1 Guile NEWS --- history of user-visible changes. -*- text -*-
2 Copyright (C) 1996, 1997, 1998 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
7 Changes since Guile 1.3:
9 * Changes to Scheme functions and syntax
11 ** New function: unread-string STRING PORT
12 Unread STRING to PORT, that is, push it back onto the port so that
13 next read operation will work on the pushed back characters.
15 ** unread-char can now be called multiple times
16 If unread-char is called multiple times, the unread characters will be
17 read again in last-in first-out order.
19 * Changes to the gh_ interface
23 Now takes a second argument which is the result array. If this
24 pointer is NULL, a new array is malloced (the old behaviour).
26 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
27 gh_scm2shorts, gh_scm2longs, gh_scm2floats
32 Changes in Guile 1.3 (released Monday, October 19, 1998):
34 * Changes to the distribution
36 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
37 To avoid conflicts, programs should name environment variables after
38 themselves, except when there's a common practice establishing some
41 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
42 giving the former precedence, and printing a warning message if the
43 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
45 ** The header files related to multi-byte characters have been removed.
46 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
47 which referred to these explicitly will probably need to be rewritten,
48 since the support for the variant string types has been removed; see
51 ** The header files append.h and sequences.h have been removed. These
52 files implemented non-R4RS operations which would encourage
53 non-portable programming style and less easy-to-read code.
55 * Changes to the stand-alone interpreter
57 ** New procedures have been added to implement a "batch mode":
59 *** Function: batch-mode?
61 Returns a boolean indicating whether the interpreter is in batch
64 *** Function: set-batch-mode?! ARG
66 If ARG is true, switches the interpreter to batch mode. The `#f'
67 case has not been implemented.
69 ** Guile now provides full command-line editing, when run interactively.
70 To use this feature, you must have the readline library installed.
71 The Guile build process will notice it, and automatically include
74 The readline library is available via anonymous FTP from any GNU
75 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
77 ** the-last-stack is now a fluid.
79 * Changes to the procedure for linking libguile with your programs
81 ** You can now use the `guile-config' utility to build programs that use Guile.
83 Guile now includes a command-line utility called `guile-config', which
84 can provide information about how to compile and link programs that
87 *** `guile-config compile' prints any C compiler flags needed to use Guile.
88 You should include this command's output on the command line you use
89 to compile C or C++ code that #includes the Guile header files. It's
90 usually just a `-I' flag to help the compiler find the Guile headers.
93 *** `guile-config link' prints any linker flags necessary to link with Guile.
95 This command writes to its standard output a list of flags which you
96 must pass to the linker to link your code against the Guile library.
97 The flags include '-lguile' itself, any other libraries the Guile
98 library depends upon, and any `-L' flags needed to help the linker
101 For example, here is a Makefile rule that builds a program named 'foo'
102 from the object files ${FOO_OBJECTS}, and links them against Guile:
105 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
107 Previous Guile releases recommended that you use autoconf to detect
108 which of a predefined set of libraries were present on your system.
109 It is more robust to use `guile-config', since it records exactly which
110 libraries the installed Guile library requires.
112 This was originally called `build-guile', but was renamed to
113 `guile-config' before Guile 1.3 was released, to be consistent with
114 the analogous script for the GTK+ GUI toolkit, which is called
118 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
120 If you are using the GNU autoconf package to configure your program,
121 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
122 (described above) and gather the necessary values for use in your
125 The GUILE_FLAGS macro expands to configure script code which runs the
126 `guile-config' script, to find out where Guile's header files and
127 libraries are installed. It sets two variables, marked for
128 substitution, as by AC_SUBST.
130 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
131 code that uses Guile header files. This is almost always just a
134 GUILE_LDFLAGS --- flags to pass to the linker to link a
135 program against Guile. This includes `-lguile' for the Guile
136 library itself, any libraries that Guile itself requires (like
137 -lqthreads), and so on. It may also include a -L flag to tell the
138 compiler where to find the libraries.
140 GUILE_FLAGS is defined in the file guile.m4, in the top-level
141 directory of the Guile distribution. You can copy it into your
142 package's aclocal.m4 file, and then use it in your configure.in file.
144 If you are using the `aclocal' program, distributed with GNU automake,
145 to maintain your aclocal.m4 file, the Guile installation process
146 installs guile.m4 where aclocal will find it. All you need to do is
147 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
148 this will copy the definition of GUILE_FLAGS into your aclocal.m4
152 * Changes to Scheme functions and syntax
154 ** Multi-byte strings have been removed, as have multi-byte and wide
155 ports. We felt that these were the wrong approach to
156 internationalization support.
158 ** New function: readline [PROMPT]
159 Read a line from the terminal, and allow the user to edit it,
160 prompting with PROMPT. READLINE provides a large set of Emacs-like
161 editing commands, lets the user recall previously typed lines, and
162 works on almost every kind of terminal, including dumb terminals.
164 READLINE assumes that the cursor is at the beginning of the line when
165 it is invoked. Thus, you can't print a prompt yourself, and then call
166 READLINE; you need to package up your prompt as a string, pass it to
167 the function, and let READLINE print the prompt itself. This is
168 because READLINE needs to know the prompt's screen width.
170 For Guile to provide this function, you must have the readline
171 library, version 2.1 or later, installed on your system. Readline is
172 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
175 See also ADD-HISTORY function.
177 ** New function: add-history STRING
178 Add STRING as the most recent line in the history used by the READLINE
179 command. READLINE does not add lines to the history itself; you must
180 call ADD-HISTORY to make previous input available to the user.
182 ** The behavior of the read-line function has changed.
184 This function now uses standard C library functions to read the line,
185 for speed. This means that it doesn not respect the value of
186 scm-line-incrementors; it assumes that lines are delimited with
189 (Note that this is read-line, the function that reads a line of text
190 from a port, not readline, the function that reads a line from a
191 terminal, providing full editing capabilities.)
193 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
195 This module provides some simple argument parsing. It exports one
198 Function: getopt-gnu-style ARG-LS
199 Parse a list of program arguments into an alist of option
202 Each item in the list of program arguments is examined to see if
203 it meets the syntax of a GNU long-named option. An argument like
204 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
205 returned alist, where MUMBLE is a keyword object with the same
206 name as the argument. An argument like `--MUMBLE=FROB' produces
207 an element of the form (MUMBLE . FROB), where FROB is a string.
209 As a special case, the returned alist also contains a pair whose
210 car is the symbol `rest'. The cdr of this pair is a list
211 containing all the items in the argument list that are not options
212 of the form mentioned above.
214 The argument `--' is treated specially: all items in the argument
215 list appearing after such an argument are not examined, and are
216 returned in the special `rest' list.
218 This function does not parse normal single-character switches.
219 You will need to parse them out of the `rest' list yourself.
221 ** The read syntax for byte vectors and short vectors has changed.
223 Instead of #bytes(...), write #y(...).
225 Instead of #short(...), write #h(...).
227 This may seem nutty, but, like the other uniform vectors, byte vectors
228 and short vectors want to have the same print and read syntax (and,
229 more basic, want to have read syntax!). Changing the read syntax to
230 use multiple characters after the hash sign breaks with the
231 conventions used in R5RS and the conventions used for the other
232 uniform vectors. It also introduces complexity in the current reader,
233 both on the C and Scheme levels. (The Right solution is probably to
234 change the syntax and prototypes for uniform vectors entirely.)
237 ** The new module (ice-9 session) provides useful interactive functions.
239 *** New procedure: (apropos REGEXP OPTION ...)
241 Display a list of top-level variables whose names match REGEXP, and
242 the modules they are imported from. Each OPTION should be one of the
245 value --- Show the value of each matching variable.
246 shadow --- Show bindings shadowed by subsequently imported modules.
247 full --- Same as both `shadow' and `value'.
251 guile> (apropos "trace" 'full)
252 debug: trace #<procedure trace args>
253 debug: untrace #<procedure untrace args>
254 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
255 the-scm-module: before-backtrace-hook ()
256 the-scm-module: backtrace #<primitive-procedure backtrace>
257 the-scm-module: after-backtrace-hook ()
258 the-scm-module: has-shown-backtrace-hint? #f
261 ** There are new functions and syntax for working with macros.
263 Guile implements macros as a special object type. Any variable whose
264 top-level binding is a macro object acts as a macro. The macro object
265 specifies how the expression should be transformed before evaluation.
267 *** Macro objects now print in a reasonable way, resembling procedures.
269 *** New function: (macro? OBJ)
270 True iff OBJ is a macro object.
272 *** New function: (primitive-macro? OBJ)
273 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
274 macro transformers, implemented in eval.c rather than Scheme code.
276 Why do we have this function?
277 - For symmetry with procedure? and primitive-procedure?,
278 - to allow custom print procedures to tell whether a macro is
279 primitive, and display it differently, and
280 - to allow compilers and user-written evaluators to distinguish
281 builtin special forms from user-defined ones, which could be
284 *** New function: (macro-type OBJ)
285 Return a value indicating what kind of macro OBJ is. Possible return
288 The symbol `syntax' --- a macro created by procedure->syntax.
289 The symbol `macro' --- a macro created by procedure->macro.
290 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
291 The boolean #f --- if OBJ is not a macro object.
293 *** New function: (macro-name MACRO)
294 Return the name of the macro object MACRO's procedure, as returned by
297 *** New function: (macro-transformer MACRO)
298 Return the transformer procedure for MACRO.
300 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
302 Specify a new macro expander to use in the current module. Each
303 MODULE is a module name, with the same meaning as in the `use-modules'
304 form; each named module's exported bindings are added to the current
305 top-level environment. TRANSFORMER is an expression evaluated in the
306 resulting environment which must yield a procedure to use as the
307 module's eval transformer: every expression evaluated in this module
308 is passed to this function, and the result passed to the Guile
311 *** macro-eval! is removed. Use local-eval instead.
313 ** Some magic has been added to the printer to better handle user
314 written printing routines (like record printers, closure printers).
316 The problem is that these user written routines must have access to
317 the current `print-state' to be able to handle fancy things like
318 detection of circular references. These print-states have to be
319 passed to the builtin printing routines (display, write, etc) to
320 properly continue the print chain.
322 We didn't want to change all existing print code so that it
323 explicitly passes thru a print state in addition to a port. Instead,
324 we extented the possible values that the builtin printing routines
325 accept as a `port'. In addition to a normal port, they now also take
326 a pair of a normal port and a print-state. Printing will go to the
327 port and the print-state will be used to control the detection of
328 circular references, etc. If the builtin function does not care for a
329 print-state, it is simply ignored.
331 User written callbacks are now called with such a pair as their
332 `port', but because every function now accepts this pair as a PORT
333 argument, you don't have to worry about that. In fact, it is probably
334 safest to not check for these pairs.
336 However, it is sometimes necessary to continue a print chain on a
337 different port, for example to get a intermediate string
338 representation of the printed value, mangle that string somehow, and
339 then to finally print the mangled string. Use the new function
341 inherit-print-state OLD-PORT NEW-PORT
343 for this. It constructs a new `port' that prints to NEW-PORT but
344 inherits the print-state of OLD-PORT.
346 ** struct-vtable-offset renamed to vtable-offset-user
348 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
350 ** There is now a fourth (optional) argument to make-vtable-vtable and
351 make-struct when constructing new types (vtables). This argument
352 initializes field vtable-index-printer of the vtable.
354 ** The detection of circular references has been extended to structs.
355 That is, a structure that -- in the process of being printed -- prints
356 itself does not lead to infinite recursion.
358 ** There is now some basic support for fluids. Please read
359 "libguile/fluid.h" to find out more. It is accessible from Scheme with
360 the following functions and macros:
364 Create a new fluid object. Fluids are not special variables or
365 some other extension to the semantics of Scheme, but rather
366 ordinary Scheme objects. You can store them into variables (that
367 are still lexically scoped, of course) or into any other place you
368 like. Every fluid has a initial value of `#f'.
372 Test whether OBJ is a fluid.
374 Function: fluid-ref FLUID
375 Function: fluid-set! FLUID VAL
377 Access/modify the fluid FLUID. Modifications are only visible
378 within the current dynamic root (that includes threads).
380 Function: with-fluids* FLUIDS VALUES THUNK
382 FLUIDS is a list of fluids and VALUES a corresponding list of
383 values for these fluids. Before THUNK gets called the values are
384 installed in the fluids and the old values of the fluids are
385 saved in the VALUES list. When the flow of control leaves THUNK
386 or reenters it, the values get swapped again. You might think of
387 this as a `safe-fluid-excursion'. Note that the VALUES list is
388 modified by `with-fluids*'.
390 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
392 The same as `with-fluids*' but with a different syntax. It looks
393 just like `let', but both FLUID and VALUE are evaluated. Remember,
394 fluids are not special variables but ordinary objects. FLUID
395 should evaluate to a fluid.
397 ** Changes to system call interfaces:
399 *** close-port, close-input-port and close-output-port now return a
400 boolean instead of an `unspecified' object. #t means that the port
401 was successfully closed, while #f means it was already closed. It is
402 also now possible for these procedures to raise an exception if an
403 error occurs (some errors from write can be delayed until close.)
405 *** the first argument to chmod, fcntl, ftell and fseek can now be a
408 *** the third argument to fcntl is now optional.
410 *** the first argument to chown can now be a file descriptor or a port.
412 *** the argument to stat can now be a port.
414 *** The following new procedures have been added (most use scsh
417 *** procedure: close PORT/FD
418 Similar to close-port (*note close-port: Closing Ports.), but also
419 works on file descriptors. A side effect of closing a file
420 descriptor is that any ports using that file descriptor are moved
421 to a different file descriptor and have their revealed counts set
424 *** procedure: port->fdes PORT
425 Returns the integer file descriptor underlying PORT. As a side
426 effect the revealed count of PORT is incremented.
428 *** procedure: fdes->ports FDES
429 Returns a list of existing ports which have FDES as an underlying
430 file descriptor, without changing their revealed counts.
432 *** procedure: fdes->inport FDES
433 Returns an existing input port which has FDES as its underlying
434 file descriptor, if one exists, and increments its revealed count.
435 Otherwise, returns a new input port with a revealed count of 1.
437 *** procedure: fdes->outport FDES
438 Returns an existing output port which has FDES as its underlying
439 file descriptor, if one exists, and increments its revealed count.
440 Otherwise, returns a new output port with a revealed count of 1.
442 The next group of procedures perform a `dup2' system call, if NEWFD
443 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
444 duplicated can be supplied as an integer or contained in a port. The
445 type of value returned varies depending on which procedure is used.
447 All procedures also have the side effect when performing `dup2' that
448 any ports using NEWFD are moved to a different file descriptor and have
449 their revealed counts set to zero.
451 *** procedure: dup->fdes PORT/FD [NEWFD]
452 Returns an integer file descriptor.
454 *** procedure: dup->inport PORT/FD [NEWFD]
455 Returns a new input port using the new file descriptor.
457 *** procedure: dup->outport PORT/FD [NEWFD]
458 Returns a new output port using the new file descriptor.
460 *** procedure: dup PORT/FD [NEWFD]
461 Returns a new port if PORT/FD is a port, with the same mode as the
462 supplied port, otherwise returns an integer file descriptor.
464 *** procedure: dup->port PORT/FD MODE [NEWFD]
465 Returns a new port using the new file descriptor. MODE supplies a
466 mode string for the port (*note open-file: File Ports.).
468 *** procedure: setenv NAME VALUE
469 Modifies the environment of the current process, which is also the
470 default environment inherited by child processes.
472 If VALUE is `#f', then NAME is removed from the environment.
473 Otherwise, the string NAME=VALUE is added to the environment,
474 replacing any existing string with name matching NAME.
476 The return value is unspecified.
478 *** procedure: truncate-file OBJ SIZE
479 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
480 can be a string containing a file name or an integer file
481 descriptor or port open for output on the file. The underlying
482 system calls are `truncate' and `ftruncate'.
484 The return value is unspecified.
486 *** procedure: setvbuf PORT MODE [SIZE]
487 Set the buffering mode for PORT. MODE can be:
495 block buffered, using a newly allocated buffer of SIZE bytes.
496 However if SIZE is zero or unspecified, the port will be made
499 This procedure should not be used after I/O has been performed with
502 Ports are usually block buffered by default, with a default buffer
503 size. Procedures e.g., *Note open-file: File Ports, which accept a
504 mode string allow `0' to be added to request an unbuffered port.
506 *** procedure: fsync PORT/FD
507 Copies any unwritten data for the specified output file descriptor
508 to disk. If PORT/FD is a port, its buffer is flushed before the
509 underlying file descriptor is fsync'd. The return value is
512 *** procedure: open-fdes PATH FLAGS [MODES]
513 Similar to `open' but returns a file descriptor instead of a port.
515 *** procedure: execle PATH ENV [ARG] ...
516 Similar to `execl', but the environment of the new process is
517 specified by ENV, which must be a list of strings as returned by
518 the `environ' procedure.
520 This procedure is currently implemented using the `execve' system
521 call, but we call it `execle' because of its Scheme calling
524 *** procedure: strerror ERRNO
525 Returns the Unix error message corresponding to ERRNO, an integer.
527 *** procedure: primitive-exit [STATUS]
528 Terminate the current process without unwinding the Scheme stack.
529 This is would typically be useful after a fork. The exit status
530 is STATUS if supplied, otherwise zero.
533 Returns an object with information about real and processor time.
534 The following procedures accept such an object as an argument and
535 return a selected component:
538 The current real time, expressed as time units relative to an
542 The CPU time units used by the calling process.
545 The CPU time units used by the system on behalf of the
549 The CPU time units used by terminated child processes of the
550 calling process, whose status has been collected (e.g., using
554 Similarly, the CPU times units used by the system on behalf of
555 terminated child processes.
557 ** Removed: list-length
558 ** Removed: list-append, list-append!
559 ** Removed: list-reverse, list-reverse!
561 ** array-map renamed to array-map!
563 ** serial-array-map renamed to serial-array-map!
565 ** catch doesn't take #f as first argument any longer
567 Previously, it was possible to pass #f instead of a key to `catch'.
568 That would cause `catch' to pass a jump buffer object to the procedure
569 passed as second argument. The procedure could then use this jump
570 buffer objekt as an argument to throw.
572 This mechanism has been removed since its utility doesn't motivate the
573 extra complexity it introduces.
575 ** The `#/' notation for lists now provokes a warning message from Guile.
576 This syntax will be removed from Guile in the near future.
578 To disable the warning message, set the GUILE_HUSH environment
579 variable to any non-empty value.
581 ** The newline character now prints as `#\newline', following the
582 normal Scheme notation, not `#\nl'.
584 * Changes to the gh_ interface
586 ** The gh_enter function now takes care of loading the Guile startup files.
587 gh_enter works by calling scm_boot_guile; see the remarks below.
589 ** Function: void gh_write (SCM x)
591 Write the printed representation of the scheme object x to the current
592 output port. Corresponds to the scheme level `write'.
594 ** gh_list_length renamed to gh_length.
596 ** vector handling routines
598 Several major changes. In particular, gh_vector() now resembles
599 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
600 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
601 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
602 vector-related gh_ functions have been implemented.
604 ** pair and list routines
606 Implemented several of the R4RS pair and list functions that were
609 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
611 New function. Converts double arrays back and forth between Scheme
614 * Changes to the scm_ interface
616 ** The function scm_boot_guile now takes care of loading the startup files.
618 Guile's primary initialization function, scm_boot_guile, now takes
619 care of loading `boot-9.scm', in the `ice-9' module, to initialize
620 Guile, define the module system, and put together some standard
621 bindings. It also loads `init.scm', which is intended to hold
622 site-specific initialization code.
624 Since Guile cannot operate properly until boot-9.scm is loaded, there
625 is no reason to separate loading boot-9.scm from Guile's other
626 initialization processes.
628 This job used to be done by scm_compile_shell_switches, which didn't
629 make much sense; in particular, it meant that people using Guile for
630 non-shell-like applications had to jump through hoops to get Guile
631 initialized properly.
633 ** The function scm_compile_shell_switches no longer loads the startup files.
634 Now, Guile always loads the startup files, whenever it is initialized;
635 see the notes above for scm_boot_guile and scm_load_startup_files.
637 ** Function: scm_load_startup_files
638 This new function takes care of loading Guile's initialization file
639 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
640 this is always called by the Guile initialization process, it's
641 probably not too useful to call this yourself, but it's there anyway.
643 ** The semantics of smob marking have changed slightly.
645 The smob marking function (the `mark' member of the scm_smobfuns
646 structure) is no longer responsible for setting the mark bit on the
647 smob. The generic smob handling code in the garbage collector will
648 set this bit. The mark function need only ensure that any other
649 objects the smob refers to get marked.
651 Note that this change means that the smob's GC8MARK bit is typically
652 already set upon entry to the mark function. Thus, marking functions
653 which look like this:
656 if (SCM_GC8MARKP (ptr))
658 SCM_SETGC8MARK (ptr);
659 ... mark objects to which the smob refers ...
662 are now incorrect, since they will return early, and fail to mark any
663 other objects the smob refers to. Some code in the Guile library used
666 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
668 If you have implemented your own I/O port type, by writing the
669 functions required by the scm_ptobfuns and then calling scm_newptob,
670 you will need to change your functions slightly.
672 The functions in a scm_ptobfuns structure now expect the port itself
673 as their argument; they used to expect the `stream' member of the
674 port's scm_port_table structure. This allows functions in an
675 scm_ptobfuns structure to easily access the port's cell (and any flags
676 it its CAR), and the port's scm_port_table structure.
678 Guile now passes the I/O port itself as the `port' argument in the
679 following scm_ptobfuns functions:
681 int (*free) (SCM port);
682 int (*fputc) (int, SCM port);
683 int (*fputs) (char *, SCM port);
684 scm_sizet (*fwrite) SCM_P ((char *ptr,
688 int (*fflush) (SCM port);
689 int (*fgetc) (SCM port);
690 int (*fclose) (SCM port);
692 The interfaces to the `mark', `print', `equalp', and `fgets' methods
695 If you have existing code which defines its own port types, it is easy
696 to convert your code to the new interface; simply apply SCM_STREAM to
697 the port argument to yield the value you code used to expect.
699 Note that since both the port and the stream have the same type in the
700 C code --- they are both SCM values --- the C compiler will not remind
701 you if you forget to update your scm_ptobfuns functions.
704 ** Function: int scm_internal_select (int fds,
708 struct timeval *timeout);
710 This is a replacement for the `select' function provided by the OS.
711 It enables I/O blocking and sleeping to happen for one cooperative
712 thread without blocking other threads. It also avoids busy-loops in
713 these situations. It is intended that all I/O blocking and sleeping
714 will finally go through this function. Currently, this function is
715 only available on systems providing `gettimeofday' and `select'.
717 ** Function: SCM scm_internal_stack_catch (SCM tag,
718 scm_catch_body_t body,
720 scm_catch_handler_t handler,
723 A new sibling to the other two C level `catch' functions
724 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
725 the stack to be saved automatically into the variable `the-last-stack'
726 (scm_the_last_stack_var) on error. This is necessary if you want to
727 use advanced error reporting, such as calling scm_display_error and
728 scm_display_backtrace. (They both take a stack object as argument.)
730 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
732 scm_catch_handler_t handler,
735 Spawns a new thread. It does a job similar to
736 scm_call_with_new_thread but takes arguments more suitable when
737 spawning threads from application C code.
739 ** The hook scm_error_callback has been removed. It was originally
740 intended as a way for the user to install his own error handler. But
741 that method works badly since it intervenes between throw and catch,
742 thereby changing the semantics of expressions like (catch #t ...).
743 The correct way to do it is to use one of the C level catch functions
744 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
746 ** Removed functions:
748 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
749 scm_list_reverse, scm_list_reverse_x
751 ** New macros: SCM_LISTn where n is one of the integers 0-9.
753 These can be used for pretty list creation from C. The idea is taken
754 from Erick Gallesio's STk.
756 ** scm_array_map renamed to scm_array_map_x
758 ** mbstrings are now removed
760 This means that the type codes scm_tc7_mb_string and
761 scm_tc7_mb_substring has been removed.
763 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
765 Since we no longer support multi-byte strings, these I/O functions
766 have been simplified, and renamed. Here are their old names, and
767 their new names and arguments:
769 scm_gen_putc -> void scm_putc (int c, SCM port);
770 scm_gen_puts -> void scm_puts (char *s, SCM port);
771 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
772 scm_gen_getc -> void scm_getc (SCM port);
775 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
777 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
779 SCM_TYP7S now masks away the bit which distinguishes substrings from
782 ** scm_catch_body_t: Backward incompatible change!
784 Body functions to scm_internal_catch and friends do not any longer
785 take a second argument. This is because it is no longer possible to
786 pass a #f arg to catch.
788 ** Calls to scm_protect_object and scm_unprotect now nest properly.
790 The function scm_protect_object protects its argument from being freed
791 by the garbage collector. scm_unprotect_object removes that
794 These functions now nest properly. That is, for every object O, there
795 is a counter which scm_protect_object(O) increments and
796 scm_unprotect_object(O) decrements, if the counter is greater than
797 zero. Every object's counter is zero when it is first created. If an
798 object's counter is greater than zero, the garbage collector will not
801 This allows you to use scm_protect_object in your code without
802 worrying that some other function you call will call
803 scm_unprotect_object, and allow it to be freed. Assuming that the
804 functions you call are well-behaved, and unprotect only those objects
805 they protect, you can follow the same rule and have confidence that
806 objects will be freed only at appropriate times.
809 Changes in Guile 1.2 (released Tuesday, June 24 1997):
811 * Changes to the distribution
813 ** Nightly snapshots are now available from ftp.red-bean.com.
814 The old server, ftp.cyclic.com, has been relinquished to its rightful
817 Nightly snapshots of the Guile development sources are now available via
818 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
820 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
821 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
823 ** To run Guile without installing it, the procedure has changed a bit.
825 If you used a separate build directory to compile Guile, you'll need
826 to include the build directory in SCHEME_LOAD_PATH, as well as the
827 source directory. See the `INSTALL' file for examples.
829 * Changes to the procedure for linking libguile with your programs
831 ** The standard Guile load path for Scheme code now includes
832 $(datadir)/guile (usually /usr/local/share/guile). This means that
833 you can install your own Scheme files there, and Guile will find them.
834 (Previous versions of Guile only checked a directory whose name
835 contained the Guile version number, so you had to re-install or move
836 your Scheme sources each time you installed a fresh version of Guile.)
838 The load path also includes $(datadir)/guile/site; we recommend
839 putting individual Scheme files there. If you want to install a
840 package with multiple source files, create a directory for them under
843 ** Guile 1.2 will now use the Rx regular expression library, if it is
844 installed on your system. When you are linking libguile into your own
845 programs, this means you will have to link against -lguile, -lqt (if
846 you configured Guile with thread support), and -lrx.
848 If you are using autoconf to generate configuration scripts for your
849 application, the following lines should suffice to add the appropriate
850 libraries to your link command:
852 ### Find Rx, quickthreads and libguile.
853 AC_CHECK_LIB(rx, main)
854 AC_CHECK_LIB(qt, main)
855 AC_CHECK_LIB(guile, scm_shell)
857 The Guile 1.2 distribution does not contain sources for the Rx
858 library, as Guile 1.0 did. If you want to use Rx, you'll need to
859 retrieve it from a GNU FTP site and install it separately.
861 * Changes to Scheme functions and syntax
863 ** The dynamic linking features of Guile are now enabled by default.
864 You can disable them by giving the `--disable-dynamic-linking' option
867 (dynamic-link FILENAME)
869 Find the object file denoted by FILENAME (a string) and link it
870 into the running Guile application. When everything works out,
871 return a Scheme object suitable for representing the linked object
872 file. Otherwise an error is thrown. How object files are
873 searched is system dependent.
875 (dynamic-object? VAL)
877 Determine whether VAL represents a dynamically linked object file.
879 (dynamic-unlink DYNOBJ)
881 Unlink the indicated object file from the application. DYNOBJ
882 should be one of the values returned by `dynamic-link'.
884 (dynamic-func FUNCTION DYNOBJ)
886 Search the C function indicated by FUNCTION (a string or symbol)
887 in DYNOBJ and return some Scheme object that can later be used
888 with `dynamic-call' to actually call this function. Right now,
889 these Scheme objects are formed by casting the address of the
890 function to `long' and converting this number to its Scheme
893 (dynamic-call FUNCTION DYNOBJ)
895 Call the C function indicated by FUNCTION and DYNOBJ. The
896 function is passed no arguments and its return value is ignored.
897 When FUNCTION is something returned by `dynamic-func', call that
898 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
899 etc.), look it up in DYNOBJ; this is equivalent to
901 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
903 Interrupts are deferred while the C function is executing (with
904 SCM_DEFER_INTS/SCM_ALLOW_INTS).
906 (dynamic-args-call FUNCTION DYNOBJ ARGS)
908 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
909 some arguments and return its return value. The C function is
910 expected to take two arguments and return an `int', just like
913 int c_func (int argc, char **argv);
915 ARGS must be a list of strings and is converted into an array of
916 `char *'. The array is passed in ARGV and its size in ARGC. The
917 return value is converted to a Scheme number and returned from the
918 call to `dynamic-args-call'.
920 When dynamic linking is disabled or not supported on your system,
921 the above functions throw errors, but they are still available.
923 Here is a small example that works on GNU/Linux:
925 (define libc-obj (dynamic-link "libc.so"))
926 (dynamic-args-call 'rand libc-obj '())
928 See the file `libguile/DYNAMIC-LINKING' for additional comments.
930 ** The #/ syntax for module names is depreciated, and will be removed
931 in a future version of Guile. Instead of
939 The latter syntax is more consistent with existing Lisp practice.
941 ** Guile now does fancier printing of structures. Structures are the
942 underlying implementation for records, which in turn are used to
943 implement modules, so all of these object now print differently and in
944 a more informative way.
946 The Scheme printer will examine the builtin variable *struct-printer*
947 whenever it needs to print a structure object. When this variable is
948 not `#f' it is deemed to be a procedure and will be applied to the
949 structure object and the output port. When *struct-printer* is `#f'
950 or the procedure return `#f' the structure object will be printed in
951 the boring #<struct 80458270> form.
953 This hook is used by some routines in ice-9/boot-9.scm to implement
954 type specific printing routines. Please read the comments there about
957 One of the more specific uses of structs are records. The printing
958 procedure that could be passed to MAKE-RECORD-TYPE is now actually
959 called. It should behave like a *struct-printer* procedure (described
962 ** Guile now supports a new R4RS-compliant syntax for keywords. A
963 token of the form #:NAME, where NAME has the same syntax as a Scheme
964 symbol, is the external representation of the keyword named NAME.
965 Keyword objects print using this syntax as well, so values containing
966 keyword objects can be read back into Guile. When used in an
967 expression, keywords are self-quoting objects.
969 Guile suports this read syntax, and uses this print syntax, regardless
970 of the current setting of the `keyword' read option. The `keyword'
971 read option only controls whether Guile recognizes the `:NAME' syntax,
972 which is incompatible with R4RS. (R4RS says such token represent
975 ** Guile has regular expression support again. Guile 1.0 included
976 functions for matching regular expressions, based on the Rx library.
977 In Guile 1.1, the Guile/Rx interface was removed to simplify the
978 distribution, and thus Guile had no regular expression support. Guile
979 1.2 again supports the most commonly used functions, and supports all
980 of SCSH's regular expression functions.
982 If your system does not include a POSIX regular expression library,
983 and you have not linked Guile with a third-party regexp library such as
984 Rx, these functions will not be available. You can tell whether your
985 Guile installation includes regular expression support by checking
986 whether the `*features*' list includes the `regex' symbol.
990 By default, Guile supports POSIX extended regular expressions. That
991 means that the characters `(', `)', `+' and `?' are special, and must
992 be escaped if you wish to match the literal characters.
994 This regular expression interface was modeled after that implemented
995 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
996 with SCSH regular expressions.
998 **** Function: string-match PATTERN STR [START]
999 Compile the string PATTERN into a regular expression and compare
1000 it with STR. The optional numeric argument START specifies the
1001 position of STR at which to begin matching.
1003 `string-match' returns a "match structure" which describes what,
1004 if anything, was matched by the regular expression. *Note Match
1005 Structures::. If STR does not match PATTERN at all,
1006 `string-match' returns `#f'.
1008 Each time `string-match' is called, it must compile its PATTERN
1009 argument into a regular expression structure. This operation is
1010 expensive, which makes `string-match' inefficient if the same regular
1011 expression is used several times (for example, in a loop). For better
1012 performance, you can compile a regular expression in advance and then
1013 match strings against the compiled regexp.
1015 **** Function: make-regexp STR [FLAGS]
1016 Compile the regular expression described by STR, and return the
1017 compiled regexp structure. If STR does not describe a legal
1018 regular expression, `make-regexp' throws a
1019 `regular-expression-syntax' error.
1021 FLAGS may be the bitwise-or of one or more of the following:
1023 **** Constant: regexp/extended
1024 Use POSIX Extended Regular Expression syntax when interpreting
1025 STR. If not set, POSIX Basic Regular Expression syntax is used.
1026 If the FLAGS argument is omitted, we assume regexp/extended.
1028 **** Constant: regexp/icase
1029 Do not differentiate case. Subsequent searches using the
1030 returned regular expression will be case insensitive.
1032 **** Constant: regexp/newline
1033 Match-any-character operators don't match a newline.
1035 A non-matching list ([^...]) not containing a newline matches a
1038 Match-beginning-of-line operator (^) matches the empty string
1039 immediately after a newline, regardless of whether the FLAGS
1040 passed to regexp-exec contain regexp/notbol.
1042 Match-end-of-line operator ($) matches the empty string
1043 immediately before a newline, regardless of whether the FLAGS
1044 passed to regexp-exec contain regexp/noteol.
1046 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
1047 Match the compiled regular expression REGEXP against `str'. If
1048 the optional integer START argument is provided, begin matching
1049 from that position in the string. Return a match structure
1050 describing the results of the match, or `#f' if no match could be
1053 FLAGS may be the bitwise-or of one or more of the following:
1055 **** Constant: regexp/notbol
1056 The match-beginning-of-line operator always fails to match (but
1057 see the compilation flag regexp/newline above) This flag may be
1058 used when different portions of a string are passed to
1059 regexp-exec and the beginning of the string should not be
1060 interpreted as the beginning of the line.
1062 **** Constant: regexp/noteol
1063 The match-end-of-line operator always fails to match (but see the
1064 compilation flag regexp/newline above)
1066 **** Function: regexp? OBJ
1067 Return `#t' if OBJ is a compiled regular expression, or `#f'
1070 Regular expressions are commonly used to find patterns in one string
1071 and replace them with the contents of another string.
1073 **** Function: regexp-substitute PORT MATCH [ITEM...]
1074 Write to the output port PORT selected contents of the match
1075 structure MATCH. Each ITEM specifies what should be written, and
1076 may be one of the following arguments:
1078 * A string. String arguments are written out verbatim.
1080 * An integer. The submatch with that number is written.
1082 * The symbol `pre'. The portion of the matched string preceding
1083 the regexp match is written.
1085 * The symbol `post'. The portion of the matched string
1086 following the regexp match is written.
1088 PORT may be `#f', in which case nothing is written; instead,
1089 `regexp-substitute' constructs a string from the specified ITEMs
1092 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
1093 Similar to `regexp-substitute', but can be used to perform global
1094 substitutions on STR. Instead of taking a match structure as an
1095 argument, `regexp-substitute/global' takes two string arguments: a
1096 REGEXP string describing a regular expression, and a TARGET string
1097 which should be matched against this regular expression.
1099 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
1102 * A function may be supplied. When this function is called, it
1103 will be passed one argument: a match structure for a given
1104 regular expression match. It should return a string to be
1105 written out to PORT.
1107 * The `post' symbol causes `regexp-substitute/global' to recurse
1108 on the unmatched portion of STR. This *must* be supplied in
1109 order to perform global search-and-replace on STR; if it is
1110 not present among the ITEMs, then `regexp-substitute/global'
1111 will return after processing a single match.
1113 *** Match Structures
1115 A "match structure" is the object returned by `string-match' and
1116 `regexp-exec'. It describes which portion of a string, if any, matched
1117 the given regular expression. Match structures include: a reference to
1118 the string that was checked for matches; the starting and ending
1119 positions of the regexp match; and, if the regexp included any
1120 parenthesized subexpressions, the starting and ending positions of each
1123 In each of the regexp match functions described below, the `match'
1124 argument must be a match structure returned by a previous call to
1125 `string-match' or `regexp-exec'. Most of these functions return some
1126 information about the original target string that was matched against a
1127 regular expression; we will call that string TARGET for easy reference.
1129 **** Function: regexp-match? OBJ
1130 Return `#t' if OBJ is a match structure returned by a previous
1131 call to `regexp-exec', or `#f' otherwise.
1133 **** Function: match:substring MATCH [N]
1134 Return the portion of TARGET matched by subexpression number N.
1135 Submatch 0 (the default) represents the entire regexp match. If
1136 the regular expression as a whole matched, but the subexpression
1137 number N did not match, return `#f'.
1139 **** Function: match:start MATCH [N]
1140 Return the starting position of submatch number N.
1142 **** Function: match:end MATCH [N]
1143 Return the ending position of submatch number N.
1145 **** Function: match:prefix MATCH
1146 Return the unmatched portion of TARGET preceding the regexp match.
1148 **** Function: match:suffix MATCH
1149 Return the unmatched portion of TARGET following the regexp match.
1151 **** Function: match:count MATCH
1152 Return the number of parenthesized subexpressions from MATCH.
1153 Note that the entire regular expression match itself counts as a
1154 subexpression, and failed submatches are included in the count.
1156 **** Function: match:string MATCH
1157 Return the original TARGET string.
1159 *** Backslash Escapes
1161 Sometimes you will want a regexp to match characters like `*' or `$'
1162 exactly. For example, to check whether a particular string represents
1163 a menu entry from an Info node, it would be useful to match it against
1164 a regexp like `^* [^:]*::'. However, this won't work; because the
1165 asterisk is a metacharacter, it won't match the `*' at the beginning of
1166 the string. In this case, we want to make the first asterisk un-magic.
1168 You can do this by preceding the metacharacter with a backslash
1169 character `\'. (This is also called "quoting" the metacharacter, and
1170 is known as a "backslash escape".) When Guile sees a backslash in a
1171 regular expression, it considers the following glyph to be an ordinary
1172 character, no matter what special meaning it would ordinarily have.
1173 Therefore, we can make the above example work by changing the regexp to
1174 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
1175 to match only a single asterisk in the target string.
1177 Since the backslash is itself a metacharacter, you may force a
1178 regexp to match a backslash in the target string by preceding the
1179 backslash with itself. For example, to find variable references in a
1180 TeX program, you might want to find occurrences of the string `\let\'
1181 followed by any number of alphabetic characters. The regular expression
1182 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
1183 each match a single backslash in the target string.
1185 **** Function: regexp-quote STR
1186 Quote each special character found in STR with a backslash, and
1187 return the resulting string.
1189 *Very important:* Using backslash escapes in Guile source code (as
1190 in Emacs Lisp or C) can be tricky, because the backslash character has
1191 special meaning for the Guile reader. For example, if Guile encounters
1192 the character sequence `\n' in the middle of a string while processing
1193 Scheme code, it replaces those characters with a newline character.
1194 Similarly, the character sequence `\t' is replaced by a horizontal tab.
1195 Several of these "escape sequences" are processed by the Guile reader
1196 before your code is executed. Unrecognized escape sequences are
1197 ignored: if the characters `\*' appear in a string, they will be
1198 translated to the single character `*'.
1200 This translation is obviously undesirable for regular expressions,
1201 since we want to be able to include backslashes in a string in order to
1202 escape regexp metacharacters. Therefore, to make sure that a backslash
1203 is preserved in a string in your Guile program, you must use *two*
1204 consecutive backslashes:
1206 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
1208 The string in this example is preprocessed by the Guile reader before
1209 any code is executed. The resulting argument to `make-regexp' is the
1210 string `^\* [^:]*', which is what we really want.
1212 This also means that in order to write a regular expression that
1213 matches a single backslash character, the regular expression string in
1214 the source code must include *four* backslashes. Each consecutive pair
1215 of backslashes gets translated by the Guile reader to a single
1216 backslash, and the resulting double-backslash is interpreted by the
1217 regexp engine as matching a single backslash character. Hence:
1219 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
1221 The reason for the unwieldiness of this syntax is historical. Both
1222 regular expression pattern matchers and Unix string processing systems
1223 have traditionally used backslashes with the special meanings described
1224 above. The POSIX regular expression specification and ANSI C standard
1225 both require these semantics. Attempting to abandon either convention
1226 would cause other kinds of compatibility problems, possibly more severe
1227 ones. Therefore, without extending the Scheme reader to support
1228 strings with different quoting conventions (an ungainly and confusing
1229 extension when implemented in other languages), we must adhere to this
1230 cumbersome escape syntax.
1232 * Changes to the gh_ interface
1234 * Changes to the scm_ interface
1236 * Changes to system call interfaces:
1238 ** The value returned by `raise' is now unspecified. It throws an exception
1241 *** A new procedure `sigaction' can be used to install signal handlers
1243 (sigaction signum [action] [flags])
1245 signum is the signal number, which can be specified using the value
1248 If action is omitted, sigaction returns a pair: the CAR is the current
1249 signal hander, which will be either an integer with the value SIG_DFL
1250 (default action) or SIG_IGN (ignore), or the Scheme procedure which
1251 handles the signal, or #f if a non-Scheme procedure handles the
1252 signal. The CDR contains the current sigaction flags for the handler.
1254 If action is provided, it is installed as the new handler for signum.
1255 action can be a Scheme procedure taking one argument, or the value of
1256 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
1257 whatever signal handler was installed before sigaction was first used.
1258 Flags can optionally be specified for the new handler (SA_RESTART is
1259 always used if the system provides it, so need not be specified.) The
1260 return value is a pair with information about the old handler as
1263 This interface does not provide access to the "signal blocking"
1264 facility. Maybe this is not needed, since the thread support may
1265 provide solutions to the problem of consistent access to data
1268 *** A new procedure `flush-all-ports' is equivalent to running
1269 `force-output' on every port open for output.
1271 ** Guile now provides information on how it was built, via the new
1272 global variable, %guile-build-info. This variable records the values
1273 of the standard GNU makefile directory variables as an assocation
1274 list, mapping variable names (symbols) onto directory paths (strings).
1275 For example, to find out where the Guile link libraries were
1276 installed, you can say:
1278 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
1281 * Changes to the scm_ interface
1283 ** The new function scm_handle_by_message_noexit is just like the
1284 existing scm_handle_by_message function, except that it doesn't call
1285 exit to terminate the process. Instead, it prints a message and just
1286 returns #f. This might be a more appropriate catch-all handler for
1287 new dynamic roots and threads.
1290 Changes in Guile 1.1 (released Friday, May 16 1997):
1292 * Changes to the distribution.
1294 The Guile 1.0 distribution has been split up into several smaller
1296 guile-core --- the Guile interpreter itself.
1297 guile-tcltk --- the interface between the Guile interpreter and
1298 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
1299 is a toolkit for building graphical user interfaces.
1300 guile-rgx-ctax --- the interface between Guile and the Rx regular
1301 expression matcher, and the translator for the Ctax
1302 programming language. These are packaged together because the
1303 Ctax translator uses Rx to parse Ctax source code.
1305 This NEWS file describes the changes made to guile-core since the 1.0
1308 We no longer distribute the documentation, since it was either out of
1309 date, or incomplete. As soon as we have current documentation, we
1314 * Changes to the stand-alone interpreter
1316 ** guile now accepts command-line arguments compatible with SCSH, Olin
1317 Shivers' Scheme Shell.
1319 In general, arguments are evaluated from left to right, but there are
1320 exceptions. The following switches stop argument processing, and
1321 stash all remaining command-line arguments as the value returned by
1322 the (command-line) function.
1323 -s SCRIPT load Scheme source code from FILE, and exit
1324 -c EXPR evalute Scheme expression EXPR, and exit
1325 -- stop scanning arguments; run interactively
1327 The switches below are processed as they are encountered.
1328 -l FILE load Scheme source code from FILE
1329 -e FUNCTION after reading script, apply FUNCTION to
1330 command line arguments
1331 -ds do -s script at this point
1332 --emacs enable Emacs protocol (experimental)
1333 -h, --help display this help and exit
1334 -v, --version display version information and exit
1335 \ read arguments from following script lines
1337 So, for example, here is a Guile script named `ekko' (thanks, Olin)
1338 which re-implements the traditional "echo" command:
1340 #!/usr/local/bin/guile -s
1343 (map (lambda (arg) (display arg) (display " "))
1347 (main (command-line))
1349 Suppose we invoke this script as follows:
1351 ekko a speckled gecko
1353 Through the magic of Unix script processing (triggered by the `#!'
1354 token at the top of the file), /usr/local/bin/guile receives the
1355 following list of command-line arguments:
1357 ("-s" "./ekko" "a" "speckled" "gecko")
1359 Unix inserts the name of the script after the argument specified on
1360 the first line of the file (in this case, "-s"), and then follows that
1361 with the arguments given to the script. Guile loads the script, which
1362 defines the `main' function, and then applies it to the list of
1363 remaining command-line arguments, ("a" "speckled" "gecko").
1365 In Unix, the first line of a script file must take the following form:
1367 #!INTERPRETER ARGUMENT
1369 where INTERPRETER is the absolute filename of the interpreter
1370 executable, and ARGUMENT is a single command-line argument to pass to
1373 You may only pass one argument to the interpreter, and its length is
1374 limited. These restrictions can be annoying to work around, so Guile
1375 provides a general mechanism (borrowed from, and compatible with,
1376 SCSH) for circumventing them.
1378 If the ARGUMENT in a Guile script is a single backslash character,
1379 `\', Guile will open the script file, parse arguments from its second
1380 and subsequent lines, and replace the `\' with them. So, for example,
1381 here is another implementation of the `ekko' script:
1383 #!/usr/local/bin/guile \
1387 (for-each (lambda (arg) (display arg) (display " "))
1391 If the user invokes this script as follows:
1393 ekko a speckled gecko
1395 Unix expands this into
1397 /usr/local/bin/guile \ ekko a speckled gecko
1399 When Guile sees the `\' argument, it replaces it with the arguments
1400 read from the second line of the script, producing:
1402 /usr/local/bin/guile -e main -s ekko a speckled gecko
1404 This tells Guile to load the `ekko' script, and apply the function
1405 `main' to the argument list ("a" "speckled" "gecko").
1407 Here is how Guile parses the command-line arguments:
1408 - Each space character terminates an argument. This means that two
1409 spaces in a row introduce an empty-string argument.
1410 - The tab character is not permitted (unless you quote it with the
1411 backslash character, as described below), to avoid confusion.
1412 - The newline character terminates the sequence of arguments, and will
1413 also terminate a final non-empty argument. (However, a newline
1414 following a space will not introduce a final empty-string argument;
1415 it only terminates the argument list.)
1416 - The backslash character is the escape character. It escapes
1417 backslash, space, tab, and newline. The ANSI C escape sequences
1418 like \n and \t are also supported. These produce argument
1419 constituents; the two-character combination \n doesn't act like a
1420 terminating newline. The escape sequence \NNN for exactly three
1421 octal digits reads as the character whose ASCII code is NNN. As
1422 above, characters produced this way are argument constituents.
1423 Backslash followed by other characters is not allowed.
1425 * Changes to the procedure for linking libguile with your programs
1427 ** Guile now builds and installs a shared guile library, if your
1428 system support shared libraries. (It still builds a static library on
1429 all systems.) Guile automatically detects whether your system
1430 supports shared libraries. To prevent Guile from buildisg shared
1431 libraries, pass the `--disable-shared' flag to the configure script.
1433 Guile takes longer to compile when it builds shared libraries, because
1434 it must compile every file twice --- once to produce position-
1435 independent object code, and once to produce normal object code.
1437 ** The libthreads library has been merged into libguile.
1439 To link a program against Guile, you now need only link against
1440 -lguile and -lqt; -lthreads is no longer needed. If you are using
1441 autoconf to generate configuration scripts for your application, the
1442 following lines should suffice to add the appropriate libraries to
1445 ### Find quickthreads and libguile.
1446 AC_CHECK_LIB(qt, main)
1447 AC_CHECK_LIB(guile, scm_shell)
1449 * Changes to Scheme functions
1451 ** Guile Scheme's special syntax for keyword objects is now optional,
1452 and disabled by default.
1454 The syntax variation from R4RS made it difficult to port some
1455 interesting packages to Guile. The routines which accepted keyword
1456 arguments (mostly in the module system) have been modified to also
1457 accept symbols whose names begin with `:'.
1459 To change the keyword syntax, you must first import the (ice-9 debug)
1461 (use-modules (ice-9 debug))
1463 Then you can enable the keyword syntax as follows:
1464 (read-set! keywords 'prefix)
1466 To disable keyword syntax, do this:
1467 (read-set! keywords #f)
1469 ** Many more primitive functions accept shared substrings as
1470 arguments. In the past, these functions required normal, mutable
1471 strings as arguments, although they never made use of this
1474 ** The uniform array functions now operate on byte vectors. These
1475 functions are `array-fill!', `serial-array-copy!', `array-copy!',
1476 `serial-array-map', `array-map', `array-for-each', and
1479 ** The new functions `trace' and `untrace' implement simple debugging
1480 support for Scheme functions.
1482 The `trace' function accepts any number of procedures as arguments,
1483 and tells the Guile interpreter to display each procedure's name and
1484 arguments each time the procedure is invoked. When invoked with no
1485 arguments, `trace' returns the list of procedures currently being
1488 The `untrace' function accepts any number of procedures as arguments,
1489 and tells the Guile interpreter not to trace them any more. When
1490 invoked with no arguments, `untrace' untraces all curretly traced
1493 The tracing in Guile has an advantage over most other systems: we
1494 don't create new procedure objects, but mark the procedure objects
1495 themselves. This means that anonymous and internal procedures can be
1498 ** The function `assert-repl-prompt' has been renamed to
1499 `set-repl-prompt!'. It takes one argument, PROMPT.
1500 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
1501 - If PROMPT is a string, we use it as a prompt.
1502 - If PROMPT is a procedure accepting no arguments, we call it, and
1503 display the result as a prompt.
1504 - Otherwise, we display "> ".
1506 ** The new function `eval-string' reads Scheme expressions from a
1507 string and evaluates them, returning the value of the last expression
1508 in the string. If the string contains no expressions, it returns an
1511 ** The new function `thunk?' returns true iff its argument is a
1512 procedure of zero arguments.
1514 ** `defined?' is now a builtin function, instead of syntax. This
1515 means that its argument should be quoted. It returns #t iff its
1516 argument is bound in the current module.
1518 ** The new syntax `use-modules' allows you to add new modules to your
1519 environment without re-typing a complete `define-module' form. It
1520 accepts any number of module names as arguments, and imports their
1521 public bindings into the current module.
1523 ** The new function (module-defined? NAME MODULE) returns true iff
1524 NAME, a symbol, is defined in MODULE, a module object.
1526 ** The new function `builtin-bindings' creates and returns a hash
1527 table containing copies of all the root module's bindings.
1529 ** The new function `builtin-weak-bindings' does the same as
1530 `builtin-bindings', but creates a doubly-weak hash table.
1532 ** The `equal?' function now considers variable objects to be
1533 equivalent if they have the same name and the same value.
1535 ** The new function `command-line' returns the command-line arguments
1536 given to Guile, as a list of strings.
1538 When using guile as a script interpreter, `command-line' returns the
1539 script's arguments; those processed by the interpreter (like `-s' or
1540 `-c') are omitted. (In other words, you get the normal, expected
1541 behavior.) Any application that uses scm_shell to process its
1542 command-line arguments gets this behavior as well.
1544 ** The new function `load-user-init' looks for a file called `.guile'
1545 in the user's home directory, and loads it if it exists. This is
1546 mostly for use by the code generated by scm_compile_shell_switches,
1547 but we thought it might also be useful in other circumstances.
1549 ** The new function `log10' returns the base-10 logarithm of its
1552 ** Changes to I/O functions
1554 *** The functions `read', `primitive-load', `read-and-eval!', and
1555 `primitive-load-path' no longer take optional arguments controlling
1556 case insensitivity and a `#' parser.
1558 Case sensitivity is now controlled by a read option called
1559 `case-insensitive'. The user can add new `#' syntaxes with the
1560 `read-hash-extend' function (see below).
1562 *** The new function `read-hash-extend' allows the user to change the
1563 syntax of Guile Scheme in a somewhat controlled way.
1565 (read-hash-extend CHAR PROC)
1566 When parsing S-expressions, if we read a `#' character followed by
1567 the character CHAR, use PROC to parse an object from the stream.
1568 If PROC is #f, remove any parsing procedure registered for CHAR.
1570 The reader applies PROC to two arguments: CHAR and an input port.
1572 *** The new functions read-delimited and read-delimited! provide a
1573 general mechanism for doing delimited input on streams.
1575 (read-delimited DELIMS [PORT HANDLE-DELIM])
1576 Read until we encounter one of the characters in DELIMS (a string),
1577 or end-of-file. PORT is the input port to read from; it defaults to
1578 the current input port. The HANDLE-DELIM parameter determines how
1579 the terminating character is handled; it should be one of the
1582 'trim omit delimiter from result
1583 'peek leave delimiter character in input stream
1584 'concat append delimiter character to returned value
1585 'split return a pair: (RESULT . TERMINATOR)
1587 HANDLE-DELIM defaults to 'peek.
1589 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
1590 A side-effecting variant of `read-delimited'.
1592 The data is written into the string BUF at the indices in the
1593 half-open interval [START, END); the default interval is the whole
1594 string: START = 0 and END = (string-length BUF). The values of
1595 START and END must specify a well-defined interval in BUF, i.e.
1596 0 <= START <= END <= (string-length BUF).
1598 It returns NBYTES, the number of bytes read. If the buffer filled
1599 up without a delimiter character being found, it returns #f. If the
1600 port is at EOF when the read starts, it returns the EOF object.
1602 If an integer is returned (i.e., the read is successfully terminated
1603 by reading a delimiter character), then the HANDLE-DELIM parameter
1604 determines how to handle the terminating character. It is described
1605 above, and defaults to 'peek.
1607 (The descriptions of these functions were borrowed from the SCSH
1608 manual, by Olin Shivers and Brian Carlstrom.)
1610 *** The `%read-delimited!' function is the primitive used to implement
1611 `read-delimited' and `read-delimited!'.
1613 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
1615 This returns a pair of values: (TERMINATOR . NUM-READ).
1616 - TERMINATOR describes why the read was terminated. If it is a
1617 character or the eof object, then that is the value that terminated
1618 the read. If it is #f, the function filled the buffer without finding
1619 a delimiting character.
1620 - NUM-READ is the number of characters read into BUF.
1622 If the read is successfully terminated by reading a delimiter
1623 character, then the gobble? parameter determines what to do with the
1624 terminating character. If true, the character is removed from the
1625 input stream; if false, the character is left in the input stream
1626 where a subsequent read operation will retrieve it. In either case,
1627 the character is also the first value returned by the procedure call.
1629 (The descriptions of this function was borrowed from the SCSH manual,
1630 by Olin Shivers and Brian Carlstrom.)
1632 *** The `read-line' and `read-line!' functions have changed; they now
1633 trim the terminator by default; previously they appended it to the
1634 returned string. For the old behavior, use (read-line PORT 'concat).
1636 *** The functions `uniform-array-read!' and `uniform-array-write!' now
1637 take new optional START and END arguments, specifying the region of
1638 the array to read and write.
1640 *** The `ungetc-char-ready?' function has been removed. We feel it's
1641 inappropriate for an interface to expose implementation details this
1644 ** Changes to the Unix library and system call interface
1646 *** The new fcntl function provides access to the Unix `fcntl' system
1649 (fcntl PORT COMMAND VALUE)
1650 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
1651 Values for COMMAND are:
1653 F_DUPFD duplicate a file descriptor
1654 F_GETFD read the descriptor's close-on-exec flag
1655 F_SETFD set the descriptor's close-on-exec flag to VALUE
1656 F_GETFL read the descriptor's flags, as set on open
1657 F_SETFL set the descriptor's flags, as set on open to VALUE
1658 F_GETOWN return the process ID of a socket's owner, for SIGIO
1659 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
1660 FD_CLOEXEC not sure what this is
1662 For details, see the documentation for the fcntl system call.
1664 *** The arguments to `select' have changed, for compatibility with
1665 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
1666 expected behavior. The MILLISECONDS parameter has been changed to
1667 MICROSECONDS, to more closely resemble the underlying system call.
1668 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
1669 corresponding return set will be the same.
1671 *** The arguments to the `mknod' system call have changed. They are
1674 (mknod PATH TYPE PERMS DEV)
1675 Create a new file (`node') in the file system. PATH is the name of
1676 the file to create. TYPE is the kind of file to create; it should
1677 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
1678 permission bits to give the newly created file. If TYPE is
1679 'block-special or 'char-special, DEV specifies which device the
1680 special file refers to; its interpretation depends on the kind of
1681 special file being created.
1683 *** The `fork' function has been renamed to `primitive-fork', to avoid
1684 clashing with various SCSH forks.
1686 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
1687 and `recvfrom!'. They no longer accept a size for a second argument;
1688 you must pass a string to hold the received value. They no longer
1689 return the buffer. Instead, `recv' returns the length of the message
1690 received, and `recvfrom' returns a pair containing the packet's length
1691 and originating address.
1693 *** The file descriptor datatype has been removed, as have the
1694 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
1695 We plan to replace these functions with a SCSH-compatible interface.
1697 *** The `create' function has been removed; it's just a special case
1700 *** There are new functions to break down process termination status
1701 values. In the descriptions below, STATUS is a value returned by
1704 (status:exit-val STATUS)
1705 If the child process exited normally, this function returns the exit
1706 code for the child process (i.e., the value passed to exit, or
1707 returned from main). If the child process did not exit normally,
1708 this function returns #f.
1710 (status:stop-sig STATUS)
1711 If the child process was suspended by a signal, this function
1712 returns the signal that suspended the child. Otherwise, it returns
1715 (status:term-sig STATUS)
1716 If the child process terminated abnormally, this function returns
1717 the signal that terminated the child. Otherwise, this function
1720 POSIX promises that exactly one of these functions will return true on
1721 a valid STATUS value.
1723 These functions are compatible with SCSH.
1725 *** There are new accessors and setters for the broken-out time vectors
1726 returned by `localtime', `gmtime', and that ilk. They are:
1728 Component Accessor Setter
1729 ========================= ============ ============
1730 seconds tm:sec set-tm:sec
1731 minutes tm:min set-tm:min
1732 hours tm:hour set-tm:hour
1733 day of the month tm:mday set-tm:mday
1734 month tm:mon set-tm:mon
1735 year tm:year set-tm:year
1736 day of the week tm:wday set-tm:wday
1737 day in the year tm:yday set-tm:yday
1738 daylight saving time tm:isdst set-tm:isdst
1739 GMT offset, seconds tm:gmtoff set-tm:gmtoff
1740 name of time zone tm:zone set-tm:zone
1742 *** There are new accessors for the vectors returned by `uname',
1743 describing the host system:
1746 ============================================== ================
1747 name of the operating system implementation utsname:sysname
1748 network name of this machine utsname:nodename
1749 release level of the operating system utsname:release
1750 version level of the operating system utsname:version
1751 machine hardware platform utsname:machine
1753 *** There are new accessors for the vectors returned by `getpw',
1754 `getpwnam', `getpwuid', and `getpwent', describing entries from the
1755 system's user database:
1758 ====================== =================
1759 user name passwd:name
1760 user password passwd:passwd
1763 real name passwd:gecos
1764 home directory passwd:dir
1765 shell program passwd:shell
1767 *** There are new accessors for the vectors returned by `getgr',
1768 `getgrnam', `getgrgid', and `getgrent', describing entries from the
1769 system's group database:
1772 ======================= ============
1773 group name group:name
1774 group password group:passwd
1776 group members group:mem
1778 *** There are new accessors for the vectors returned by `gethost',
1779 `gethostbyaddr', `gethostbyname', and `gethostent', describing
1783 ========================= ===============
1784 official name of host hostent:name
1785 alias list hostent:aliases
1786 host address type hostent:addrtype
1787 length of address hostent:length
1788 list of addresses hostent:addr-list
1790 *** There are new accessors for the vectors returned by `getnet',
1791 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
1795 ========================= ===============
1796 official name of net netent:name
1797 alias list netent:aliases
1798 net number type netent:addrtype
1799 net number netent:net
1801 *** There are new accessors for the vectors returned by `getproto',
1802 `getprotobyname', `getprotobynumber', and `getprotoent', describing
1806 ========================= ===============
1807 official protocol name protoent:name
1808 alias list protoent:aliases
1809 protocol number protoent:proto
1811 *** There are new accessors for the vectors returned by `getserv',
1812 `getservbyname', `getservbyport', and `getservent', describing
1816 ========================= ===============
1817 official service name servent:name
1818 alias list servent:aliases
1819 port number servent:port
1820 protocol to use servent:proto
1822 *** There are new accessors for the sockaddr structures returned by
1823 `accept', `getsockname', `getpeername', `recvfrom!':
1826 ======================================== ===============
1827 address format (`family') sockaddr:fam
1828 path, for file domain addresses sockaddr:path
1829 address, for internet domain addresses sockaddr:addr
1830 TCP or UDP port, for internet sockaddr:port
1832 *** The `getpwent', `getgrent', `gethostent', `getnetent',
1833 `getprotoent', and `getservent' functions now return #f at the end of
1834 the user database. (They used to throw an exception.)
1836 Note that calling MUMBLEent function is equivalent to calling the
1837 corresponding MUMBLE function with no arguments.
1839 *** The `setpwent', `setgrent', `sethostent', `setnetent',
1840 `setprotoent', and `setservent' routines now take no arguments.
1842 *** The `gethost', `getproto', `getnet', and `getserv' functions now
1843 provide more useful information when they throw an exception.
1845 *** The `lnaof' function has been renamed to `inet-lnaof'.
1847 *** Guile now claims to have the `current-time' feature.
1849 *** The `mktime' function now takes an optional second argument ZONE,
1850 giving the time zone to use for the conversion. ZONE should be a
1851 string, in the same format as expected for the "TZ" environment variable.
1853 *** The `strptime' function now returns a pair (TIME . COUNT), where
1854 TIME is the parsed time as a vector, and COUNT is the number of
1855 characters from the string left unparsed. This function used to
1856 return the remaining characters as a string.
1858 *** The `gettimeofday' function has replaced the old `time+ticks' function.
1859 The return value is now (SECONDS . MICROSECONDS); the fractional
1860 component is no longer expressed in "ticks".
1862 *** The `ticks/sec' constant has been removed, in light of the above change.
1864 * Changes to the gh_ interface
1866 ** gh_eval_str() now returns an SCM object which is the result of the
1869 ** gh_scm2str() now copies the Scheme data to a caller-provided C
1872 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
1873 and returns the array
1875 ** gh_scm2str0() is gone: there is no need to distinguish
1876 null-terminated from non-null-terminated, since gh_scm2newstr() allows
1877 the user to interpret the data both ways.
1879 * Changes to the scm_ interface
1881 ** The new function scm_symbol_value0 provides an easy way to get a
1882 symbol's value from C code:
1884 SCM scm_symbol_value0 (char *NAME)
1885 Return the value of the symbol named by the null-terminated string
1886 NAME in the current module. If the symbol named NAME is unbound in
1887 the current module, return SCM_UNDEFINED.
1889 ** The new function scm_sysintern0 creates new top-level variables,
1890 without assigning them a value.
1892 SCM scm_sysintern0 (char *NAME)
1893 Create a new Scheme top-level variable named NAME. NAME is a
1894 null-terminated string. Return the variable's value cell.
1896 ** The function scm_internal_catch is the guts of catch. It handles
1897 all the mechanics of setting up a catch target, invoking the catch
1898 body, and perhaps invoking the handler if the body does a throw.
1900 The function is designed to be usable from C code, but is general
1901 enough to implement all the semantics Guile Scheme expects from throw.
1903 TAG is the catch tag. Typically, this is a symbol, but this function
1904 doesn't actually care about that.
1906 BODY is a pointer to a C function which runs the body of the catch;
1907 this is the code you can throw from. We call it like this:
1908 BODY (BODY_DATA, JMPBUF)
1910 BODY_DATA is just the BODY_DATA argument we received; we pass it
1911 through to BODY as its first argument. The caller can make
1912 BODY_DATA point to anything useful that BODY might need.
1913 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
1914 which we have just created and initialized.
1916 HANDLER is a pointer to a C function to deal with a throw to TAG,
1917 should one occur. We call it like this:
1918 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
1920 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
1921 same idea as BODY_DATA above.
1922 THROWN_TAG is the tag that the user threw to; usually this is
1923 TAG, but it could be something else if TAG was #t (i.e., a
1924 catch-all), or the user threw to a jmpbuf.
1925 THROW_ARGS is the list of arguments the user passed to the THROW
1928 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
1929 is just a pointer we pass through to HANDLER. We don't actually
1930 use either of those pointers otherwise ourselves. The idea is
1931 that, if our caller wants to communicate something to BODY or
1932 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
1933 HANDLER can then use. Think of it as a way to make BODY and
1934 HANDLER closures, not just functions; MUMBLE_DATA points to the
1937 Of course, it's up to the caller to make sure that any data a
1938 MUMBLE_DATA needs is protected from GC. A common way to do this is
1939 to make MUMBLE_DATA a pointer to data stored in an automatic
1940 structure variable; since the collector must scan the stack for
1941 references anyway, this assures that any references in MUMBLE_DATA
1944 ** The new function scm_internal_lazy_catch is exactly like
1945 scm_internal_catch, except:
1947 - It does not unwind the stack (this is the major difference).
1948 - If handler returns, its value is returned from the throw.
1949 - BODY always receives #f as its JMPBUF argument (since there's no
1950 jmpbuf associated with a lazy catch, because we don't unwind the
1953 ** scm_body_thunk is a new body function you can pass to
1954 scm_internal_catch if you want the body to be like Scheme's `catch'
1955 --- a thunk, or a function of one argument if the tag is #f.
1957 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
1958 contains the Scheme procedure to invoke as the body, and the tag
1959 we're catching. If the tag is #f, then we pass JMPBUF (created by
1960 scm_internal_catch) to the body procedure; otherwise, the body gets
1963 ** scm_handle_by_proc is a new handler function you can pass to
1964 scm_internal_catch if you want the handler to act like Scheme's catch
1965 --- call a procedure with the tag and the throw arguments.
1967 If the user does a throw to this catch, this function runs a handler
1968 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
1969 variable holding the Scheme procedure object to invoke. It ought to
1970 be a pointer to an automatic variable (i.e., one living on the stack),
1971 or the procedure object should be otherwise protected from GC.
1973 ** scm_handle_by_message is a new handler function to use with
1974 `scm_internal_catch' if you want Guile to print a message and die.
1975 It's useful for dealing with throws to uncaught keys at the top level.
1977 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
1978 message header to print; if zero, we use "guile" instead. That
1979 text is followed by a colon, then the message described by ARGS.
1981 ** The return type of scm_boot_guile is now void; the function does
1982 not return a value, and indeed, never returns at all.
1984 ** The new function scm_shell makes it easy for user applications to
1985 process command-line arguments in a way that is compatible with the
1986 stand-alone guile interpreter (which is in turn compatible with SCSH,
1989 To use the scm_shell function, first initialize any guile modules
1990 linked into your application, and then call scm_shell with the values
1991 of ARGC and ARGV your `main' function received. scm_shell will add
1992 any SCSH-style meta-arguments from the top of the script file to the
1993 argument vector, and then process the command-line arguments. This
1994 generally means loading a script file or starting up an interactive
1995 command interpreter. For details, see "Changes to the stand-alone
1998 ** The new functions scm_get_meta_args and scm_count_argv help you
1999 implement the SCSH-style meta-argument, `\'.
2001 char **scm_get_meta_args (int ARGC, char **ARGV)
2002 If the second element of ARGV is a string consisting of a single
2003 backslash character (i.e. "\\" in Scheme notation), open the file
2004 named by the following argument, parse arguments from it, and return
2005 the spliced command line. The returned array is terminated by a
2008 For details of argument parsing, see above, under "guile now accepts
2009 command-line arguments compatible with SCSH..."
2011 int scm_count_argv (char **ARGV)
2012 Count the arguments in ARGV, assuming it is terminated by a null
2015 For an example of how these functions might be used, see the source
2016 code for the function scm_shell in libguile/script.c.
2018 You will usually want to use scm_shell instead of calling this
2021 ** The new function scm_compile_shell_switches turns an array of
2022 command-line arguments into Scheme code to carry out the actions they
2023 describe. Given ARGC and ARGV, it returns a Scheme expression to
2024 evaluate, and calls scm_set_program_arguments to make any remaining
2025 command-line arguments available to the Scheme code. For example,
2026 given the following arguments:
2028 -e main -s ekko a speckled gecko
2030 scm_set_program_arguments will return the following expression:
2032 (begin (load "ekko") (main (command-line)) (quit))
2034 You will usually want to use scm_shell instead of calling this
2037 ** The function scm_shell_usage prints a usage message appropriate for
2038 an interpreter that uses scm_compile_shell_switches to handle its
2039 command-line arguments.
2041 void scm_shell_usage (int FATAL, char *MESSAGE)
2042 Print a usage message to the standard error output. If MESSAGE is
2043 non-zero, write it before the usage message, followed by a newline.
2044 If FATAL is non-zero, exit the process, using FATAL as the
2045 termination status. (If you want to be compatible with Guile,
2046 always use 1 as the exit status when terminating due to command-line
2049 You will usually want to use scm_shell instead of calling this
2052 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
2053 expressions. It used to return SCM_EOL. Earth-shattering.
2055 ** The macros for declaring scheme objects in C code have been
2056 rearranged slightly. They are now:
2058 SCM_SYMBOL (C_NAME, SCHEME_NAME)
2059 Declare a static SCM variable named C_NAME, and initialize it to
2060 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
2061 be a C identifier, and SCHEME_NAME should be a C string.
2063 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
2064 Just like SCM_SYMBOL, but make C_NAME globally visible.
2066 SCM_VCELL (C_NAME, SCHEME_NAME)
2067 Create a global variable at the Scheme level named SCHEME_NAME.
2068 Declare a static SCM variable named C_NAME, and initialize it to
2069 point to the Scheme variable's value cell.
2071 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
2072 Just like SCM_VCELL, but make C_NAME globally visible.
2074 The `guile-snarf' script writes initialization code for these macros
2075 to its standard output, given C source code as input.
2077 The SCM_GLOBAL macro is gone.
2079 ** The scm_read_line and scm_read_line_x functions have been replaced
2080 by Scheme code based on the %read-delimited! procedure (known to C
2081 code as scm_read_delimited_x). See its description above for more
2084 ** The function scm_sys_open has been renamed to scm_open. It now
2085 returns a port instead of an FD object.
2087 * The dynamic linking support has changed. For more information, see
2088 libguile/DYNAMIC-LINKING.
2093 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
2096 * Changes to the 'guile' program:
2098 ** Guile now loads some new files when it starts up. Guile first
2099 searches the load path for init.scm, and loads it if found. Then, if
2100 Guile is not being used to execute a script, and the user's home
2101 directory contains a file named `.guile', Guile loads that.
2103 ** You can now use Guile as a shell script interpreter.
2105 To paraphrase the SCSH manual:
2107 When Unix tries to execute an executable file whose first two
2108 characters are the `#!', it treats the file not as machine code to
2109 be directly executed by the native processor, but as source code
2110 to be executed by some interpreter. The interpreter to use is
2111 specified immediately after the #! sequence on the first line of
2112 the source file. The kernel reads in the name of the interpreter,
2113 and executes that instead. It passes the interpreter the source
2114 filename as its first argument, with the original arguments
2115 following. Consult the Unix man page for the `exec' system call
2116 for more information.
2118 Now you can use Guile as an interpreter, using a mechanism which is a
2119 compatible subset of that provided by SCSH.
2121 Guile now recognizes a '-s' command line switch, whose argument is the
2122 name of a file of Scheme code to load. It also treats the two
2123 characters `#!' as the start of a comment, terminated by `!#'. Thus,
2124 to make a file of Scheme code directly executable by Unix, insert the
2125 following two lines at the top of the file:
2127 #!/usr/local/bin/guile -s
2130 Guile treats the argument of the `-s' command-line switch as the name
2131 of a file of Scheme code to load, and treats the sequence `#!' as the
2132 start of a block comment, terminated by `!#'.
2134 For example, here's a version of 'echo' written in Scheme:
2136 #!/usr/local/bin/guile -s
2138 (let loop ((args (cdr (program-arguments))))
2141 (display (car args))
2142 (if (pair? (cdr args))
2144 (loop (cdr args)))))
2147 Why does `#!' start a block comment terminated by `!#', instead of the
2148 end of the line? That is the notation SCSH uses, and although we
2149 don't yet support the other SCSH features that motivate that choice,
2150 we would like to be backward-compatible with any existing Guile
2151 scripts once we do. Furthermore, if the path to Guile on your system
2152 is too long for your kernel, you can start the script with this
2156 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
2159 Note that some very old Unix systems don't support the `#!' syntax.
2162 ** You can now run Guile without installing it.
2164 Previous versions of the interactive Guile interpreter (`guile')
2165 couldn't start up unless Guile's Scheme library had been installed;
2166 they used the value of the environment variable `SCHEME_LOAD_PATH'
2167 later on in the startup process, but not to find the startup code
2168 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
2171 To run Guile without installing it, build it in the normal way, and
2172 then set the environment variable `SCHEME_LOAD_PATH' to a
2173 colon-separated list of directories, including the top-level directory
2174 of the Guile sources. For example, if you unpacked Guile so that the
2175 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
2178 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
2181 ** Guile's read-eval-print loop no longer prints #<unspecified>
2182 results. If the user wants to see this, she can evaluate the
2183 expression (assert-repl-print-unspecified #t), perhaps in her startup
2186 ** Guile no longer shows backtraces by default when an error occurs;
2187 however, it does display a message saying how to get one, and how to
2188 request that they be displayed by default. After an error, evaluate
2190 to see a backtrace, and
2191 (debug-enable 'backtrace)
2192 to see them by default.
2196 * Changes to Guile Scheme:
2198 ** Guile now distinguishes between #f and the empty list.
2200 This is for compatibility with the IEEE standard, the (possibly)
2201 upcoming Revised^5 Report on Scheme, and many extant Scheme
2204 Guile used to have #f and '() denote the same object, to make Scheme's
2205 type system more compatible with Emacs Lisp's. However, the change
2206 caused too much trouble for Scheme programmers, and we found another
2207 way to reconcile Emacs Lisp with Scheme that didn't require this.
2210 ** Guile's delq, delv, delete functions, and their destructive
2211 counterparts, delq!, delv!, and delete!, now remove all matching
2212 elements from the list, not just the first. This matches the behavior
2213 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
2214 functions which inspired them.
2216 I recognize that this change may break code in subtle ways, but it
2217 seems best to make the change before the FSF's first Guile release,
2221 ** The compiled-library-path function has been deleted from libguile.
2223 ** The facilities for loading Scheme source files have changed.
2225 *** The variable %load-path now tells Guile which directories to search
2226 for Scheme code. Its value is a list of strings, each of which names
2229 *** The variable %load-extensions now tells Guile which extensions to
2230 try appending to a filename when searching the load path. Its value
2231 is a list of strings. Its default value is ("" ".scm").
2233 *** (%search-load-path FILENAME) searches the directories listed in the
2234 value of the %load-path variable for a Scheme file named FILENAME,
2235 with all the extensions listed in %load-extensions. If it finds a
2236 match, then it returns its full filename. If FILENAME is absolute, it
2237 returns it unchanged. Otherwise, it returns #f.
2239 %search-load-path will not return matches that refer to directories.
2241 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
2242 uses %seach-load-path to find a file named FILENAME, and loads it if
2243 it finds it. If it can't read FILENAME for any reason, it throws an
2246 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
2249 *** load uses the same searching semantics as primitive-load.
2251 *** The functions %try-load, try-load-with-path, %load, load-with-path,
2252 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
2253 path, and load-module-with-path have been deleted. The functions
2254 above should serve their purposes.
2256 *** If the value of the variable %load-hook is a procedure,
2257 `primitive-load' applies its value to the name of the file being
2258 loaded (without the load path directory name prepended). If its value
2259 is #f, it is ignored. Otherwise, an error occurs.
2261 This is mostly useful for printing load notification messages.
2264 ** The function `eval!' is no longer accessible from the scheme level.
2265 We can't allow operations which introduce glocs into the scheme level,
2266 because Guile's type system can't handle these as data. Use `eval' or
2267 `read-and-eval!' (see below) as replacement.
2269 ** The new function read-and-eval! reads an expression from PORT,
2270 evaluates it, and returns the result. This is more efficient than
2271 simply calling `read' and `eval', since it is not necessary to make a
2272 copy of the expression for the evaluator to munge.
2274 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
2275 for the `read' function.
2278 ** The function `int?' has been removed; its definition was identical
2279 to that of `integer?'.
2281 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
2282 use the R4RS names for these functions.
2284 ** The function object-properties no longer returns the hash handle;
2285 it simply returns the object's property list.
2287 ** Many functions have been changed to throw errors, instead of
2288 returning #f on failure. The point of providing exception handling in
2289 the language is to simplify the logic of user code, but this is less
2290 useful if Guile's primitives don't throw exceptions.
2292 ** The function `fileno' has been renamed from `%fileno'.
2294 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
2297 * Changes to Guile's C interface:
2299 ** The library's initialization procedure has been simplified.
2300 scm_boot_guile now has the prototype:
2302 void scm_boot_guile (int ARGC,
2304 void (*main_func) (),
2307 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
2308 MAIN_FUNC should do all the work of the program (initializing other
2309 packages, reading user input, etc.) before returning. When MAIN_FUNC
2310 returns, call exit (0); this function never returns. If you want some
2311 other exit value, MAIN_FUNC may call exit itself.
2313 scm_boot_guile arranges for program-arguments to return the strings
2314 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
2315 scm_set_program_arguments with the final list, so Scheme code will
2316 know which arguments have been processed.
2318 scm_boot_guile establishes a catch-all catch handler which prints an
2319 error message and exits the process. This means that Guile exits in a
2320 coherent way when system errors occur and the user isn't prepared to
2321 handle it. If the user doesn't like this behavior, they can establish
2322 their own universal catcher in MAIN_FUNC to shadow this one.
2324 Why must the caller do all the real work from MAIN_FUNC? The garbage
2325 collector assumes that all local variables of type SCM will be above
2326 scm_boot_guile's stack frame on the stack. If you try to manipulate
2327 SCM values after this function returns, it's the luck of the draw
2328 whether the GC will be able to find the objects you allocate. So,
2329 scm_boot_guile function exits, rather than returning, to discourage
2330 people from making that mistake.
2332 The IN, OUT, and ERR arguments were removed; there are other
2333 convenient ways to override these when desired.
2335 The RESULT argument was deleted; this function should never return.
2337 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
2341 ** Guile's header files should no longer conflict with your system's
2344 In order to compile code which #included <libguile.h>, previous
2345 versions of Guile required you to add a directory containing all the
2346 Guile header files to your #include path. This was a problem, since
2347 Guile's header files have names which conflict with many systems'
2350 Now only <libguile.h> need appear in your #include path; you must
2351 refer to all Guile's other header files as <libguile/mumble.h>.
2352 Guile's installation procedure puts libguile.h in $(includedir), and
2353 the rest in $(includedir)/libguile.
2356 ** Two new C functions, scm_protect_object and scm_unprotect_object,
2357 have been added to the Guile library.
2359 scm_protect_object (OBJ) protects OBJ from the garbage collector.
2360 OBJ will not be freed, even if all other references are dropped,
2361 until someone does scm_unprotect_object (OBJ). Both functions
2364 Note that calls to scm_protect_object do not nest. You can call
2365 scm_protect_object any number of times on a given object, and the
2366 next call to scm_unprotect_object will unprotect it completely.
2368 Basically, scm_protect_object and scm_unprotect_object just
2369 maintain a list of references to things. Since the GC knows about
2370 this list, all objects it mentions stay alive. scm_protect_object
2371 adds its argument to the list; scm_unprotect_object remove its
2372 argument from the list.
2375 ** scm_eval_0str now returns the value of the last expression
2378 ** The new function scm_read_0str reads an s-expression from a
2379 null-terminated string, and returns it.
2381 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
2382 to a Scheme port object.
2384 ** The new function `scm_set_program_arguments' allows C code to set
2385 the value returned by the Scheme `program-arguments' function.
2390 * Guile no longer includes sophisticated Tcl/Tk support.
2392 The old Tcl/Tk support was unsatisfying to us, because it required the
2393 user to link against the Tcl library, as well as Tk and Guile. The
2394 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
2395 referring to widgets by names, rather than exporting widgets to Scheme
2396 code as a special datatype.
2398 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
2399 maintainers described some very interesting changes in progress to the
2400 Tcl/Tk internals, which would facilitate clean interfaces between lone
2401 Tk and other interpreters --- even for garbage-collected languages
2402 like Scheme. They expected the new Tk to be publicly available in the
2405 Since it seems that Guile might soon have a new, cleaner interface to
2406 lone Tk, and that the old Guile/Tk glue code would probably need to be
2407 completely rewritten, we (Jim Blandy and Richard Stallman) have
2408 decided not to support the old code. We'll spend the time instead on
2409 a good interface to the newer Tk, as soon as it is available.
2411 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
2414 Copyright information:
2416 Copyright (C) 1996,1997 Free Software Foundation, Inc.
2418 Permission is granted to anyone to make or distribute verbatim copies
2419 of this document as received, in any medium, provided that the
2420 copyright notice and this permission notice are preserved,
2421 thus giving the recipient permission to redistribute in turn.
2423 Permission is granted to distribute modified versions
2424 of this document, or of portions of it,
2425 under the above conditions, provided also that they
2426 carry prominent notices stating who last changed them.
2431 paragraph-separate: "[
\f]*$"