above.
4211 The argument `--' is treated specially: all items in the argument
4212 list appearing after such an argument are not examined, and are
4213 returned in the special `rest' list.
4215 This function does not parse normal single-character switches.
4216 You will need to parse them out of the `rest' list yourself.
4218 ** The read syntax for byte vectors and short vectors has changed.
4220 Instead of #bytes(...), write #y(...).
4222 Instead of #short(...), write #h(...).
4224 This may seem nutty, but, like the other uniform vectors, byte vectors
4225 and short vectors want to have the same print and read syntax (and,
4226 more basic, want to have read syntax!). Changing the read syntax to
4227 use multiple characters after the hash sign breaks with the
4228 conventions used in R5RS and the conventions used for the other
4229 uniform vectors. It also introduces complexity in the current reader,
4230 both on the C and Scheme levels. (The Right solution is probably to
4231 change the syntax and prototypes for uniform vectors entirely.)
4234 ** The new module (ice-9 session) provides useful interactive functions.
4236 *** New procedure: (apropos REGEXP OPTION ...)
4238 Display a list of top-level variables whose names match REGEXP, and
4239 the modules they are imported from. Each OPTION should be one of the
4242 value --- Show the value of each matching variable.
4243 shadow --- Show bindings shadowed by subsequently imported modules.
4244 full --- Same as both `shadow' and `value'.
4248 guile> (apropos "trace" 'full)
4249 debug: trace #<procedure trace args>
4250 debug: untrace #<procedure untrace args>
4251 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4252 the-scm-module: before-backtrace-hook ()
4253 the-scm-module: backtrace #<primitive-procedure backtrace>
4254 the-scm-module: after-backtrace-hook ()
4255 the-scm-module: has-shown-backtrace-hint? #f
4258 ** There are new functions and syntax for working with macros.
4260 Guile implements macros as a special object type. Any variable whose
4261 top-level binding is a macro object acts as a macro. The macro object
4262 specifies how the expression should be transformed before evaluation.
4264 *** Macro objects now print in a reasonable way, resembling procedures.
4266 *** New function: (macro? OBJ)
4267 True iff OBJ is a macro object.
4269 *** New function: (primitive-macro? OBJ)
4270 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4271 macro transformers, implemented in eval.c rather than Scheme code.
4273 Why do we have this function?
4274 - For symmetry with procedure? and primitive-procedure?,
4275 - to allow custom print procedures to tell whether a macro is
4276 primitive, and display it differently, and
4277 - to allow compilers and user-written evaluators to distinguish
4278 builtin special forms from user-defined ones, which could be
4281 *** New function: (macro-type OBJ)
4282 Return a value indicating what kind of macro OBJ is. Possible return
4285 The symbol `syntax' --- a macro created by procedure->syntax.
4286 The symbol `macro' --- a macro created by procedure->macro.
4287 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4288 The boolean #f --- if OBJ is not a macro object.
4290 *** New function: (macro-name MACRO)
4291 Return the name of the macro object MACRO's procedure, as returned by
4294 *** New function: (macro-transformer MACRO)
4295 Return the transformer procedure for MACRO.
4297 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4299 Specify a new macro expander to use in the current module. Each
4300 MODULE is a module name, with the same meaning as in the `use-modules'
4301 form; each named module's exported bindings are added to the current
4302 top-level environment. TRANSFORMER is an expression evaluated in the
4303 resulting environment which must yield a procedure to use as the
4304 module's eval transformer: every expression evaluated in this module
4305 is passed to this function, and the result passed to the Guile
4308 *** macro-eval! is removed. Use local-eval instead.
4310 ** Some magic has been added to the printer to better handle user
4311 written printing routines (like record printers, closure printers).
4313 The problem is that these user written routines must have access to
4314 the current `print-state' to be able to handle fancy things like
4315 detection of circular references. These print-states have to be
4316 passed to the builtin printing routines (display, write, etc) to
4317 properly continue the print chain.
4319 We didn't want to change all existing print code so that it
4320 explicitly passes thru a print state in addition to a port. Instead,
4321 we extented the possible values that the builtin printing routines
4322 accept as a `port'. In addition to a normal port, they now also take
4323 a pair of a normal port and a print-state. Printing will go to the
4324 port and the print-state will be used to control the detection of
4325 circular references, etc. If the builtin function does not care for a
4326 print-state, it is simply ignored.
4328 User written callbacks are now called with such a pair as their
4329 `port', but because every function now accepts this pair as a PORT
4330 argument, you don't have to worry about that. In fact, it is probably
4331 safest to not check for these pairs.
4333 However, it is sometimes necessary to continue a print chain on a
4334 different port, for example to get a intermediate string
4335 representation of the printed value, mangle that string somehow, and
4336 then to finally print the mangled string. Use the new function
4338 inherit-print-state OLD-PORT NEW-PORT
4340 for this. It constructs a new `port' that prints to NEW-PORT but
4341 inherits the print-state of OLD-PORT.
4343 ** struct-vtable-offset renamed to vtable-offset-user
4345 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
4347 ** There is now a third optional argument to make-vtable-vtable
4348 (and fourth to make-struct) when constructing new types (vtables).
4349 This argument initializes field vtable-index-printer of the vtable.
4351 ** The detection of circular references has been extended to structs.
4352 That is, a structure that -- in the process of being printed -- prints
4353 itself does not lead to infinite recursion.
4355 ** There is now some basic support for fluids. Please read
4356 "libguile/fluid.h" to find out more. It is accessible from Scheme with
4357 the following functions and macros:
4359 Function: make-fluid
4361 Create a new fluid object. Fluids are not special variables or
4362 some other extension to the semantics of Scheme, but rather
4363 ordinary Scheme objects. You can store them into variables (that
4364 are still lexically scoped, of course) or into any other place you
4365 like. Every fluid has a initial value of `#f'.
4367 Function: fluid? OBJ
4369 Test whether OBJ is a fluid.
4371 Function: fluid-ref FLUID
4372 Function: fluid-set! FLUID VAL
4374 Access/modify the fluid FLUID. Modifications are only visible
4375 within the current dynamic root (that includes threads).
4377 Function: with-fluids* FLUIDS VALUES THUNK
4379 FLUIDS is a list of fluids and VALUES a corresponding list of
4380 values for these fluids. Before THUNK gets called the values are
4381 installed in the fluids and the old values of the fluids are
4382 saved in the VALUES list. When the flow of control leaves THUNK
4383 or reenters it, the values get swapped again. You might think of
4384 this as a `safe-fluid-excursion'. Note that the VALUES list is
4385 modified by `with-fluids*'.
4387 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
4389 The same as `with-fluids*' but with a different syntax. It looks
4390 just like `let', but both FLUID and VALUE are evaluated. Remember,
4391 fluids are not special variables but ordinary objects. FLUID
4392 should evaluate to a fluid.
4394 ** Changes to system call interfaces:
4396 *** close-port, close-input-port and close-output-port now return a
4397 boolean instead of an `unspecified' object. #t means that the port
4398 was successfully closed, while #f means it was already closed. It is
4399 also now possible for these procedures to raise an exception if an
4400 error occurs (some errors from write can be delayed until close.)
4402 *** the first argument to chmod, fcntl, ftell and fseek can now be a
4405 *** the third argument to fcntl is now optional.
4407 *** the first argument to chown can now be a file descriptor or a port.
4409 *** the argument to stat can now be a port.
4411 *** The following new procedures have been added (most use scsh
4414 *** procedure: close PORT/FD
4415 Similar to close-port (*note close-port: Closing Ports.), but also
4416 works on file descriptors. A side effect of closing a file
4417 descriptor is that any ports using that file descriptor are moved
4418 to a different file descriptor and have their revealed counts set
4421 *** procedure: port->fdes PORT
4422 Returns the integer file descriptor underlying PORT. As a side
4423 effect the revealed count of PORT is incremented.
4425 *** procedure: fdes->ports FDES
4426 Returns a list of existing ports which have FDES as an underlying
4427 file descriptor, without changing their revealed counts.
4429 *** procedure: fdes->inport FDES
4430 Returns an existing input port which has FDES as its underlying
4431 file descriptor, if one exists, and increments its revealed count.
4432 Otherwise, returns a new input port with a revealed count of 1.
4434 *** procedure: fdes->outport FDES
4435 Returns an existing output port which has FDES as its underlying
4436 file descriptor, if one exists, and increments its revealed count.
4437 Otherwise, returns a new output port with a revealed count of 1.
4439 The next group of procedures perform a `dup2' system call, if NEWFD
4440 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
4441 duplicated can be supplied as an integer or contained in a port. The
4442 type of value returned varies depending on which procedure is used.
4444 All procedures also have the side effect when performing `dup2' that
4445 any ports using NEWFD are moved to a different file descriptor and have
4446 their revealed counts set to zero.
4448 *** procedure: dup->fdes PORT/FD [NEWFD]
4449 Returns an integer file descriptor.
4451 *** procedure: dup->inport PORT/FD [NEWFD]
4452 Returns a new input port using the new file descriptor.
4454 *** procedure: dup->outport PORT/FD [NEWFD]
4455 Returns a new output port using the new file descriptor.
4457 *** procedure: dup PORT/FD [NEWFD]
4458 Returns a new port if PORT/FD is a port, with the same mode as the
4459 supplied port, otherwise returns an integer file descriptor.
4461 *** procedure: dup->port PORT/FD MODE [NEWFD]
4462 Returns a new port using the new file descriptor. MODE supplies a
4463 mode string for the port (*note open-file: File Ports.).
4465 *** procedure: setenv NAME VALUE
4466 Modifies the environment of the current process, which is also the
4467 default environment inherited by child processes.
4469 If VALUE is `#f', then NAME is removed from the environment.
4470 Otherwise, the string NAME=VALUE is added to the environment,
4471 replacing any existing string with name matching NAME.
4473 The return value is unspecified.
4475 *** procedure: truncate-file OBJ SIZE
4476 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
4477 can be a string containing a file name or an integer file
4478 descriptor or port open for output on the file. The underlying
4479 system calls are `truncate' and `ftruncate'.
4481 The return value is unspecified.
4483 *** procedure: setvbuf PORT MODE [SIZE]
4484 Set the buffering mode for PORT. MODE can be:
4492 block buffered, using a newly allocated buffer of SIZE bytes.
4493 However if SIZE is zero or unspecified, the port will be made
4496 This procedure should not be used after I/O has been performed with
4499 Ports are usually block buffered by default, with a default buffer
4500 size. Procedures e.g., *Note open-file: File Ports, which accept a
4501 mode string allow `0' to be added to request an unbuffered port.
4503 *** procedure: fsync PORT/FD
4504 Copies any unwritten data for the specified output file descriptor
4505 to disk. If PORT/FD is a port, its buffer is flushed before the
4506 underlying file descriptor is fsync'd. The return value is
4509 *** procedure: open-fdes PATH FLAGS [MODES]
4510 Similar to `open' but returns a file descriptor instead of a port.
4512 *** procedure: execle PATH ENV [ARG] ...
4513 Similar to `execl', but the environment of the new process is
4514 specified by ENV, which must be a list of strings as returned by
4515 the `environ' procedure.
4517 This procedure is currently implemented using the `execve' system
4518 call, but we call it `execle' because of its Scheme calling
4521 *** procedure: strerror ERRNO
4522 Returns the Unix error message corresponding to ERRNO, an integer.
4524 *** procedure: primitive-exit [STATUS]
4525 Terminate the current process without unwinding the Scheme stack.
4526 This is would typically be useful after a fork. The exit status
4527 is STATUS if supplied, otherwise zero.
4529 *** procedure: times
4530 Returns an object with information about real and processor time.
4531 The following procedures accept such an object as an argument and
4532 return a selected component:
4535 The current real time, expressed as time units relative to an
4539 The CPU time units used by the calling process.
4542 The CPU time units used by the system on behalf of the
4546 The CPU time units used by terminated child processes of the
4547 calling process, whose status has been collected (e.g., using
4551 Similarly, the CPU times units used by the system on behalf of
4552 terminated child processes.
4554 ** Removed: list-length
4555 ** Removed: list-append, list-append!
4556 ** Removed: list-reverse, list-reverse!
4558 ** array-map renamed to array-map!
4560 ** serial-array-map renamed to serial-array-map!
4562 ** catch doesn't take #f as first argument any longer
4564 Previously, it was possible to pass #f instead of a key to `catch'.
4565 That would cause `catch' to pass a jump buffer object to the procedure
4566 passed as second argument. The procedure could then use this jump
4567 buffer objekt as an argument to throw.
4569 This mechanism has been removed since its utility doesn't motivate the
4570 extra complexity it introduces.
4572 ** The `#/' notation for lists now provokes a warning message from Guile.
4573 This syntax will be removed from Guile in the near future.
4575 To disable the warning message, set the GUILE_HUSH environment
4576 variable to any non-empty value.
4578 ** The newline character now prints as `#\newline', following the
4579 normal Scheme notation, not `#\nl'.
4581 * Changes to the gh_ interface
4583 ** The gh_enter function now takes care of loading the Guile startup files.
4584 gh_enter works by calling scm_boot_guile; see the remarks below.
4586 ** Function: void gh_write (SCM x)
4588 Write the printed representation of the scheme object x to the current
4589 output port. Corresponds to the scheme level `write'.
4591 ** gh_list_length renamed to gh_length.
4593 ** vector handling routines
4595 Several major changes. In particular, gh_vector() now resembles
4596 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
4597 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
4598 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
4599 vector-related gh_ functions have been implemented.
4601 ** pair and list routines
4603 Implemented several of the R4RS pair and list functions that were
4606 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
4608 New function. Converts double arrays back and forth between Scheme
4611 * Changes to the scm_ interface
4613 ** The function scm_boot_guile now takes care of loading the startup files.
4615 Guile's primary initialization function, scm_boot_guile, now takes
4616 care of loading `boot-9.scm', in the `ice-9' module, to initialize
4617 Guile, define the module system, and put together some standard
4618 bindings. It also loads `init.scm', which is intended to hold
4619 site-specific initialization code.
4621 Since Guile cannot operate properly until boot-9.scm is loaded, there
4622 is no reason to separate loading boot-9.scm from Guile's other
4623 initialization processes.
4625 This job used to be done by scm_compile_shell_switches, which didn't
4626 make much sense; in particular, it meant that people using Guile for
4627 non-shell-like applications had to jump through hoops to get Guile
4628 initialized properly.
4630 ** The function scm_compile_shell_switches no longer loads the startup files.
4631 Now, Guile always loads the startup files, whenever it is initialized;
4632 see the notes above for scm_boot_guile and scm_load_startup_files.
4634 ** Function: scm_load_startup_files
4635 This new function takes care of loading Guile's initialization file
4636 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
4637 this is always called by the Guile initialization process, it's
4638 probably not too useful to call this yourself, but it's there anyway.
4640 ** The semantics of smob marking have changed slightly.
4642 The smob marking function (the `mark' member of the scm_smobfuns
4643 structure) is no longer responsible for setting the mark bit on the
4644 smob. The generic smob handling code in the garbage collector will
4645 set this bit. The mark function need only ensure that any other
4646 objects the smob refers to get marked.
4648 Note that this change means that the smob's GC8MARK bit is typically
4649 already set upon entry to the mark function. Thus, marking functions
4650 which look like this:
4653 if (SCM_GC8MARKP (ptr))
4655 SCM_SETGC8MARK (ptr);
4656 ... mark objects to which the smob refers ...
4659 are now incorrect, since they will return early, and fail to mark any
4660 other objects the smob refers to. Some code in the Guile library used
4663 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
4665 If you have implemented your own I/O port type, by writing the
4666 functions required by the scm_ptobfuns and then calling scm_newptob,
4667 you will need to change your functions slightly.
4669 The functions in a scm_ptobfuns structure now expect the port itself
4670 as their argument; they used to expect the `stream' member of the
4671 port's scm_port_table structure. This allows functions in an
4672 scm_ptobfuns structure to easily access the port's cell (and any flags
4673 it its CAR), and the port's scm_port_table structure.
4675 Guile now passes the I/O port itself as the `port' argument in the
4676 following scm_ptobfuns functions:
4678 int (*free) (SCM port);
4679 int (*fputc) (int, SCM port);
4680 int (*fputs) (char *, SCM port);
4681 scm_sizet (*fwrite) SCM_P ((char *ptr,
4685 int (*fflush) (SCM port);
4686 int (*fgetc) (SCM port);
4687 int (*fclose) (SCM port);
4689 The interfaces to the `mark', `print', `equalp', and `fgets' methods
4692 If you have existing code which defines its own port types, it is easy
4693 to convert your code to the new interface; simply apply SCM_STREAM to
4694 the port argument to yield the value you code used to expect.
4696 Note that since both the port and the stream have the same type in the
4697 C code --- they are both SCM values --- the C compiler will not remind
4698 you if you forget to update your scm_ptobfuns functions.
4701 ** Function: int scm_internal_select (int fds,
4705 struct timeval *timeout);
4707 This is a replacement for the `select' function provided by the OS.
4708 It enables I/O blocking and sleeping to happen for one cooperative
4709 thread without blocking other threads. It also avoids busy-loops in
4710 these situations. It is intended that all I/O blocking and sleeping
4711 will finally go through this function. Currently, this function is
4712 only available on systems providing `gettimeofday' and `select'.
4714 ** Function: SCM scm_internal_stack_catch (SCM tag,
4715 scm_catch_body_t body,
4717 scm_catch_handler_t handler,
4720 A new sibling to the other two C level `catch' functions
4721 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
4722 the stack to be saved automatically into the variable `the-last-stack'
4723 (scm_the_last_stack_var) on error. This is necessary if you want to
4724 use advanced error reporting, such as calling scm_display_error and
4725 scm_display_backtrace. (They both take a stack object as argument.)
4727 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
4729 scm_catch_handler_t handler,
4732 Spawns a new thread. It does a job similar to
4733 scm_call_with_new_thread but takes arguments more suitable when
4734 spawning threads from application C code.
4736 ** The hook scm_error_callback has been removed. It was originally
4737 intended as a way for the user to install his own error handler. But
4738 that method works badly since it intervenes between throw and catch,
4739 thereby changing the semantics of expressions like (catch #t ...).
4740 The correct way to do it is to use one of the C level catch functions
4741 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
4743 ** Removed functions:
4745 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
4746 scm_list_reverse, scm_list_reverse_x
4748 ** New macros: SCM_LISTn where n is one of the integers 0-9.
4750 These can be used for pretty list creation from C. The idea is taken
4751 from Erick Gallesio's STk.
4753 ** scm_array_map renamed to scm_array_map_x
4755 ** mbstrings are now removed
4757 This means that the type codes scm_tc7_mb_string and
4758 scm_tc7_mb_substring has been removed.
4760 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
4762 Since we no longer support multi-byte strings, these I/O functions
4763 have been simplified, and renamed. Here are their old names, and
4764 their new names and arguments:
4766 scm_gen_putc -> void scm_putc (int c, SCM port);
4767 scm_gen_puts -> void scm_puts (char *s, SCM port);
4768 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
4769 scm_gen_getc -> void scm_getc (SCM port);
4772 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
4774 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
4776 SCM_TYP7S now masks away the bit which distinguishes substrings from
4779 ** scm_catch_body_t: Backward incompatible change!
4781 Body functions to scm_internal_catch and friends do not any longer
4782 take a second argument. This is because it is no longer possible to
4783 pass a #f arg to catch.
4785 ** Calls to scm_protect_object and scm_unprotect now nest properly.
4787 The function scm_protect_object protects its argument from being freed
4788 by the garbage collector. scm_unprotect_object removes that
4791 These functions now nest properly. That is, for every object O, there
4792 is a counter which scm_protect_object(O) increments and
4793 scm_unprotect_object(O) decrements, if the counter is greater than
4794 zero. Every object's counter is zero when it is first created. If an
4795 object's counter is greater than zero, the garbage collector will not
4796 reclaim its storage.
4798 This allows you to use scm_protect_object in your code without
4799 worrying that some other function you call will call
4800 scm_unprotect_object, and allow it to be freed. Assuming that the
4801 functions you call are well-behaved, and unprotect only those objects
4802 they protect, you can follow the same rule and have confidence that
4803 objects will be freed only at appropriate times.
4806 Changes in Guile 1.2 (released Tuesday, June 24 1997):
4808 * Changes to the distribution
4810 ** Nightly snapshots are now available from ftp.red-bean.com.
4811 The old server, ftp.cyclic.com, has been relinquished to its rightful
4814 Nightly snapshots of the Guile development sources are now available via
4815 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
4817 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
4818 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
4820 ** To run Guile without installing it, the procedure has changed a bit.
4822 If you used a separate build directory to compile Guile, you'll need
4823 to include the build directory in SCHEME_LOAD_PATH, as well as the
4824 source directory. See the `INSTALL' file for examples.
4826 * Changes to the procedure for linking libguile with your programs
4828 ** The standard Guile load path for Scheme code now includes
4829 $(datadir)/guile (usually /usr/local/share/guile). This means that
4830 you can install your own Scheme files there, and Guile will find them.
4831 (Previous versions of Guile only checked a directory whose name
4832 contained the Guile version number, so you had to re-install or move
4833 your Scheme sources each time you installed a fresh version of Guile.)
4835 The load path also includes $(datadir)/guile/site; we recommend
4836 putting individual Scheme files there. If you want to install a
4837 package with multiple source files, create a directory for them under
4840 ** Guile 1.2 will now use the Rx regular expression library, if it is
4841 installed on your system. When you are linking libguile into your own
4842 programs, this means you will have to link against -lguile, -lqt (if
4843 you configured Guile with thread support), and -lrx.
4845 If you are using autoconf to generate configuration scripts for your
4846 application, the following lines should suffice to add the appropriate
4847 libraries to your link command:
4849 ### Find Rx, quickthreads and libguile.
4850 AC_CHECK_LIB(rx, main)
4851 AC_CHECK_LIB(qt, main)
4852 AC_CHECK_LIB(guile, scm_shell)
4854 The Guile 1.2 distribution does not contain sources for the Rx
4855 library, as Guile 1.0 did. If you want to use Rx, you'll need to
4856 retrieve it from a GNU FTP site and install it separately.
4858 * Changes to Scheme functions and syntax
4860 ** The dynamic linking features of Guile are now enabled by default.
4861 You can disable them by giving the `--disable-dynamic-linking' option
4864 (dynamic-link FILENAME)
4866 Find the object file denoted by FILENAME (a string) and link it
4867 into the running Guile application. When everything works out,
4868 return a Scheme object suitable for representing the linked object
4869 file. Otherwise an error is thrown. How object files are
4870 searched is system dependent.
4872 (dynamic-object? VAL)
4874 Determine whether VAL represents a dynamically linked object file.
4876 (dynamic-unlink DYNOBJ)
4878 Unlink the indicated object file from the application. DYNOBJ
4879 should be one of the values returned by `dynamic-link'.
4881 (dynamic-func FUNCTION DYNOBJ)
4883 Search the C function indicated by FUNCTION (a string or symbol)
4884 in DYNOBJ and return some Scheme object that can later be used
4885 with `dynamic-call' to actually call this function. Right now,
4886 these Scheme objects are formed by casting the address of the
4887 function to `long' and converting this number to its Scheme
4890 (dynamic-call FUNCTION DYNOBJ)
4892 Call the C function indicated by FUNCTION and DYNOBJ. The
4893 function is passed no arguments and its return value is ignored.
4894 When FUNCTION is something returned by `dynamic-func', call that
4895 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
4896 etc.), look it up in DYNOBJ; this is equivalent to
4898 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
4900 Interrupts are deferred while the C function is executing (with
4901 SCM_DEFER_INTS/SCM_ALLOW_INTS).
4903 (dynamic-args-call FUNCTION DYNOBJ ARGS)
4905 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
4906 some arguments and return its return value. The C function is
4907 expected to take two arguments and return an `int', just like
4910 int c_func (int argc, char **argv);
4912 ARGS must be a list of strings and is converted into an array of
4913 `char *'. The array is passed in ARGV and its size in ARGC. The
4914 return value is converted to a Scheme number and returned from the
4915 call to `dynamic-args-call'.
4917 When dynamic linking is disabled or not supported on your system,
4918 the above functions throw errors, but they are still available.
4920 Here is a small example that works on GNU/Linux:
4922 (define libc-obj (dynamic-link "libc.so"))
4923 (dynamic-args-call 'rand libc-obj '())
4925 See the file `libguile/DYNAMIC-LINKING' for additional comments.
4927 ** The #/ syntax for module names is depreciated, and will be removed
4928 in a future version of Guile. Instead of
4936 The latter syntax is more consistent with existing Lisp practice.
4938 ** Guile now does fancier printing of structures. Structures are the
4939 underlying implementation for records, which in turn are used to
4940 implement modules, so all of these object now print differently and in
4941 a more informative way.
4943 The Scheme printer will examine the builtin variable *struct-printer*
4944 whenever it needs to print a structure object. When this variable is
4945 not `#f' it is deemed to be a procedure and will be applied to the
4946 structure object and the output port. When *struct-printer* is `#f'
4947 or the procedure return `#f' the structure object will be printed in
4948 the boring #<struct 80458270> form.
4950 This hook is used by some routines in ice-9/boot-9.scm to implement
4951 type specific printing routines. Please read the comments there about
4954 One of the more specific uses of structs are records. The printing
4955 procedure that could be passed to MAKE-RECORD-TYPE is now actually
4956 called. It should behave like a *struct-printer* procedure (described
4959 ** Guile now supports a new R4RS-compliant syntax for keywords. A
4960 token of the form #:NAME, where NAME has the same syntax as a Scheme
4961 symbol, is the external representation of the keyword named NAME.
4962 Keyword objects print using this syntax as well, so values containing
4963 keyword objects can be read back into Guile. When used in an
4964 expression, keywords are self-quoting objects.
4966 Guile suports this read syntax, and uses this print syntax, regardless
4967 of the current setting of the `keyword' read option. The `keyword'
4968 read option only controls whether Guile recognizes the `:NAME' syntax,
4969 which is incompatible with R4RS. (R4RS says such token represent
4972 ** Guile has regular expression support again. Guile 1.0 included
4973 functions for matching regular expressions, based on the Rx library.
4974 In Guile 1.1, the Guile/Rx interface was removed to simplify the
4975 distribution, and thus Guile had no regular expression support. Guile
4976 1.2 again supports the most commonly used functions, and supports all
4977 of SCSH's regular expression functions.
4979 If your system does not include a POSIX regular expression library,
4980 and you have not linked Guile with a third-party regexp library such as
4981 Rx, these functions will not be available. You can tell whether your
4982 Guile installation includes regular expression support by checking
4983 whether the `*features*' list includes the `regex' symbol.
4985 *** regexp functions
4987 By default, Guile supports POSIX extended regular expressions. That
4988 means that the characters `(', `)', `+' and `?' are special, and must
4989 be escaped if you wish to match the literal characters.
4991 This regular expression interface was modeled after that implemented
4992 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
4993 with SCSH regular expressions.
4995 **** Function: string-match PATTERN STR [START]
4996 Compile the string PATTERN into a regular expression and compare
4997 it with STR. The optional numeric argument START specifies the
4998 position of STR at which to begin matching.
5000 `string-match' returns a "match structure" which describes what,
5001 if anything, was matched by the regular expression. *Note Match
5002 Structures::. If STR does not match PATTERN at all,
5003 `string-match' returns `#f'.
5005 Each time `string-match' is called, it must compile its PATTERN
5006 argument into a regular expression structure. This operation is
5007 expensive, which makes `string-match' inefficient if the same regular
5008 expression is used several times (for example, in a loop). For better
5009 performance, you can compile a regular expression in advance and then
5010 match strings against the compiled regexp.
5012 **** Function: make-regexp STR [FLAGS]
5013 Compile the regular expression described by STR, and return the
5014 compiled regexp structure. If STR does not describe a legal
5015 regular expression, `make-regexp' throws a
5016 `regular-expression-syntax' error.
5018 FLAGS may be the bitwise-or of one or more of the following:
5020 **** Constant: regexp/extended
5021 Use POSIX Extended Regular Expression syntax when interpreting
5022 STR. If not set, POSIX Basic Regular Expression syntax is used.
5023 If the FLAGS argument is omitted, we assume regexp/extended.
5025 **** Constant: regexp/icase
5026 Do not differentiate case. Subsequent searches using the
5027 returned regular expression will be case insensitive.
5029 **** Constant: regexp/newline
5030 Match-any-character operators don't match a newline.
5032 A non-matching list ([^...]) not containing a newline matches a
5035 Match-beginning-of-line operator (^) matches the empty string
5036 immediately after a newline, regardless of whether the FLAGS
5037 passed to regexp-exec contain regexp/notbol.
5039 Match-end-of-line operator ($) matches the empty string
5040 immediately before a newline, regardless of whether the FLAGS
5041 passed to regexp-exec contain regexp/noteol.
5043 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5044 Match the compiled regular expression REGEXP against `str'. If
5045 the optional integer START argument is provided, begin matching
5046 from that position in the string. Return a match structure
5047 describing the results of the match, or `#f' if no match could be
5050 FLAGS may be the bitwise-or of one or more of the following:
5052 **** Constant: regexp/notbol
5053 The match-beginning-of-line operator always fails to match (but
5054 see the compilation flag regexp/newline above) This flag may be
5055 used when different portions of a string are passed to
5056 regexp-exec and the beginning of the string should not be
5057 interpreted as the beginning of the line.
5059 **** Constant: regexp/noteol
5060 The match-end-of-line operator always fails to match (but see the
5061 compilation flag regexp/newline above)
5063 **** Function: regexp? OBJ
5064 Return `#t' if OBJ is a compiled regular expression, or `#f'
5067 Regular expressions are commonly used to find patterns in one string
5068 and replace them with the contents of another string.
5070 **** Function: regexp-substitute PORT MATCH [ITEM...]
5071 Write to the output port PORT selected contents of the match
5072 structure MATCH. Each ITEM specifies what should be written, and
5073 may be one of the following arguments:
5075 * A string. String arguments are written out verbatim.
5077 * An integer. The submatch with that number is written.
5079 * The symbol `pre'. The portion of the matched string preceding
5080 the regexp match is written.
5082 * The symbol `post'. The portion of the matched string
5083 following the regexp match is written.
5085 PORT may be `#f', in which case nothing is written; instead,
5086 `regexp-substitute' constructs a string from the specified ITEMs
5089 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5090 Similar to `regexp-substitute', but can be used to perform global
5091 substitutions on STR. Instead of taking a match structure as an
5092 argument, `regexp-substitute/global' takes two string arguments: a
5093 REGEXP string describing a regular expression, and a TARGET string
5094 which should be matched against this regular expression.
5096 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5099 * A function may be supplied. When this function is called, it
5100 will be passed one argument: a match structure for a given
5101 regular expression match. It should return a string to be
5102 written out to PORT.
5104 * The `post' symbol causes `regexp-substitute/global' to recurse
5105 on the unmatched portion of STR. This *must* be supplied in
5106 order to perform global search-and-replace on STR; if it is
5107 not present among the ITEMs, then `regexp-substitute/global'
5108 will return after processing a single match.
5110 *** Match Structures
5112 A "match structure" is the object returned by `string-match' and
5113 `regexp-exec'. It describes which portion of a string, if any, matched
5114 the given regular expression. Match structures include: a reference to
5115 the string that was checked for matches; the starting and ending
5116 positions of the regexp match; and, if the regexp included any
5117 parenthesized subexpressions, the starting and ending positions of each
5120 In each of the regexp match functions described below, the `match'
5121 argument must be a match structure returned by a previous call to
5122 `string-match' or `regexp-exec'. Most of these functions return some
5123 information about the original target string that was matched against a
5124 regular expression; we will call that string TARGET for easy reference.
5126 **** Function: regexp-match? OBJ
5127 Return `#t' if OBJ is a match structure returned by a previous
5128 call to `regexp-exec', or `#f' otherwise.
5130 **** Function: match:substring MATCH [N]
5131 Return the portion of TARGET matched by subexpression number N.
5132 Submatch 0 (the default) represents the entire regexp match. If
5133 the regular expression as a whole matched, but the subexpression
5134 number N did not match, return `#f'.
5136 **** Function: match:start MATCH [N]
5137 Return the starting position of submatch number N.
5139 **** Function: match:end MATCH [N]
5140 Return the ending position of submatch number N.
5142 **** Function: match:prefix MATCH
5143 Return the unmatched portion of TARGET preceding the regexp match.
5145 **** Function: match:suffix MATCH
5146 Return the unmatched portion of TARGET following the regexp match.
5148 **** Function: match:count MATCH
5149 Return the number of parenthesized subexpressions from MATCH.
5150 Note that the entire regular expression match itself counts as a
5151 subexpression, and failed submatches are included in the count.
5153 **** Function: match:string MATCH
5154 Return the original TARGET string.
5156 *** Backslash Escapes
5158 Sometimes you will want a regexp to match characters like `*' or `$'
5159 exactly. For example, to check whether a particular string represents
5160 a menu entry from an Info node, it would be useful to match it against
5161 a regexp like `^* [^:]*::'. However, this won't work; because the
5162 asterisk is a metacharacter, it won't match the `*' at the beginning of
5163 the string. In this case, we want to make the first asterisk un-magic.
5165 You can do this by preceding the metacharacter with a backslash
5166 character `\'. (This is also called "quoting" the metacharacter, and
5167 is known as a "backslash escape".) When Guile sees a backslash in a
5168 regular expression, it considers the following glyph to be an ordinary
5169 character, no matter what special meaning it would ordinarily have.
5170 Therefore, we can make the above example work by changing the regexp to
5171 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5172 to match only a single asterisk in the target string.
5174 Since the backslash is itself a metacharacter, you may force a
5175 regexp to match a backslash in the target string by preceding the
5176 backslash with itself. For example, to find variable references in a
5177 TeX program, you might want to find occurrences of the string `\let\'
5178 followed by any number of alphabetic characters. The regular expression
5179 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5180 each match a single backslash in the target string.
5182 **** Function: regexp-quote STR
5183 Quote each special character found in STR with a backslash, and
5184 return the resulting string.
5186 *Very important:* Using backslash escapes in Guile source code (as
5187 in Emacs Lisp or C) can be tricky, because the backslash character has
5188 special meaning for the Guile reader. For example, if Guile encounters
5189 the character sequence `\n' in the middle of a string while processing
5190 Scheme code, it replaces those characters with a newline character.
5191 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5192 Several of these "escape sequences" are processed by the Guile reader
5193 before your code is executed. Unrecognized escape sequences are
5194 ignored: if the characters `\*' appear in a string, they will be
5195 translated to the single character `*'.
5197 This translation is obviously undesirable for regular expressions,
5198 since we want to be able to include backslashes in a string in order to
5199 escape regexp metacharacters. Therefore, to make sure that a backslash
5200 is preserved in a string in your Guile program, you must use *two*
5201 consecutive backslashes:
5203 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5205 The string in this example is preprocessed by the Guile reader before
5206 any code is executed. The resulting argument to `make-regexp' is the
5207 string `^\* [^:]*', which is what we really want.
5209 This also means that in order to write a regular expression that
5210 matches a single backslash character, the regular expression string in
5211 the source code must include *four* backslashes. Each consecutive pair
5212 of backslashes gets translated by the Guile reader to a single
5213 backslash, and the resulting double-backslash is interpreted by the
5214 regexp engine as matching a single backslash character. Hence:
5216 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5218 The reason for the unwieldiness of this syntax is historical. Both
5219 regular expression pattern matchers and Unix string processing systems
5220 have traditionally used backslashes with the special meanings described
5221 above. The POSIX regular expression specification and ANSI C standard
5222 both require these semantics. Attempting to abandon either convention
5223 would cause other kinds of compatibility problems, possibly more severe
5224 ones. Therefore, without extending the Scheme reader to support
5225 strings with different quoting conventions (an ungainly and confusing
5226 extension when implemented in other languages), we must adhere to this
5227 cumbersome escape syntax.
5229 * Changes to the gh_ interface
5231 * Changes to the scm_ interface
5233 * Changes to system call interfaces:
5235 ** The value returned by `raise' is now unspecified. It throws an exception
5238 *** A new procedure `sigaction' can be used to install signal handlers
5240 (sigaction signum [action] [flags])
5242 signum is the signal number, which can be specified using the value
5245 If action is omitted, sigaction returns a pair: the CAR is the current
5246 signal hander, which will be either an integer with the value SIG_DFL
5247 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5248 handles the signal, or #f if a non-Scheme procedure handles the
5249 signal. The CDR contains the current sigaction flags for the handler.
5251 If action is provided, it is installed as the new handler for signum.
5252 action can be a Scheme procedure taking one argument, or the value of
5253 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5254 whatever signal handler was installed before sigaction was first used.
5255 Flags can optionally be specified for the new handler (SA_RESTART is
5256 always used if the system provides it, so need not be specified.) The
5257 return value is a pair with information about the old handler as
5260 This interface does not provide access to the "signal blocking"
5261 facility. Maybe this is not needed, since the thread support may
5262 provide solutions to the problem of consistent access to data
5265 *** A new procedure `flush-all-ports' is equivalent to running
5266 `force-output' on every port open for output.
5268 ** Guile now provides information on how it was built, via the new
5269 global variable, %guile-build-info. This variable records the values
5270 of the standard GNU makefile directory variables as an assocation
5271 list, mapping variable names (symbols) onto directory paths (strings).
5272 For example, to find out where the Guile link libraries were
5273 installed, you can say:
5275 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5278 * Changes to the scm_ interface
5280 ** The new function scm_handle_by_message_noexit is just like the
5281 existing scm_handle_by_message function, except that it doesn't call
5282 exit to terminate the process. Instead, it prints a message and just
5283 returns #f. This might be a more appropriate catch-all handler for
5284 new dynamic roots and threads.
5287 Changes in Guile 1.1 (released Friday, May 16 1997):
5289 * Changes to the distribution.
5291 The Guile 1.0 distribution has been split up into several smaller
5293 guile-core --- the Guile interpreter itself.
5294 guile-tcltk --- the interface between the Guile interpreter and
5295 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5296 is a toolkit for building graphical user interfaces.
5297 guile-rgx-ctax --- the interface between Guile and the Rx regular
5298 expression matcher, and the translator for the Ctax
5299 programming language. These are packaged together because the
5300 Ctax translator uses Rx to parse Ctax source code.
5302 This NEWS file describes the changes made to guile-core since the 1.0
5305 We no longer distribute the documentation, since it was either out of
5306 date, or incomplete. As soon as we have current documentation, we
5311 * Changes to the stand-alone interpreter
5313 ** guile now accepts command-line arguments compatible with SCSH, Olin
5314 Shivers' Scheme Shell.
5316 In general, arguments are evaluated from left to right, but there are
5317 exceptions. The following switches stop argument processing, and
5318 stash all remaining command-line arguments as the value returned by
5319 the (command-line) function.
5320 -s SCRIPT load Scheme source code from FILE, and exit
5321 -c EXPR evalute Scheme expression EXPR, and exit
5322 -- stop scanning arguments; run interactively
5324 The switches below are processed as they are encountered.
5325 -l FILE load Scheme source code from FILE
5326 -e FUNCTION after reading script, apply FUNCTION to
5327 command line arguments
5328 -ds do -s script at this point
5329 --emacs enable Emacs protocol (experimental)
5330 -h, --help display this help and exit
5331 -v, --version display version information and exit
5332 \ read arguments from following script lines
5334 So, for example, here is a Guile script named `ekko' (thanks, Olin)
5335 which re-implements the traditional "echo" command:
5337 #!/usr/local/bin/guile -s
5340 (map (lambda (arg) (display arg) (display " "))
5344 (main (command-line))
5346 Suppose we invoke this script as follows:
5348 ekko a speckled gecko
5350 Through the magic of Unix script processing (triggered by the `#!'
5351 token at the top of the file), /usr/local/bin/guile receives the
5352 following list of command-line arguments:
5354 ("-s" "./ekko" "a" "speckled" "gecko")
5356 Unix inserts the name of the script after the argument specified on
5357 the first line of the file (in this case, "-s"), and then follows that
5358 with the arguments given to the script. Guile loads the script, which
5359 defines the `main' function, and then applies it to the list of
5360 remaining command-line arguments, ("a" "speckled" "gecko").
5362 In Unix, the first line of a script file must take the following form:
5364 #!INTERPRETER ARGUMENT
5366 where INTERPRETER is the absolute filename of the interpreter
5367 executable, and ARGUMENT is a single command-line argument to pass to
5370 You may only pass one argument to the interpreter, and its length is
5371 limited. These restrictions can be annoying to work around, so Guile
5372 provides a general mechanism (borrowed from, and compatible with,
5373 SCSH) for circumventing them.
5375 If the ARGUMENT in a Guile script is a single backslash character,
5376 `\', Guile will open the script file, parse arguments from its second
5377 and subsequent lines, and replace the `\' with them. So, for example,
5378 here is another implementation of the `ekko' script:
5380 #!/usr/local/bin/guile \
5384 (for-each (lambda (arg) (display arg) (display " "))
5388 If the user invokes this script as follows:
5390 ekko a speckled gecko
5392 Unix expands this into
5394 /usr/local/bin/guile \ ekko a speckled gecko
5396 When Guile sees the `\' argument, it replaces it with the arguments
5397 read from the second line of the script, producing:
5399 /usr/local/bin/guile -e main -s ekko a speckled gecko
5401 This tells Guile to load the `ekko' script, and apply the function
5402 `main' to the argument list ("a" "speckled" "gecko").
5404 Here is how Guile parses the command-line arguments:
5405 - Each space character terminates an argument. This means that two
5406 spaces in a row introduce an empty-string argument.
5407 - The tab character is not permitted (unless you quote it with the
5408 backslash character, as described below), to avoid confusion.
5409 - The newline character terminates the sequence of arguments, and will
5410 also terminate a final non-empty argument. (However, a newline
5411 following a space will not introduce a final empty-string argument;
5412 it only terminates the argument list.)
5413 - The backslash character is the escape character. It escapes
5414 backslash, space, tab, and newline. The ANSI C escape sequences
5415 like \n and \t are also supported. These produce argument
5416 constituents; the two-character combination \n doesn't act like a
5417 terminating newline. The escape sequence \NNN for exactly three
5418 octal digits reads as the character whose ASCII code is NNN. As
5419 above, characters produced this way are argument constituents.
5420 Backslash followed by other characters is not allowed.
5422 * Changes to the procedure for linking libguile with your programs
5424 ** Guile now builds and installs a shared guile library, if your
5425 system support shared libraries. (It still builds a static library on
5426 all systems.) Guile automatically detects whether your system
5427 supports shared libraries. To prevent Guile from buildisg shared
5428 libraries, pass the `--disable-shared' flag to the configure script.
5430 Guile takes longer to compile when it builds shared libraries, because
5431 it must compile every file twice --- once to produce position-
5432 independent object code, and once to produce normal object code.
5434 ** The libthreads library has been merged into libguile.
5436 To link a program against Guile, you now need only link against
5437 -lguile and -lqt; -lthreads is no longer needed. If you are using
5438 autoconf to generate configuration scripts for your application, the
5439 following lines should suffice to add the appropriate libraries to
5442 ### Find quickthreads and libguile.
5443 AC_CHECK_LIB(qt, main)
5444 AC_CHECK_LIB(guile, scm_shell)
5446 * Changes to Scheme functions
5448 ** Guile Scheme's special syntax for keyword objects is now optional,
5449 and disabled by default.
5451 The syntax variation from R4RS made it difficult to port some
5452 interesting packages to Guile. The routines which accepted keyword
5453 arguments (mostly in the module system) have been modified to also
5454 accept symbols whose names begin with `:'.
5456 To change the keyword syntax, you must first import the (ice-9 debug)
5458 (use-modules (ice-9 debug))
5460 Then you can enable the keyword syntax as follows:
5461 (read-set! keywords 'prefix)
5463 To disable keyword syntax, do this:
5464 (read-set! keywords #f)
5466 ** Many more primitive functions accept shared substrings as
5467 arguments. In the past, these functions required normal, mutable
5468 strings as arguments, although they never made use of this
5471 ** The uniform array functions now operate on byte vectors. These
5472 functions are `array-fill!', `serial-array-copy!', `array-copy!',
5473 `serial-array-map', `array-map', `array-for-each', and
5476 ** The new functions `trace' and `untrace' implement simple debugging
5477 support for Scheme functions.
5479 The `trace' function accepts any number of procedures as arguments,
5480 and tells the Guile interpreter to display each procedure's name and
5481 arguments each time the procedure is invoked. When invoked with no
5482 arguments, `trace' returns the list of procedures currently being
5485 The `untrace' function accepts any number of procedures as arguments,
5486 and tells the Guile interpreter not to trace them any more. When
5487 invoked with no arguments, `untrace' untraces all curretly traced
5490 The tracing in Guile has an advantage over most other systems: we
5491 don't create new procedure objects, but mark the procedure objects
5492 themselves. This means that anonymous and internal procedures can be
5495 ** The function `assert-repl-prompt' has been renamed to
5496 `set-repl-prompt!'. It takes one argument, PROMPT.
5497 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
5498 - If PROMPT is a string, we use it as a prompt.
5499 - If PROMPT is a procedure accepting no arguments, we call it, and
5500 display the result as a prompt.
5501 - Otherwise, we display "> ".
5503 ** The new function `eval-string' reads Scheme expressions from a
5504 string and evaluates them, returning the value of the last expression
5505 in the string. If the string contains no expressions, it returns an
5508 ** The new function `thunk?' returns true iff its argument is a
5509 procedure of zero arguments.
5511 ** `defined?' is now a builtin function, instead of syntax. This
5512 means that its argument should be quoted. It returns #t iff its
5513 argument is bound in the current module.
5515 ** The new syntax `use-modules' allows you to add new modules to your
5516 environment without re-typing a complete `define-module' form. It
5517 accepts any number of module names as arguments, and imports their
5518 public bindings into the current module.
5520 ** The new function (module-defined? NAME MODULE) returns true iff
5521 NAME, a symbol, is defined in MODULE, a module object.
5523 ** The new function `builtin-bindings' creates and returns a hash
5524 table containing copies of all the root module's bindings.
5526 ** The new function `builtin-weak-bindings' does the same as
5527 `builtin-bindings', but creates a doubly-weak hash table.
5529 ** The `equal?' function now considers variable objects to be
5530 equivalent if they have the same name and the same value.
5532 ** The new function `command-line' returns the command-line arguments
5533 given to Guile, as a list of strings.
5535 When using guile as a script interpreter, `command-line' returns the
5536 script's arguments; those processed by the interpreter (like `-s' or
5537 `-c') are omitted. (In other words, you get the normal, expected
5538 behavior.) Any application that uses scm_shell to process its
5539 command-line arguments gets this behavior as well.
5541 ** The new function `load-user-init' looks for a file called `.guile'
5542 in the user's home directory, and loads it if it exists. This is
5543 mostly for use by the code generated by scm_compile_shell_switches,
5544 but we thought it might also be useful in other circumstances.
5546 ** The new function `log10' returns the base-10 logarithm of its
5549 ** Changes to I/O functions
5551 *** The functions `read', `primitive-load', `read-and-eval!', and
5552 `primitive-load-path' no longer take optional arguments controlling
5553 case insensitivity and a `#' parser.
5555 Case sensitivity is now controlled by a read option called
5556 `case-insensitive'. The user can add new `#' syntaxes with the
5557 `read-hash-extend' function (see below).
5559 *** The new function `read-hash-extend' allows the user to change the
5560 syntax of Guile Scheme in a somewhat controlled way.
5562 (read-hash-extend CHAR PROC)
5563 When parsing S-expressions, if we read a `#' character followed by
5564 the character CHAR, use PROC to parse an object from the stream.
5565 If PROC is #f, remove any parsing procedure registered for CHAR.
5567 The reader applies PROC to two arguments: CHAR and an input port.
5569 *** The new functions read-delimited and read-delimited! provide a
5570 general mechanism for doing delimited input on streams.
5572 (read-delimited DELIMS [PORT HANDLE-DELIM])
5573 Read until we encounter one of the characters in DELIMS (a string),
5574 or end-of-file. PORT is the input port to read from; it defaults to
5575 the current input port. The HANDLE-DELIM parameter determines how
5576 the terminating character is handled; it should be one of the
5579 'trim omit delimiter from result
5580 'peek leave delimiter character in input stream
5581 'concat append delimiter character to returned value
5582 'split return a pair: (RESULT . TERMINATOR)
5584 HANDLE-DELIM defaults to 'peek.
5586 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
5587 A side-effecting variant of `read-delimited'.
5589 The data is written into the string BUF at the indices in the
5590 half-open interval [START, END); the default interval is the whole
5591 string: START = 0 and END = (string-length BUF). The values of
5592 START and END must specify a well-defined interval in BUF, i.e.
5593 0 <= START <= END <= (string-length BUF).
5595 It returns NBYTES, the number of bytes read. If the buffer filled
5596 up without a delimiter character being found, it returns #f. If the
5597 port is at EOF when the read starts, it returns the EOF object.
5599 If an integer is returned (i.e., the read is successfully terminated
5600 by reading a delimiter character), then the HANDLE-DELIM parameter
5601 determines how to handle the terminating character. It is described
5602 above, and defaults to 'peek.
5604 (The descriptions of these functions were borrowed from the SCSH
5605 manual, by Olin Shivers and Brian Carlstrom.)
5607 *** The `%read-delimited!' function is the primitive used to implement
5608 `read-delimited' and `read-delimited!'.
5610 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
5612 This returns a pair of values: (TERMINATOR . NUM-READ).
5613 - TERMINATOR describes why the read was terminated. If it is a
5614 character or the eof object, then that is the value that terminated
5615 the read. If it is #f, the function filled the buffer without finding
5616 a delimiting character.
5617 - NUM-READ is the number of characters read into BUF.
5619 If the read is successfully terminated by reading a delimiter
5620 character, then the gobble? parameter determines what to do with the
5621 terminating character. If true, the character is removed from the
5622 input stream; if false, the character is left in the input stream
5623 where a subsequent read operation will retrieve it. In either case,
5624 the character is also the first value returned by the procedure call.
5626 (The descriptions of this function was borrowed from the SCSH manual,
5627 by Olin Shivers and Brian Carlstrom.)
5629 *** The `read-line' and `read-line!' functions have changed; they now
5630 trim the terminator by default; previously they appended it to the
5631 returned string. For the old behavior, use (read-line PORT 'concat).
5633 *** The functions `uniform-array-read!' and `uniform-array-write!' now
5634 take new optional START and END arguments, specifying the region of
5635 the array to read and write.
5637 *** The `ungetc-char-ready?' function has been removed. We feel it's
5638 inappropriate for an interface to expose implementation details this
5641 ** Changes to the Unix library and system call interface
5643 *** The new fcntl function provides access to the Unix `fcntl' system
5646 (fcntl PORT COMMAND VALUE)
5647 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
5648 Values for COMMAND are:
5650 F_DUPFD duplicate a file descriptor
5651 F_GETFD read the descriptor's close-on-exec flag
5652 F_SETFD set the descriptor's close-on-exec flag to VALUE
5653 F_GETFL read the descriptor's flags, as set on open
5654 F_SETFL set the descriptor's flags, as set on open to VALUE
5655 F_GETOWN return the process ID of a socket's owner, for SIGIO
5656 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
5657 FD_CLOEXEC not sure what this is
5659 For details, see the documentation for the fcntl system call.
5661 *** The arguments to `select' have changed, for compatibility with
5662 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
5663 expected behavior. The MILLISECONDS parameter has been changed to
5664 MICROSECONDS, to more closely resemble the underlying system call.
5665 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
5666 corresponding return set will be the same.
5668 *** The arguments to the `mknod' system call have changed. They are
5671 (mknod PATH TYPE PERMS DEV)
5672 Create a new file (`node') in the file system. PATH is the name of
5673 the file to create. TYPE is the kind of file to create; it should
5674 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
5675 permission bits to give the newly created file. If TYPE is
5676 'block-special or 'char-special, DEV specifies which device the
5677 special file refers to; its interpretation depends on the kind of
5678 special file being created.
5680 *** The `fork' function has been renamed to `primitive-fork', to avoid
5681 clashing with various SCSH forks.
5683 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
5684 and `recvfrom!'. They no longer accept a size for a second argument;
5685 you must pass a string to hold the received value. They no longer
5686 return the buffer. Instead, `recv' returns the length of the message
5687 received, and `recvfrom' returns a pair containing the packet's length
5688 and originating address.
5690 *** The file descriptor datatype has been removed, as have the
5691 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
5692 We plan to replace these functions with a SCSH-compatible interface.
5694 *** The `create' function has been removed; it's just a special case
5697 *** There are new functions to break down process termination status
5698 values. In the descriptions below, STATUS is a value returned by
5701 (status:exit-val STATUS)
5702 If the child process exited normally, this function returns the exit
5703 code for the child process (i.e., the value passed to exit, or
5704 returned from main). If the child process did not exit normally,
5705 this function returns #f.
5707 (status:stop-sig STATUS)
5708 If the child process was suspended by a signal, this function
5709 returns the signal that suspended the child. Otherwise, it returns
5712 (status:term-sig STATUS)
5713 If the child process terminated abnormally, this function returns
5714 the signal that terminated the child. Otherwise, this function
5717 POSIX promises that exactly one of these functions will return true on
5718 a valid STATUS value.
5720 These functions are compatible with SCSH.
5722 *** There are new accessors and setters for the broken-out time vectors
5723 returned by `localtime', `gmtime', and that ilk. They are:
5725 Component Accessor Setter
5726 ========================= ============ ============
5727 seconds tm:sec set-tm:sec
5728 minutes tm:min set-tm:min
5729 hours tm:hour set-tm:hour
5730 day of the month tm:mday set-tm:mday
5731 month tm:mon set-tm:mon
5732 year tm:year set-tm:year
5733 day of the week tm:wday set-tm:wday
5734 day in the year tm:yday set-tm:yday
5735 daylight saving time tm:isdst set-tm:isdst
5736 GMT offset, seconds tm:gmtoff set-tm:gmtoff
5737 name of time zone tm:zone set-tm:zone
5739 *** There are new accessors for the vectors returned by `uname',
5740 describing the host system:
5743 ============================================== ================
5744 name of the operating system implementation utsname:sysname
5745 network name of this machine utsname:nodename
5746 release level of the operating system utsname:release
5747 version level of the operating system utsname:version
5748 machine hardware platform utsname:machine
5750 *** There are new accessors for the vectors returned by `getpw',
5751 `getpwnam', `getpwuid', and `getpwent', describing entries from the
5752 system's user database:
5755 ====================== =================
5756 user name passwd:name
5757 user password passwd:passwd
5760 real name passwd:gecos
5761 home directory passwd:dir
5762 shell program passwd:shell
5764 *** There are new accessors for the vectors returned by `getgr',
5765 `getgrnam', `getgrgid', and `getgrent', describing entries from the
5766 system's group database:
5769 ======================= ============
5770 group name group:name
5771 group password group:passwd
5773 group members group:mem
5775 *** There are new accessors for the vectors returned by `gethost',
5776 `gethostbyaddr', `gethostbyname', and `gethostent', describing
5780 ========================= ===============
5781 official name of host hostent:name
5782 alias list hostent:aliases
5783 host address type hostent:addrtype
5784 length of address hostent:length
5785 list of addresses hostent:addr-list
5787 *** There are new accessors for the vectors returned by `getnet',
5788 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
5792 ========================= ===============
5793 official name of net netent:name
5794 alias list netent:aliases
5795 net number type netent:addrtype
5796 net number netent:net
5798 *** There are new accessors for the vectors returned by `getproto',
5799 `getprotobyname', `getprotobynumber', and `getprotoent', describing
5803 ========================= ===============
5804 official protocol name protoent:name
5805 alias list protoent:aliases
5806 protocol number protoent:proto
5808 *** There are new accessors for the vectors returned by `getserv',
5809 `getservbyname', `getservbyport', and `getservent', describing
5813 ========================= ===============
5814 official service name servent:name
5815 alias list servent:aliases
5816 port number servent:port
5817 protocol to use servent:proto
5819 *** There are new accessors for the sockaddr structures returned by
5820 `accept', `getsockname', `getpeername', `recvfrom!':
5823 ======================================== ===============
5824 address format (`family') sockaddr:fam
5825 path, for file domain addresses sockaddr:path
5826 address, for internet domain addresses sockaddr:addr
5827 TCP or UDP port, for internet sockaddr:port
5829 *** The `getpwent', `getgrent', `gethostent', `getnetent',
5830 `getprotoent', and `getservent' functions now return #f at the end of
5831 the user database. (They used to throw an exception.)
5833 Note that calling MUMBLEent function is equivalent to calling the
5834 corresponding MUMBLE function with no arguments.
5836 *** The `setpwent', `setgrent', `sethostent', `setnetent',
5837 `setprotoent', and `setservent' routines now take no arguments.
5839 *** The `gethost', `getproto', `getnet', and `getserv' functions now
5840 provide more useful information when they throw an exception.
5842 *** The `lnaof' function has been renamed to `inet-lnaof'.
5844 *** Guile now claims to have the `current-time' feature.
5846 *** The `mktime' function now takes an optional second argument ZONE,
5847 giving the time zone to use for the conversion. ZONE should be a
5848 string, in the same format as expected for the "TZ" environment variable.
5850 *** The `strptime' function now returns a pair (TIME . COUNT), where
5851 TIME is the parsed time as a vector, and COUNT is the number of
5852 characters from the string left unparsed. This function used to
5853 return the remaining characters as a string.
5855 *** The `gettimeofday' function has replaced the old `time+ticks' function.
5856 The return value is now (SECONDS . MICROSECONDS); the fractional
5857 component is no longer expressed in "ticks".
5859 *** The `ticks/sec' constant has been removed, in light of the above change.
5861 * Changes to the gh_ interface
5863 ** gh_eval_str() now returns an SCM object which is the result of the
5866 ** gh_scm2str() now copies the Scheme data to a caller-provided C
5869 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
5870 and returns the array
5872 ** gh_scm2str0() is gone: there is no need to distinguish
5873 null-terminated from non-null-terminated, since gh_scm2newstr() allows
5874 the user to interpret the data both ways.
5876 * Changes to the scm_ interface
5878 ** The new function scm_symbol_value0 provides an easy way to get a
5879 symbol's value from C code:
5881 SCM scm_symbol_value0 (char *NAME)
5882 Return the value of the symbol named by the null-terminated string
5883 NAME in the current module. If the symbol named NAME is unbound in
5884 the current module, return SCM_UNDEFINED.
5886 ** The new function scm_sysintern0 creates new top-level variables,
5887 without assigning them a value.
5889 SCM scm_sysintern0 (char *NAME)
5890 Create a new Scheme top-level variable named NAME. NAME is a
5891 null-terminated string. Return the variable's value cell.
5893 ** The function scm_internal_catch is the guts of catch. It handles
5894 all the mechanics of setting up a catch target, invoking the catch
5895 body, and perhaps invoking the handler if the body does a throw.
5897 The function is designed to be usable from C code, but is general
5898 enough to implement all the semantics Guile Scheme expects from throw.
5900 TAG is the catch tag. Typically, this is a symbol, but this function
5901 doesn't actually care about that.
5903 BODY is a pointer to a C function which runs the body of the catch;
5904 this is the code you can throw from. We call it like this:
5905 BODY (BODY_DATA, JMPBUF)
5907 BODY_DATA is just the BODY_DATA argument we received; we pass it
5908 through to BODY as its first argument. The caller can make
5909 BODY_DATA point to anything useful that BODY might need.
5910 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
5911 which we have just created and initialized.
5913 HANDLER is a pointer to a C function to deal with a throw to TAG,
5914 should one occur. We call it like this:
5915 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
5917 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
5918 same idea as BODY_DATA above.
5919 THROWN_TAG is the tag that the user threw to; usually this is
5920 TAG, but it could be something else if TAG was #t (i.e., a
5921 catch-all), or the user threw to a jmpbuf.
5922 THROW_ARGS is the list of arguments the user passed to the THROW
5925 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
5926 is just a pointer we pass through to HANDLER. We don't actually
5927 use either of those pointers otherwise ourselves. The idea is
5928 that, if our caller wants to communicate something to BODY or
5929 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
5930 HANDLER can then use. Think of it as a way to make BODY and
5931 HANDLER closures, not just functions; MUMBLE_DATA points to the
5934 Of course, it's up to the caller to make sure that any data a
5935 MUMBLE_DATA needs is protected from GC. A common way to do this is
5936 to make MUMBLE_DATA a pointer to data stored in an automatic
5937 structure variable; since the collector must scan the stack for
5938 references anyway, this assures that any references in MUMBLE_DATA
5941 ** The new function scm_internal_lazy_catch is exactly like
5942 scm_internal_catch, except:
5944 - It does not unwind the stack (this is the major difference).
5945 - If handler returns, its value is returned from the throw.
5946 - BODY always receives #f as its JMPBUF argument (since there's no
5947 jmpbuf associated with a lazy catch, because we don't unwind the
5950 ** scm_body_thunk is a new body function you can pass to
5951 scm_internal_catch if you want the body to be like Scheme's `catch'
5952 --- a thunk, or a function of one argument if the tag is #f.
5954 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
5955 contains the Scheme procedure to invoke as the body, and the tag
5956 we're catching. If the tag is #f, then we pass JMPBUF (created by
5957 scm_internal_catch) to the body procedure; otherwise, the body gets
5960 ** scm_handle_by_proc is a new handler function you can pass to
5961 scm_internal_catch if you want the handler to act like Scheme's catch
5962 --- call a procedure with the tag and the throw arguments.
5964 If the user does a throw to this catch, this function runs a handler
5965 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
5966 variable holding the Scheme procedure object to invoke. It ought to
5967 be a pointer to an automatic variable (i.e., one living on the stack),
5968 or the procedure object should be otherwise protected from GC.
5970 ** scm_handle_by_message is a new handler function to use with
5971 `scm_internal_catch' if you want Guile to print a message and die.
5972 It's useful for dealing with throws to uncaught keys at the top level.
5974 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
5975 message header to print; if zero, we use "guile" instead. That
5976 text is followed by a colon, then the message described by ARGS.
5978 ** The return type of scm_boot_guile is now void; the function does
5979 not return a value, and indeed, never returns at all.
5981 ** The new function scm_shell makes it easy for user applications to
5982 process command-line arguments in a way that is compatible with the
5983 stand-alone guile interpreter (which is in turn compatible with SCSH,
5986 To use the scm_shell function, first initialize any guile modules
5987 linked into your application, and then call scm_shell with the values
5988 of ARGC and ARGV your `main' function received. scm_shell will add
5989 any SCSH-style meta-arguments from the top of the script file to the
5990 argument vector, and then process the command-line arguments. This
5991 generally means loading a script file or starting up an interactive
5992 command interpreter. For details, see "Changes to the stand-alone
5995 ** The new functions scm_get_meta_args and scm_count_argv help you
5996 implement the SCSH-style meta-argument, `\'.
5998 char **scm_get_meta_args (int ARGC, char **ARGV)
5999 If the second element of ARGV is a string consisting of a single
6000 backslash character (i.e. "\\" in Scheme notation), open the file
6001 named by the following argument, parse arguments from it, and return
6002 the spliced command line. The returned array is terminated by a
6005 For details of argument parsing, see above, under "guile now accepts
6006 command-line arguments compatible with SCSH..."
6008 int scm_count_argv (char **ARGV)
6009 Count the arguments in ARGV, assuming it is terminated by a null
6012 For an example of how these functions might be used, see the source
6013 code for the function scm_shell in libguile/script.c.
6015 You will usually want to use scm_shell instead of calling this
6018 ** The new function scm_compile_shell_switches turns an array of
6019 command-line arguments into Scheme code to carry out the actions they
6020 describe. Given ARGC and ARGV, it returns a Scheme expression to
6021 evaluate, and calls scm_set_program_arguments to make any remaining
6022 command-line arguments available to the Scheme code. For example,
6023 given the following arguments:
6025 -e main -s ekko a speckled gecko
6027 scm_set_program_arguments will return the following expression:
6029 (begin (load "ekko") (main (command-line)) (quit))
6031 You will usually want to use scm_shell instead of calling this
6034 ** The function scm_shell_usage prints a usage message appropriate for
6035 an interpreter that uses scm_compile_shell_switches to handle its
6036 command-line arguments.
6038 void scm_shell_usage (int FATAL, char *MESSAGE)
6039 Print a usage message to the standard error output. If MESSAGE is
6040 non-zero, write it before the usage message, followed by a newline.
6041 If FATAL is non-zero, exit the process, using FATAL as the
6042 termination status. (If you want to be compatible with Guile,
6043 always use 1 as the exit status when terminating due to command-line
6046 You will usually want to use scm_shell instead of calling this
6049 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6050 expressions. It used to return SCM_EOL. Earth-shattering.
6052 ** The macros for declaring scheme objects in C code have been
6053 rearranged slightly. They are now:
6055 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6056 Declare a static SCM variable named C_NAME, and initialize it to
6057 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6058 be a C identifier, and SCHEME_NAME should be a C string.
6060 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6061 Just like SCM_SYMBOL, but make C_NAME globally visible.
6063 SCM_VCELL (C_NAME, SCHEME_NAME)
6064 Create a global variable at the Scheme level named SCHEME_NAME.
6065 Declare a static SCM variable named C_NAME, and initialize it to
6066 point to the Scheme variable's value cell.
6068 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6069 Just like SCM_VCELL, but make C_NAME globally visible.
6071 The `guile-snarf' script writes initialization code for these macros
6072 to its standard output, given C source code as input.
6074 The SCM_GLOBAL macro is gone.
6076 ** The scm_read_line and scm_read_line_x functions have been replaced
6077 by Scheme code based on the %read-delimited! procedure (known to C
6078 code as scm_read_delimited_x). See its description above for more
6081 ** The function scm_sys_open has been renamed to scm_open. It now
6082 returns a port instead of an FD object.
6084 * The dynamic linking support has changed. For more information, see
6085 libguile/DYNAMIC-LINKING.
6090 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6093 * Changes to the 'guile' program:
6095 ** Guile now loads some new files when it starts up. Guile first
6096 searches the load path for init.scm, and loads it if found. Then, if
6097 Guile is not being used to execute a script, and the user's home
6098 directory contains a file named `.guile', Guile loads that.
6100 ** You can now use Guile as a shell script interpreter.
6102 To paraphrase the SCSH manual:
6104 When Unix tries to execute an executable file whose first two
6105 characters are the `#!', it treats the file not as machine code to
6106 be directly executed by the native processor, but as source code
6107 to be executed by some interpreter. The interpreter to use is
6108 specified immediately after the #! sequence on the first line of
6109 the source file. The kernel reads in the name of the interpreter,
6110 and executes that instead. It passes the interpreter the source
6111 filename as its first argument, with the original arguments
6112 following. Consult the Unix man page for the `exec' system call
6113 for more information.
6115 Now you can use Guile as an interpreter, using a mechanism which is a
6116 compatible subset of that provided by SCSH.
6118 Guile now recognizes a '-s' command line switch, whose argument is the
6119 name of a file of Scheme code to load. It also treats the two
6120 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6121 to make a file of Scheme code directly executable by Unix, insert the
6122 following two lines at the top of the file:
6124 #!/usr/local/bin/guile -s
6127 Guile treats the argument of the `-s' command-line switch as the name
6128 of a file of Scheme code to load, and treats the sequence `#!' as the
6129 start of a block comment, terminated by `!#'.
6131 For example, here's a version of 'echo' written in Scheme:
6133 #!/usr/local/bin/guile -s
6135 (let loop ((args (cdr (program-arguments))))
6138 (display (car args))
6139 (if (pair? (cdr args))
6141 (loop (cdr args)))))
6144 Why does `#!' start a block comment terminated by `!#', instead of the
6145 end of the line? That is the notation SCSH uses, and although we
6146 don't yet support the other SCSH features that motivate that choice,
6147 we would like to be backward-compatible with any existing Guile
6148 scripts once we do. Furthermore, if the path to Guile on your system
6149 is too long for your kernel, you can start the script with this
6153 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6156 Note that some very old Unix systems don't support the `#!' syntax.
6159 ** You can now run Guile without installing it.
6161 Previous versions of the interactive Guile interpreter (`guile')
6162 couldn't start up unless Guile's Scheme library had been installed;
6163 they used the value of the environment variable `SCHEME_LOAD_PATH'
6164 later on in the startup process, but not to find the startup code
6165 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6168 To run Guile without installing it, build it in the normal way, and
6169 then set the environment variable `SCHEME_LOAD_PATH' to a
6170 colon-separated list of directories, including the top-level directory
6171 of the Guile sources. For example, if you unpacked Guile so that the
6172 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6175 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6178 ** Guile's read-eval-print loop no longer prints #<unspecified>
6179 results. If the user wants to see this, she can evaluate the
6180 expression (assert-repl-print-unspecified #t), perhaps in her startup
6183 ** Guile no longer shows backtraces by default when an error occurs;
6184 however, it does display a message saying how to get one, and how to
6185 request that they be displayed by default. After an error, evaluate
6187 to see a backtrace, and
6188 (debug-enable 'backtrace)
6189 to see them by default.
6193 * Changes to Guile Scheme:
6195 ** Guile now distinguishes between #f and the empty list.
6197 This is for compatibility with the IEEE standard, the (possibly)
6198 upcoming Revised^5 Report on Scheme, and many extant Scheme
6201 Guile used to have #f and '() denote the same object, to make Scheme's
6202 type system more compatible with Emacs Lisp's. However, the change
6203 caused too much trouble for Scheme programmers, and we found another
6204 way to reconcile Emacs Lisp with Scheme that didn't require this.
6207 ** Guile's delq, delv, delete functions, and their destructive
6208 counterparts, delq!, delv!, and delete!, now remove all matching
6209 elements from the list, not just the first. This matches the behavior
6210 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6211 functions which inspired them.
6213 I recognize that this change may break code in subtle ways, but it
6214 seems best to make the change before the FSF's first Guile release,
6218 ** The compiled-library-path function has been deleted from libguile.
6220 ** The facilities for loading Scheme source files have changed.
6222 *** The variable %load-path now tells Guile which directories to search
6223 for Scheme code. Its value is a list of strings, each of which names
6226 *** The variable %load-extensions now tells Guile which extensions to
6227 try appending to a filename when searching the load path. Its value
6228 is a list of strings. Its default value is ("" ".scm").
6230 *** (%search-load-path FILENAME) searches the directories listed in the
6231 value of the %load-path variable for a Scheme file named FILENAME,
6232 with all the extensions listed in %load-extensions. If it finds a
6233 match, then it returns its full filename. If FILENAME is absolute, it
6234 returns it unchanged. Otherwise, it returns #f.
6236 %search-load-path will not return matches that refer to directories.
6238 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6239 uses %seach-load-path to find a file named FILENAME, and loads it if
6240 it finds it. If it can't read FILENAME for any reason, it throws an
6243 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6246 *** load uses the same searching semantics as primitive-load.
6248 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6249 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6250 path, and load-module-with-path have been deleted. The functions
6251 above should serve their purposes.
6253 *** If the value of the variable %load-hook is a procedure,
6254 `primitive-load' applies its value to the name of the file being
6255 loaded (without the load path directory name prepended). If its value
6256 is #f, it is ignored. Otherwise, an error occurs.
6258 This is mostly useful for printing load notification messages.
6261 ** The function `eval!' is no longer accessible from the scheme level.
6262 We can't allow operations which introduce glocs into the scheme level,
6263 because Guile's type system can't handle these as data. Use `eval' or
6264 `read-and-eval!' (see below) as replacement.
6266 ** The new function read-and-eval! reads an expression from PORT,
6267 evaluates it, and returns the result. This is more efficient than
6268 simply calling `read' and `eval', since it is not necessary to make a
6269 copy of the expression for the evaluator to munge.
6271 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6272 for the `read' function.
6275 ** The function `int?' has been removed; its definition was identical
6276 to that of `integer?'.
6278 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6279 use the R4RS names for these functions.
6281 ** The function object-properties no longer returns the hash handle;
6282 it simply returns the object's property list.
6284 ** Many functions have been changed to throw errors, instead of
6285 returning #f on failure. The point of providing exception handling in
6286 the language is to simplify the logic of user code, but this is less
6287 useful if Guile's primitives don't throw exceptions.
6289 ** The function `fileno' has been renamed from `%fileno'.
6291 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6294 * Changes to Guile's C interface:
6296 ** The library's initialization procedure has been simplified.
6297 scm_boot_guile now has the prototype:
6299 void scm_boot_guile (int ARGC,
6301 void (*main_func) (),
6304 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6305 MAIN_FUNC should do all the work of the program (initializing other
6306 packages, reading user input, etc.) before returning. When MAIN_FUNC
6307 returns, call exit (0); this function never returns. If you want some
6308 other exit value, MAIN_FUNC may call exit itself.
6310 scm_boot_guile arranges for program-arguments to return the strings
6311 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6312 scm_set_program_arguments with the final list, so Scheme code will
6313 know which arguments have been processed.
6315 scm_boot_guile establishes a catch-all catch handler which prints an
6316 error message and exits the process. This means that Guile exits in a
6317 coherent way when system errors occur and the user isn't prepared to
6318 handle it. If the user doesn't like this behavior, they can establish
6319 their own universal catcher in MAIN_FUNC to shadow this one.
6321 Why must the caller do all the real work from MAIN_FUNC? The garbage
6322 collector assumes that all local variables of type SCM will be above
6323 scm_boot_guile's stack frame on the stack. If you try to manipulate
6324 SCM values after this function returns, it's the luck of the draw
6325 whether the GC will be able to find the objects you allocate. So,
6326 scm_boot_guile function exits, rather than returning, to discourage
6327 people from making that mistake.
6329 The IN, OUT, and ERR arguments were removed; there are other
6330 convenient ways to override these when desired.
6332 The RESULT argument was deleted; this function should never return.
6334 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
6338 ** Guile's header files should no longer conflict with your system's
6341 In order to compile code which #included <libguile.h>, previous
6342 versions of Guile required you to add a directory containing all the
6343 Guile header files to your #include path. This was a problem, since
6344 Guile's header files have names which conflict with many systems'
6347 Now only <libguile.h> need appear in your #include path; you must
6348 refer to all Guile's other header files as <libguile/mumble.h>.
6349 Guile's installation procedure puts libguile.h in $(includedir), and
6350 the rest in $(includedir)/libguile.
6353 ** Two new C functions, scm_protect_object and scm_unprotect_object,
6354 have been added to the Guile library.
6356 scm_protect_object (OBJ) protects OBJ from the garbage collector.
6357 OBJ will not be freed, even if all other references are dropped,
6358 until someone does scm_unprotect_object (OBJ). Both functions
6361 Note that calls to scm_protect_object do not nest. You can call
6362 scm_protect_object any number of times on a given object, and the
6363 next call to scm_unprotect_object will unprotect it completely.
6365 Basically, scm_protect_object and scm_unprotect_object just
6366 maintain a list of references to things. Since the GC knows about
6367 this list, all objects it mentions stay alive. scm_protect_object
6368 adds its argument to the list; scm_unprotect_object remove its
6369 argument from the list.
6372 ** scm_eval_0str now returns the value of the last expression
6375 ** The new function scm_read_0str reads an s-expression from a
6376 null-terminated string, and returns it.
6378 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
6379 to a Scheme port object.
6381 ** The new function `scm_set_program_arguments' allows C code to set
6382 the value returned by the Scheme `program-arguments' function.
6387 * Guile no longer includes sophisticated Tcl/Tk support.
6389 The old Tcl/Tk support was unsatisfying to us, because it required the
6390 user to link against the Tcl library, as well as Tk and Guile. The
6391 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
6392 referring to widgets by names, rather than exporting widgets to Scheme
6393 code as a special datatype.
6395 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
6396 maintainers described some very interesting changes in progress to the
6397 Tcl/Tk internals, which would facilitate clean interfaces between lone
6398 Tk and other interpreters --- even for garbage-collected languages
6399 like Scheme. They expected the new Tk to be publicly available in the
6402 Since it seems that Guile might soon have a new, cleaner interface to
6403 lone Tk, and that the old Guile/Tk glue code would probably need to be
6404 completely rewritten, we (Jim Blandy and Richard Stallman) have
6405 decided not to support the old code. We'll spend the time instead on
6406 a good interface to the newer Tk, as soon as it is available.
6408 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
6411 Copyright information:
6413 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
6415 Permission is granted to anyone to make or distribute verbatim copies
6416 of this document as received, in any medium, provided that the
6417 copyright notice and this permission notice are preserved,
6418 thus giving the recipient permission to redistribute in turn.
6420 Permission is granted to distribute modified versions
6421 of this document, or of portions of it,
6422 under the above conditions, provided also that they
6423 carry prominent notices stating who last changed them.
6428 paragraph-separate: "[
\f]*$"