1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003 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 ** Checking for duplicate bindings in module system
66 The module system now can check for duplicate imported bindings.
67 The syntax to enable this feature is:
74 This will report an error if both (bar) and (baz) exports a binding
77 The syntax for the :duplicates option is:
79 :duplicates HANDLER-NAME | (HANDLER1-NAME HANDLER2-NAME ...)
81 Specifying multiple handlers is useful since some handlers (such as
82 merge-generics) can defer conflict resolution to others.
84 Currently available duplicates handlers are:
86 check report an error for bindings with a common name
87 first select the first encountered binding (override)
88 last select the last encountered binding (override)
89 merge-generics merge generic functions with a common name
90 into an <extended-generic>
92 ** Merging generic functions
94 It is sometimes tempting to use GOOPS accessors with short names.
95 For example, it is tempting to use the name `x' for the x-coordinate
98 Assume that we work with a graphical package which needs to use two
99 independent vector packages for 2D and 3D vectors respectively. If
100 both packages export `x' we will encounter a name collision.
102 This can now be resolved with the duplicates handler `merge-generics'
103 which merges all generic functions with a common name:
105 (define-module (math 2D-vectors)
106 :use-module (oop goops)
109 (define-module (math 3D-vectors)
110 :use-module (oop goops)
113 (define-module (my-module)
114 :use-module (math 2D-vectors)
115 :use-module (math 3D-vectors)
116 :duplicates merge-generics)
118 x in (my-module) will now share methods with x in both imported
121 The detailed rule for method visibility is this:
123 Let's call the imported generic functions the "ancestor functions".
124 x in (my-module) is, in turn, a "descendant function" of the imported
125 functions. For any generic function gf, the applicable methods are
126 selected from the union of the methods of the descendant functions,
127 the methods of gf and the methods of the ancestor functions.
129 This implies that x in (math 2D-vectors) can see the methods of x in
130 (my-module) and vice versa, while x in (math 2D-vectors) doesn't see
131 the methods of x in (math 3D-vectors), thus preserving modularity.
133 If duplicates checking is desired in the above example, the following
134 form of the :duplicates option can be used instead:
136 :duplicates (merge-generics check)
138 ** New function: effective-version
140 Returns the "effective" version number. This is just the normal full
141 version string without the final micro-version number. See "Changes
142 to the distribution" above.
146 Futures is a way of providing an alternative evaluation policy, very
147 similar in principle to "promises". Like promises, futures allow the
148 main process to continue instantly, but while promises postpone
149 evaluation ("lazy" evaluation) until the value is requested, futures
150 immediately starts evaluation in a parallel thread.
152 Futures are good when you want to express that "I'll need the value of
153 this computation sometime soon" and want to allow processing to go on
154 in the background until that time arrives.
156 ** New syntax: future FORM
158 Begin evaluation of FORM in a parallel thread and return the future
159 immediately. (Akin to 'delay'.)
161 ** New procedure: future-ref FUTURE
163 Return the computed value of the future. Wait if the computation is
164 not finished. (Akin to 'force'.)
166 ** New syntax: parallel FORM ...
168 Compute the results of FORM ... in parallel (in a separate thread for
169 each form) and return them as multiple values.
171 ** New syntax: letpar ((VAR EXP) ...) BODYFORM ...
173 Like 'let' but evaluates the binding expressions EXP ... in parallel.
175 ** New functions: par-map, par-for-each PROC ARGLIST ...
177 Like 'map' and 'for-each' but evaluate the procedure PROC in a
178 separate thread for each (set of) argument(s). All applications are
179 guaranteed to be completed before the procedure returns.
181 ** New functions: n-par-map, n-par-for-each N PROC ARGLIST ...
183 Like 'par-map' and 'par-for-each' but evaluate the procedure PROC in N
184 threads. This is useful when PROC uses large amounts of resources
185 and/or the argument list(s) is/are long so that one thread per (set
186 of) argument(s) would consume too much system resources. On a
187 dual-CPU system, N = 4 would often be a good choice.
189 ** Fair mutexes and condition variables
191 Fair mutexes and condition variables have been added. The fairness
192 means that scheduling is arranged to give as equal time shares as
193 possible and that threads are awakened in a first-in-first-out
194 manner. This is not guaranteed with standard mutexes and condition
197 In addition, fair mutexes are recursive. Locking a fair mutex that
198 you have already locked will succeed. Every call to lock-mutex must
199 be matched with a call to unlock-mutex. Only the last call to
200 unlock-mutex will actually unlock the mutex.
202 A fair condition variable must be used together with a fair mutex,
203 just as a standard condition variable must be used together with a
206 ** New functions: make-fair-mutex, make-fair-condition-variable'
208 Make a new fair mutex and a new fair condition variable respectively.
210 ** New function 'try-mutex'.
212 This function will attempt to lock a mutex but will return immediately
213 instead if blocking and indicate failure.
215 ** Waiting on a condition variable can have a timeout.
217 The funtion 'wait-condition-variable' now takes a third, optional
218 argument that specifies the point in time where the waiting should be
221 ** New function 'broadcast-condition-variable'.
223 ** New functions 'all-threads' and 'current-thread'.
225 ** Signals and system asyncs work better with threads.
227 The function 'sigaction' now takes a fourth, optional, argument that
228 specifies the thread that the handler should run in. When the
229 argument is omitted, the handler will run in the thread that called
232 Likewise, 'system-async-mark' takes a second, optional, argument that
233 specifies the thread that the async should run in. When it is
234 omitted, the async will run in the thread that called
237 C code can use the new functions scm_sigaction_for_thread and
238 scm_system_async_mark_for_thread to pass the new thread argument.
240 ** The function 'system-async' is deprecated.
242 You can now pass any zero-argument procedure to 'system-async-mark'.
243 The function 'system-async' will just return its argument unchanged
246 ** New functions 'call-with-blocked-asyncs' and
247 'call-with-unblocked-asyncs'
249 The expression (call-with-blocked-asyncs PROC) will call PROC and will
250 block execution of system asyncs for the current thread by one level
251 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
252 procedure and will unblock the execution of system asyncs by one
253 level for the current thread.
255 Only system asyncs are affected by these functions.
257 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
259 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
260 instead. Those functions are easier to use correctly and can be
263 ** New function 'unsetenv'.
265 ** New macro 'define-syntax-public'.
267 It works like 'define-syntax' and also exports the defined macro (but
270 ** There is support for Infinity and NaNs.
272 Following PLT Scheme, Guile can now work with infinite numbers, and
275 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
276 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
277 "+nan.0"). These numbers are inexact and have no exact counterpart.
279 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
280 sign of the dividend. The infinities are integers, and they answer #t
281 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
282 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
293 ERROR: Numerical overflow
295 Two new predicates 'inf?' and 'nan?' can be used to test for the
298 ** Inexact zero can have a sign.
300 Guile can now distinguish between plus and minus inexact zero, if your
301 platform supports this, too. The two zeros are equal according to
302 '=', but not according to 'eqv?'. For example
313 ** We now have uninterned symbols.
315 The new function 'make-symbol' will return a uninterned symbol. This
316 is a symbol that is unique and is guaranteed to remain unique.
317 However, uninterned symbols can not yet be read back in.
319 Use the new function 'symbol-interned?' to check whether a symbol is
322 ** pretty-print has more options.
324 The function pretty-print from the (ice-9 pretty-print) module can now
325 also be invoked with keyword arguments that control things like
326 maximum output width. See its online documentation.
328 ** Variables have no longer a special behavior for `equal?'.
330 Previously, comparing two variables with `equal?' would recursivly
331 compare their values. This is no longer done. Variables are now only
332 `equal?' if they are `eq?'.
334 ** `(begin)' is now valid.
336 You can now use an empty `begin' form. It will yield #<unspecified>
337 when evaluated and simply be ignored in a definition context.
339 ** Removed: substring-move-left!, substring-move-right!
341 Use `substring-move!' instead.
343 ** Deprecated: procedure->macro
345 Change your code to use either procedure->memoizing-macro or, probably better,
346 to use r5rs macros. Also, be aware that macro expansion will not be done
347 during evaluation, but prior to evaluation.
349 ** Soft ports now allow a `char-ready?' procedure
351 The vector argument to `make-soft-port' can now have a length of
352 either 5 or 6. (Previously the length had to be 5.) The optional 6th
353 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
354 that returns the number of characters that can be read immediately
355 without the soft port blocking.
357 ** New debugging feature: breakpoints.
359 Guile now has breakpoints. For details see the `Debugging Features'
360 chapter in the reference manual.
362 ** Deprecated: undefine
364 There is no replacement for undefine.
366 * Changes to the C interface
368 ** New function: scm_effective_version
370 Returns the "effective" version number. This is just the normal full
371 version string without the final micro-version number. See "Changes
372 to the distribution" above.
374 ** The function scm_call_with_new_thread has a new prototype.
376 Instead of taking a list with the thunk and handler, these two
377 arguments are now passed directly:
379 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
381 This is an incompatible change.
383 ** The value 'scm_mask_ints' is no longer writable.
385 Previously, you could set scm_mask_ints directly. This is no longer
386 possible. Use scm_c_call_with_blocked_asyncs and
387 scm_c_call_with_unblocked_asyncs instead.
389 ** New functions scm_c_call_with_blocked_asyncs and
390 scm_c_call_with_unblocked_asyncs
392 Like scm_call_with_blocked_asyncs etc. but for C functions.
394 ** New snarfer macro SCM_DEFINE_PUBLIC.
396 This is like SCM_DEFINE, but also calls scm_c_export for the defined
397 function in the init section.
399 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
401 ** New macros SCM_VECTOR_REF and SCM_VECTOR_SET.
403 Use these in preference to SCM_VELTS.
405 ** The SCM_VELTS macros now returns a read-only vector. For writing,
406 use the new macros SCM_WRITABLE_VELTS or SCM_VECTOR_SET. The use of
407 SCM_WRITABLE_VELTS is discouraged, though.
409 ** Garbage collector rewrite.
411 The garbage collector is cleaned up a lot, and now uses lazy
412 sweeping. This is reflected in the output of (gc-stats); since cells
413 are being freed when they are allocated, the cells-allocated field
414 stays roughly constant.
416 For malloc related triggers, the behavior is changed. It uses the same
417 heuristic as the cell-triggered collections. It may be tuned with the
418 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
419 for minimum yield of malloc related triggers. The default is 40.
420 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
423 Debugging operations for the freelist have been deprecated, along with
424 the C variables that control garbage collection. The environment
425 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
426 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
428 ** The function scm_definedp has been renamed to scm_defined_p
430 The name scm_definedp is deprecated.
432 ** The struct scm_cell has been renamed to scm_t_cell
434 This is in accordance to Guile's naming scheme for types. Note that
435 the name scm_cell is now used for a function that allocates and
436 initializes a new cell (see below).
438 ** New functions for memory management
440 A new set of functions for memory management has been added since the
441 old way (scm_must_malloc, scm_must_free, etc) was error prone and
442 indeed, Guile itself contained some long standing bugs that could
443 cause aborts in long running programs.
445 The new functions are more symmetrical and do not need cooperation
446 from smob free routines, among other improvements.
448 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
449 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
450 scm_gc_free, scm_gc_register_collectable_memory, and
451 scm_gc_unregister_collectable_memory. Refer to the manual for more
452 details and for upgrading instructions.
454 The old functions for memory management have been deprecated. They
455 are: scm_must_malloc, scm_must_realloc, scm_must_free,
456 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
458 ** New function: scm_str2string
460 This function creates a scheme string from a 0-terminated C string. The input
463 ** Declarations of exported features are marked with SCM_API.
465 Every declaration of a feature that belongs to the exported Guile API
466 has been marked by adding the macro "SCM_API" to the start of the
467 declaration. This macro can expand into different things, the most
468 common of which is just "extern" for Unix platforms. On Win32, it can
469 be used to control which symbols are exported from a DLL.
471 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
472 will expand into "__declspec (dllimport) extern", which is needed for
473 linking to the Guile DLL in Windows.
475 There are also SCM_RL_IMPORT, QT_IMPORT, SCM_SRFI1314_IMPORT, and
476 SCM_SRFI4_IMPORT, for the corresponding libraries.
478 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
480 Use the new functions scm_cell and scm_double_cell instead. The old macros
481 had problems because with them allocation and initialization was separated and
482 the GC could sometimes observe half initialized cells. Only careful coding by
483 the user of SCM_NEWCELL and SCM_NEWCELL2 could make this safe and efficient.
485 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
487 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
490 ** SRCBRKP has been deprecated.
492 Use scm_c_source_property_breakpoint_p instead.
494 ** Deprecated: scm_makmacro
496 Change your code to use either scm_makmmacro or, probably better, to use r5rs
497 macros. Also, be aware that macro expansion will not be done during
498 evaluation, but prior to evaluation.
500 ** Removed from scm_root_state: def_inp, def_outp, def_errp, together
501 with corresponding macros scm_def_inp, scm_def_outp and scm_def_errp.
502 These were undocumented and unused copies of the standard ports at the
503 time that Guile was initialised. Normally the current ports should be
504 used instead, obtained from scm_current_input_port () etc. If an
505 application needs to retain earlier ports, it should save them in a
506 gc-protected location.
508 ** Removed compile time option MEMOIZE_LOCALS
510 Now, caching of local variable positions during memoization is mandatory.
511 However, the option to disable the caching has most probably not been used
514 ** Removed compile time option SCM_RECKLESS
516 Full number of arguments checking of closures is mandatory now. However, the
517 option to disable the checking has most probably not been used anyway.
519 ** Removed compile time option SCM_CAUTIOUS
521 Full number of arguments checking of closures is mandatory now. However, the
522 option to disable the checking has most probably not been used anyway.
524 ** Deprecated configure flags USE_THREADS and GUILE_ISELECT
526 Previously, when the C preprocessor macro USE_THREADS was defined,
527 libguile included a thread API. This API is now always included, even
528 when threads are not really supported. Thus, you don't need to test
531 Analogously, GUILE_ISELECT was defined when the function
532 scm_internal_select was provided by Guile. This function is now
533 always defined, and GUILE_ISELECT with it.
535 ** Removed definitions: scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify,
536 s_t_ify, scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
537 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2, scm_tc16_allocated,
538 SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY, SCM_IM_0_COND,
539 SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED, scm_debug_newcell,
540 scm_debug_newcell2, scm_substring_move_left_x, scm_substring_move_right_x,
541 long_long, ulong_long, scm_sizet, SCM_WNA, SCM_OUTOFRANGE, SCM_NALLOC,
542 SCM_HUP_SIGNAL, SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL,
543 SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL,
544 SCM_SIG_ORD, SCM_ORD_SIG, SCM_NUM_SIGS, moddata, registered_mods,
545 scm_register_module_xxx, scm_registered_modules,
546 scm_clear_registered_modules, scm_wta, *top-level-lookup-closure*,
547 scm_top_level_lookup_closure_var, scm_system_transformer, scm_eval_3,
548 scm_eval2, SCM_SETAND_CAR, SCM_SETOR_CAR, SCM_SETAND_CDR, SCM_SETOR_CDR,
549 SCM_FREEP, SCM_NFREEP, SCM_GC8MARKP, SCM_SETGC8MARK, SCM_CLRGC8MARK,
550 SCM_GCTYP16, SCM_GCCDR, scm_remember, scm_protect_object,
551 scm_unprotect_object, root_module_lookup_closure, scm_sym_app,
552 scm_sym_modules, module_prefix, make_modules_in_var,
553 beautify_user_module_x_var, try_module_autoload_var, scm_module_full_name,
554 scm_the_root_module, scm_make_module, scm_ensure_user_module,
555 scm_load_scheme_module, scm_port, scm_ptob_descriptor, scm_port_rw_active,
556 scm_close_all_ports_except, scm_rstate, scm_rng, scm_i_rstate,
557 SCM_SLOPPY_STRINGP, SCM_RWSTRINGP, SCM_STRING_UCHARS, SCM_STRING_CHARS,
558 scm_read_only_string_p, scm_makstr, scm_makfromstr,
559 scm_make_shared_substring, scm_tc7_substring, SCM_SLOPPY_CONSP,
560 SCM_SLOPPY_NCONSP, scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
561 sym_huh, scm_variable_set_name_hint, scm_builtin_variable, SCM_VARVCELL,
562 SCM_UDVARIABLEP, SCM_DEFVARIABLEP, scm_internal_with_fluids,
563 scm_make_gsubr, scm_make_gsubr_with_generic, scm_create_hook, list*,
564 SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4, SCM_LIST5,
565 SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9, scm_listify, scm_sloppy_memq,
566 scm_sloppy_memv, scm_sloppy_member, scm_end_of_file_key,
567 scm_read_and_eval_x, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
568 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
569 scm_subr_entry, SCM_SUBR_DOC, scm_make_subr_opt, scm_make_subr,
570 scm_make_subr_with_generic, setjmp_type, setjmp_type,
571 scm_call_catching_errors, scm_make_smob_type_mfpe, scm_set_smob_mfpe,
572 scm_strprint_obj, scm_read_0str, scm_eval_0str, SCM_CHARS, SCM_UCHARS,
573 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
574 SCM_LENGTH_MAX, SCM_LENGTH, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
575 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR, scm_strhash,
576 scm_sym2vcell, scm_sym2ovcell_soft, scm_sym2ovcell,
577 scm_intern_obarray_soft, scm_intern_obarray, scm_intern, scm_intern0,
578 scm_sysintern, scm_sysintern0, scm_sysintern0_no_module_lookup,
579 scm_symbol_value0, scm_string_to_obarray_symbol, scm_intern_symbol,
580 scm_unintern_symbol, scm_symbol_binding, scm_symbol_interned_p,
581 scm_symbol_bound_p, scm_symbol_set_x, scm_gentemp,
582 scm_init_symbols_deprecated, s_vector_set_length_x, scm_vector_set_length_x,
583 scm_contregs, scm_debug_info, scm_debug_frame, SCM_DSIDEVAL, SCM_OPDIRP,
584 scm_fport, scm_option, SCM_CONST_LONG, SCM_VCELL, SCM_GLOBAL_VCELL,
585 SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT, scm_srcprops, scm_srcprops_chunk,
586 scm_info_frame, scm_stack, scm_array, scm_array_dim, SCM_ARRAY_CONTIGUOUS,
587 SCM_HUGE_LENGTH, SCM_FUNC_NAME, SCM_WTA, RETURN_SCM_WTA,
588 SCM_VALIDATE_NUMBER_COPY, SCM_VALIDATE_NUMBER_DEF_COPY,
589 SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
590 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
591 SCM_VALIDATE_RWSTRING, SCM_VALIDATE_OPDIR, DIGITS, scm_small_istr2int,
592 scm_istr2int, scm_istr2flo, scm_istring2number, scm_istr2int,
593 scm_istr2flo, scm_istring2number, scm_vtable_index_vcell, scm_si_vcell,
594 SCM_ECONSP, SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
595 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable
597 Changes since Guile 1.4:
599 * Changes to the distribution
601 ** A top-level TODO file is included.
603 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
605 Guile now always uses three numbers to represent the version,
606 i.e. "1.6.5". The first number, 1, is the major version number, the
607 second number, 6, is the minor version number, and the third number,
608 5, is the micro version number. Changes in major version number
609 indicate major changes in Guile.
611 Minor version numbers that are even denote stable releases, and odd
612 minor version numbers denote development versions (which may be
613 unstable). The micro version number indicates a minor sub-revision of
614 a given MAJOR.MINOR release.
616 In keeping with the new scheme, (minor-version) and scm_minor_version
617 no longer return everything but the major version number. They now
618 just return the minor version number. Two new functions
619 (micro-version) and scm_micro_version have been added to report the
620 micro version number.
622 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
624 ** New preprocessor definitions are available for checking versions.
626 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
627 SCM_MICRO_VERSION to the appropriate integer values.
629 ** Guile now actively warns about deprecated features.
631 The new configure option `--enable-deprecated=LEVEL' and the
632 environment variable GUILE_WARN_DEPRECATED control this mechanism.
633 See INSTALL and README for more information.
635 ** Guile is much more likely to work on 64-bit architectures.
637 Guile now compiles and passes "make check" with only two UNRESOLVED GC
638 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
639 for the use of a test machine, and thanks to Stefan Jahn for ia64
642 ** New functions: setitimer and getitimer.
644 These implement a fairly direct interface to the libc functions of the
647 ** The #. reader extension is now disabled by default.
649 For safety reasons, #. evaluation is disabled by default. To
650 re-enable it, set the fluid read-eval? to #t. For example:
652 (fluid-set! read-eval? #t)
654 but make sure you realize the potential security risks involved. With
655 read-eval? enabled, reading a data file from an untrusted source can
658 ** New SRFI modules have been added:
660 SRFI-0 `cond-expand' is now supported in Guile, without requiring
663 (srfi srfi-1) is a library containing many useful pair- and list-processing
666 (srfi srfi-2) exports and-let*.
668 (srfi srfi-4) implements homogeneous numeric vector datatypes.
670 (srfi srfi-6) is a dummy module for now, since guile already provides
671 all of the srfi-6 procedures by default: open-input-string,
672 open-output-string, get-output-string.
674 (srfi srfi-8) exports receive.
676 (srfi srfi-9) exports define-record-type.
678 (srfi srfi-10) exports define-reader-ctor and implements the reader
681 (srfi srfi-11) exports let-values and let*-values.
683 (srfi srfi-13) implements the SRFI String Library.
685 (srfi srfi-14) implements the SRFI Character-Set Library.
687 (srfi srfi-17) implements setter and getter-with-setter and redefines
688 some accessor procedures as procedures with getters. (such as car,
689 cdr, vector-ref etc.)
691 (srfi srfi-19) implements the SRFI Time/Date Library.
693 ** New scripts / "executable modules"
695 Subdirectory "scripts" contains Scheme modules that are packaged to
696 also be executable as scripts. At this time, these scripts are available:
705 See README there for more info.
707 These scripts can be invoked from the shell with the new program
708 "guile-tools", which keeps track of installation directory for you.
711 $ guile-tools display-commentary srfi/*.scm
713 guile-tools is copied to the standard $bindir on "make install".
715 ** New module (ice-9 stack-catch):
717 stack-catch is like catch, but saves the current state of the stack in
718 the fluid the-last-stack. This fluid can be useful when using the
719 debugger and when re-throwing an error.
721 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
723 This has been done to prevent problems on lesser operating systems
724 that can't tolerate `*'s in file names. The exported macro continues
725 to be named `and-let*', of course.
727 On systems that support it, there is also a compatibility module named
728 (ice-9 and-let*). It will go away in the next release.
730 ** New modules (oop goops) etc.:
735 (oop goops active-slot)
736 (oop goops composite-slot)
738 The Guile Object Oriented Programming System (GOOPS) has been
739 integrated into Guile. For further information, consult the GOOPS
740 manual and tutorial in the `doc' directory.
742 ** New module (ice-9 rdelim).
744 This exports the following procedures which were previously defined
745 in the default environment:
747 read-line read-line! read-delimited read-delimited! %read-delimited!
748 %read-line write-line
750 For backwards compatibility the definitions are still imported into the
751 default environment in this version of Guile. However you should add:
753 (use-modules (ice-9 rdelim))
755 to any program which uses the definitions, since this may change in
758 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
759 can be used for similar functionality.
761 ** New module (ice-9 rw)
763 This is a subset of the (scsh rw) module from guile-scsh. Currently
764 it defines two procedures:
766 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
768 Read characters from a port or file descriptor into a string STR.
769 A port must have an underlying file descriptor -- a so-called
770 fport. This procedure is scsh-compatible and can efficiently read
773 *** New function: write-string/partial str [port_or_fdes [start [end]]]
775 Write characters from a string STR to a port or file descriptor.
776 A port must have an underlying file descriptor -- a so-called
777 fport. This procedure is mostly compatible and can efficiently
780 ** New module (ice-9 match)
782 This module includes Andrew K. Wright's pattern matcher. See
783 ice-9/match.scm for brief description or
785 http://www.star-lab.com/wright/code.html
787 for complete documentation.
789 ** New module (ice-9 buffered-input)
791 This module provides procedures to construct an input port from an
792 underlying source of input that reads and returns its input in chunks.
793 The underlying input source is a Scheme procedure, specified by the
794 caller, which the port invokes whenever it needs more input.
796 This is useful when building an input port whose back end is Readline
797 or a UI element such as the GtkEntry widget.
801 The reference and tutorial documentation that was previously
802 distributed separately, as `guile-doc', is now included in the core
803 Guile distribution. The documentation consists of the following
806 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
809 - The Guile Reference Manual (guile.texi) contains (or is intended to
810 contain) reference documentation on all aspects of Guile.
812 - The GOOPS Manual (goops.texi) contains both tutorial-style and
813 reference documentation for using GOOPS, Guile's Object Oriented
816 - The Revised^5 Report on the Algorithmic Language Scheme
819 See the README file in the `doc' directory for more details.
821 ** There are a couple of examples in the examples/ directory now.
823 * Changes to the stand-alone interpreter
825 ** New command line option `--use-srfi'
827 Using this option, SRFI modules can be loaded on startup and be
828 available right from the beginning. This makes programming portable
829 Scheme programs easier.
831 The option `--use-srfi' expects a comma-separated list of numbers,
832 each representing a SRFI number to be loaded into the interpreter
833 before starting evaluating a script file or the REPL. Additionally,
834 the feature identifier for the loaded SRFIs is recognized by
835 `cond-expand' when using this option.
838 $ guile --use-srfi=8,13
839 guile> (receive (x z) (values 1 2) (+ 1 2))
841 guile> (string-pad "bla" 20)
844 ** Guile now always starts up in the `(guile-user)' module.
846 Previously, scripts executed via the `-s' option would run in the
847 `(guile)' module and the repl would run in the `(guile-user)' module.
848 Now every user action takes place in the `(guile-user)' module by
851 * Changes to Scheme functions and syntax
853 ** Character classifiers work for non-ASCII characters.
855 The predicates `char-alphabetic?', `char-numeric?',
856 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
857 no longer check whether their arguments are ASCII characters.
858 Previously, a character would only be considered alphabetic when it
859 was also ASCII, for example.
861 ** Previously deprecated Scheme functions have been removed:
863 tag - no replacement.
864 fseek - replaced by seek.
865 list* - replaced by cons*.
867 ** It's now possible to create modules with controlled environments
871 (use-modules (ice-9 safe))
872 (define m (make-safe-module))
873 ;;; m will now be a module containing only a safe subset of R5RS
874 (eval '(+ 1 2) m) --> 3
875 (eval 'load m) --> ERROR: Unbound variable: load
877 ** Evaluation of "()", the empty list, is now an error.
879 Previously, the expression "()" evaluated to the empty list. This has
880 been changed to signal a "missing expression" error. The correct way
881 to write the empty list as a literal constant is to use quote: "'()".
883 ** New concept of `Guile Extensions'.
885 A Guile Extension is just a ordinary shared library that can be linked
886 at run-time. We found it advantageous to give this simple concept a
887 dedicated name to distinguish the issues related to shared libraries
888 from the issues related to the module system.
890 *** New function: load-extension
892 Executing (load-extension lib init) is mostly equivalent to
894 (dynamic-call init (dynamic-link lib))
896 except when scm_register_extension has been called previously.
897 Whenever appropriate, you should use `load-extension' instead of
898 dynamic-link and dynamic-call.
900 *** New C function: scm_c_register_extension
902 This function registers a initialization function for use by
903 `load-extension'. Use it when you don't want specific extensions to
904 be loaded as shared libraries (for example on platforms that don't
905 support dynamic linking).
907 ** Auto-loading of compiled-code modules is deprecated.
909 Guile used to be able to automatically find and link a shared
910 library to satisfy requests for a module. For example, the module
911 `(foo bar)' could be implemented by placing a shared library named
912 "foo/libbar.so" (or with a different extension) in a directory on the
915 This has been found to be too tricky, and is no longer supported. The
916 shared libraries are now called "extensions". You should now write a
917 small Scheme file that calls `load-extension' to load the shared
918 library and initialize it explicitely.
920 The shared libraries themselves should be installed in the usual
921 places for shared libraries, with names like "libguile-foo-bar".
923 For example, place this into a file "foo/bar.scm"
925 (define-module (foo bar))
927 (load-extension "libguile-foo-bar" "foobar_init")
929 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
931 `eval' is now R5RS, that is it takes two arguments.
932 The second argument is an environment specifier, i.e. either
934 (scheme-report-environment 5)
936 (interaction-environment)
942 ** The module system has been made more disciplined.
944 The function `eval' will save and restore the current module around
945 the evaluation of the specified expression. While this expression is
946 evaluated, `(current-module)' will now return the right module, which
947 is the module specified as the second argument to `eval'.
949 A consequence of this change is that `eval' is not particularly
950 useful when you want allow the evaluated code to change what module is
951 designated as the current module and have this change persist from one
952 call to `eval' to the next. The read-eval-print-loop is an example
953 where `eval' is now inadequate. To compensate, there is a new
954 function `primitive-eval' that does not take a module specifier and
955 that does not save/restore the current module. You should use this
956 function together with `set-current-module', `current-module', etc
957 when you want to have more control over the state that is carried from
958 one eval to the next.
960 Additionally, it has been made sure that forms that are evaluated at
961 the top level are always evaluated with respect to the current module.
962 Previously, subforms of top-level forms such as `begin', `case',
963 etc. did not respect changes to the current module although these
964 subforms are at the top-level as well.
966 To prevent strange behavior, the forms `define-module',
967 `use-modules', `use-syntax', and `export' have been restricted to only
968 work on the top level. The forms `define-public' and
969 `defmacro-public' only export the new binding on the top level. They
970 behave just like `define' and `defmacro', respectively, when they are
971 used in a lexical environment.
973 Also, `export' will no longer silently re-export bindings imported
974 from a used module. It will emit a `deprecation' warning and will
975 cease to perform any re-export in the next version. If you actually
976 want to re-export bindings, use the new `re-export' in place of
977 `export'. The new `re-export' will not make copies of variables when
978 rexporting them, as `export' did wrongly.
980 ** Module system now allows selection and renaming of imported bindings
982 Previously, when using `use-modules' or the `#:use-module' clause in
983 the `define-module' form, all the bindings (association of symbols to
984 values) for imported modules were added to the "current module" on an
985 as-is basis. This has been changed to allow finer control through two
986 new facilities: selection and renaming.
988 You can now select which of the imported module's bindings are to be
989 visible in the current module by using the `:select' clause. This
990 clause also can be used to rename individual bindings. For example:
992 ;; import all bindings no questions asked
993 (use-modules (ice-9 common-list))
995 ;; import four bindings, renaming two of them;
996 ;; the current module sees: every some zonk-y zonk-n
997 (use-modules ((ice-9 common-list)
1000 (remove-if-not . zonk-n))))
1002 You can also programmatically rename all selected bindings using the
1003 `:renamer' clause, which specifies a proc that takes a symbol and
1004 returns another symbol. Because it is common practice to use a prefix,
1005 we now provide the convenience procedure `symbol-prefix-proc'. For
1008 ;; import four bindings, renaming two of them specifically,
1009 ;; and all four w/ prefix "CL:";
1010 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1011 (use-modules ((ice-9 common-list)
1013 (remove-if . zonk-y)
1014 (remove-if-not . zonk-n))
1015 :renamer (symbol-prefix-proc 'CL:)))
1017 ;; import four bindings, renaming two of them specifically,
1018 ;; and all four by upcasing.
1019 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1020 (define (upcase-symbol sym)
1021 (string->symbol (string-upcase (symbol->string sym))))
1023 (use-modules ((ice-9 common-list)
1025 (remove-if . zonk-y)
1026 (remove-if-not . zonk-n))
1027 :renamer upcase-symbol))
1029 Note that programmatic renaming is done *after* individual renaming.
1030 Also, the above examples show `use-modules', but the same facilities are
1031 available for the `#:use-module' clause of `define-module'.
1033 See manual for more info.
1035 ** The semantics of guardians have changed.
1037 The changes are for the most part compatible. An important criterion
1038 was to keep the typical usage of guardians as simple as before, but to
1039 make the semantics safer and (as a result) more useful.
1041 *** All objects returned from guardians are now properly alive.
1043 It is now guaranteed that any object referenced by an object returned
1044 from a guardian is alive. It's now impossible for a guardian to
1045 return a "contained" object before its "containing" object.
1047 One incompatible (but probably not very important) change resulting
1048 from this is that it is no longer possible to guard objects that
1049 indirectly reference themselves (i.e. are parts of cycles). If you do
1050 so accidentally, you'll get a warning.
1052 *** There are now two types of guardians: greedy and sharing.
1054 If you call (make-guardian #t) or just (make-guardian), you'll get a
1055 greedy guardian, and for (make-guardian #f) a sharing guardian.
1057 Greedy guardians are the default because they are more "defensive".
1058 You can only greedily guard an object once. If you guard an object
1059 more than once, once in a greedy guardian and the rest of times in
1060 sharing guardians, then it is guaranteed that the object won't be
1061 returned from sharing guardians as long as it is greedily guarded
1064 Guardians returned by calls to `make-guardian' can now take one more
1065 optional parameter, which says whether to throw an error in case an
1066 attempt is made to greedily guard an object that is already greedily
1067 guarded. The default is true, i.e. throw an error. If the parameter
1068 is false, the guardian invocation returns #t if guarding was
1069 successful and #f if it wasn't.
1071 Also, since greedy guarding is, in effect, a side-effecting operation
1072 on objects, a new function is introduced: `destroy-guardian!'.
1073 Invoking this function on a guardian renders it unoperative and, if
1074 the guardian is greedy, clears the "greedily guarded" property of the
1075 objects that were guarded by it, thus undoing the side effect.
1077 Note that all this hair is hardly very important, since guardian
1078 objects are usually permanent.
1080 ** Continuations created by call-with-current-continuation now accept
1081 any number of arguments, as required by R5RS.
1083 ** New function `issue-deprecation-warning'
1085 This function is used to display the deprecation messages that are
1086 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1089 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1093 ;; `id' is deprecated. Use `identity' instead.
1098 ** New syntax `begin-deprecated'
1100 When deprecated features are included (as determined by the configure
1101 option --enable-deprecated), `begin-deprecated' is identical to
1102 `begin'. When deprecated features are excluded, it always evaluates
1103 to `#f', ignoring the body forms.
1105 ** New function `make-object-property'
1107 This function returns a new `procedure with setter' P that can be used
1108 to attach a property to objects. When calling P as
1112 where `obj' is any kind of object, it attaches `val' to `obj' in such
1113 a way that it can be retrieved by calling P as
1117 This function will replace procedure properties, symbol properties and
1118 source properties eventually.
1120 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1122 Instead of #&optional, #&key, etc you should now use #:optional,
1123 #:key, etc. Since #:optional is a keyword, you can write it as just
1124 :optional when (read-set! keywords 'prefix) is active.
1126 The old reader syntax `#&' is still supported, but deprecated. It
1127 will be removed in the next release.
1129 ** New define-module option: pure
1131 Tells the module system not to include any bindings from the root
1136 (define-module (totally-empty-module)
1139 ** New define-module option: export NAME1 ...
1141 Export names NAME1 ...
1143 This option is required if you want to be able to export bindings from
1144 a module which doesn't import one of `define-public' or `export'.
1148 (define-module (foo)
1150 :use-module (ice-9 r5rs)
1153 ;;; Note that we're pure R5RS below this point!
1158 ** New function: object->string OBJ
1160 Return a Scheme string obtained by printing a given object.
1162 ** New function: port? X
1164 Returns a boolean indicating whether X is a port. Equivalent to
1165 `(or (input-port? X) (output-port? X))'.
1167 ** New function: file-port?
1169 Determines whether a given object is a port that is related to a file.
1171 ** New function: port-for-each proc
1173 Apply PROC to each port in the Guile port table in turn. The return
1174 value is unspecified. More specifically, PROC is applied exactly once
1175 to every port that exists in the system at the time PORT-FOR-EACH is
1176 invoked. Changes to the port table while PORT-FOR-EACH is running
1177 have no effect as far as PORT-FOR-EACH is concerned.
1179 ** New function: dup2 oldfd newfd
1181 A simple wrapper for the `dup2' system call. Copies the file
1182 descriptor OLDFD to descriptor number NEWFD, replacing the
1183 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1184 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1185 to move away ports which are using NEWFD. The return value is
1188 ** New function: close-fdes fd
1190 A simple wrapper for the `close' system call. Close file
1191 descriptor FD, which must be an integer. Unlike close (*note
1192 close: Ports and File Descriptors.), the file descriptor will be
1193 closed even if a port is using it. The return value is
1196 ** New function: crypt password salt
1198 Encrypts `password' using the standard unix password encryption
1201 ** New function: chroot path
1203 Change the root directory of the running process to `path'.
1205 ** New functions: getlogin, cuserid
1207 Return the login name or the user name of the current effective user
1210 ** New functions: getpriority which who, setpriority which who prio
1212 Get or set the priority of the running process.
1214 ** New function: getpass prompt
1216 Read a password from the terminal, first displaying `prompt' and
1219 ** New function: flock file operation
1221 Set/remove an advisory shared or exclusive lock on `file'.
1223 ** New functions: sethostname name, gethostname
1225 Set or get the hostname of the machine the current process is running
1228 ** New function: mkstemp! tmpl
1230 mkstemp creates a new unique file in the file system and returns a
1231 new buffered port open for reading and writing to the file. TMPL
1232 is a string specifying where the file should be created: it must
1233 end with `XXXXXX' and will be changed in place to return the name
1234 of the temporary file.
1236 ** New function: open-input-string string
1238 Return an input string port which delivers the characters from
1239 `string'. This procedure, together with `open-output-string' and
1240 `get-output-string' implements SRFI-6.
1242 ** New function: open-output-string
1244 Return an output string port which collects all data written to it.
1245 The data can then be retrieved by `get-output-string'.
1247 ** New function: get-output-string
1249 Return the contents of an output string port.
1251 ** New function: identity
1253 Return the argument.
1255 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1256 are represented in Scheme as integers with normal host byte ordering.
1258 ** New function: inet-pton family address
1260 Convert a printable string network address into an integer. Note that
1261 unlike the C version of this function, the result is an integer with
1262 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1265 (inet-pton AF_INET "127.0.0.1") => 2130706433
1266 (inet-pton AF_INET6 "::1") => 1
1268 ** New function: inet-ntop family address
1270 Convert an integer network address into a printable string. Note that
1271 unlike the C version of this function, the input is an integer with
1272 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1275 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1276 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1277 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1281 Use `identity' instead.
1287 ** Deprecated: return-it
1291 ** Deprecated: string-character-length
1293 Use `string-length' instead.
1295 ** Deprecated: flags
1297 Use `logior' instead.
1299 ** Deprecated: close-all-ports-except.
1301 This was intended for closing ports in a child process after a fork,
1302 but it has the undesirable side effect of flushing buffers.
1303 port-for-each is more flexible.
1305 ** The (ice-9 popen) module now attempts to set up file descriptors in
1306 the child process from the current Scheme ports, instead of using the
1307 current values of file descriptors 0, 1, and 2 in the parent process.
1309 ** Removed function: builtin-weak-bindings
1311 There is no such concept as a weak binding any more.
1313 ** Removed constants: bignum-radix, scm-line-incrementors
1315 ** define-method: New syntax mandatory.
1317 The new method syntax is now mandatory:
1319 (define-method (NAME ARG-SPEC ...) BODY ...)
1320 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1322 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1323 REST-ARG ::= ARG-NAME
1325 If you have old code using the old syntax, import
1326 (oop goops old-define-method) before (oop goops) as in:
1328 (use-modules (oop goops old-define-method) (oop goops))
1330 ** Deprecated function: builtin-variable
1331 Removed function: builtin-bindings
1333 There is no longer a distinction between builtin or other variables.
1334 Use module system operations for all variables.
1336 ** Lazy-catch handlers are no longer allowed to return.
1338 That is, a call to `throw', `error', etc is now guaranteed to not
1341 ** Bugfixes for (ice-9 getopt-long)
1343 This module is now tested using test-suite/tests/getopt-long.test.
1344 The following bugs have been fixed:
1346 *** Parsing for options that are specified to have `optional' args now checks
1347 if the next element is an option instead of unconditionally taking it as the
1350 *** An error is now thrown for `--opt=val' when the option description
1351 does not specify `(value #t)' or `(value optional)'. This condition used to
1352 be accepted w/o error, contrary to the documentation.
1354 *** The error message for unrecognized options is now more informative.
1355 It used to be "not a record", an artifact of the implementation.
1357 *** The error message for `--opt' terminating the arg list (no value), when
1358 `(value #t)' is specified, is now more informative. It used to be "not enough
1361 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1362 The expansion used to be like so:
1364 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1366 Note that the "5d" is dropped. Now it is like so:
1368 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1370 This enables single-char options to have adjoining arguments as long as their
1371 constituent characters are not potential single-char options.
1373 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1375 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1376 property `arglist', which can be retrieved by `arity'. The result is that
1377 `arity' can give more detailed information than before:
1381 guile> (use-modules (ice-9 optargs))
1382 guile> (define* (foo #:optional a b c) a)
1384 0 or more arguments in `lambda*:G0'.
1389 3 optional arguments: `a', `b' and `c'.
1390 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
1392 2 required arguments: `a' and `b', 2 keyword arguments: `c'
1393 and `d', other keywords allowed.
1394 guile> (define* (baz a b #:optional c #:rest r) a)
1396 2 required arguments: `a' and `b', 1 optional argument: `c',
1399 * Changes to the C interface
1401 ** Types have been renamed from scm_*_t to scm_t_*.
1403 This has been done for POSIX sake. It reserves identifiers ending
1404 with "_t". What a concept.
1406 The old names are still available with status `deprecated'.
1408 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
1410 ** Deprecated features have been removed.
1414 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
1415 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
1417 *** C Functions removed
1419 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
1420 scm_fseek - replaced by scm_seek.
1421 gc-thunk - replaced by after-gc-hook.
1422 gh_int2scmb - replaced by gh_bool2scm.
1423 scm_tc_dblr - replaced by scm_tc16_real.
1424 scm_tc_dblc - replaced by scm_tc16_complex.
1425 scm_list_star - replaced by scm_cons_star.
1427 ** Deprecated: scm_makfromstr
1429 Use scm_mem2string instead.
1431 ** Deprecated: scm_make_shared_substring
1433 Explicit shared substrings will disappear from Guile.
1435 Instead, "normal" strings will be implemented using sharing
1436 internally, combined with a copy-on-write strategy.
1438 ** Deprecated: scm_read_only_string_p
1440 The concept of read-only strings will disappear in next release of
1443 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
1445 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
1447 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
1449 Call a procedure with the indicated number of arguments.
1453 scm_call_1 (proc, arg1);
1455 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
1457 Call a procedure with the indicated number of arguments and a list
1462 scm_apply_1 (proc, arg1, args);
1464 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
1466 Create a list of the given number of elements.
1468 ** Renamed function: scm_listify has been replaced by scm_list_n.
1470 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
1471 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
1473 Use functions scm_list_N instead.
1475 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
1477 Used by an application to read arbitrary number of bytes from a port.
1478 Same semantics as libc read, except that scm_c_read only returns less
1479 than SIZE bytes if at end-of-file.
1481 Warning: Doesn't update port line and column counts!
1483 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
1485 Used by an application to write arbitrary number of bytes to an SCM
1486 port. Similar semantics as libc write. However, unlike libc
1487 write, scm_c_write writes the requested number of bytes and has no
1490 Warning: Doesn't update port line and column counts!
1492 ** New function: scm_init_guile ()
1494 In contrast to scm_boot_guile, scm_init_guile will return normally
1495 after initializing Guile. It is not available on all systems, tho.
1497 ** New functions: scm_str2symbol, scm_mem2symbol
1499 The function scm_str2symbol takes a const char* pointing to a zero-terminated
1500 field of characters and creates a scheme symbol object from that C string.
1501 The function scm_mem2symbol takes a const char* and a number of characters and
1502 creates a symbol from the characters in that memory area.
1504 ** New functions: scm_primitive_make_property
1505 scm_primitive_property_ref
1506 scm_primitive_property_set_x
1507 scm_primitive_property_del_x
1509 These functions implement a new way to deal with object properties.
1510 See libguile/properties.c for their documentation.
1512 ** New function: scm_done_free (long size)
1514 This function is the inverse of scm_done_malloc. Use it to report the
1515 amount of smob memory you free. The previous method, which involved
1516 calling scm_done_malloc with negative argument, was somewhat
1517 unintuitive (and is still available, of course).
1519 ** New function: scm_c_memq (SCM obj, SCM list)
1521 This function provides a fast C level alternative for scm_memq for the case
1522 that the list parameter is known to be a proper list. The function is a
1523 replacement for scm_sloppy_memq, but is stricter in its requirements on its
1524 list input parameter, since for anything else but a proper list the function's
1525 behaviour is undefined - it may even crash or loop endlessly. Further, for
1526 the case that the object is not found in the list, scm_c_memq returns #f which
1527 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
1529 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
1530 scm_remember_upto_here
1532 These functions replace the function scm_remember.
1534 ** Deprecated function: scm_remember
1536 Use one of the new functions scm_remember_upto_here_1,
1537 scm_remember_upto_here_2 or scm_remember_upto_here instead.
1539 ** New function: scm_allocate_string
1541 This function replaces the function scm_makstr.
1543 ** Deprecated function: scm_makstr
1545 Use the new function scm_allocate_string instead.
1547 ** New global variable scm_gc_running_p introduced.
1549 Use this variable to find out if garbage collection is being executed. Up to
1550 now applications have used scm_gc_heap_lock to test if garbage collection was
1551 running, which also works because of the fact that up to know only the garbage
1552 collector has set this variable. But, this is an implementation detail that
1553 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
1554 of this variable is (and has been) not fully safe anyway.
1556 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
1558 Use these instead of SCM_LENGTH_MAX.
1560 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
1561 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
1562 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
1564 Use these instead of SCM_LENGTH.
1566 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
1567 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
1568 SCM_SET_BITVECTOR_LENGTH
1570 Use these instead of SCM_SETLENGTH
1572 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
1573 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
1576 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
1579 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
1580 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
1583 Use these instead of SCM_SETCHARS.
1585 ** New macro: SCM_BITVECTOR_P
1587 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
1589 Use instead of SCM_COERCE_SUBSTR.
1591 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
1593 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
1595 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
1596 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
1597 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
1598 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
1599 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
1600 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
1601 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
1602 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
1603 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
1604 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
1605 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
1606 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
1607 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
1608 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
1609 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
1611 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
1612 Use scm_memory_error instead of SCM_NALLOC.
1613 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
1614 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
1615 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
1616 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
1617 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
1618 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
1619 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
1620 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
1621 Use SCM_STRINGP instead of SCM_RWSTRINGP.
1622 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
1623 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
1624 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
1625 Use a type specific setter macro instead of SCM_SETLENGTH.
1626 Use a type specific setter macro instead of SCM_SETCHARS.
1627 Use a type specific length macro instead of SCM_LENGTH_MAX.
1628 Use SCM_GCMARKP instead of SCM_GC8MARKP.
1629 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
1630 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
1631 Use SCM_TYP16 instead of SCM_GCTYP16.
1632 Use SCM_CDR instead of SCM_GCCDR.
1633 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
1634 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
1635 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
1636 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
1637 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
1638 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
1639 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
1641 ** Removed function: scm_struct_init
1643 ** Removed variable: scm_symhash_dim
1645 ** Renamed function: scm_make_cont has been replaced by
1646 scm_make_continuation, which has a different interface.
1648 ** Deprecated function: scm_call_catching_errors
1650 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
1652 ** Deprecated function: scm_strhash
1654 Use scm_string_hash instead.
1656 ** Deprecated function: scm_vector_set_length_x
1658 Instead, create a fresh vector of the desired size and copy the contents.
1660 ** scm_gensym has changed prototype
1662 scm_gensym now only takes one argument.
1664 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
1667 There is now only a single symbol type scm_tc7_symbol.
1668 The tag scm_tc7_lvector was not used anyway.
1670 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
1672 Use scm_make_smob_type and scm_set_smob_XXX instead.
1674 ** New function scm_set_smob_apply.
1676 This can be used to set an apply function to a smob type.
1678 ** Deprecated function: scm_strprint_obj
1680 Use scm_object_to_string instead.
1682 ** Deprecated function: scm_wta
1684 Use scm_wrong_type_arg, or another appropriate error signalling function
1687 ** Explicit support for obarrays has been deprecated.
1689 Use `scm_str2symbol' and the generic hashtable functions instead.
1691 ** The concept of `vcells' has been deprecated.
1693 The data type `variable' is now used exclusively. `Vcells' have been
1694 a low-level concept so you are likely not affected by this change.
1696 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
1697 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
1699 Use scm_c_define or scm_c_lookup instead, as appropriate.
1701 *** New functions: scm_c_module_lookup, scm_c_lookup,
1702 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
1703 scm_module_define, scm_define.
1705 These functions work with variables instead of with vcells.
1707 ** New functions for creating and defining `subr's and `gsubr's.
1709 The new functions more clearly distinguish between creating a subr (or
1710 gsubr) object and adding it to the current module.
1712 These new functions are available: scm_c_make_subr, scm_c_define_subr,
1713 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
1714 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
1715 scm_c_define_gsubr_with_generic.
1717 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
1718 scm_make_subr_with_generic, scm_make_gsubr,
1719 scm_make_gsubr_with_generic.
1721 Use the new ones from above instead.
1723 ** C interface to the module system has changed.
1725 While we suggest that you avoid as many explicit module system
1726 operations from C as possible for the time being, the C interface has
1727 been made more similar to the high-level Scheme module system.
1729 *** New functions: scm_c_define_module, scm_c_use_module,
1730 scm_c_export, scm_c_resolve_module.
1732 They mostly work like their Scheme namesakes. scm_c_define_module
1733 takes a function that is called a context where the new module is
1736 *** Deprecated functions: scm_the_root_module, scm_make_module,
1737 scm_ensure_user_module, scm_load_scheme_module.
1739 Use the new functions instead.
1741 ** Renamed function: scm_internal_with_fluids becomes
1744 scm_internal_with_fluids is available as a deprecated function.
1746 ** New function: scm_c_with_fluid.
1748 Just like scm_c_with_fluids, but takes one fluid and one value instead
1751 ** Deprecated typedefs: long_long, ulong_long.
1753 They are of questionable utility and they pollute the global
1756 ** Deprecated typedef: scm_sizet
1758 It is of questionable utility now that Guile requires ANSI C, and is
1761 ** Deprecated typedefs: scm_port_rw_active, scm_port,
1762 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
1763 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
1765 Made more compliant with the naming policy by adding a _t at the end.
1767 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
1768 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
1770 With the exception of the mysterious scm_2ulong2big, they are still
1771 available under new names (scm_i_mkbig etc). These functions are not
1772 intended to be used in user code. You should avoid dealing with
1773 bignums directly, and should deal with numbers in general (which can
1776 ** Change in behavior: scm_num2long, scm_num2ulong
1778 The scm_num2[u]long functions don't any longer accept an inexact
1779 argument. This change in behavior is motivated by concordance with
1780 R5RS: It is more common that a primitive doesn't want to accept an
1781 inexact for an exact.
1783 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
1784 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
1785 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
1788 These are conversion functions between the various ANSI C integral
1789 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
1790 accept an inexact argument.
1792 ** New functions: scm_float2num, scm_double2num,
1793 scm_num2float, scm_num2double.
1795 These are conversion functions between the two ANSI C float types and
1798 ** New number validation macros:
1799 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
1803 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
1805 These are just nicer-named old scm_protect_object and
1806 scm_unprotect_object.
1808 ** Deprecated functions: scm_protect_object, scm_unprotect_object
1810 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
1812 These functions can be used to register pointers to locations that
1815 ** Deprecated function: scm_create_hook.
1817 Its sins are: misleading name, non-modularity and lack of general
1821 Changes since Guile 1.3.4:
1823 * Changes to the distribution
1825 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
1827 We've changed the way we handle generated files in the Guile source
1828 repository. As a result, the procedure for building trees obtained
1829 from the nightly FTP snapshots or via CVS has changed:
1830 - You must have appropriate versions of autoconf, automake, and
1831 libtool installed on your system. See README for info on how to
1832 obtain these programs.
1833 - Before configuring the tree, you must first run the script
1834 `autogen.sh' at the top of the source tree.
1836 The Guile repository used to contain not only source files, written by
1837 humans, but also some generated files, like configure scripts and
1838 Makefile.in files. Even though the contents of these files could be
1839 derived mechanically from other files present, we thought it would
1840 make the tree easier to build if we checked them into CVS.
1842 However, this approach means that minor differences between
1843 developer's installed tools and habits affected the whole team.
1844 So we have removed the generated files from the repository, and
1845 added the autogen.sh script, which will reconstruct them
1849 ** configure now has experimental options to remove support for certain
1852 --disable-arrays omit array and uniform array support
1853 --disable-posix omit posix interfaces
1854 --disable-networking omit networking interfaces
1855 --disable-regex omit regular expression interfaces
1857 These are likely to become separate modules some day.
1859 ** New configure option --enable-debug-freelist
1861 This enables a debugging version of SCM_NEWCELL(), and also registers
1862 an extra primitive, the setter `gc-set-debug-check-freelist!'.
1864 Configure with the --enable-debug-freelist option to enable
1865 the gc-set-debug-check-freelist! primitive, and then use:
1867 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
1868 (gc-set-debug-check-freelist! #f) # turn off checking
1870 Checking of the freelist forces a traversal of the freelist and
1871 a garbage collection before each allocation of a cell. This can
1872 slow down the interpreter dramatically, so the setter should be used to
1873 turn on this extra processing only when necessary.
1875 ** New configure option --enable-debug-malloc
1877 Include code for debugging of calls to scm_must_malloc/realloc/free.
1881 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
1882 2. objects reallocated by scm_must_realloc has been allocated by
1884 3. reallocated objects are reallocated with the same what string
1886 But, most importantly, it records the number of allocated objects of
1887 each kind. This is useful when searching for memory leaks.
1889 A Guile compiled with this option provides the primitive
1890 `malloc-stats' which returns an alist with pairs of kind and the
1891 number of objects of that kind.
1893 ** All includes are now referenced relative to the root directory
1895 Since some users have had problems with mixups between Guile and
1896 system headers, we have decided to always refer to Guile headers via
1897 their parent directories. This essentially creates a "private name
1898 space" for Guile headers. This means that the compiler only is given
1899 -I options for the root build and root source directory.
1901 ** Header files kw.h and genio.h have been removed.
1903 ** The module (ice-9 getopt-gnu-style) has been removed.
1905 ** New module (ice-9 documentation)
1907 Implements the interface to documentation strings associated with
1910 ** New module (ice-9 time)
1912 Provides a macro `time', which displays execution time of a given form.
1914 ** New module (ice-9 history)
1916 Loading this module enables value history in the repl.
1918 * Changes to the stand-alone interpreter
1920 ** New command line option --debug
1922 Start Guile with debugging evaluator and backtraces enabled.
1924 This is useful when debugging your .guile init file or scripts.
1926 ** New help facility
1928 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
1929 (help REGEXP) ditto for objects with names matching REGEXP (a string)
1930 (help 'NAME) gives documentation for NAME, even if it is not an object
1931 (help ,EXPR) gives documentation for object returned by EXPR
1932 (help (my module)) gives module commentary for `(my module)'
1933 (help) gives this text
1935 `help' searches among bindings exported from loaded modules, while
1936 `apropos' searches among bindings visible from the "current" module.
1938 Examples: (help help)
1940 (help "output-string")
1942 ** `help' and `apropos' now prints full module names
1944 ** Dynamic linking now uses libltdl from the libtool package.
1946 The old system dependent code for doing dynamic linking has been
1947 replaced with calls to the libltdl functions which do all the hairy
1950 The major improvement is that you can now directly pass libtool
1951 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
1952 will be able to do the best shared library job you can get, via
1955 The way dynamic libraries are found has changed and is not really
1956 portable across platforms, probably. It is therefore recommended to
1957 use absolute filenames when possible.
1959 If you pass a filename without an extension to `dynamic-link', it will
1960 try a few appropriate ones. Thus, the most platform ignorant way is
1961 to specify a name like "libfoo", without any directories and
1964 ** Guile COOP threads are now compatible with LinuxThreads
1966 Previously, COOP threading wasn't possible in applications linked with
1967 Linux POSIX threads due to their use of the stack pointer to find the
1968 thread context. This has now been fixed with a workaround which uses
1969 the pthreads to allocate the stack.
1971 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
1973 ** Positions of erring expression in scripts
1975 With version 1.3.4, the location of the erring expression in Guile
1976 scipts is no longer automatically reported. (This should have been
1977 documented before the 1.3.4 release.)
1979 You can get this information by enabling recording of positions of
1980 source expressions and running the debugging evaluator. Put this at
1981 the top of your script (or in your "site" file):
1983 (read-enable 'positions)
1984 (debug-enable 'debug)
1986 ** Backtraces in scripts
1988 It is now possible to get backtraces in scripts.
1992 (debug-enable 'debug 'backtrace)
1994 at the top of the script.
1996 (The first options enables the debugging evaluator.
1997 The second enables backtraces.)
1999 ** Part of module system symbol lookup now implemented in C
2001 The eval closure of most modules is now implemented in C. Since this
2002 was one of the bottlenecks for loading speed, Guile now loads code
2003 substantially faster than before.
2005 ** Attempting to get the value of an unbound variable now produces
2006 an exception with a key of 'unbound-variable instead of 'misc-error.
2008 ** The initial default output port is now unbuffered if it's using a
2009 tty device. Previously in this situation it was line-buffered.
2011 ** New hook: after-gc-hook
2013 after-gc-hook takes over the role of gc-thunk. This hook is run at
2014 the first SCM_TICK after a GC. (Thus, the code is run at the same
2015 point during evaluation as signal handlers.)
2017 Note that this hook should be used only for diagnostic and debugging
2018 purposes. It is not certain that it will continue to be well-defined
2019 when this hook is run in the future.
2021 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2022 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2024 ** Improvements to garbage collector
2026 Guile 1.4 has a new policy for triggering heap allocation and
2027 determining the sizes of heap segments. It fixes a number of problems
2030 1. The new policy can handle two separate pools of cells
2031 (2-word/4-word) better. (The old policy would run wild, allocating
2032 more and more memory for certain programs.)
2034 2. The old code would sometimes allocate far too much heap so that the
2035 Guile process became gigantic. The new code avoids this.
2037 3. The old code would sometimes allocate too little so that few cells
2038 were freed at GC so that, in turn, too much time was spent in GC.
2040 4. The old code would often trigger heap allocation several times in a
2041 row. (The new scheme predicts how large the segments needs to be
2042 in order not to need further allocation.)
2044 All in all, the new GC policy will make larger applications more
2047 The new GC scheme also is prepared for POSIX threading. Threads can
2048 allocate private pools of cells ("clusters") with just a single
2049 function call. Allocation of single cells from such a cluster can
2050 then proceed without any need of inter-thread synchronization.
2052 ** New environment variables controlling GC parameters
2054 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2057 Allocation of 2-word cell heaps:
2059 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2062 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2063 GC in percent of total heap size
2066 Allocation of 4-word cell heaps
2067 (used for real numbers and misc other objects):
2069 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2071 (See entry "Way for application to customize GC parameters" under
2072 section "Changes to the scm_ interface" below.)
2074 ** Guile now implements reals using 4-word cells
2076 This speeds up computation with reals. (They were earlier allocated
2077 with `malloc'.) There is still some room for optimizations, however.
2079 ** Some further steps toward POSIX thread support have been taken
2081 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2082 don't have much effect any longer, and many of them will be removed in
2086 are only handled at the top of the evaluator loop, immediately after
2087 I/O, and in scm_equalp.
2089 *** The GC can allocate thread private pools of pairs.
2091 * Changes to Scheme functions and syntax
2093 ** close-input-port and close-output-port are now R5RS
2095 These procedures have been turned into primitives and have R5RS behaviour.
2097 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2099 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2100 extended by the more sophisticated version in (ice-9 format)
2102 (simple-format port message . args)
2103 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2104 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2105 the escapes are replaced with corresponding members of ARGS:
2106 ~A formats using `display' and ~S formats using `write'.
2107 If DESTINATION is #t, then use the `current-output-port',
2108 if DESTINATION is #f, then return a string containing the formatted text.
2109 Does not add a trailing newline."
2111 ** string-ref: the second argument is no longer optional.
2113 ** string, list->string: no longer accept strings in their arguments,
2114 only characters, for compatibility with R5RS.
2116 ** New procedure: port-closed? PORT
2117 Returns #t if PORT is closed or #f if it is open.
2119 ** Deprecated: list*
2121 The list* functionality is now provided by cons* (SRFI-1 compliant)
2123 ** New procedure: cons* ARG1 ARG2 ... ARGn
2125 Like `list', but the last arg provides the tail of the constructed list,
2126 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2128 Requires at least one argument. If given one argument, that argument
2129 is returned as result.
2131 This function is called `list*' in some other Schemes and in Common LISP.
2133 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2135 ** New procedure: object-documentation OBJECT
2137 Returns the documentation string associated with OBJECT. The
2138 procedure uses a caching mechanism so that subsequent lookups are
2141 Exported by (ice-9 documentation).
2143 ** module-name now returns full names of modules
2145 Previously, only the last part of the name was returned (`session' for
2146 `(ice-9 session)'). Ex: `(ice-9 session)'.
2148 * Changes to the gh_ interface
2150 ** Deprecated: gh_int2scmb
2152 Use gh_bool2scm instead.
2154 * Changes to the scm_ interface
2156 ** Guile primitives now carry docstrings!
2158 Thanks to Greg Badros!
2160 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2162 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2163 macros and must contain a docstring that is extracted into foo.doc using a new
2164 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2166 However, a major overhaul of these macros is scheduled for the next release of
2169 ** Guile primitives use a new technique for validation of arguments
2171 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2172 the readability of argument checking.
2174 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2176 ** New macros: SCM_PACK, SCM_UNPACK
2178 Compose/decompose an SCM value.
2180 The SCM type is now treated as an abstract data type and may be defined as a
2181 long, a void* or as a struct, depending on the architecture and compile time
2182 options. This makes it easier to find several types of bugs, for example when
2183 SCM values are treated as integers without conversion. Values of the SCM type
2184 should be treated as "atomic" values. These macros are used when
2185 composing/decomposing an SCM value, either because you want to access
2186 individual bits, or because you want to treat it as an integer value.
2188 E.g., in order to set bit 7 in an SCM value x, use the expression
2190 SCM_PACK (SCM_UNPACK (x) | 0x80)
2192 ** The name property of hooks is deprecated.
2193 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2195 You can emulate this feature by using object properties.
2197 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2198 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2201 These macros will be removed in a future release of Guile.
2203 ** The following types, functions and macros from numbers.h are deprecated:
2204 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2205 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2207 Further, it is recommended not to rely on implementation details for guile's
2208 current implementation of bignums. It is planned to replace this
2209 implementation with gmp in the future.
2211 ** Port internals: the rw_random variable in the scm_port structure
2212 must be set to non-zero in any random access port. In recent Guile
2213 releases it was only set for bidirectional random-access ports.
2215 ** Port internals: the seek ptob procedure is now responsible for
2216 resetting the buffers if required. The change was made so that in the
2217 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2218 the fport and strport ptobs can avoid resetting the buffers,
2219 in particular to avoid discarding unread chars. An existing port
2220 type can be fixed by adding something like the following to the
2221 beginning of the ptob seek procedure:
2223 if (pt->rw_active == SCM_PORT_READ)
2224 scm_end_input (object);
2225 else if (pt->rw_active == SCM_PORT_WRITE)
2226 ptob->flush (object);
2228 although to actually avoid resetting the buffers and discard unread
2229 chars requires further hacking that depends on the characteristics
2232 ** Deprecated functions: scm_fseek, scm_tag
2234 These functions are no longer used and will be removed in a future version.
2236 ** The scm_sysmissing procedure is no longer used in libguile.
2237 Unless it turns out to be unexpectedly useful to somebody, it will be
2238 removed in a future version.
2240 ** The format of error message strings has changed
2242 The two C procedures: scm_display_error and scm_error, as well as the
2243 primitive `scm-error', now use scm_simple_format to do their work.
2244 This means that the message strings of all code must be updated to use
2245 ~A where %s was used before, and ~S where %S was used before.
2247 During the period when there still are a lot of old Guiles out there,
2248 you might want to support both old and new versions of Guile.
2250 There are basically two methods to achieve this. Both methods use
2253 AC_CHECK_FUNCS(scm_simple_format)
2255 in your configure.in.
2257 Method 1: Use the string concatenation features of ANSI C's
2262 #ifdef HAVE_SCM_SIMPLE_FORMAT
2268 Then represent each of your error messages using a preprocessor macro:
2270 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2274 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2275 (define make-message string-append)
2277 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2279 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2283 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2288 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2292 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2294 Don't use the functions coop_mutex_init and
2295 coop_condition_variable_init. They will change.
2297 Use scm_mutex_init and scm_cond_init instead.
2299 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2300 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2301 COND, as `scm_cond_wait' does, but it also bounds the duration
2302 of the wait. If COND has not been signaled before time ABSTIME,
2303 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2304 returns the error code `ETIMEDOUT'.
2306 The ABSTIME parameter specifies an absolute time, with the same
2307 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2308 to 00:00:00 GMT, January 1, 1970.
2310 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2311 `scm_cond_broadcast' restarts all the threads that are waiting
2312 on the condition variable COND. Nothing happens if no threads are
2315 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2316 `scm_key_create' allocates a new TSD key. The key is stored in
2317 the location pointed to by KEY. There is no limit on the number
2318 of keys allocated at a given time. The value initially associated
2319 with the returned key is `NULL' in all currently executing threads.
2321 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2322 function associated with the key. When a thread terminates,
2323 DESTR_FUNCTION is called on the value associated with the key in
2324 that thread. The DESTR_FUNCTION is not called if a key is deleted
2325 with `scm_key_delete' or a value is changed with
2326 `scm_setspecific'. The order in which destructor functions are
2327 called at thread termination time is unspecified.
2329 Destructors are not yet implemented.
2331 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2332 `scm_setspecific' changes the value associated with KEY in the
2333 calling thread, storing the given POINTER instead.
2335 ** New function: scm_getspecific (scm_key_t KEY)
2336 `scm_getspecific' returns the value currently associated with
2337 KEY in the calling thread.
2339 ** New function: scm_key_delete (scm_key_t KEY)
2340 `scm_key_delete' deallocates a TSD key. It does not check
2341 whether non-`NULL' values are associated with that key in the
2342 currently executing threads, nor call the destructor function
2343 associated with the key.
2345 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2347 Initialize a C level hook HOOK with associated HOOK_DATA and type
2348 TYPE. (See scm_c_hook_run ().)
2350 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2352 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2353 is true, add it last, otherwise first. The same FUNC can be added
2354 multiple times if FUNC_DATA differ and vice versa.
2356 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2358 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2359 function is only removed if both FUNC and FUNC_DATA matches.
2361 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2363 Run hook HOOK passing DATA to the hook functions.
2365 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2366 returned is undefined.
2368 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2369 returns a non-NULL value. This value is returned as the result of
2370 scm_c_hook_run. If all functions return NULL, NULL is returned.
2372 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2373 returns a NULL value, and NULL is returned. If all functions returns
2374 a non-NULL value, the last value is returned.
2376 ** New C level GC hooks
2378 Five new C level hooks has been added to the garbage collector.
2380 scm_before_gc_c_hook
2383 are run before locking and after unlocking the heap. The system is
2384 thus in a mode where evaluation can take place. (Except that
2385 scm_before_gc_c_hook must not allocate new cells.)
2387 scm_before_mark_c_hook
2388 scm_before_sweep_c_hook
2389 scm_after_sweep_c_hook
2391 are run when the heap is locked. These are intended for extension of
2392 the GC in a modular fashion. Examples are the weaks and guardians
2395 ** Way for application to customize GC parameters
2397 The application can set up other default values for the GC heap
2398 allocation parameters
2400 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
2401 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
2402 GUILE_MAX_SEGMENT_SIZE,
2406 scm_default_init_heap_size_1, scm_default_min_yield_1,
2407 scm_default_init_heap_size_2, scm_default_min_yield_2,
2408 scm_default_max_segment_size
2410 respectively before callong scm_boot_guile.
2412 (See entry "New environment variables ..." in section
2413 "Changes to the stand-alone interpreter" above.)
2415 ** scm_protect_object/scm_unprotect_object now nest
2417 This means that you can call scm_protect_object multiple times on an
2418 object and count on the object being protected until
2419 scm_unprotect_object has been call the same number of times.
2421 The functions also have better time complexity.
2423 Still, it is usually possible to structure the application in a way
2424 that you don't need to use these functions. For example, if you use a
2425 protected standard Guile list to keep track of live objects rather
2426 than some custom data type, objects will die a natural death when they
2427 are no longer needed.
2429 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
2431 Guile does not provide the float representation for inexact real numbers any
2432 more. Now, only doubles are used to represent inexact real numbers. Further,
2433 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
2434 and scm_tc16_complex, respectively.
2436 ** Removed deprecated type scm_smobfuns
2438 ** Removed deprecated function scm_newsmob
2440 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
2442 There is an ongoing discussion among the developers whether to
2443 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
2444 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
2445 until this issue has been settled.
2447 ** Removed deprecated type tag scm_tc16_kw
2449 ** Added type tag scm_tc16_keyword
2451 (This was introduced already in release 1.3.4 but was not documented
2454 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
2456 * Changes to system call interfaces:
2458 ** The "select" procedure now tests port buffers for the ability to
2459 provide input or accept output. Previously only the underlying file
2460 descriptors were checked.
2462 ** New variable PIPE_BUF: the maximum number of bytes that can be
2463 atomically written to a pipe.
2465 ** If a facility is not available on the system when Guile is
2466 compiled, the corresponding primitive procedure will not be defined.
2467 Previously it would have been defined but would throw a system-error
2468 exception if called. Exception handlers which catch this case may
2469 need minor modification: an error will be thrown with key
2470 'unbound-variable instead of 'system-error. Alternatively it's
2471 now possible to use `defined?' to check whether the facility is
2474 ** Procedures which depend on the timezone should now give the correct
2475 result on systems which cache the TZ environment variable, even if TZ
2476 is changed without calling tzset.
2478 * Changes to the networking interfaces:
2480 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
2481 long integers between network and host format. For now, it's not
2482 particularly convenient to do this kind of thing, but consider:
2484 (define write-network-long
2485 (lambda (value port)
2486 (let ((v (make-uniform-vector 1 1 0)))
2487 (uniform-vector-set! v 0 (htonl value))
2488 (uniform-vector-write v port))))
2490 (define read-network-long
2492 (let ((v (make-uniform-vector 1 1 0)))
2493 (uniform-vector-read! v port)
2494 (ntohl (uniform-vector-ref v 0)))))
2496 ** If inet-aton fails, it now throws an error with key 'misc-error
2497 instead of 'system-error, since errno is not relevant.
2499 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
2500 specific keys instead of 'system-error. The latter is inappropriate
2501 since errno will not have been set. The keys are:
2502 'host-not-found, 'try-again, 'no-recovery and 'no-data.
2504 ** sethostent, setnetent, setprotoent, setservent: now take an
2505 optional argument STAYOPEN, which specifies whether the database
2506 remains open after a database entry is accessed randomly (e.g., using
2507 gethostbyname for the hosts database.) The default is #f. Previously
2511 Changes since Guile 1.3.2:
2513 * Changes to the stand-alone interpreter
2517 An initial version of the Guile debugger written by Chris Hanson has
2518 been added. The debugger is still under development but is included
2519 in the distribution anyway since it is already quite useful.
2525 after an error to enter the debugger. Type `help' inside the debugger
2526 for a description of available commands.
2528 If you prefer to have stack frames numbered and printed in
2529 anti-chronological order and prefer up in the stack to be down on the
2530 screen as is the case in gdb, you can put
2532 (debug-enable 'backwards)
2534 in your .guile startup file. (However, this means that Guile can't
2535 use indentation to indicate stack level.)
2537 The debugger is autoloaded into Guile at the first use.
2539 ** Further enhancements to backtraces
2541 There is a new debug option `width' which controls the maximum width
2542 on the screen of printed stack frames. Fancy printing parameters
2543 ("level" and "length" as in Common LISP) are adaptively adjusted for
2544 each stack frame to give maximum information while still fitting
2545 within the bounds. If the stack frame can't be made to fit by
2546 adjusting parameters, it is simply cut off at the end. This is marked
2549 ** Some modules are now only loaded when the repl is started
2551 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
2552 regex) are now loaded into (guile-user) only if the repl has been
2553 started. The effect is that the startup time for scripts has been
2554 reduced to 30% of what it was previously.
2556 Correctly written scripts load the modules they require at the top of
2557 the file and should not be affected by this change.
2559 ** Hooks are now represented as smobs
2561 * Changes to Scheme functions and syntax
2563 ** Readline support has changed again.
2565 The old (readline-activator) module is gone. Use (ice-9 readline)
2566 instead, which now contains all readline functionality. So the code
2567 to activate readline is now
2569 (use-modules (ice-9 readline))
2572 This should work at any time, including from the guile prompt.
2574 To avoid confusion about the terms of Guile's license, please only
2575 enable readline for your personal use; please don't make it the
2576 default for others. Here is why we make this rather odd-sounding
2579 Guile is normally licensed under a weakened form of the GNU General
2580 Public License, which allows you to link code with Guile without
2581 placing that code under the GPL. This exception is important to some
2584 However, since readline is distributed under the GNU General Public
2585 License, when you link Guile with readline, either statically or
2586 dynamically, you effectively change Guile's license to the strict GPL.
2587 Whenever you link any strictly GPL'd code into Guile, uses of Guile
2588 which are normally permitted become forbidden. This is a rather
2589 non-obvious consequence of the licensing terms.
2591 So, to make sure things remain clear, please let people choose for
2592 themselves whether to link GPL'd libraries like readline with Guile.
2594 ** regexp-substitute/global has changed slightly, but incompatibly.
2596 If you include a function in the item list, the string of the match
2597 object it receives is the same string passed to
2598 regexp-substitute/global, not some suffix of that string.
2599 Correspondingly, the match's positions are relative to the entire
2600 string, not the suffix.
2602 If the regexp can match the empty string, the way matches are chosen
2603 from the string has changed. regexp-substitute/global recognizes the
2604 same set of matches that list-matches does; see below.
2606 ** New function: list-matches REGEXP STRING [FLAGS]
2608 Return a list of match objects, one for every non-overlapping, maximal
2609 match of REGEXP in STRING. The matches appear in left-to-right order.
2610 list-matches only reports matches of the empty string if there are no
2611 other matches which begin on, end at, or include the empty match's
2614 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2616 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
2618 For each match of REGEXP in STRING, apply PROC to the match object,
2619 and the last value PROC returned, or INIT for the first call. Return
2620 the last value returned by PROC. We apply PROC to the matches as they
2621 appear from left to right.
2623 This function recognizes matches according to the same criteria as
2626 Thus, you could define list-matches like this:
2628 (define (list-matches regexp string . flags)
2629 (reverse! (apply fold-matches regexp string '() cons flags)))
2631 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2635 *** New function: hook? OBJ
2637 Return #t if OBJ is a hook, otherwise #f.
2639 *** New function: make-hook-with-name NAME [ARITY]
2641 Return a hook with name NAME and arity ARITY. The default value for
2642 ARITY is 0. The only effect of NAME is that it will appear when the
2643 hook object is printed to ease debugging.
2645 *** New function: hook-empty? HOOK
2647 Return #t if HOOK doesn't contain any procedures, otherwise #f.
2649 *** New function: hook->list HOOK
2651 Return a list of the procedures that are called when run-hook is
2654 ** `map' signals an error if its argument lists are not all the same length.
2656 This is the behavior required by R5RS, so this change is really a bug
2657 fix. But it seems to affect a lot of people's code, so we're
2658 mentioning it here anyway.
2660 ** Print-state handling has been made more transparent
2662 Under certain circumstances, ports are represented as a port with an
2663 associated print state. Earlier, this pair was represented as a pair
2664 (see "Some magic has been added to the printer" below). It is now
2665 indistinguishable (almost; see `get-print-state') from a port on the
2668 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
2670 Return a new port with the associated print state PRINT-STATE.
2672 *** New function: get-print-state OUTPUT-PORT
2674 Return the print state associated with this port if it exists,
2675 otherwise return #f.
2677 *** New function: directory-stream? OBJECT
2679 Returns true iff OBJECT is a directory stream --- the sort of object
2680 returned by `opendir'.
2682 ** New function: using-readline?
2684 Return #t if readline is in use in the current repl.
2686 ** structs will be removed in 1.4
2688 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
2689 and use GOOPS objects as the fundamental record type.
2691 * Changes to the scm_ interface
2693 ** structs will be removed in 1.4
2695 The entire current struct interface (struct.c, struct.h) will be
2696 replaced in Guile 1.4. We will merge GOOPS into libguile and use
2697 GOOPS objects as the fundamental record type.
2699 ** The internal representation of subr's has changed
2701 Instead of giving a hint to the subr name, the CAR field of the subr
2702 now contains an index to a subr entry in scm_subr_table.
2704 *** New variable: scm_subr_table
2706 An array of subr entries. A subr entry contains the name, properties
2707 and documentation associated with the subr. The properties and
2708 documentation slots are not yet used.
2710 ** A new scheme for "forwarding" calls to a builtin to a generic function
2712 It is now possible to extend the functionality of some Guile
2713 primitives by letting them defer a call to a GOOPS generic function on
2714 argument mismatch. This means that there is no loss of efficiency in
2719 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
2720 (define-method + ((x <string>) (y <string>))
2721 (string-append x y))
2723 + will still be as efficient as usual in numerical calculations, but
2724 can also be used for concatenating strings.
2726 Who will be the first one to extend Guile's numerical tower to
2727 rationals? :) [OK, there a few other things to fix before this can
2728 be made in a clean way.]
2730 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
2732 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
2734 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
2736 These do the same job as SCM_PROC and SCM_PROC1, but they also define
2737 a variable GENERIC which can be used by the dispatch macros below.
2739 [This is experimental code which may change soon.]
2741 *** New macros for forwarding control to a generic on arg type error
2743 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
2745 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
2747 These correspond to the scm_wta function call, and have the same
2748 behaviour until the user has called the GOOPS primitive
2749 `enable-primitive-generic!'. After that, these macros will apply the
2750 generic function GENERIC to the argument(s) instead of calling
2753 [This is experimental code which may change soon.]
2755 *** New macros for argument testing with generic dispatch
2757 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
2759 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
2761 These correspond to the SCM_ASSERT macro, but will defer control to
2762 GENERIC on error after `enable-primitive-generic!' has been called.
2764 [This is experimental code which may change soon.]
2766 ** New function: SCM scm_eval_body (SCM body, SCM env)
2768 Evaluates the body of a special form.
2770 ** The internal representation of struct's has changed
2772 Previously, four slots were allocated for the procedure(s) of entities
2773 and operators. The motivation for this representation had to do with
2774 the structure of the evaluator, the wish to support tail-recursive
2775 generic functions, and efficiency. Since the generic function
2776 dispatch mechanism has changed, there is no longer a need for such an
2777 expensive representation, and the representation has been simplified.
2779 This should not make any difference for most users.
2781 ** GOOPS support has been cleaned up.
2783 Some code has been moved from eval.c to objects.c and code in both of
2784 these compilation units has been cleaned up and better structured.
2786 *** New functions for applying generic functions
2788 New function: SCM scm_apply_generic (GENERIC, ARGS)
2789 New function: SCM scm_call_generic_0 (GENERIC)
2790 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
2791 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
2792 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
2794 ** Deprecated function: scm_make_named_hook
2796 It is now replaced by:
2798 ** New function: SCM scm_create_hook (const char *name, int arity)
2800 Creates a hook in the same way as make-hook above but also
2801 binds a variable named NAME to it.
2803 This is the typical way of creating a hook from C code.
2805 Currently, the variable is created in the "current" module.
2806 This might change when we get the new module system.
2808 [The behaviour is identical to scm_make_named_hook.]
2812 Changes since Guile 1.3:
2814 * Changes to mailing lists
2816 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
2818 See the README file to find current addresses for all the Guile
2821 * Changes to the distribution
2823 ** Readline support is no longer included with Guile by default.
2825 Based on the different license terms of Guile and Readline, we
2826 concluded that Guile should not *by default* cause the linking of
2827 Readline into an application program. Readline support is now offered
2828 as a separate module, which is linked into an application only when
2829 you explicitly specify it.
2831 Although Guile is GNU software, its distribution terms add a special
2832 exception to the usual GNU General Public License (GPL). Guile's
2833 license includes a clause that allows you to link Guile with non-free
2834 programs. We add this exception so as not to put Guile at a
2835 disadvantage vis-a-vis other extensibility packages that support other
2838 In contrast, the GNU Readline library is distributed under the GNU
2839 General Public License pure and simple. This means that you may not
2840 link Readline, even dynamically, into an application unless it is
2841 distributed under a free software license that is compatible the GPL.
2843 Because of this difference in distribution terms, an application that
2844 can use Guile may not be able to use Readline. Now users will be
2845 explicitly offered two independent decisions about the use of these
2848 You can activate the readline support by issuing
2850 (use-modules (readline-activator))
2853 from your ".guile" file, for example.
2855 * Changes to the stand-alone interpreter
2857 ** All builtins now print as primitives.
2858 Previously builtin procedures not belonging to the fundamental subr
2859 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
2860 Now, they print as #<primitive-procedure NAME>.
2862 ** Backtraces slightly more intelligible.
2863 gsubr-apply and macro transformer application frames no longer appear
2866 * Changes to Scheme functions and syntax
2868 ** Guile now correctly handles internal defines by rewriting them into
2869 their equivalent letrec. Previously, internal defines would
2870 incrementally add to the innermost environment, without checking
2871 whether the restrictions specified in RnRS were met. This lead to the
2872 correct behaviour when these restriction actually were met, but didn't
2873 catch all illegal uses. Such an illegal use could lead to crashes of
2874 the Guile interpreter or or other unwanted results. An example of
2875 incorrect internal defines that made Guile behave erratically:
2887 The problem with this example is that the definition of `c' uses the
2888 value of `b' directly. This confuses the meoization machine of Guile
2889 so that the second call of `b' (this time in a larger environment that
2890 also contains bindings for `c' and `d') refers to the binding of `c'
2891 instead of `a'. You could also make Guile crash with a variation on
2896 (define (b flag) (if flag a 1))
2897 (define c (1+ (b flag)))
2905 From now on, Guile will issue an `Unbound variable: b' error message
2910 A hook contains a list of functions which should be called on
2911 particular occasions in an existing program. Hooks are used for
2914 A window manager might have a hook before-window-map-hook. The window
2915 manager uses the function run-hooks to call all functions stored in
2916 before-window-map-hook each time a window is mapped. The user can
2917 store functions in the hook using add-hook!.
2919 In Guile, hooks are first class objects.
2921 *** New function: make-hook [N_ARGS]
2923 Return a hook for hook functions which can take N_ARGS arguments.
2924 The default value for N_ARGS is 0.
2926 (See also scm_make_named_hook below.)
2928 *** New function: add-hook! HOOK PROC [APPEND_P]
2930 Put PROC at the beginning of the list of functions stored in HOOK.
2931 If APPEND_P is supplied, and non-false, put PROC at the end instead.
2933 PROC must be able to take the number of arguments specified when the
2936 If PROC already exists in HOOK, then remove it first.
2938 *** New function: remove-hook! HOOK PROC
2940 Remove PROC from the list of functions in HOOK.
2942 *** New function: reset-hook! HOOK
2944 Clear the list of hook functions stored in HOOK.
2946 *** New function: run-hook HOOK ARG1 ...
2948 Run all hook functions stored in HOOK with arguments ARG1 ... .
2949 The number of arguments supplied must correspond to the number given
2950 when the hook was created.
2952 ** The function `dynamic-link' now takes optional keyword arguments.
2953 The only keyword argument that is currently defined is `:global
2954 BOOL'. With it, you can control whether the shared library will be
2955 linked in global mode or not. In global mode, the symbols from the
2956 linked library can be used to resolve references from other
2957 dynamically linked libraries. In non-global mode, the linked
2958 library is essentially invisible and can only be accessed via
2959 `dynamic-func', etc. The default is now to link in global mode.
2960 Previously, the default has been non-global mode.
2962 The `#:global' keyword is only effective on platforms that support
2963 the dlopen family of functions.
2965 ** New function `provided?'
2967 - Function: provided? FEATURE
2968 Return true iff FEATURE is supported by this installation of
2969 Guile. FEATURE must be a symbol naming a feature; the global
2970 variable `*features*' is a list of available features.
2972 ** Changes to the module (ice-9 expect):
2974 *** The expect-strings macro now matches `$' in a regular expression
2975 only at a line-break or end-of-file by default. Previously it would
2976 match the end of the string accumulated so far. The old behaviour
2977 can be obtained by setting the variable `expect-strings-exec-flags'
2980 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
2981 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
2982 in a regular expression will still match before a line-break or
2983 end-of-file. The default is `regexp/noteol'.
2985 *** The expect-strings macro now uses a variable
2986 `expect-strings-compile-flags' for the flags to be supplied to
2987 `make-regexp'. The default is `regexp/newline', which was previously
2990 *** The expect macro now supplies two arguments to a match procedure:
2991 the current accumulated string and a flag to indicate whether
2992 end-of-file has been reached. Previously only the string was supplied.
2993 If end-of-file is reached, the match procedure will be called an
2994 additional time with the same accumulated string as the previous call
2995 but with the flag set.
2997 ** New module (ice-9 format), implementing the Common Lisp `format' function.
2999 This code, and the documentation for it that appears here, was
3000 borrowed from SLIB, with minor adaptations for Guile.
3002 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3003 An almost complete implementation of Common LISP format description
3004 according to the CL reference book `Common LISP' from Guy L.
3005 Steele, Digital Press. Backward compatible to most of the
3006 available Scheme format implementations.
3008 Returns `#t', `#f' or a string; has side effect of printing
3009 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3010 to the current output port and `#t' is returned. If DESTINATION
3011 is `#f', a formatted string is returned as the result of the call.
3012 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3013 format string; FORMAT-STRING is then the first argument and the
3014 output is returned as a string. If DESTINATION is a number, the
3015 output is to the current error port if available by the
3016 implementation. Otherwise DESTINATION must be an output port and
3019 FORMAT-STRING must be a string. In case of a formatting error
3020 format returns `#f' and prints a message on the current output or
3021 error port. Characters are output as if the string were output by
3022 the `display' function with the exception of those prefixed by a
3023 tilde (~). For a detailed description of the FORMAT-STRING syntax
3024 please consult a Common LISP format reference manual. For a test
3025 suite to verify this format implementation load `formatst.scm'.
3026 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3028 Note: `format' is not reentrant, i.e. only one `format'-call may
3029 be executed at a time.
3032 *** Format Specification (Format version 3.0)
3034 Please consult a Common LISP format reference manual for a detailed
3035 description of the format string syntax. For a demonstration of the
3036 implemented directives see `formatst.scm'.
3038 This implementation supports directive parameters and modifiers (`:'
3039 and `@' characters). Multiple parameters must be separated by a comma
3040 (`,'). Parameters can be numerical parameters (positive or negative),
3041 character parameters (prefixed by a quote character (`''), variable
3042 parameters (`v'), number of rest arguments parameter (`#'), empty and
3043 default parameters. Directive characters are case independent. The
3044 general form of a directive is:
3046 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3048 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3050 *** Implemented CL Format Control Directives
3052 Documentation syntax: Uppercase characters represent the
3053 corresponding control directive characters. Lowercase characters
3054 represent control directive parameter descriptions.
3057 Any (print as `display' does).
3061 `~MINCOL,COLINC,MINPAD,PADCHARA'
3065 S-expression (print as `write' does).
3069 `~MINCOL,COLINC,MINPAD,PADCHARS'
3075 print number sign always.
3078 print comma separated.
3080 `~MINCOL,PADCHAR,COMMACHARD'
3086 print number sign always.
3089 print comma separated.
3091 `~MINCOL,PADCHAR,COMMACHARX'
3097 print number sign always.
3100 print comma separated.
3102 `~MINCOL,PADCHAR,COMMACHARO'
3108 print number sign always.
3111 print comma separated.
3113 `~MINCOL,PADCHAR,COMMACHARB'
3118 `~N,MINCOL,PADCHAR,COMMACHARR'
3122 print a number as a Roman numeral.
3125 print a number as an "old fashioned" Roman numeral.
3128 print a number as an ordinal English number.
3131 print a number as a cardinal English number.
3136 prints `y' and `ies'.
3139 as `~P but jumps 1 argument backward.'
3142 as `~@P but jumps 1 argument backward.'
3147 prints a character as the reader can understand it (i.e. `#\'
3151 prints a character as emacs does (eg. `^C' for ASCII 03).
3154 Fixed-format floating-point (prints a flonum like MMM.NNN).
3155 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3157 If the number is positive a plus sign is printed.
3160 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3161 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3163 If the number is positive a plus sign is printed.
3166 General floating-point (prints a flonum either fixed or
3168 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3170 If the number is positive a plus sign is printed.
3173 Dollars floating-point (prints a flonum in fixed with signs
3175 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3177 If the number is positive a plus sign is printed.
3180 A sign is always printed and appears before the padding.
3183 The sign appears before the padding.
3191 print newline if not at the beginning of the output line.
3193 prints `~&' and then N-1 newlines.
3198 print N page separators.
3208 newline is ignored, white space left.
3211 newline is left, white space ignored.
3216 relative tabulation.
3222 Indirection (expects indirect arguments as a list).
3224 extracts indirect arguments from format arguments.
3227 Case conversion (converts by `string-downcase').
3229 converts by `string-capitalize'.
3232 converts by `string-capitalize-first'.
3235 converts by `string-upcase'.
3238 Argument Jumping (jumps 1 argument forward).
3240 jumps N arguments forward.
3243 jumps 1 argument backward.
3246 jumps N arguments backward.
3249 jumps to the 0th argument.
3252 jumps to the Nth argument (beginning from 0)
3254 `~[STR0~;STR1~;...~;STRN~]'
3255 Conditional Expression (numerical clause conditional).
3257 take argument from N.
3260 true test conditional.
3263 if-else-then conditional.
3269 default clause follows.
3272 Iteration (args come from the next argument (a list)).
3274 at most N iterations.
3277 args from next arg (a list of lists).
3280 args from the rest of arguments.
3283 args from the rest args (lists).
3294 aborts if N <= M <= K
3296 *** Not Implemented CL Format Control Directives
3299 print `#f' as an empty list (see below).
3302 print `#f' as an empty list (see below).
3308 (sorry I don't understand its semantics completely)
3310 *** Extended, Replaced and Additional Control Directives
3312 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3313 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3314 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3315 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3316 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3317 COMMAWIDTH is the number of characters between two comma
3321 print a R4RS complex number as `~F~@Fi' with passed parameters for
3325 Pretty print formatting of an argument for scheme code lists.
3331 Flushes the output if format DESTINATION is a port.
3334 Print a `#\space' character
3336 print N `#\space' characters.
3339 Print a `#\tab' character
3341 print N `#\tab' characters.
3344 Takes N as an integer representation for a character. No arguments
3345 are consumed. N is converted to a character by `integer->char'. N
3346 must be a positive decimal number.
3349 Print out readproof. Prints out internal objects represented as
3350 `#<...>' as strings `"#<...>"' so that the format output can always
3351 be processed by `read'.
3354 Print out readproof. Prints out internal objects represented as
3355 `#<...>' as strings `"#<...>"' so that the format output can always
3356 be processed by `read'.
3359 Prints information and a copyright notice on the format
3362 prints format version.
3365 may also print number strings, i.e. passing a number as a string
3366 and format it accordingly.
3368 *** Configuration Variables
3370 The format module exports some configuration variables to suit the
3371 systems and users needs. There should be no modification necessary for
3372 the configuration that comes with Guile. Format detects automatically
3373 if the running scheme system implements floating point numbers and
3376 format:symbol-case-conv
3377 Symbols are converted by `symbol->string' so the case type of the
3378 printed symbols is implementation dependent.
3379 `format:symbol-case-conv' is a one arg closure which is either
3380 `#f' (no conversion), `string-upcase', `string-downcase' or
3381 `string-capitalize'. (default `#f')
3383 format:iobj-case-conv
3384 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3385 implementation internal objects. (default `#f')
3388 The character prefixing the exponent value in `~E' printing.
3391 *** Compatibility With Other Format Implementations
3397 Downward compatible except for padding support and `~A', `~S',
3398 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
3399 `printf' padding support which is completely replaced by the CL
3400 `format' padding style.
3403 Downward compatible except for `~', which is not documented
3404 (ignores all characters inside the format string up to a newline
3405 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
3406 numerical and variable parameters and `:/@' modifiers in the CL
3410 Downward compatible except for `~A' and `~S' which print in
3411 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
3412 directive parameters or modifiers)).
3415 Downward compatible except for an optional destination parameter:
3416 S2C accepts a format call without a destination which returns a
3417 formatted string. This is equivalent to a #f destination in S2C.
3418 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
3419 parameters or modifiers)).
3422 ** Changes to string-handling functions.
3424 These functions were added to support the (ice-9 format) module, above.
3426 *** New function: string-upcase STRING
3427 *** New function: string-downcase STRING
3429 These are non-destructive versions of the existing string-upcase! and
3430 string-downcase! functions.
3432 *** New function: string-capitalize! STRING
3433 *** New function: string-capitalize STRING
3435 These functions convert the first letter of each word in the string to
3438 (string-capitalize "howdy there")
3441 As with the other functions, string-capitalize! modifies the string in
3442 place, while string-capitalize returns a modified copy of its argument.
3444 *** New function: string-ci->symbol STRING
3446 Return a symbol whose name is STRING, but having the same case as if
3447 the symbol had be read by `read'.
3449 Guile can be configured to be sensitive or insensitive to case
3450 differences in Scheme identifiers. If Guile is case-insensitive, all
3451 symbols are converted to lower case on input. The `string-ci->symbol'
3452 function returns a symbol whose name in STRING, transformed as Guile
3453 would if STRING were input.
3455 *** New function: substring-move! STRING1 START END STRING2 START
3457 Copy the substring of STRING1 from START (inclusive) to END
3458 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
3459 string, and the source and destination areas may overlap; in all
3460 cases, the function behaves as if all the characters were copied
3463 *** Extended functions: substring-move-left! substring-move-right!
3465 These functions now correctly copy arbitrarily overlapping substrings;
3466 they are both synonyms for substring-move!.
3469 ** New module (ice-9 getopt-long), with the function `getopt-long'.
3471 getopt-long is a function for parsing command-line arguments in a
3472 manner consistent with other GNU programs.
3474 (getopt-long ARGS GRAMMAR)
3475 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
3477 ARGS should be a list of strings. Its first element should be the
3478 name of the program; subsequent elements should be the arguments
3479 that were passed to the program on the command line. The
3480 `program-arguments' procedure returns a list of this form.
3482 GRAMMAR is a list of the form:
3483 ((OPTION (PROPERTY VALUE) ...) ...)
3485 Each OPTION should be a symbol. `getopt-long' will accept a
3486 command-line option named `--OPTION'.
3487 Each option can have the following (PROPERTY VALUE) pairs:
3489 (single-char CHAR) --- Accept `-CHAR' as a single-character
3490 equivalent to `--OPTION'. This is how to specify traditional
3492 (required? BOOL) --- If BOOL is true, the option is required.
3493 getopt-long will raise an error if it is not found in ARGS.
3494 (value BOOL) --- If BOOL is #t, the option accepts a value; if
3495 it is #f, it does not; and if it is the symbol
3496 `optional', the option may appear in ARGS with or
3498 (predicate FUNC) --- If the option accepts a value (i.e. you
3499 specified `(value #t)' for this option), then getopt
3500 will apply FUNC to the value, and throw an exception
3501 if it returns #f. FUNC should be a procedure which
3502 accepts a string and returns a boolean value; you may
3503 need to use quasiquotes to get it into GRAMMAR.
3505 The (PROPERTY VALUE) pairs may occur in any order, but each
3506 property may occur only once. By default, options do not have
3507 single-character equivalents, are not required, and do not take
3510 In ARGS, single-character options may be combined, in the usual
3511 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
3512 accepts values, then it must be the last option in the
3513 combination; the value is the next argument. So, for example, using
3514 the following grammar:
3515 ((apples (single-char #\a))
3516 (blimps (single-char #\b) (value #t))
3517 (catalexis (single-char #\c) (value #t)))
3518 the following argument lists would be acceptable:
3519 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
3520 for "blimps" and "catalexis")
3521 ("-ab" "bang" "-c" "couth") (same)
3522 ("-ac" "couth" "-b" "bang") (same)
3523 ("-abc" "couth" "bang") (an error, since `-b' is not the
3524 last option in its combination)
3526 If an option's value is optional, then `getopt-long' decides
3527 whether it has a value by looking at what follows it in ARGS. If
3528 the next element is a string, and it does not appear to be an
3529 option itself, then that string is the option's value.
3531 The value of a long option can appear as the next element in ARGS,
3532 or it can follow the option name, separated by an `=' character.
3533 Thus, using the same grammar as above, the following argument lists
3535 ("--apples" "Braeburn" "--blimps" "Goodyear")
3536 ("--apples=Braeburn" "--blimps" "Goodyear")
3537 ("--blimps" "Goodyear" "--apples=Braeburn")
3539 If the option "--" appears in ARGS, argument parsing stops there;
3540 subsequent arguments are returned as ordinary arguments, even if
3541 they resemble options. So, in the argument list:
3542 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
3543 `getopt-long' will recognize the `apples' option as having the
3544 value "Granny Smith", but it will not recognize the `blimp'
3545 option; it will return the strings "--blimp" and "Goodyear" as
3546 ordinary argument strings.
3548 The `getopt-long' function returns the parsed argument list as an
3549 assocation list, mapping option names --- the symbols from GRAMMAR
3550 --- onto their values, or #t if the option does not accept a value.
3551 Unused options do not appear in the alist.
3553 All arguments that are not the value of any option are returned
3554 as a list, associated with the empty list.
3556 `getopt-long' throws an exception if:
3557 - it finds an unrecognized option in ARGS
3558 - a required option is omitted
3559 - an option that requires an argument doesn't get one
3560 - an option that doesn't accept an argument does get one (this can
3561 only happen using the long option `--opt=value' syntax)
3562 - an option predicate fails
3567 `((lockfile-dir (required? #t)
3570 (predicate ,file-is-directory?))
3571 (verbose (required? #f)
3574 (x-includes (single-char #\x))
3575 (rnet-server (single-char #\y)
3576 (predicate ,string?))))
3578 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
3579 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
3581 => ((() "foo1" "-fred" "foo2" "foo3")
3582 (rnet-server . "lamprod")
3583 (x-includes . "/usr/include")
3584 (lockfile-dir . "/tmp")
3587 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
3589 It will be removed in a few releases.
3591 ** New syntax: lambda*
3592 ** New syntax: define*
3593 ** New syntax: define*-public
3594 ** New syntax: defmacro*
3595 ** New syntax: defmacro*-public
3596 Guile now supports optional arguments.
3598 `lambda*', `define*', `define*-public', `defmacro*' and
3599 `defmacro*-public' are identical to the non-* versions except that
3600 they use an extended type of parameter list that has the following BNF
3601 syntax (parentheses are literal, square brackets indicate grouping,
3602 and `*', `+' and `?' have the usual meaning):
3604 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
3605 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
3606 [[#&rest identifier]|[. identifier]]? ) | [identifier]
3608 ext-var-decl ::= identifier | ( identifier expression )
3610 The semantics are best illustrated with the following documentation
3611 and examples for `lambda*':
3614 lambda extended for optional and keyword arguments
3616 lambda* creates a procedure that takes optional arguments. These
3617 are specified by putting them inside brackets at the end of the
3618 paramater list, but before any dotted rest argument. For example,
3619 (lambda* (a b #&optional c d . e) '())
3620 creates a procedure with fixed arguments a and b, optional arguments c
3621 and d, and rest argument e. If the optional arguments are omitted
3622 in a call, the variables for them are unbound in the procedure. This
3623 can be checked with the bound? macro.
3625 lambda* can also take keyword arguments. For example, a procedure
3627 (lambda* (#&key xyzzy larch) '())
3628 can be called with any of the argument lists (#:xyzzy 11)
3629 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
3630 are given as keywords are bound to values.
3632 Optional and keyword arguments can also be given default values
3633 which they take on when they are not present in a call, by giving a
3634 two-item list in place of an optional argument, for example in:
3635 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
3636 foo is a fixed argument, bar is an optional argument with default
3637 value 42, and baz is a keyword argument with default value 73.
3638 Default value expressions are not evaluated unless they are needed
3639 and until the procedure is called.
3641 lambda* now supports two more special parameter list keywords.
3643 lambda*-defined procedures now throw an error by default if a
3644 keyword other than one of those specified is found in the actual
3645 passed arguments. However, specifying #&allow-other-keys
3646 immediately after the kyword argument declarations restores the
3647 previous behavior of ignoring unknown keywords. lambda* also now
3648 guarantees that if the same keyword is passed more than once, the
3649 last one passed is the one that takes effect. For example,
3650 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
3651 #:heads 37 #:tails 42 #:heads 99)
3652 would result in (99 47) being displayed.
3654 #&rest is also now provided as a synonym for the dotted syntax rest
3655 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
3656 all respects to lambda*. This is provided for more similarity to DSSSL,
3657 MIT-Scheme and Kawa among others, as well as for refugees from other
3660 Further documentation may be found in the optargs.scm file itself.
3662 The optional argument module also exports the macros `let-optional',
3663 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
3664 are not documented here because they may be removed in the future, but
3665 full documentation is still available in optargs.scm.
3667 ** New syntax: and-let*
3668 Guile now supports the `and-let*' form, described in the draft SRFI-2.
3670 Syntax: (land* (<clause> ...) <body> ...)
3671 Each <clause> should have one of the following forms:
3672 (<variable> <expression>)
3675 Each <variable> or <bound-variable> should be an identifier. Each
3676 <expression> should be a valid expression. The <body> should be a
3677 possibly empty sequence of expressions, like the <body> of a
3680 Semantics: A LAND* expression is evaluated by evaluating the
3681 <expression> or <bound-variable> of each of the <clause>s from
3682 left to right. The value of the first <expression> or
3683 <bound-variable> that evaluates to a false value is returned; the
3684 remaining <expression>s and <bound-variable>s are not evaluated.
3685 The <body> forms are evaluated iff all the <expression>s and
3686 <bound-variable>s evaluate to true values.
3688 The <expression>s and the <body> are evaluated in an environment
3689 binding each <variable> of the preceding (<variable> <expression>)
3690 clauses to the value of the <expression>. Later bindings
3691 shadow earlier bindings.
3693 Guile's and-let* macro was contributed by Michael Livshin.
3695 ** New sorting functions
3697 *** New function: sorted? SEQUENCE LESS?
3698 Returns `#t' when the sequence argument is in non-decreasing order
3699 according to LESS? (that is, there is no adjacent pair `... x y
3700 ...' for which `(less? y x)').
3702 Returns `#f' when the sequence contains at least one out-of-order
3703 pair. It is an error if the sequence is neither a list nor a
3706 *** New function: merge LIST1 LIST2 LESS?
3707 LIST1 and LIST2 are sorted lists.
3708 Returns the sorted list of all elements in LIST1 and LIST2.
3710 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
3711 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
3712 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
3713 (Here "<" should read "comes before".)
3715 *** New procedure: merge! LIST1 LIST2 LESS?
3716 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
3717 the result. If the code is compiled, and LESS? constructs no new
3718 pairs, no pairs at all will be allocated. The first pair of the
3719 result will be either the first pair of LIST1 or the first pair of
3722 *** New function: sort SEQUENCE LESS?
3723 Accepts either a list or a vector, and returns a new sequence
3724 which is sorted. The new sequence is the same type as the input.
3725 Always `(sorted? (sort sequence less?) less?)'. The original
3726 sequence is not altered in any way. The new sequence shares its
3727 elements with the old one; no elements are copied.
3729 *** New procedure: sort! SEQUENCE LESS
3730 Returns its sorted result in the original boxes. No new storage is
3731 allocated at all. Proper usage: (set! slist (sort! slist <))
3733 *** New function: stable-sort SEQUENCE LESS?
3734 Similar to `sort' but stable. That is, if "equal" elements are
3735 ordered a < b in the original sequence, they will have the same order
3738 *** New function: stable-sort! SEQUENCE LESS?
3739 Similar to `sort!' but stable.
3740 Uses temporary storage when sorting vectors.
3742 *** New functions: sort-list, sort-list!
3743 Added for compatibility with scsh.
3745 ** New built-in random number support
3747 *** New function: random N [STATE]
3748 Accepts a positive integer or real N and returns a number of the
3749 same type between zero (inclusive) and N (exclusive). The values
3750 returned have a uniform distribution.
3752 The optional argument STATE must be of the type produced by
3753 `copy-random-state' or `seed->random-state'. It defaults to the value
3754 of the variable `*random-state*'. This object is used to maintain the
3755 state of the pseudo-random-number generator and is altered as a side
3756 effect of the `random' operation.
3758 *** New variable: *random-state*
3759 Holds a data structure that encodes the internal state of the
3760 random-number generator that `random' uses by default. The nature
3761 of this data structure is implementation-dependent. It may be
3762 printed out and successfully read back in, but may or may not
3763 function correctly as a random-number state object in another
3766 *** New function: copy-random-state [STATE]
3767 Returns a new object of type suitable for use as the value of the
3768 variable `*random-state*' and as a second argument to `random'.
3769 If argument STATE is given, a copy of it is returned. Otherwise a
3770 copy of `*random-state*' is returned.
3772 *** New function: seed->random-state SEED
3773 Returns a new object of type suitable for use as the value of the
3774 variable `*random-state*' and as a second argument to `random'.
3775 SEED is a string or a number. A new state is generated and
3776 initialized using SEED.
3778 *** New function: random:uniform [STATE]
3779 Returns an uniformly distributed inexact real random number in the
3780 range between 0 and 1.
3782 *** New procedure: random:solid-sphere! VECT [STATE]
3783 Fills VECT with inexact real random numbers the sum of whose
3784 squares is less than 1.0. Thinking of VECT as coordinates in
3785 space of dimension N = `(vector-length VECT)', the coordinates are
3786 uniformly distributed within the unit N-shere. The sum of the
3787 squares of the numbers is returned. VECT can be either a vector
3788 or a uniform vector of doubles.
3790 *** New procedure: random:hollow-sphere! VECT [STATE]
3791 Fills VECT with inexact real random numbers the sum of whose squares
3792 is equal to 1.0. Thinking of VECT as coordinates in space of
3793 dimension n = `(vector-length VECT)', the coordinates are uniformly
3794 distributed over the surface of the unit n-shere. VECT can be either
3795 a vector or a uniform vector of doubles.
3797 *** New function: random:normal [STATE]
3798 Returns an inexact real in a normal distribution with mean 0 and
3799 standard deviation 1. For a normal distribution with mean M and
3800 standard deviation D use `(+ M (* D (random:normal)))'.
3802 *** New procedure: random:normal-vector! VECT [STATE]
3803 Fills VECT with inexact real random numbers which are independent and
3804 standard normally distributed (i.e., with mean 0 and variance 1).
3805 VECT can be either a vector or a uniform vector of doubles.
3807 *** New function: random:exp STATE
3808 Returns an inexact real in an exponential distribution with mean 1.
3809 For an exponential distribution with mean U use (* U (random:exp)).
3811 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
3813 These functions now operate on numbers in the range of a C unsigned
3816 These functions used to operate on numbers in the range of a C signed
3817 long; however, this seems inappropriate, because Guile integers don't
3820 ** New function: make-guardian
3821 This is an implementation of guardians as described in
3822 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
3823 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
3824 Programming Language Design and Implementation, June 1993
3825 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
3827 ** New functions: delq1!, delv1!, delete1!
3828 These procedures behave similar to delq! and friends but delete only
3829 one object if at all.
3831 ** New function: unread-string STRING PORT
3832 Unread STRING to PORT, that is, push it back onto the port so that
3833 next read operation will work on the pushed back characters.
3835 ** unread-char can now be called multiple times
3836 If unread-char is called multiple times, the unread characters will be
3837 read again in last-in first-out order.
3839 ** the procedures uniform-array-read! and uniform-array-write! now
3840 work on any kind of port, not just ports which are open on a file.
3842 ** Now 'l' in a port mode requests line buffering.
3844 ** The procedure truncate-file now works on string ports as well
3845 as file ports. If the size argument is omitted, the current
3846 file position is used.
3848 ** new procedure: seek PORT/FDES OFFSET WHENCE
3849 The arguments are the same as for the old fseek procedure, but it
3850 works on string ports as well as random-access file ports.
3852 ** the fseek procedure now works on string ports, since it has been
3853 redefined using seek.
3855 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
3856 size is not supplied.
3858 ** the newline procedure no longer flushes the port if it's not
3859 line-buffered: previously it did if it was the current output port.
3861 ** open-pipe and close-pipe are no longer primitive procedures, but
3862 an emulation can be obtained using `(use-modules (ice-9 popen))'.
3864 ** the freopen procedure has been removed.
3866 ** new procedure: drain-input PORT
3867 Drains PORT's read buffers (including any pushed-back characters)
3868 and returns the contents as a single string.
3870 ** New function: map-in-order PROC LIST1 LIST2 ...
3871 Version of `map' which guarantees that the procedure is applied to the
3872 lists in serial order.
3874 ** Renamed `serial-array-copy!' and `serial-array-map!' to
3875 `array-copy-in-order!' and `array-map-in-order!'. The old names are
3876 now obsolete and will go away in release 1.5.
3878 ** New syntax: collect BODY1 ...
3879 Version of `begin' which returns a list of the results of the body
3880 forms instead of the result of the last body form. In contrast to
3881 `begin', `collect' allows an empty body.
3883 ** New functions: read-history FILENAME, write-history FILENAME
3884 Read/write command line history from/to file. Returns #t on success
3885 and #f if an error occured.
3887 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
3889 These procedures return a list of definitions available in the specified
3890 argument, a relative module reference. In the case of no argument,
3891 `(current-module)' is now consulted for definitions to return, instead
3892 of simply returning #f, the former behavior.
3894 ** The #/ syntax for lists is no longer supported.
3896 Earlier versions of Scheme accepted this syntax, but printed a
3899 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
3901 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
3904 * Changes to the gh_ interface
3908 Now takes a second argument which is the result array. If this
3909 pointer is NULL, a new array is malloced (the old behaviour).
3911 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
3912 gh_scm2shorts, gh_scm2longs, gh_scm2floats
3916 * Changes to the scm_ interface
3918 ** Function: scm_make_named_hook (char* name, int n_args)
3920 Creates a hook in the same way as make-hook above but also
3921 binds a variable named NAME to it.
3923 This is the typical way of creating a hook from C code.
3925 Currently, the variable is created in the "current" module. This
3926 might change when we get the new module system.
3928 ** The smob interface
3930 The interface for creating smobs has changed. For documentation, see
3931 data-rep.info (made from guile-core/doc/data-rep.texi).
3933 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
3935 >>> This function will be removed in 1.3.4. <<<
3939 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
3940 This function adds a new smob type, named NAME, with instance size
3941 SIZE to the system. The return value is a tag that is used in
3942 creating instances of the type. If SIZE is 0, then no memory will
3943 be allocated when instances of the smob are created, and nothing
3944 will be freed by the default free function.
3946 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
3947 This function sets the smob marking procedure for the smob type
3948 specified by the tag TC. TC is the tag returned by
3949 `scm_make_smob_type'.
3951 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
3952 This function sets the smob freeing procedure for the smob type
3953 specified by the tag TC. TC is the tag returned by
3954 `scm_make_smob_type'.
3956 *** Function: void scm_set_smob_print (tc, print)
3958 - Function: void scm_set_smob_print (long tc,
3959 scm_sizet (*print) (SCM,
3963 This function sets the smob printing procedure for the smob type
3964 specified by the tag TC. TC is the tag returned by
3965 `scm_make_smob_type'.
3967 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
3968 This function sets the smob equality-testing predicate for the
3969 smob type specified by the tag TC. TC is the tag returned by
3970 `scm_make_smob_type'.
3972 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
3973 Make VALUE contain a smob instance of the type with type code TC and
3974 smob data DATA. VALUE must be previously declared as C type `SCM'.
3976 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
3977 This macro expands to a block of code that creates a smob instance
3978 of the type with type code TC and smob data DATA, and returns that
3979 `SCM' value. It should be the last piece of code in a block.
3981 ** The interfaces for using I/O ports and implementing port types
3982 (ptobs) have changed significantly. The new interface is based on
3983 shared access to buffers and a new set of ptob procedures.
3985 *** scm_newptob has been removed
3989 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
3991 - Function: SCM scm_make_port_type (char *type_name,
3992 int (*fill_buffer) (SCM port),
3993 void (*write_flush) (SCM port));
3995 Similarly to the new smob interface, there is a set of function
3996 setters by which the user can customize the behaviour of his port
3997 type. See ports.h (scm_set_port_XXX).
3999 ** scm_strport_to_string: New function: creates a new string from
4000 a string port's buffer.
4002 ** Plug in interface for random number generators
4003 The variable `scm_the_rng' in random.c contains a value and three
4004 function pointers which together define the current random number
4005 generator being used by the Scheme level interface and the random
4006 number library functions.
4008 The user is free to replace the default generator with the generator
4011 *** Variable: size_t scm_the_rng.rstate_size
4012 The size of the random state type used by the current RNG
4015 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4016 Given the random STATE, return 32 random bits.
4018 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4019 Seed random state STATE using string S of length N.
4021 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4022 Given random state STATE, return a malloced copy.
4025 The default RNG is the MWC (Multiply With Carry) random number
4026 generator described by George Marsaglia at the Department of
4027 Statistics and Supercomputer Computations Research Institute, The
4028 Florida State University (http://stat.fsu.edu/~geo).
4030 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4031 passes all tests in the DIEHARD test suite
4032 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4033 costs one multiply and one add on platforms which either supports long
4034 longs (gcc does this on most systems) or have 64 bit longs. The cost
4035 is four multiply on other systems but this can be optimized by writing
4036 scm_i_uniform32 in assembler.
4038 These functions are provided through the scm_the_rng interface for use
4039 by libguile and the application.
4041 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4042 Given the random STATE, return 32 random bits.
4043 Don't use this function directly. Instead go through the plugin
4044 interface (see "Plug in interface" above).
4046 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4047 Initialize STATE using SEED of length N.
4049 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4050 Return a malloc:ed copy of STATE. This function can easily be re-used
4051 in the interfaces to other RNGs.
4053 ** Random number library functions
4054 These functions use the current RNG through the scm_the_rng interface.
4055 It might be a good idea to use these functions from your C code so
4056 that only one random generator is used by all code in your program.
4058 The default random state is stored in:
4060 *** Variable: SCM scm_var_random_state
4061 Contains the vcell of the Scheme variable "*random-state*" which is
4062 used as default state by all random number functions in the Scheme
4067 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4069 *** Function: scm_rstate *scm_c_default_rstate (void)
4070 This is a convenience function which returns the value of
4071 scm_var_random_state. An error message is generated if this value
4072 isn't a random state.
4074 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4075 Make a new random state from the string SEED of length LENGTH.
4077 It is generally not a good idea to use multiple random states in a
4078 program. While subsequent random numbers generated from one random
4079 state are guaranteed to be reasonably independent, there is no such
4080 guarantee for numbers generated from different random states.
4082 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4083 Return 32 random bits.
4085 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4086 Return a sample from the uniform(0,1) distribution.
4088 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4089 Return a sample from the normal(0,1) distribution.
4091 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4092 Return a sample from the exp(1) distribution.
4094 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4095 Return a sample from the discrete uniform(0,M) distribution.
4097 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4098 Return a sample from the discrete uniform(0,M) distribution.
4099 M must be a bignum object. The returned value may be an INUM.
4103 Changes in Guile 1.3 (released Monday, October 19, 1998):
4105 * Changes to the distribution
4107 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4108 To avoid conflicts, programs should name environment variables after
4109 themselves, except when there's a common practice establishing some
4112 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4113 giving the former precedence, and printing a warning message if the
4114 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4116 ** The header files related to multi-byte characters have been removed.
4117 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4118 which referred to these explicitly will probably need to be rewritten,
4119 since the support for the variant string types has been removed; see
4122 ** The header files append.h and sequences.h have been removed. These
4123 files implemented non-R4RS operations which would encourage
4124 non-portable programming style and less easy-to-read code.
4126 * Changes to the stand-alone interpreter
4128 ** New procedures have been added to implement a "batch mode":
4130 *** Function: batch-mode?
4132 Returns a boolean indicating whether the interpreter is in batch
4135 *** Function: set-batch-mode?! ARG
4137 If ARG is true, switches the interpreter to batch mode. The `#f'
4138 case has not been implemented.
4140 ** Guile now provides full command-line editing, when run interactively.
4141 To use this feature, you must have the readline library installed.
4142 The Guile build process will notice it, and automatically include
4145 The readline library is available via anonymous FTP from any GNU
4146 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4148 ** the-last-stack is now a fluid.
4150 * Changes to the procedure for linking libguile with your programs
4152 ** You can now use the `guile-config' utility to build programs that use Guile.
4154 Guile now includes a command-line utility called `guile-config', which
4155 can provide information about how to compile and link programs that
4158 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4159 You should include this command's output on the command line you use
4160 to compile C or C++ code that #includes the Guile header files. It's
4161 usually just a `-I' flag to help the compiler find the Guile headers.
4164 *** `guile-config link' prints any linker flags necessary to link with Guile.
4166 This command writes to its standard output a list of flags which you
4167 must pass to the linker to link your code against the Guile library.
4168 The flags include '-lguile' itself, any other libraries the Guile
4169 library depends upon, and any `-L' flags needed to help the linker
4170 find those libraries.
4172 For example, here is a Makefile rule that builds a program named 'foo'
4173 from the object files ${FOO_OBJECTS}, and links them against Guile:
4176 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4178 Previous Guile releases recommended that you use autoconf to detect
4179 which of a predefined set of libraries were present on your system.
4180 It is more robust to use `guile-config', since it records exactly which
4181 libraries the installed Guile library requires.
4183 This was originally called `build-guile', but was renamed to
4184 `guile-config' before Guile 1.3 was released, to be consistent with
4185 the analogous script for the GTK+ GUI toolkit, which is called
4189 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4191 If you are using the GNU autoconf package to configure your program,
4192 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4193 (described above) and gather the necessary values for use in your
4196 The GUILE_FLAGS macro expands to configure script code which runs the
4197 `guile-config' script, to find out where Guile's header files and
4198 libraries are installed. It sets two variables, marked for
4199 substitution, as by AC_SUBST.
4201 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4202 code that uses Guile header files. This is almost always just a
4205 GUILE_LDFLAGS --- flags to pass to the linker to link a
4206 program against Guile. This includes `-lguile' for the Guile
4207 library itself, any libraries that Guile itself requires (like
4208 -lqthreads), and so on. It may also include a -L flag to tell the
4209 compiler where to find the libraries.
4211 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4212 directory of the Guile distribution. You can copy it into your
4213 package's aclocal.m4 file, and then use it in your configure.in file.
4215 If you are using the `aclocal' program, distributed with GNU automake,
4216 to maintain your aclocal.m4 file, the Guile installation process
4217 installs guile.m4 where aclocal will find it. All you need to do is
4218 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4219 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4223 * Changes to Scheme functions and syntax
4225 ** Multi-byte strings have been removed, as have multi-byte and wide
4226 ports. We felt that these were the wrong approach to
4227 internationalization support.
4229 ** New function: readline [PROMPT]
4230 Read a line from the terminal, and allow the user to edit it,
4231 prompting with PROMPT. READLINE provides a large set of Emacs-like
4232 editing commands, lets the user recall previously typed lines, and
4233 works on almost every kind of terminal, including dumb terminals.
4235 READLINE assumes that the cursor is at the beginning of the line when
4236 it is invoked. Thus, you can't print a prompt yourself, and then call
4237 READLINE; you need to package up your prompt as a string, pass it to
4238 the function, and let READLINE print the prompt itself. This is
4239 because READLINE needs to know the prompt's screen width.
4241 For Guile to provide this function, you must have the readline
4242 library, version 2.1 or later, installed on your system. Readline is
4243 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4244 any GNU mirror site.
4246 See also ADD-HISTORY function.
4248 ** New function: add-history STRING
4249 Add STRING as the most recent line in the history used by the READLINE
4250 command. READLINE does not add lines to the history itself; you must
4251 call ADD-HISTORY to make previous input available to the user.
4253 ** The behavior of the read-line function has changed.
4255 This function now uses standard C library functions to read the line,
4256 for speed. This means that it doesn not respect the value of
4257 scm-line-incrementors; it assumes that lines are delimited with
4260 (Note that this is read-line, the function that reads a line of text
4261 from a port, not readline, the function that reads a line from a
4262 terminal, providing full editing capabilities.)
4264 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4266 This module provides some simple argument parsing. It exports one
4269 Function: getopt-gnu-style ARG-LS
4270 Parse a list of program arguments into an alist of option
4273 Each item in the list of program arguments is examined to see if
4274 it meets the syntax of a GNU long-named option. An argument like
4275 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4276 returned alist, where MUMBLE is a keyword object with the same
4277 name as the argument. An argument like `--MUMBLE=FROB' produces
4278 an element of the form (MUMBLE . FROB), where FROB is a string.
4280 As a special case, the returned alist also contains a pair whose
4281 car is the symbol `rest'. The cdr of this pair is a list
4282 containing all the items in the argument list that are not options
4283 of the form mentioned above.
4285 The argument `--' is treated specially: all items in the argument
4286 list appearing after such an argument are not examined, and are
4287 returned in the special `rest' list.
4289 This function does not parse normal single-character switches.
4290 You will need to parse them out of the `rest' list yourself.
4292 ** The read syntax for byte vectors and short vectors has changed.
4294 Instead of #bytes(...), write #y(...).
4296 Instead of #short(...), write #h(...).
4298 This may seem nutty, but, like the other uniform vectors, byte vectors
4299 and short vectors want to have the same print and read syntax (and,
4300 more basic, want to have read syntax!). Changing the read syntax to
4301 use multiple characters after the hash sign breaks with the
4302 conventions used in R5RS and the conventions used for the other
4303 uniform vectors. It also introduces complexity in the current reader,
4304 both on the C and Scheme levels. (The Right solution is probably to
4305 change the syntax and prototypes for uniform vectors entirely.)
4308 ** The new module (ice-9 session) provides useful interactive functions.
4310 *** New procedure: (apropos REGEXP OPTION ...)
4312 Display a list of top-level variables whose names match REGEXP, and
4313 the modules they are imported from. Each OPTION should be one of the
4316 value --- Show the value of each matching variable.
4317 shadow --- Show bindings shadowed by subsequently imported modules.
4318 full --- Same as both `shadow' and `value'.
4322 guile> (apropos "trace" 'full)
4323 debug: trace #<procedure trace args>
4324 debug: untrace #<procedure untrace args>
4325 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4326 the-scm-module: before-backtrace-hook ()
4327 the-scm-module: backtrace #<primitive-procedure backtrace>
4328 the-scm-module: after-backtrace-hook ()
4329 the-scm-module: has-shown-backtrace-hint? #f
4332 ** There are new functions and syntax for working with macros.
4334 Guile implements macros as a special object type. Any variable whose
4335 top-level binding is a macro object acts as a macro. The macro object
4336 specifies how the expression should be transformed before evaluation.
4338 *** Macro objects now print in a reasonable way, resembling procedures.
4340 *** New function: (macro? OBJ)
4341 True iff OBJ is a macro object.
4343 *** New function: (primitive-macro? OBJ)
4344 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4345 macro transformers, implemented in eval.c rather than Scheme code.
4347 Why do we have this function?
4348 - For symmetry with procedure? and primitive-procedure?,
4349 - to allow custom print procedures to tell whether a macro is
4350 primitive, and display it differently, and
4351 - to allow compilers and user-written evaluators to distinguish
4352 builtin special forms from user-defined ones, which could be
4355 *** New function: (macro-type OBJ)
4356 Return a value indicating what kind of macro OBJ is. Possible return
4359 The symbol `syntax' --- a macro created by procedure->syntax.
4360 The symbol `macro' --- a macro created by procedure->macro.
4361 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4362 The boolean #f --- if OBJ is not a macro object.
4364 *** New function: (macro-name MACRO)
4365 Return the name of the macro object MACRO's procedure, as returned by
4368 *** New function: (macro-transformer MACRO)
4369 Return the transformer procedure for MACRO.
4371 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4373 Specify a new macro expander to use in the current module. Each
4374 MODULE is a module name, with the same meaning as in the `use-modules'
4375 form; each named module's exported bindings are added to the current
4376 top-level environment. TRANSFORMER is an expression evaluated in the
4377 resulting environment which must yield a procedure to use as the
4378 module's eval transformer: every expression evaluated in this module
4379 is passed to this function, and the result passed to the Guile
4382 *** macro-eval! is removed. Use local-eval instead.
4384 ** Some magic has been added to the printer to better handle user
4385 written printing routines (like record printers, closure printers).
4387 The problem is that these user written routines must have access to
4388 the current `print-state' to be able to handle fancy things like
4389 detection of circular references. These print-states have to be
4390 passed to the builtin printing routines (display, write, etc) to
4391 properly continue the print chain.
4393 We didn't want to change all existing print code so that it
4394 explicitly passes thru a print state in addition to a port. Instead,
4395 we extented the possible values that the builtin printing routines
4396 accept as a `port'. In addition to a normal port, they now also take
4397 a pair of a normal port and a print-state. Printing will go to the
4398 port and the print-state will be used to control the detection of
4399 circular references, etc. If the builtin function does not care for a
4400 print-state, it is simply ignored.
4402 User written callbacks are now called with such a pair as their
4403 `port', but because every function now accepts this pair as a PORT
4404 argument, you don't have to worry about that. In fact, it is probably
4405 safest to not check for these pairs.
4407 However, it is sometimes necessary to continue a print chain on a
4408 different port, for example to get a intermediate string
4409 representation of the printed value, mangle that string somehow, and
4410 then to finally print the mangled string. Use the new function
4412 inherit-print-state OLD-PORT NEW-PORT
4414 for this. It constructs a new `port' that prints to NEW-PORT but
4415 inherits the print-state of OLD-PORT.
4417 ** struct-vtable-offset renamed to vtable-offset-user
4419 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
4421 ** There is now a third optional argument to make-vtable-vtable
4422 (and fourth to make-struct) when constructing new types (vtables).
4423 This argument initializes field vtable-index-printer of the vtable.
4425 ** The detection of circular references has been extended to structs.
4426 That is, a structure that -- in the process of being printed -- prints
4427 itself does not lead to infinite recursion.
4429 ** There is now some basic support for fluids. Please read
4430 "libguile/fluid.h" to find out more. It is accessible from Scheme with
4431 the following functions and macros:
4433 Function: make-fluid
4435 Create a new fluid object. Fluids are not special variables or
4436 some other extension to the semantics of Scheme, but rather
4437 ordinary Scheme objects. You can store them into variables (that
4438 are still lexically scoped, of course) or into any other place you
4439 like. Every fluid has a initial value of `#f'.
4441 Function: fluid? OBJ
4443 Test whether OBJ is a fluid.
4445 Function: fluid-ref FLUID
4446 Function: fluid-set! FLUID VAL
4448 Access/modify the fluid FLUID. Modifications are only visible
4449 within the current dynamic root (that includes threads).
4451 Function: with-fluids* FLUIDS VALUES THUNK
4453 FLUIDS is a list of fluids and VALUES a corresponding list of
4454 values for these fluids. Before THUNK gets called the values are
4455 installed in the fluids and the old values of the fluids are
4456 saved in the VALUES list. When the flow of control leaves THUNK
4457 or reenters it, the values get swapped again. You might think of
4458 this as a `safe-fluid-excursion'. Note that the VALUES list is
4459 modified by `with-fluids*'.
4461 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
4463 The same as `with-fluids*' but with a different syntax. It looks
4464 just like `let', but both FLUID and VALUE are evaluated. Remember,
4465 fluids are not special variables but ordinary objects. FLUID
4466 should evaluate to a fluid.
4468 ** Changes to system call interfaces:
4470 *** close-port, close-input-port and close-output-port now return a
4471 boolean instead of an `unspecified' object. #t means that the port
4472 was successfully closed, while #f means it was already closed. It is
4473 also now possible for these procedures to raise an exception if an
4474 error occurs (some errors from write can be delayed until close.)
4476 *** the first argument to chmod, fcntl, ftell and fseek can now be a
4479 *** the third argument to fcntl is now optional.
4481 *** the first argument to chown can now be a file descriptor or a port.
4483 *** the argument to stat can now be a port.
4485 *** The following new procedures have been added (most use scsh
4488 *** procedure: close PORT/FD
4489 Similar to close-port (*note close-port: Closing Ports.), but also
4490 works on file descriptors. A side effect of closing a file
4491 descriptor is that any ports using that file descriptor are moved
4492 to a different file descriptor and have their revealed counts set
4495 *** procedure: port->fdes PORT
4496 Returns the integer file descriptor underlying PORT. As a side
4497 effect the revealed count of PORT is incremented.
4499 *** procedure: fdes->ports FDES
4500 Returns a list of existing ports which have FDES as an underlying
4501 file descriptor, without changing their revealed counts.
4503 *** procedure: fdes->inport FDES
4504 Returns an existing input port which has FDES as its underlying
4505 file descriptor, if one exists, and increments its revealed count.
4506 Otherwise, returns a new input port with a revealed count of 1.
4508 *** procedure: fdes->outport FDES
4509 Returns an existing output port which has FDES as its underlying
4510 file descriptor, if one exists, and increments its revealed count.
4511 Otherwise, returns a new output port with a revealed count of 1.
4513 The next group of procedures perform a `dup2' system call, if NEWFD
4514 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
4515 duplicated can be supplied as an integer or contained in a port. The
4516 type of value returned varies depending on which procedure is used.
4518 All procedures also have the side effect when performing `dup2' that
4519 any ports using NEWFD are moved to a different file descriptor and have
4520 their revealed counts set to zero.
4522 *** procedure: dup->fdes PORT/FD [NEWFD]
4523 Returns an integer file descriptor.
4525 *** procedure: dup->inport PORT/FD [NEWFD]
4526 Returns a new input port using the new file descriptor.
4528 *** procedure: dup->outport PORT/FD [NEWFD]
4529 Returns a new output port using the new file descriptor.
4531 *** procedure: dup PORT/FD [NEWFD]
4532 Returns a new port if PORT/FD is a port, with the same mode as the
4533 supplied port, otherwise returns an integer file descriptor.
4535 *** procedure: dup->port PORT/FD MODE [NEWFD]
4536 Returns a new port using the new file descriptor. MODE supplies a
4537 mode string for the port (*note open-file: File Ports.).
4539 *** procedure: setenv NAME VALUE
4540 Modifies the environment of the current process, which is also the
4541 default environment inherited by child processes.
4543 If VALUE is `#f', then NAME is removed from the environment.
4544 Otherwise, the string NAME=VALUE is added to the environment,
4545 replacing any existing string with name matching NAME.
4547 The return value is unspecified.
4549 *** procedure: truncate-file OBJ SIZE
4550 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
4551 can be a string containing a file name or an integer file
4552 descriptor or port open for output on the file. The underlying
4553 system calls are `truncate' and `ftruncate'.
4555 The return value is unspecified.
4557 *** procedure: setvbuf PORT MODE [SIZE]
4558 Set the buffering mode for PORT. MODE can be:
4566 block buffered, using a newly allocated buffer of SIZE bytes.
4567 However if SIZE is zero or unspecified, the port will be made
4570 This procedure should not be used after I/O has been performed with
4573 Ports are usually block buffered by default, with a default buffer
4574 size. Procedures e.g., *Note open-file: File Ports, which accept a
4575 mode string allow `0' to be added to request an unbuffered port.
4577 *** procedure: fsync PORT/FD
4578 Copies any unwritten data for the specified output file descriptor
4579 to disk. If PORT/FD is a port, its buffer is flushed before the
4580 underlying file descriptor is fsync'd. The return value is
4583 *** procedure: open-fdes PATH FLAGS [MODES]
4584 Similar to `open' but returns a file descriptor instead of a port.
4586 *** procedure: execle PATH ENV [ARG] ...
4587 Similar to `execl', but the environment of the new process is
4588 specified by ENV, which must be a list of strings as returned by
4589 the `environ' procedure.
4591 This procedure is currently implemented using the `execve' system
4592 call, but we call it `execle' because of its Scheme calling
4595 *** procedure: strerror ERRNO
4596 Returns the Unix error message corresponding to ERRNO, an integer.
4598 *** procedure: primitive-exit [STATUS]
4599 Terminate the current process without unwinding the Scheme stack.
4600 This is would typically be useful after a fork. The exit status
4601 is STATUS if supplied, otherwise zero.
4603 *** procedure: times
4604 Returns an object with information about real and processor time.
4605 The following procedures accept such an object as an argument and
4606 return a selected component:
4609 The current real time, expressed as time units relative to an
4613 The CPU time units used by the calling process.
4616 The CPU time units used by the system on behalf of the
4620 The CPU time units used by terminated child processes of the
4621 calling process, whose status has been collected (e.g., using
4625 Similarly, the CPU times units used by the system on behalf of
4626 terminated child processes.
4628 ** Removed: list-length
4629 ** Removed: list-append, list-append!
4630 ** Removed: list-reverse, list-reverse!
4632 ** array-map renamed to array-map!
4634 ** serial-array-map renamed to serial-array-map!
4636 ** catch doesn't take #f as first argument any longer
4638 Previously, it was possible to pass #f instead of a key to `catch'.
4639 That would cause `catch' to pass a jump buffer object to the procedure
4640 passed as second argument. The procedure could then use this jump
4641 buffer objekt as an argument to throw.
4643 This mechanism has been removed since its utility doesn't motivate the
4644 extra complexity it introduces.
4646 ** The `#/' notation for lists now provokes a warning message from Guile.
4647 This syntax will be removed from Guile in the near future.
4649 To disable the warning message, set the GUILE_HUSH environment
4650 variable to any non-empty value.
4652 ** The newline character now prints as `#\newline', following the
4653 normal Scheme notation, not `#\nl'.
4655 * Changes to the gh_ interface
4657 ** The gh_enter function now takes care of loading the Guile startup files.
4658 gh_enter works by calling scm_boot_guile; see the remarks below.
4660 ** Function: void gh_write (SCM x)
4662 Write the printed representation of the scheme object x to the current
4663 output port. Corresponds to the scheme level `write'.
4665 ** gh_list_length renamed to gh_length.
4667 ** vector handling routines
4669 Several major changes. In particular, gh_vector() now resembles
4670 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
4671 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
4672 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
4673 vector-related gh_ functions have been implemented.
4675 ** pair and list routines
4677 Implemented several of the R4RS pair and list functions that were
4680 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
4682 New function. Converts double arrays back and forth between Scheme
4685 * Changes to the scm_ interface
4687 ** The function scm_boot_guile now takes care of loading the startup files.
4689 Guile's primary initialization function, scm_boot_guile, now takes
4690 care of loading `boot-9.scm', in the `ice-9' module, to initialize
4691 Guile, define the module system, and put together some standard
4692 bindings. It also loads `init.scm', which is intended to hold
4693 site-specific initialization code.
4695 Since Guile cannot operate properly until boot-9.scm is loaded, there
4696 is no reason to separate loading boot-9.scm from Guile's other
4697 initialization processes.
4699 This job used to be done by scm_compile_shell_switches, which didn't
4700 make much sense; in particular, it meant that people using Guile for
4701 non-shell-like applications had to jump through hoops to get Guile
4702 initialized properly.
4704 ** The function scm_compile_shell_switches no longer loads the startup files.
4705 Now, Guile always loads the startup files, whenever it is initialized;
4706 see the notes above for scm_boot_guile and scm_load_startup_files.
4708 ** Function: scm_load_startup_files
4709 This new function takes care of loading Guile's initialization file
4710 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
4711 this is always called by the Guile initialization process, it's
4712 probably not too useful to call this yourself, but it's there anyway.
4714 ** The semantics of smob marking have changed slightly.
4716 The smob marking function (the `mark' member of the scm_smobfuns
4717 structure) is no longer responsible for setting the mark bit on the
4718 smob. The generic smob handling code in the garbage collector will
4719 set this bit. The mark function need only ensure that any other
4720 objects the smob refers to get marked.
4722 Note that this change means that the smob's GC8MARK bit is typically
4723 already set upon entry to the mark function. Thus, marking functions
4724 which look like this:
4727 if (SCM_GC8MARKP (ptr))
4729 SCM_SETGC8MARK (ptr);
4730 ... mark objects to which the smob refers ...
4733 are now incorrect, since they will return early, and fail to mark any
4734 other objects the smob refers to. Some code in the Guile library used
4737 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
4739 If you have implemented your own I/O port type, by writing the
4740 functions required by the scm_ptobfuns and then calling scm_newptob,
4741 you will need to change your functions slightly.
4743 The functions in a scm_ptobfuns structure now expect the port itself
4744 as their argument; they used to expect the `stream' member of the
4745 port's scm_port_table structure. This allows functions in an
4746 scm_ptobfuns structure to easily access the port's cell (and any flags
4747 it its CAR), and the port's scm_port_table structure.
4749 Guile now passes the I/O port itself as the `port' argument in the
4750 following scm_ptobfuns functions:
4752 int (*free) (SCM port);
4753 int (*fputc) (int, SCM port);
4754 int (*fputs) (char *, SCM port);
4755 scm_sizet (*fwrite) SCM_P ((char *ptr,
4759 int (*fflush) (SCM port);
4760 int (*fgetc) (SCM port);
4761 int (*fclose) (SCM port);
4763 The interfaces to the `mark', `print', `equalp', and `fgets' methods
4766 If you have existing code which defines its own port types, it is easy
4767 to convert your code to the new interface; simply apply SCM_STREAM to
4768 the port argument to yield the value you code used to expect.
4770 Note that since both the port and the stream have the same type in the
4771 C code --- they are both SCM values --- the C compiler will not remind
4772 you if you forget to update your scm_ptobfuns functions.
4775 ** Function: int scm_internal_select (int fds,
4779 struct timeval *timeout);
4781 This is a replacement for the `select' function provided by the OS.
4782 It enables I/O blocking and sleeping to happen for one cooperative
4783 thread without blocking other threads. It also avoids busy-loops in
4784 these situations. It is intended that all I/O blocking and sleeping
4785 will finally go through this function. Currently, this function is
4786 only available on systems providing `gettimeofday' and `select'.
4788 ** Function: SCM scm_internal_stack_catch (SCM tag,
4789 scm_catch_body_t body,
4791 scm_catch_handler_t handler,
4794 A new sibling to the other two C level `catch' functions
4795 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
4796 the stack to be saved automatically into the variable `the-last-stack'
4797 (scm_the_last_stack_var) on error. This is necessary if you want to
4798 use advanced error reporting, such as calling scm_display_error and
4799 scm_display_backtrace. (They both take a stack object as argument.)
4801 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
4803 scm_catch_handler_t handler,
4806 Spawns a new thread. It does a job similar to
4807 scm_call_with_new_thread but takes arguments more suitable when
4808 spawning threads from application C code.
4810 ** The hook scm_error_callback has been removed. It was originally
4811 intended as a way for the user to install his own error handler. But
4812 that method works badly since it intervenes between throw and catch,
4813 thereby changing the semantics of expressions like (catch #t ...).
4814 The correct way to do it is to use one of the C level catch functions
4815 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
4817 ** Removed functions:
4819 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
4820 scm_list_reverse, scm_list_reverse_x
4822 ** New macros: SCM_LISTn where n is one of the integers 0-9.
4824 These can be used for pretty list creation from C. The idea is taken
4825 from Erick Gallesio's STk.
4827 ** scm_array_map renamed to scm_array_map_x
4829 ** mbstrings are now removed
4831 This means that the type codes scm_tc7_mb_string and
4832 scm_tc7_mb_substring has been removed.
4834 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
4836 Since we no longer support multi-byte strings, these I/O functions
4837 have been simplified, and renamed. Here are their old names, and
4838 their new names and arguments:
4840 scm_gen_putc -> void scm_putc (int c, SCM port);
4841 scm_gen_puts -> void scm_puts (char *s, SCM port);
4842 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
4843 scm_gen_getc -> void scm_getc (SCM port);
4846 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
4848 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
4850 SCM_TYP7S now masks away the bit which distinguishes substrings from
4853 ** scm_catch_body_t: Backward incompatible change!
4855 Body functions to scm_internal_catch and friends do not any longer
4856 take a second argument. This is because it is no longer possible to
4857 pass a #f arg to catch.
4859 ** Calls to scm_protect_object and scm_unprotect now nest properly.
4861 The function scm_protect_object protects its argument from being freed
4862 by the garbage collector. scm_unprotect_object removes that
4865 These functions now nest properly. That is, for every object O, there
4866 is a counter which scm_protect_object(O) increments and
4867 scm_unprotect_object(O) decrements, if the counter is greater than
4868 zero. Every object's counter is zero when it is first created. If an
4869 object's counter is greater than zero, the garbage collector will not
4870 reclaim its storage.
4872 This allows you to use scm_protect_object in your code without
4873 worrying that some other function you call will call
4874 scm_unprotect_object, and allow it to be freed. Assuming that the
4875 functions you call are well-behaved, and unprotect only those objects
4876 they protect, you can follow the same rule and have confidence that
4877 objects will be freed only at appropriate times.
4880 Changes in Guile 1.2 (released Tuesday, June 24 1997):
4882 * Changes to the distribution
4884 ** Nightly snapshots are now available from ftp.red-bean.com.
4885 The old server, ftp.cyclic.com, has been relinquished to its rightful
4888 Nightly snapshots of the Guile development sources are now available via
4889 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
4891 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
4892 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
4894 ** To run Guile without installing it, the procedure has changed a bit.
4896 If you used a separate build directory to compile Guile, you'll need
4897 to include the build directory in SCHEME_LOAD_PATH, as well as the
4898 source directory. See the `INSTALL' file for examples.
4900 * Changes to the procedure for linking libguile with your programs
4902 ** The standard Guile load path for Scheme code now includes
4903 $(datadir)/guile (usually /usr/local/share/guile). This means that
4904 you can install your own Scheme files there, and Guile will find them.
4905 (Previous versions of Guile only checked a directory whose name
4906 contained the Guile version number, so you had to re-install or move
4907 your Scheme sources each time you installed a fresh version of Guile.)
4909 The load path also includes $(datadir)/guile/site; we recommend
4910 putting individual Scheme files there. If you want to install a
4911 package with multiple source files, create a directory for them under
4914 ** Guile 1.2 will now use the Rx regular expression library, if it is
4915 installed on your system. When you are linking libguile into your own
4916 programs, this means you will have to link against -lguile, -lqt (if
4917 you configured Guile with thread support), and -lrx.
4919 If you are using autoconf to generate configuration scripts for your
4920 application, the following lines should suffice to add the appropriate
4921 libraries to your link command:
4923 ### Find Rx, quickthreads and libguile.
4924 AC_CHECK_LIB(rx, main)
4925 AC_CHECK_LIB(qt, main)
4926 AC_CHECK_LIB(guile, scm_shell)
4928 The Guile 1.2 distribution does not contain sources for the Rx
4929 library, as Guile 1.0 did. If you want to use Rx, you'll need to
4930 retrieve it from a GNU FTP site and install it separately.
4932 * Changes to Scheme functions and syntax
4934 ** The dynamic linking features of Guile are now enabled by default.
4935 You can disable them by giving the `--disable-dynamic-linking' option
4938 (dynamic-link FILENAME)
4940 Find the object file denoted by FILENAME (a string) and link it
4941 into the running Guile application. When everything works out,
4942 return a Scheme object suitable for representing the linked object
4943 file. Otherwise an error is thrown. How object files are
4944 searched is system dependent.
4946 (dynamic-object? VAL)
4948 Determine whether VAL represents a dynamically linked object file.
4950 (dynamic-unlink DYNOBJ)
4952 Unlink the indicated object file from the application. DYNOBJ
4953 should be one of the values returned by `dynamic-link'.
4955 (dynamic-func FUNCTION DYNOBJ)
4957 Search the C function indicated by FUNCTION (a string or symbol)
4958 in DYNOBJ and return some Scheme object that can later be used
4959 with `dynamic-call' to actually call this function. Right now,
4960 these Scheme objects are formed by casting the address of the
4961 function to `long' and converting this number to its Scheme
4964 (dynamic-call FUNCTION DYNOBJ)
4966 Call the C function indicated by FUNCTION and DYNOBJ. The
4967 function is passed no arguments and its return value is ignored.
4968 When FUNCTION is something returned by `dynamic-func', call that
4969 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
4970 etc.), look it up in DYNOBJ; this is equivalent to
4972 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
4974 Interrupts are deferred while the C function is executing (with
4975 SCM_DEFER_INTS/SCM_ALLOW_INTS).
4977 (dynamic-args-call FUNCTION DYNOBJ ARGS)
4979 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
4980 some arguments and return its return value. The C function is
4981 expected to take two arguments and return an `int', just like
4984 int c_func (int argc, char **argv);
4986 ARGS must be a list of strings and is converted into an array of
4987 `char *'. The array is passed in ARGV and its size in ARGC. The
4988 return value is converted to a Scheme number and returned from the
4989 call to `dynamic-args-call'.
4991 When dynamic linking is disabled or not supported on your system,
4992 the above functions throw errors, but they are still available.
4994 Here is a small example that works on GNU/Linux:
4996 (define libc-obj (dynamic-link "libc.so"))
4997 (dynamic-args-call 'rand libc-obj '())
4999 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5001 ** The #/ syntax for module names is depreciated, and will be removed
5002 in a future version of Guile. Instead of
5010 The latter syntax is more consistent with existing Lisp practice.
5012 ** Guile now does fancier printing of structures. Structures are the
5013 underlying implementation for records, which in turn are used to
5014 implement modules, so all of these object now print differently and in
5015 a more informative way.
5017 The Scheme printer will examine the builtin variable *struct-printer*
5018 whenever it needs to print a structure object. When this variable is
5019 not `#f' it is deemed to be a procedure and will be applied to the
5020 structure object and the output port. When *struct-printer* is `#f'
5021 or the procedure return `#f' the structure object will be printed in
5022 the boring #<struct 80458270> form.
5024 This hook is used by some routines in ice-9/boot-9.scm to implement
5025 type specific printing routines. Please read the comments there about
5028 One of the more specific uses of structs are records. The printing
5029 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5030 called. It should behave like a *struct-printer* procedure (described
5033 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5034 token of the form #:NAME, where NAME has the same syntax as a Scheme
5035 symbol, is the external representation of the keyword named NAME.
5036 Keyword objects print using this syntax as well, so values containing
5037 keyword objects can be read back into Guile. When used in an
5038 expression, keywords are self-quoting objects.
5040 Guile suports this read syntax, and uses this print syntax, regardless
5041 of the current setting of the `keyword' read option. The `keyword'
5042 read option only controls whether Guile recognizes the `:NAME' syntax,
5043 which is incompatible with R4RS. (R4RS says such token represent
5046 ** Guile has regular expression support again. Guile 1.0 included
5047 functions for matching regular expressions, based on the Rx library.
5048 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5049 distribution, and thus Guile had no regular expression support. Guile
5050 1.2 again supports the most commonly used functions, and supports all
5051 of SCSH's regular expression functions.
5053 If your system does not include a POSIX regular expression library,
5054 and you have not linked Guile with a third-party regexp library such as
5055 Rx, these functions will not be available. You can tell whether your
5056 Guile installation includes regular expression support by checking
5057 whether the `*features*' list includes the `regex' symbol.
5059 *** regexp functions
5061 By default, Guile supports POSIX extended regular expressions. That
5062 means that the characters `(', `)', `+' and `?' are special, and must
5063 be escaped if you wish to match the literal characters.
5065 This regular expression interface was modeled after that implemented
5066 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5067 with SCSH regular expressions.
5069 **** Function: string-match PATTERN STR [START]
5070 Compile the string PATTERN into a regular expression and compare
5071 it with STR. The optional numeric argument START specifies the
5072 position of STR at which to begin matching.
5074 `string-match' returns a "match structure" which describes what,
5075 if anything, was matched by the regular expression. *Note Match
5076 Structures::. If STR does not match PATTERN at all,
5077 `string-match' returns `#f'.
5079 Each time `string-match' is called, it must compile its PATTERN
5080 argument into a regular expression structure. This operation is
5081 expensive, which makes `string-match' inefficient if the same regular
5082 expression is used several times (for example, in a loop). For better
5083 performance, you can compile a regular expression in advance and then
5084 match strings against the compiled regexp.
5086 **** Function: make-regexp STR [FLAGS]
5087 Compile the regular expression described by STR, and return the
5088 compiled regexp structure. If STR does not describe a legal
5089 regular expression, `make-regexp' throws a
5090 `regular-expression-syntax' error.
5092 FLAGS may be the bitwise-or of one or more of the following:
5094 **** Constant: regexp/extended
5095 Use POSIX Extended Regular Expression syntax when interpreting
5096 STR. If not set, POSIX Basic Regular Expression syntax is used.
5097 If the FLAGS argument is omitted, we assume regexp/extended.
5099 **** Constant: regexp/icase
5100 Do not differentiate case. Subsequent searches using the
5101 returned regular expression will be case insensitive.
5103 **** Constant: regexp/newline
5104 Match-any-character operators don't match a newline.
5106 A non-matching list ([^...]) not containing a newline matches a
5109 Match-beginning-of-line operator (^) matches the empty string
5110 immediately after a newline, regardless of whether the FLAGS
5111 passed to regexp-exec contain regexp/notbol.
5113 Match-end-of-line operator ($) matches the empty string
5114 immediately before a newline, regardless of whether the FLAGS
5115 passed to regexp-exec contain regexp/noteol.
5117 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5118 Match the compiled regular expression REGEXP against `str'. If
5119 the optional integer START argument is provided, begin matching
5120 from that position in the string. Return a match structure
5121 describing the results of the match, or `#f' if no match could be
5124 FLAGS may be the bitwise-or of one or more of the following:
5126 **** Constant: regexp/notbol
5127 The match-beginning-of-line operator always fails to match (but
5128 see the compilation flag regexp/newline above) This flag may be
5129 used when different portions of a string are passed to
5130 regexp-exec and the beginning of the string should not be
5131 interpreted as the beginning of the line.
5133 **** Constant: regexp/noteol
5134 The match-end-of-line operator always fails to match (but see the
5135 compilation flag regexp/newline above)
5137 **** Function: regexp? OBJ
5138 Return `#t' if OBJ is a compiled regular expression, or `#f'
5141 Regular expressions are commonly used to find patterns in one string
5142 and replace them with the contents of another string.
5144 **** Function: regexp-substitute PORT MATCH [ITEM...]
5145 Write to the output port PORT selected contents of the match
5146 structure MATCH. Each ITEM specifies what should be written, and
5147 may be one of the following arguments:
5149 * A string. String arguments are written out verbatim.
5151 * An integer. The submatch with that number is written.
5153 * The symbol `pre'. The portion of the matched string preceding
5154 the regexp match is written.
5156 * The symbol `post'. The portion of the matched string
5157 following the regexp match is written.
5159 PORT may be `#f', in which case nothing is written; instead,
5160 `regexp-substitute' constructs a string from the specified ITEMs
5163 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5164 Similar to `regexp-substitute', but can be used to perform global
5165 substitutions on STR. Instead of taking a match structure as an
5166 argument, `regexp-substitute/global' takes two string arguments: a
5167 REGEXP string describing a regular expression, and a TARGET string
5168 which should be matched against this regular expression.
5170 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5173 * A function may be supplied. When this function is called, it
5174 will be passed one argument: a match structure for a given
5175 regular expression match. It should return a string to be
5176 written out to PORT.
5178 * The `post' symbol causes `regexp-substitute/global' to recurse
5179 on the unmatched portion of STR. This *must* be supplied in
5180 order to perform global search-and-replace on STR; if it is
5181 not present among the ITEMs, then `regexp-substitute/global'
5182 will return after processing a single match.
5184 *** Match Structures
5186 A "match structure" is the object returned by `string-match' and
5187 `regexp-exec'. It describes which portion of a string, if any, matched
5188 the given regular expression. Match structures include: a reference to
5189 the string that was checked for matches; the starting and ending
5190 positions of the regexp match; and, if the regexp included any
5191 parenthesized subexpressions, the starting and ending positions of each
5194 In each of the regexp match functions described below, the `match'
5195 argument must be a match structure returned by a previous call to
5196 `string-match' or `regexp-exec'. Most of these functions return some
5197 information about the original target string that was matched against a
5198 regular expression; we will call that string TARGET for easy reference.
5200 **** Function: regexp-match? OBJ
5201 Return `#t' if OBJ is a match structure returned by a previous
5202 call to `regexp-exec', or `#f' otherwise.
5204 **** Function: match:substring MATCH [N]
5205 Return the portion of TARGET matched by subexpression number N.
5206 Submatch 0 (the default) represents the entire regexp match. If
5207 the regular expression as a whole matched, but the subexpression
5208 number N did not match, return `#f'.
5210 **** Function: match:start MATCH [N]
5211 Return the starting position of submatch number N.
5213 **** Function: match:end MATCH [N]
5214 Return the ending position of submatch number N.
5216 **** Function: match:prefix MATCH
5217 Return the unmatched portion of TARGET preceding the regexp match.
5219 **** Function: match:suffix MATCH
5220 Return the unmatched portion of TARGET following the regexp match.
5222 **** Function: match:count MATCH
5223 Return the number of parenthesized subexpressions from MATCH.
5224 Note that the entire regular expression match itself counts as a
5225 subexpression, and failed submatches are included in the count.
5227 **** Function: match:string MATCH
5228 Return the original TARGET string.
5230 *** Backslash Escapes
5232 Sometimes you will want a regexp to match characters like `*' or `$'
5233 exactly. For example, to check whether a particular string represents
5234 a menu entry from an Info node, it would be useful to match it against
5235 a regexp like `^* [^:]*::'. However, this won't work; because the
5236 asterisk is a metacharacter, it won't match the `*' at the beginning of
5237 the string. In this case, we want to make the first asterisk un-magic.
5239 You can do this by preceding the metacharacter with a backslash
5240 character `\'. (This is also called "quoting" the metacharacter, and
5241 is known as a "backslash escape".) When Guile sees a backslash in a
5242 regular expression, it considers the following glyph to be an ordinary
5243 character, no matter what special meaning it would ordinarily have.
5244 Therefore, we can make the above example work by changing the regexp to
5245 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5246 to match only a single asterisk in the target string.
5248 Since the backslash is itself a metacharacter, you may force a
5249 regexp to match a backslash in the target string by preceding the
5250 backslash with itself. For example, to find variable references in a
5251 TeX program, you might want to find occurrences of the string `\let\'
5252 followed by any number of alphabetic characters. The regular expression
5253 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5254 each match a single backslash in the target string.
5256 **** Function: regexp-quote STR
5257 Quote each special character found in STR with a backslash, and
5258 return the resulting string.
5260 *Very important:* Using backslash escapes in Guile source code (as
5261 in Emacs Lisp or C) can be tricky, because the backslash character has
5262 special meaning for the Guile reader. For example, if Guile encounters
5263 the character sequence `\n' in the middle of a string while processing
5264 Scheme code, it replaces those characters with a newline character.
5265 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5266 Several of these "escape sequences" are processed by the Guile reader
5267 before your code is executed. Unrecognized escape sequences are
5268 ignored: if the characters `\*' appear in a string, they will be
5269 translated to the single character `*'.
5271 This translation is obviously undesirable for regular expressions,
5272 since we want to be able to include backslashes in a string in order to
5273 escape regexp metacharacters. Therefore, to make sure that a backslash
5274 is preserved in a string in your Guile program, you must use *two*
5275 consecutive backslashes:
5277 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5279 The string in this example is preprocessed by the Guile reader before
5280 any code is executed. The resulting argument to `make-regexp' is the
5281 string `^\* [^:]*', which is what we really want.
5283 This also means that in order to write a regular expression that
5284 matches a single backslash character, the regular expression string in
5285 the source code must include *four* backslashes. Each consecutive pair
5286 of backslashes gets translated by the Guile reader to a single
5287 backslash, and the resulting double-backslash is interpreted by the
5288 regexp engine as matching a single backslash character. Hence:
5290 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5292 The reason for the unwieldiness of this syntax is historical. Both
5293 regular expression pattern matchers and Unix string processing systems
5294 have traditionally used backslashes with the special meanings described
5295 above. The POSIX regular expression specification and ANSI C standard
5296 both require these semantics. Attempting to abandon either convention
5297 would cause other kinds of compatibility problems, possibly more severe
5298 ones. Therefore, without extending the Scheme reader to support
5299 strings with different quoting conventions (an ungainly and confusing
5300 extension when implemented in other languages), we must adhere to this
5301 cumbersome escape syntax.
5303 * Changes to the gh_ interface
5305 * Changes to the scm_ interface
5307 * Changes to system call interfaces:
5309 ** The value returned by `raise' is now unspecified. It throws an exception
5312 *** A new procedure `sigaction' can be used to install signal handlers
5314 (sigaction signum [action] [flags])
5316 signum is the signal number, which can be specified using the value
5319 If action is omitted, sigaction returns a pair: the CAR is the current
5320 signal hander, which will be either an integer with the value SIG_DFL
5321 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5322 handles the signal, or #f if a non-Scheme procedure handles the
5323 signal. The CDR contains the current sigaction flags for the handler.
5325 If action is provided, it is installed as the new handler for signum.
5326 action can be a Scheme procedure taking one argument, or the value of
5327 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5328 whatever signal handler was installed before sigaction was first used.
5329 Flags can optionally be specified for the new handler (SA_RESTART is
5330 always used if the system provides it, so need not be specified.) The
5331 return value is a pair with information about the old handler as
5334 This interface does not provide access to the "signal blocking"
5335 facility. Maybe this is not needed, since the thread support may
5336 provide solutions to the problem of consistent access to data
5339 *** A new procedure `flush-all-ports' is equivalent to running
5340 `force-output' on every port open for output.
5342 ** Guile now provides information on how it was built, via the new
5343 global variable, %guile-build-info. This variable records the values
5344 of the standard GNU makefile directory variables as an assocation
5345 list, mapping variable names (symbols) onto directory paths (strings).
5346 For example, to find out where the Guile link libraries were
5347 installed, you can say:
5349 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5352 * Changes to the scm_ interface
5354 ** The new function scm_handle_by_message_noexit is just like the
5355 existing scm_handle_by_message function, except that it doesn't call
5356 exit to terminate the process. Instead, it prints a message and just
5357 returns #f. This might be a more appropriate catch-all handler for
5358 new dynamic roots and threads.
5361 Changes in Guile 1.1 (released Friday, May 16 1997):
5363 * Changes to the distribution.
5365 The Guile 1.0 distribution has been split up into several smaller
5367 guile-core --- the Guile interpreter itself.
5368 guile-tcltk --- the interface between the Guile interpreter and
5369 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5370 is a toolkit for building graphical user interfaces.
5371 guile-rgx-ctax --- the interface between Guile and the Rx regular
5372 expression matcher, and the translator for the Ctax
5373 programming language. These are packaged together because the
5374 Ctax translator uses Rx to parse Ctax source code.
5376 This NEWS file describes the changes made to guile-core since the 1.0
5379 We no longer distribute the documentation, since it was either out of
5380 date, or incomplete. As soon as we have current documentation, we
5385 * Changes to the stand-alone interpreter
5387 ** guile now accepts command-line arguments compatible with SCSH, Olin
5388 Shivers' Scheme Shell.
5390 In general, arguments are evaluated from left to right, but there are
5391 exceptions. The following switches stop argument processing, and
5392 stash all remaining command-line arguments as the value returned by
5393 the (command-line) function.
5394 -s SCRIPT load Scheme source code from FILE, and exit
5395 -c EXPR evalute Scheme expression EXPR, and exit
5396 -- stop scanning arguments; run interactively
5398 The switches below are processed as they are encountered.
5399 -l FILE load Scheme source code from FILE
5400 -e FUNCTION after reading script, apply FUNCTION to
5401 command line arguments
5402 -ds do -s script at this point
5403 --emacs enable Emacs protocol (experimental)
5404 -h, --help display this help and exit
5405 -v, --version display version information and exit
5406 \ read arguments from following script lines
5408 So, for example, here is a Guile script named `ekko' (thanks, Olin)
5409 which re-implements the traditional "echo" command:
5411 #!/usr/local/bin/guile -s
5414 (map (lambda (arg) (display arg) (display " "))
5418 (main (command-line))
5420 Suppose we invoke this script as follows:
5422 ekko a speckled gecko
5424 Through the magic of Unix script processing (triggered by the `#!'
5425 token at the top of the file), /usr/local/bin/guile receives the
5426 following list of command-line arguments:
5428 ("-s" "./ekko" "a" "speckled" "gecko")
5430 Unix inserts the name of the script after the argument specified on
5431 the first line of the file (in this case, "-s"), and then follows that
5432 with the arguments given to the script. Guile loads the script, which
5433 defines the `main' function, and then applies it to the list of
5434 remaining command-line arguments, ("a" "speckled" "gecko").
5436 In Unix, the first line of a script file must take the following form:
5438 #!INTERPRETER ARGUMENT
5440 where INTERPRETER is the absolute filename of the interpreter
5441 executable, and ARGUMENT is a single command-line argument to pass to
5444 You may only pass one argument to the interpreter, and its length is
5445 limited. These restrictions can be annoying to work around, so Guile
5446 provides a general mechanism (borrowed from, and compatible with,
5447 SCSH) for circumventing them.
5449 If the ARGUMENT in a Guile script is a single backslash character,
5450 `\', Guile will open the script file, parse arguments from its second
5451 and subsequent lines, and replace the `\' with them. So, for example,
5452 here is another implementation of the `ekko' script:
5454 #!/usr/local/bin/guile \
5458 (for-each (lambda (arg) (display arg) (display " "))
5462 If the user invokes this script as follows:
5464 ekko a speckled gecko
5466 Unix expands this into
5468 /usr/local/bin/guile \ ekko a speckled gecko
5470 When Guile sees the `\' argument, it replaces it with the arguments
5471 read from the second line of the script, producing:
5473 /usr/local/bin/guile -e main -s ekko a speckled gecko
5475 This tells Guile to load the `ekko' script, and apply the function
5476 `main' to the argument list ("a" "speckled" "gecko").
5478 Here is how Guile parses the command-line arguments:
5479 - Each space character terminates an argument. This means that two
5480 spaces in a row introduce an empty-string argument.
5481 - The tab character is not permitted (unless you quote it with the
5482 backslash character, as described below), to avoid confusion.
5483 - The newline character terminates the sequence of arguments, and will
5484 also terminate a final non-empty argument. (However, a newline
5485 following a space will not introduce a final empty-string argument;
5486 it only terminates the argument list.)
5487 - The backslash character is the escape character. It escapes
5488 backslash, space, tab, and newline. The ANSI C escape sequences
5489 like \n and \t are also supported. These produce argument
5490 constituents; the two-character combination \n doesn't act like a
5491 terminating newline. The escape sequence \NNN for exactly three
5492 octal digits reads as the character whose ASCII code is NNN. As
5493 above, characters produced this way are argument constituents.
5494 Backslash followed by other characters is not allowed.
5496 * Changes to the procedure for linking libguile with your programs
5498 ** Guile now builds and installs a shared guile library, if your
5499 system support shared libraries. (It still builds a static library on
5500 all systems.) Guile automatically detects whether your system
5501 supports shared libraries. To prevent Guile from buildisg shared
5502 libraries, pass the `--disable-shared' flag to the configure script.
5504 Guile takes longer to compile when it builds shared libraries, because
5505 it must compile every file twice --- once to produce position-
5506 independent object code, and once to produce normal object code.
5508 ** The libthreads library has been merged into libguile.
5510 To link a program against Guile, you now need only link against
5511 -lguile and -lqt; -lthreads is no longer needed. If you are using
5512 autoconf to generate configuration scripts for your application, the
5513 following lines should suffice to add the appropriate libraries to
5516 ### Find quickthreads and libguile.
5517 AC_CHECK_LIB(qt, main)
5518 AC_CHECK_LIB(guile, scm_shell)
5520 * Changes to Scheme functions
5522 ** Guile Scheme's special syntax for keyword objects is now optional,
5523 and disabled by default.
5525 The syntax variation from R4RS made it difficult to port some
5526 interesting packages to Guile. The routines which accepted keyword
5527 arguments (mostly in the module system) have been modified to also
5528 accept symbols whose names begin with `:'.
5530 To change the keyword syntax, you must first import the (ice-9 debug)
5532 (use-modules (ice-9 debug))
5534 Then you can enable the keyword syntax as follows:
5535 (read-set! keywords 'prefix)
5537 To disable keyword syntax, do this:
5538 (read-set! keywords #f)
5540 ** Many more primitive functions accept shared substrings as
5541 arguments. In the past, these functions required normal, mutable
5542 strings as arguments, although they never made use of this
5545 ** The uniform array functions now operate on byte vectors. These
5546 functions are `array-fill!', `serial-array-copy!', `array-copy!',
5547 `serial-array-map', `array-map', `array-for-each', and
5550 ** The new functions `trace' and `untrace' implement simple debugging
5551 support for Scheme functions.
5553 The `trace' function accepts any number of procedures as arguments,
5554 and tells the Guile interpreter to display each procedure's name and
5555 arguments each time the procedure is invoked. When invoked with no
5556 arguments, `trace' returns the list of procedures currently being
5559 The `untrace' function accepts any number of procedures as arguments,
5560 and tells the Guile interpreter not to trace them any more. When
5561 invoked with no arguments, `untrace' untraces all curretly traced
5564 The tracing in Guile has an advantage over most other systems: we
5565 don't create new procedure objects, but mark the procedure objects
5566 themselves. This means that anonymous and internal procedures can be
5569 ** The function `assert-repl-prompt' has been renamed to
5570 `set-repl-prompt!'. It takes one argument, PROMPT.
5571 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
5572 - If PROMPT is a string, we use it as a prompt.
5573 - If PROMPT is a procedure accepting no arguments, we call it, and
5574 display the result as a prompt.
5575 - Otherwise, we display "> ".
5577 ** The new function `eval-string' reads Scheme expressions from a
5578 string and evaluates them, returning the value of the last expression
5579 in the string. If the string contains no expressions, it returns an
5582 ** The new function `thunk?' returns true iff its argument is a
5583 procedure of zero arguments.
5585 ** `defined?' is now a builtin function, instead of syntax. This
5586 means that its argument should be quoted. It returns #t iff its
5587 argument is bound in the current module.
5589 ** The new syntax `use-modules' allows you to add new modules to your
5590 environment without re-typing a complete `define-module' form. It
5591 accepts any number of module names as arguments, and imports their
5592 public bindings into the current module.
5594 ** The new function (module-defined? NAME MODULE) returns true iff
5595 NAME, a symbol, is defined in MODULE, a module object.
5597 ** The new function `builtin-bindings' creates and returns a hash
5598 table containing copies of all the root module's bindings.
5600 ** The new function `builtin-weak-bindings' does the same as
5601 `builtin-bindings', but creates a doubly-weak hash table.
5603 ** The `equal?' function now considers variable objects to be
5604 equivalent if they have the same name and the same value.
5606 ** The new function `command-line' returns the command-line arguments
5607 given to Guile, as a list of strings.
5609 When using guile as a script interpreter, `command-line' returns the
5610 script's arguments; those processed by the interpreter (like `-s' or
5611 `-c') are omitted. (In other words, you get the normal, expected
5612 behavior.) Any application that uses scm_shell to process its
5613 command-line arguments gets this behavior as well.
5615 ** The new function `load-user-init' looks for a file called `.guile'
5616 in the user's home directory, and loads it if it exists. This is
5617 mostly for use by the code generated by scm_compile_shell_switches,
5618 but we thought it might also be useful in other circumstances.
5620 ** The new function `log10' returns the base-10 logarithm of its
5623 ** Changes to I/O functions
5625 *** The functions `read', `primitive-load', `read-and-eval!', and
5626 `primitive-load-path' no longer take optional arguments controlling
5627 case insensitivity and a `#' parser.
5629 Case sensitivity is now controlled by a read option called
5630 `case-insensitive'. The user can add new `#' syntaxes with the
5631 `read-hash-extend' function (see below).
5633 *** The new function `read-hash-extend' allows the user to change the
5634 syntax of Guile Scheme in a somewhat controlled way.
5636 (read-hash-extend CHAR PROC)
5637 When parsing S-expressions, if we read a `#' character followed by
5638 the character CHAR, use PROC to parse an object from the stream.
5639 If PROC is #f, remove any parsing procedure registered for CHAR.
5641 The reader applies PROC to two arguments: CHAR and an input port.
5643 *** The new functions read-delimited and read-delimited! provide a
5644 general mechanism for doing delimited input on streams.
5646 (read-delimited DELIMS [PORT HANDLE-DELIM])
5647 Read until we encounter one of the characters in DELIMS (a string),
5648 or end-of-file. PORT is the input port to read from; it defaults to
5649 the current input port. The HANDLE-DELIM parameter determines how
5650 the terminating character is handled; it should be one of the
5653 'trim omit delimiter from result
5654 'peek leave delimiter character in input stream
5655 'concat append delimiter character to returned value
5656 'split return a pair: (RESULT . TERMINATOR)
5658 HANDLE-DELIM defaults to 'peek.
5660 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
5661 A side-effecting variant of `read-delimited'.
5663 The data is written into the string BUF at the indices in the
5664 half-open interval [START, END); the default interval is the whole
5665 string: START = 0 and END = (string-length BUF). The values of
5666 START and END must specify a well-defined interval in BUF, i.e.
5667 0 <= START <= END <= (string-length BUF).
5669 It returns NBYTES, the number of bytes read. If the buffer filled
5670 up without a delimiter character being found, it returns #f. If the
5671 port is at EOF when the read starts, it returns the EOF object.
5673 If an integer is returned (i.e., the read is successfully terminated
5674 by reading a delimiter character), then the HANDLE-DELIM parameter
5675 determines how to handle the terminating character. It is described
5676 above, and defaults to 'peek.
5678 (The descriptions of these functions were borrowed from the SCSH
5679 manual, by Olin Shivers and Brian Carlstrom.)
5681 *** The `%read-delimited!' function is the primitive used to implement
5682 `read-delimited' and `read-delimited!'.
5684 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
5686 This returns a pair of values: (TERMINATOR . NUM-READ).
5687 - TERMINATOR describes why the read was terminated. If it is a
5688 character or the eof object, then that is the value that terminated
5689 the read. If it is #f, the function filled the buffer without finding
5690 a delimiting character.
5691 - NUM-READ is the number of characters read into BUF.
5693 If the read is successfully terminated by reading a delimiter
5694 character, then the gobble? parameter determines what to do with the
5695 terminating character. If true, the character is removed from the
5696 input stream; if false, the character is left in the input stream
5697 where a subsequent read operation will retrieve it. In either case,
5698 the character is also the first value returned by the procedure call.
5700 (The descriptions of this function was borrowed from the SCSH manual,
5701 by Olin Shivers and Brian Carlstrom.)
5703 *** The `read-line' and `read-line!' functions have changed; they now
5704 trim the terminator by default; previously they appended it to the
5705 returned string. For the old behavior, use (read-line PORT 'concat).
5707 *** The functions `uniform-array-read!' and `uniform-array-write!' now
5708 take new optional START and END arguments, specifying the region of
5709 the array to read and write.
5711 *** The `ungetc-char-ready?' function has been removed. We feel it's
5712 inappropriate for an interface to expose implementation details this
5715 ** Changes to the Unix library and system call interface
5717 *** The new fcntl function provides access to the Unix `fcntl' system
5720 (fcntl PORT COMMAND VALUE)
5721 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
5722 Values for COMMAND are:
5724 F_DUPFD duplicate a file descriptor
5725 F_GETFD read the descriptor's close-on-exec flag
5726 F_SETFD set the descriptor's close-on-exec flag to VALUE
5727 F_GETFL read the descriptor's flags, as set on open
5728 F_SETFL set the descriptor's flags, as set on open to VALUE
5729 F_GETOWN return the process ID of a socket's owner, for SIGIO
5730 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
5731 FD_CLOEXEC not sure what this is
5733 For details, see the documentation for the fcntl system call.
5735 *** The arguments to `select' have changed, for compatibility with
5736 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
5737 expected behavior. The MILLISECONDS parameter has been changed to
5738 MICROSECONDS, to more closely resemble the underlying system call.
5739 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
5740 corresponding return set will be the same.
5742 *** The arguments to the `mknod' system call have changed. They are
5745 (mknod PATH TYPE PERMS DEV)
5746 Create a new file (`node') in the file system. PATH is the name of
5747 the file to create. TYPE is the kind of file to create; it should
5748 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
5749 permission bits to give the newly created file. If TYPE is
5750 'block-special or 'char-special, DEV specifies which device the
5751 special file refers to; its interpretation depends on the kind of
5752 special file being created.
5754 *** The `fork' function has been renamed to `primitive-fork', to avoid
5755 clashing with various SCSH forks.
5757 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
5758 and `recvfrom!'. They no longer accept a size for a second argument;
5759 you must pass a string to hold the received value. They no longer
5760 return the buffer. Instead, `recv' returns the length of the message
5761 received, and `recvfrom' returns a pair containing the packet's length
5762 and originating address.
5764 *** The file descriptor datatype has been removed, as have the
5765 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
5766 We plan to replace these functions with a SCSH-compatible interface.
5768 *** The `create' function has been removed; it's just a special case
5771 *** There are new functions to break down process termination status
5772 values. In the descriptions below, STATUS is a value returned by
5775 (status:exit-val STATUS)
5776 If the child process exited normally, this function returns the exit
5777 code for the child process (i.e., the value passed to exit, or
5778 returned from main). If the child process did not exit normally,
5779 this function returns #f.
5781 (status:stop-sig STATUS)
5782 If the child process was suspended by a signal, this function
5783 returns the signal that suspended the child. Otherwise, it returns
5786 (status:term-sig STATUS)
5787 If the child process terminated abnormally, this function returns
5788 the signal that terminated the child. Otherwise, this function
5791 POSIX promises that exactly one of these functions will return true on
5792 a valid STATUS value.
5794 These functions are compatible with SCSH.
5796 *** There are new accessors and setters for the broken-out time vectors
5797 returned by `localtime', `gmtime', and that ilk. They are:
5799 Component Accessor Setter
5800 ========================= ============ ============
5801 seconds tm:sec set-tm:sec
5802 minutes tm:min set-tm:min
5803 hours tm:hour set-tm:hour
5804 day of the month tm:mday set-tm:mday
5805 month tm:mon set-tm:mon
5806 year tm:year set-tm:year
5807 day of the week tm:wday set-tm:wday
5808 day in the year tm:yday set-tm:yday
5809 daylight saving time tm:isdst set-tm:isdst
5810 GMT offset, seconds tm:gmtoff set-tm:gmtoff
5811 name of time zone tm:zone set-tm:zone
5813 *** There are new accessors for the vectors returned by `uname',
5814 describing the host system:
5817 ============================================== ================
5818 name of the operating system implementation utsname:sysname
5819 network name of this machine utsname:nodename
5820 release level of the operating system utsname:release
5821 version level of the operating system utsname:version
5822 machine hardware platform utsname:machine
5824 *** There are new accessors for the vectors returned by `getpw',
5825 `getpwnam', `getpwuid', and `getpwent', describing entries from the
5826 system's user database:
5829 ====================== =================
5830 user name passwd:name
5831 user password passwd:passwd
5834 real name passwd:gecos
5835 home directory passwd:dir
5836 shell program passwd:shell
5838 *** There are new accessors for the vectors returned by `getgr',
5839 `getgrnam', `getgrgid', and `getgrent', describing entries from the
5840 system's group database:
5843 ======================= ============
5844 group name group:name
5845 group password group:passwd
5847 group members group:mem
5849 *** There are new accessors for the vectors returned by `gethost',
5850 `gethostbyaddr', `gethostbyname', and `gethostent', describing
5854 ========================= ===============
5855 official name of host hostent:name
5856 alias list hostent:aliases
5857 host address type hostent:addrtype
5858 length of address hostent:length
5859 list of addresses hostent:addr-list
5861 *** There are new accessors for the vectors returned by `getnet',
5862 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
5866 ========================= ===============
5867 official name of net netent:name
5868 alias list netent:aliases
5869 net number type netent:addrtype
5870 net number netent:net
5872 *** There are new accessors for the vectors returned by `getproto',
5873 `getprotobyname', `getprotobynumber', and `getprotoent', describing
5877 ========================= ===============
5878 official protocol name protoent:name
5879 alias list protoent:aliases
5880 protocol number protoent:proto
5882 *** There are new accessors for the vectors returned by `getserv',
5883 `getservbyname', `getservbyport', and `getservent', describing
5887 ========================= ===============
5888 official service name servent:name
5889 alias list servent:aliases
5890 port number servent:port
5891 protocol to use servent:proto
5893 *** There are new accessors for the sockaddr structures returned by
5894 `accept', `getsockname', `getpeername', `recvfrom!':
5897 ======================================== ===============
5898 address format (`family') sockaddr:fam
5899 path, for file domain addresses sockaddr:path
5900 address, for internet domain addresses sockaddr:addr
5901 TCP or UDP port, for internet sockaddr:port
5903 *** The `getpwent', `getgrent', `gethostent', `getnetent',
5904 `getprotoent', and `getservent' functions now return #f at the end of
5905 the user database. (They used to throw an exception.)
5907 Note that calling MUMBLEent function is equivalent to calling the
5908 corresponding MUMBLE function with no arguments.
5910 *** The `setpwent', `setgrent', `sethostent', `setnetent',
5911 `setprotoent', and `setservent' routines now take no arguments.
5913 *** The `gethost', `getproto', `getnet', and `getserv' functions now
5914 provide more useful information when they throw an exception.
5916 *** The `lnaof' function has been renamed to `inet-lnaof'.
5918 *** Guile now claims to have the `current-time' feature.
5920 *** The `mktime' function now takes an optional second argument ZONE,
5921 giving the time zone to use for the conversion. ZONE should be a
5922 string, in the same format as expected for the "TZ" environment variable.
5924 *** The `strptime' function now returns a pair (TIME . COUNT), where
5925 TIME is the parsed time as a vector, and COUNT is the number of
5926 characters from the string left unparsed. This function used to
5927 return the remaining characters as a string.
5929 *** The `gettimeofday' function has replaced the old `time+ticks' function.
5930 The return value is now (SECONDS . MICROSECONDS); the fractional
5931 component is no longer expressed in "ticks".
5933 *** The `ticks/sec' constant has been removed, in light of the above change.
5935 * Changes to the gh_ interface
5937 ** gh_eval_str() now returns an SCM object which is the result of the
5940 ** gh_scm2str() now copies the Scheme data to a caller-provided C
5943 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
5944 and returns the array
5946 ** gh_scm2str0() is gone: there is no need to distinguish
5947 null-terminated from non-null-terminated, since gh_scm2newstr() allows
5948 the user to interpret the data both ways.
5950 * Changes to the scm_ interface
5952 ** The new function scm_symbol_value0 provides an easy way to get a
5953 symbol's value from C code:
5955 SCM scm_symbol_value0 (char *NAME)
5956 Return the value of the symbol named by the null-terminated string
5957 NAME in the current module. If the symbol named NAME is unbound in
5958 the current module, return SCM_UNDEFINED.
5960 ** The new function scm_sysintern0 creates new top-level variables,
5961 without assigning them a value.
5963 SCM scm_sysintern0 (char *NAME)
5964 Create a new Scheme top-level variable named NAME. NAME is a
5965 null-terminated string. Return the variable's value cell.
5967 ** The function scm_internal_catch is the guts of catch. It handles
5968 all the mechanics of setting up a catch target, invoking the catch
5969 body, and perhaps invoking the handler if the body does a throw.
5971 The function is designed to be usable from C code, but is general
5972 enough to implement all the semantics Guile Scheme expects from throw.
5974 TAG is the catch tag. Typically, this is a symbol, but this function
5975 doesn't actually care about that.
5977 BODY is a pointer to a C function which runs the body of the catch;
5978 this is the code you can throw from. We call it like this:
5979 BODY (BODY_DATA, JMPBUF)
5981 BODY_DATA is just the BODY_DATA argument we received; we pass it
5982 through to BODY as its first argument. The caller can make
5983 BODY_DATA point to anything useful that BODY might need.
5984 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
5985 which we have just created and initialized.
5987 HANDLER is a pointer to a C function to deal with a throw to TAG,
5988 should one occur. We call it like this:
5989 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
5991 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
5992 same idea as BODY_DATA above.
5993 THROWN_TAG is the tag that the user threw to; usually this is
5994 TAG, but it could be something else if TAG was #t (i.e., a
5995 catch-all), or the user threw to a jmpbuf.
5996 THROW_ARGS is the list of arguments the user passed to the THROW
5999 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6000 is just a pointer we pass through to HANDLER. We don't actually
6001 use either of those pointers otherwise ourselves. The idea is
6002 that, if our caller wants to communicate something to BODY or
6003 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6004 HANDLER can then use. Think of it as a way to make BODY and
6005 HANDLER closures, not just functions; MUMBLE_DATA points to the
6008 Of course, it's up to the caller to make sure that any data a
6009 MUMBLE_DATA needs is protected from GC. A common way to do this is
6010 to make MUMBLE_DATA a pointer to data stored in an automatic
6011 structure variable; since the collector must scan the stack for
6012 references anyway, this assures that any references in MUMBLE_DATA
6015 ** The new function scm_internal_lazy_catch is exactly like
6016 scm_internal_catch, except:
6018 - It does not unwind the stack (this is the major difference).
6019 - If handler returns, its value is returned from the throw.
6020 - BODY always receives #f as its JMPBUF argument (since there's no
6021 jmpbuf associated with a lazy catch, because we don't unwind the
6024 ** scm_body_thunk is a new body function you can pass to
6025 scm_internal_catch if you want the body to be like Scheme's `catch'
6026 --- a thunk, or a function of one argument if the tag is #f.
6028 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6029 contains the Scheme procedure to invoke as the body, and the tag
6030 we're catching. If the tag is #f, then we pass JMPBUF (created by
6031 scm_internal_catch) to the body procedure; otherwise, the body gets
6034 ** scm_handle_by_proc is a new handler function you can pass to
6035 scm_internal_catch if you want the handler to act like Scheme's catch
6036 --- call a procedure with the tag and the throw arguments.
6038 If the user does a throw to this catch, this function runs a handler
6039 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6040 variable holding the Scheme procedure object to invoke. It ought to
6041 be a pointer to an automatic variable (i.e., one living on the stack),
6042 or the procedure object should be otherwise protected from GC.
6044 ** scm_handle_by_message is a new handler function to use with
6045 `scm_internal_catch' if you want Guile to print a message and die.
6046 It's useful for dealing with throws to uncaught keys at the top level.
6048 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6049 message header to print; if zero, we use "guile" instead. That
6050 text is followed by a colon, then the message described by ARGS.
6052 ** The return type of scm_boot_guile is now void; the function does
6053 not return a value, and indeed, never returns at all.
6055 ** The new function scm_shell makes it easy for user applications to
6056 process command-line arguments in a way that is compatible with the
6057 stand-alone guile interpreter (which is in turn compatible with SCSH,
6060 To use the scm_shell function, first initialize any guile modules
6061 linked into your application, and then call scm_shell with the values
6062 of ARGC and ARGV your `main' function received. scm_shell will add
6063 any SCSH-style meta-arguments from the top of the script file to the
6064 argument vector, and then process the command-line arguments. This
6065 generally means loading a script file or starting up an interactive
6066 command interpreter. For details, see "Changes to the stand-alone
6069 ** The new functions scm_get_meta_args and scm_count_argv help you
6070 implement the SCSH-style meta-argument, `\'.
6072 char **scm_get_meta_args (int ARGC, char **ARGV)
6073 If the second element of ARGV is a string consisting of a single
6074 backslash character (i.e. "\\" in Scheme notation), open the file
6075 named by the following argument, parse arguments from it, and return
6076 the spliced command line. The returned array is terminated by a
6079 For details of argument parsing, see above, under "guile now accepts
6080 command-line arguments compatible with SCSH..."
6082 int scm_count_argv (char **ARGV)
6083 Count the arguments in ARGV, assuming it is terminated by a null
6086 For an example of how these functions might be used, see the source
6087 code for the function scm_shell in libguile/script.c.
6089 You will usually want to use scm_shell instead of calling this
6092 ** The new function scm_compile_shell_switches turns an array of
6093 command-line arguments into Scheme code to carry out the actions they
6094 describe. Given ARGC and ARGV, it returns a Scheme expression to
6095 evaluate, and calls scm_set_program_arguments to make any remaining
6096 command-line arguments available to the Scheme code. For example,
6097 given the following arguments:
6099 -e main -s ekko a speckled gecko
6101 scm_set_program_arguments will return the following expression:
6103 (begin (load "ekko") (main (command-line)) (quit))
6105 You will usually want to use scm_shell instead of calling this
6108 ** The function scm_shell_usage prints a usage message appropriate for
6109 an interpreter that uses scm_compile_shell_switches to handle its
6110 command-line arguments.
6112 void scm_shell_usage (int FATAL, char *MESSAGE)
6113 Print a usage message to the standard error output. If MESSAGE is
6114 non-zero, write it before the usage message, followed by a newline.
6115 If FATAL is non-zero, exit the process, using FATAL as the
6116 termination status. (If you want to be compatible with Guile,
6117 always use 1 as the exit status when terminating due to command-line
6120 You will usually want to use scm_shell instead of calling this
6123 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6124 expressions. It used to return SCM_EOL. Earth-shattering.
6126 ** The macros for declaring scheme objects in C code have been
6127 rearranged slightly. They are now:
6129 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6130 Declare a static SCM variable named C_NAME, and initialize it to
6131 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6132 be a C identifier, and SCHEME_NAME should be a C string.
6134 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6135 Just like SCM_SYMBOL, but make C_NAME globally visible.
6137 SCM_VCELL (C_NAME, SCHEME_NAME)
6138 Create a global variable at the Scheme level named SCHEME_NAME.
6139 Declare a static SCM variable named C_NAME, and initialize it to
6140 point to the Scheme variable's value cell.
6142 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6143 Just like SCM_VCELL, but make C_NAME globally visible.
6145 The `guile-snarf' script writes initialization code for these macros
6146 to its standard output, given C source code as input.
6148 The SCM_GLOBAL macro is gone.
6150 ** The scm_read_line and scm_read_line_x functions have been replaced
6151 by Scheme code based on the %read-delimited! procedure (known to C
6152 code as scm_read_delimited_x). See its description above for more
6155 ** The function scm_sys_open has been renamed to scm_open. It now
6156 returns a port instead of an FD object.
6158 * The dynamic linking support has changed. For more information, see
6159 libguile/DYNAMIC-LINKING.
6164 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6167 * Changes to the 'guile' program:
6169 ** Guile now loads some new files when it starts up. Guile first
6170 searches the load path for init.scm, and loads it if found. Then, if
6171 Guile is not being used to execute a script, and the user's home
6172 directory contains a file named `.guile', Guile loads that.
6174 ** You can now use Guile as a shell script interpreter.
6176 To paraphrase the SCSH manual:
6178 When Unix tries to execute an executable file whose first two
6179 characters are the `#!', it treats the file not as machine code to
6180 be directly executed by the native processor, but as source code
6181 to be executed by some interpreter. The interpreter to use is
6182 specified immediately after the #! sequence on the first line of
6183 the source file. The kernel reads in the name of the interpreter,
6184 and executes that instead. It passes the interpreter the source
6185 filename as its first argument, with the original arguments
6186 following. Consult the Unix man page for the `exec' system call
6187 for more information.
6189 Now you can use Guile as an interpreter, using a mechanism which is a
6190 compatible subset of that provided by SCSH.
6192 Guile now recognizes a '-s' command line switch, whose argument is the
6193 name of a file of Scheme code to load. It also treats the two
6194 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6195 to make a file of Scheme code directly executable by Unix, insert the
6196 following two lines at the top of the file:
6198 #!/usr/local/bin/guile -s
6201 Guile treats the argument of the `-s' command-line switch as the name
6202 of a file of Scheme code to load, and treats the sequence `#!' as the
6203 start of a block comment, terminated by `!#'.
6205 For example, here's a version of 'echo' written in Scheme:
6207 #!/usr/local/bin/guile -s
6209 (let loop ((args (cdr (program-arguments))))
6212 (display (car args))
6213 (if (pair? (cdr args))
6215 (loop (cdr args)))))
6218 Why does `#!' start a block comment terminated by `!#', instead of the
6219 end of the line? That is the notation SCSH uses, and although we
6220 don't yet support the other SCSH features that motivate that choice,
6221 we would like to be backward-compatible with any existing Guile
6222 scripts once we do. Furthermore, if the path to Guile on your system
6223 is too long for your kernel, you can start the script with this
6227 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6230 Note that some very old Unix systems don't support the `#!' syntax.
6233 ** You can now run Guile without installing it.
6235 Previous versions of the interactive Guile interpreter (`guile')
6236 couldn't start up unless Guile's Scheme library had been installed;
6237 they used the value of the environment variable `SCHEME_LOAD_PATH'
6238 later on in the startup process, but not to find the startup code
6239 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6242 To run Guile without installing it, build it in the normal way, and
6243 then set the environment variable `SCHEME_LOAD_PATH' to a
6244 colon-separated list of directories, including the top-level directory
6245 of the Guile sources. For example, if you unpacked Guile so that the
6246 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6249 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6252 ** Guile's read-eval-print loop no longer prints #<unspecified>
6253 results. If the user wants to see this, she can evaluate the
6254 expression (assert-repl-print-unspecified #t), perhaps in her startup
6257 ** Guile no longer shows backtraces by default when an error occurs;
6258 however, it does display a message saying how to get one, and how to
6259 request that they be displayed by default. After an error, evaluate
6261 to see a backtrace, and
6262 (debug-enable 'backtrace)
6263 to see them by default.
6267 * Changes to Guile Scheme:
6269 ** Guile now distinguishes between #f and the empty list.
6271 This is for compatibility with the IEEE standard, the (possibly)
6272 upcoming Revised^5 Report on Scheme, and many extant Scheme
6275 Guile used to have #f and '() denote the same object, to make Scheme's
6276 type system more compatible with Emacs Lisp's. However, the change
6277 caused too much trouble for Scheme programmers, and we found another
6278 way to reconcile Emacs Lisp with Scheme that didn't require this.
6281 ** Guile's delq, delv, delete functions, and their destructive
6282 counterparts, delq!, delv!, and delete!, now remove all matching
6283 elements from the list, not just the first. This matches the behavior
6284 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6285 functions which inspired them.
6287 I recognize that this change may break code in subtle ways, but it
6288 seems best to make the change before the FSF's first Guile release,
6292 ** The compiled-library-path function has been deleted from libguile.
6294 ** The facilities for loading Scheme source files have changed.
6296 *** The variable %load-path now tells Guile which directories to search
6297 for Scheme code. Its value is a list of strings, each of which names
6300 *** The variable %load-extensions now tells Guile which extensions to
6301 try appending to a filename when searching the load path. Its value
6302 is a list of strings. Its default value is ("" ".scm").
6304 *** (%search-load-path FILENAME) searches the directories listed in the
6305 value of the %load-path variable for a Scheme file named FILENAME,
6306 with all the extensions listed in %load-extensions. If it finds a
6307 match, then it returns its full filename. If FILENAME is absolute, it
6308 returns it unchanged. Otherwise, it returns #f.
6310 %search-load-path will not return matches that refer to directories.
6312 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6313 uses %seach-load-path to find a file named FILENAME, and loads it if
6314 it finds it. If it can't read FILENAME for any reason, it throws an
6317 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6320 *** load uses the same searching semantics as primitive-load.
6322 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6323 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6324 path, and load-module-with-path have been deleted. The functions
6325 above should serve their purposes.
6327 *** If the value of the variable %load-hook is a procedure,
6328 `primitive-load' applies its value to the name of the file being
6329 loaded (without the load path directory name prepended). If its value
6330 is #f, it is ignored. Otherwise, an error occurs.
6332 This is mostly useful for printing load notification messages.
6335 ** The function `eval!' is no longer accessible from the scheme level.
6336 We can't allow operations which introduce glocs into the scheme level,
6337 because Guile's type system can't handle these as data. Use `eval' or
6338 `read-and-eval!' (see below) as replacement.
6340 ** The new function read-and-eval! reads an expression from PORT,
6341 evaluates it, and returns the result. This is more efficient than
6342 simply calling `read' and `eval', since it is not necessary to make a
6343 copy of the expression for the evaluator to munge.
6345 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6346 for the `read' function.
6349 ** The function `int?' has been removed; its definition was identical
6350 to that of `integer?'.
6352 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6353 use the R4RS names for these functions.
6355 ** The function object-properties no longer returns the hash handle;
6356 it simply returns the object's property list.
6358 ** Many functions have been changed to throw errors, instead of
6359 returning #f on failure. The point of providing exception handling in
6360 the language is to simplify the logic of user code, but this is less
6361 useful if Guile's primitives don't throw exceptions.
6363 ** The function `fileno' has been renamed from `%fileno'.
6365 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6368 * Changes to Guile's C interface:
6370 ** The library's initialization procedure has been simplified.
6371 scm_boot_guile now has the prototype:
6373 void scm_boot_guile (int ARGC,
6375 void (*main_func) (),
6378 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6379 MAIN_FUNC should do all the work of the program (initializing other
6380 packages, reading user input, etc.) before returning. When MAIN_FUNC
6381 returns, call exit (0); this function never returns. If you want some
6382 other exit value, MAIN_FUNC may call exit itself.
6384 scm_boot_guile arranges for program-arguments to return the strings
6385 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6386 scm_set_program_arguments with the final list, so Scheme code will
6387 know which arguments have been processed.
6389 scm_boot_guile establishes a catch-all catch handler which prints an
6390 error message and exits the process. This means that Guile exits in a
6391 coherent way when system errors occur and the user isn't prepared to
6392 handle it. If the user doesn't like this behavior, they can establish
6393 their own universal catcher in MAIN_FUNC to shadow this one.
6395 Why must the caller do all the real work from MAIN_FUNC? The garbage
6396 collector assumes that all local variables of type SCM will be above
6397 scm_boot_guile's stack frame on the stack. If you try to manipulate
6398 SCM values after this function returns, it's the luck of the draw
6399 whether the GC will be able to find the objects you allocate. So,
6400 scm_boot_guile function exits, rather than returning, to discourage
6401 people from making that mistake.
6403 The IN, OUT, and ERR arguments were removed; there are other
6404 convenient ways to override these when desired.
6406 The RESULT argument was deleted; this function should never return.
6408 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
6412 ** Guile's header files should no longer conflict with your system's
6415 In order to compile code which #included <libguile.h>, previous
6416 versions of Guile required you to add a directory containing all the
6417 Guile header files to your #include path. This was a problem, since
6418 Guile's header files have names which conflict with many systems'
6421 Now only <libguile.h> need appear in your #include path; you must
6422 refer to all Guile's other header files as <libguile/mumble.h>.
6423 Guile's installation procedure puts libguile.h in $(includedir), and
6424 the rest in $(includedir)/libguile.
6427 ** Two new C functions, scm_protect_object and scm_unprotect_object,
6428 have been added to the Guile library.
6430 scm_protect_object (OBJ) protects OBJ from the garbage collector.
6431 OBJ will not be freed, even if all other references are dropped,
6432 until someone does scm_unprotect_object (OBJ). Both functions
6435 Note that calls to scm_protect_object do not nest. You can call
6436 scm_protect_object any number of times on a given object, and the
6437 next call to scm_unprotect_object will unprotect it completely.
6439 Basically, scm_protect_object and scm_unprotect_object just
6440 maintain a list of references to things. Since the GC knows about
6441 this list, all objects it mentions stay alive. scm_protect_object
6442 adds its argument to the list; scm_unprotect_object remove its
6443 argument from the list.
6446 ** scm_eval_0str now returns the value of the last expression
6449 ** The new function scm_read_0str reads an s-expression from a
6450 null-terminated string, and returns it.
6452 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
6453 to a Scheme port object.
6455 ** The new function `scm_set_program_arguments' allows C code to set
6456 the value returned by the Scheme `program-arguments' function.
6461 * Guile no longer includes sophisticated Tcl/Tk support.
6463 The old Tcl/Tk support was unsatisfying to us, because it required the
6464 user to link against the Tcl library, as well as Tk and Guile. The
6465 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
6466 referring to widgets by names, rather than exporting widgets to Scheme
6467 code as a special datatype.
6469 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
6470 maintainers described some very interesting changes in progress to the
6471 Tcl/Tk internals, which would facilitate clean interfaces between lone
6472 Tk and other interpreters --- even for garbage-collected languages
6473 like Scheme. They expected the new Tk to be publicly available in the
6476 Since it seems that Guile might soon have a new, cleaner interface to
6477 lone Tk, and that the old Guile/Tk glue code would probably need to be
6478 completely rewritten, we (Jim Blandy and Richard Stallman) have
6479 decided not to support the old code. We'll spend the time instead on
6480 a good interface to the newer Tk, as soon as it is available.
6482 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
6485 Copyright information:
6487 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
6489 Permission is granted to anyone to make or distribute verbatim copies
6490 of this document as received, in any medium, provided that the
6491 copyright notice and this permission notice are preserved,
6492 thus giving the recipient permission to redistribute in turn.
6494 Permission is granted to distribute modified versions
6495 of this document, or of portions of it,
6496 under the above conditions, provided also that they
6497 carry prominent notices stating who last changed them.
6502 paragraph-separate: "[
\f]*$"