1 Guile NEWS --- history of user-visible changes. -*- text -*-
2 Copyright (C) 1996, 1997 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@prep.ai.mit.edu.
7 Changes since Guile 1.1:
11 Changes in Guile 1.1 (Fri May 16 1997):
13 * Changes to the distribution.
15 The Guile 1.0 distribution has been split up into several smaller
17 guile-core --- the Guile interpreter itself.
18 guile-tcltk --- the interface between the Guile interpreter and
19 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
20 is a toolkit for building graphical user interfaces.
21 guile-rgx-ctax --- the interface between Guile and the Rx regular
22 expression matcher, and the translator for the Ctax
23 programming language. These are packaged together because the
24 Ctax translator uses Rx to parse Ctax source code.
26 This NEWS file describes the changes made to guile-core since the 1.0
29 We no longer distribute the documentation, since it was either out of
30 date, or incomplete. As soon as we have current documentation, we
33 * Changes to the stand-alone interpreter
35 ** guile now accepts command-line arguments compatible with SCSH, Olin
36 Shivers' Scheme Shell.
38 In general, arguments are evaluated from left to right, but there are
39 exceptions. The following switches stop argument processing, and
40 stash all remaining command-line arguments as the value returned by
41 the (command-line) function.
42 -s SCRIPT load Scheme source code from FILE, and exit
43 -c EXPR evalute Scheme expression EXPR, and exit
44 -- stop scanning arguments; run interactively
46 The switches below are processed as they are encountered.
47 -l FILE load Scheme source code from FILE
48 -e FUNCTION after reading script, apply FUNCTION to
49 command line arguments
50 -ds do -s script at this point
51 --emacs enable Emacs protocol (experimental)
52 -h, --help display this help and exit
53 -v, --version display version information and exit
54 \ read arguments from following script lines
56 So, for example, here is a Guile script named `ekko' (thanks, Olin)
57 which re-implements the traditional "echo" command:
59 #!/usr/local/bin/guile -s
62 (map (lambda (arg) (display arg) (display " "))
68 Suppose we invoke this script as follows:
72 Through the magic of Unix script processing (triggered by the `#!'
73 token at the top of the file), /usr/local/bin/guile receives the
74 following list of command-line arguments:
76 ("-s" "./ekko" "a" "speckled" "gecko")
78 Unix inserts the name of the script after the argument specified on
79 the first line of the file (in this case, "-s"), and then follows that
80 with the arguments given to the script. Guile loads the script, which
81 defines the `main' function, and then applies it to the list of
82 remaining command-line arguments, ("a" "speckled" "gecko").
84 In Unix, the first line of a script file must take the following form:
86 #!INTERPRETER ARGUMENT
88 where INTERPRETER is the absolute filename of the interpreter
89 executable, and ARGUMENT is a single command-line argument to pass to
92 You may only pass one argument to the interpreter, and its length is
93 limited. These restrictions can be annoying to work around, so Guile
94 provides a general mechanism (borrowed from, and compatible with,
95 SCSH) for circumventing them.
97 If the ARGUMENT in a Guile script is a single backslash character,
98 `\', Guile will open the script file, parse arguments from its second
99 and subsequent lines, and replace the `\' with them. So, for example,
100 here is another implementation of the `ekko' script:
102 #!/usr/local/bin/guile \
106 (for-each (lambda (arg) (display arg) (display " "))
110 If the user invokes this script as follows:
112 ekko a speckled gecko
114 Unix expands this into
116 /usr/local/bin/guile \ ekko a speckled gecko
118 When Guile sees the `\' argument, it replaces it with the arguments
119 read from the second line of the script, producing:
121 /usr/local/bin/guile -e main -s ekko a speckled gecko
123 This tells Guile to load the `ekko' script, and apply the function
124 `main' to the argument list ("a" "speckled" "gecko").
126 Here is how Guile parses the command-line arguments:
127 - Each space character terminates an argument. This means that two
128 spaces in a row introduce an empty-string argument.
129 - The tab character is not permitted (unless you quote it with the
130 backslash character, as described below), to avoid confusion.
131 - The newline character terminates the sequence of arguments, and will
132 also terminate a final non-empty argument. (However, a newline
133 following a space will not introduce a final empty-string argument;
134 it only terminates the argument list.)
135 - The backslash character is the escape character. It escapes
136 backslash, space, tab, and newline. The ANSI C escape sequences
137 like \n and \t are also supported. These produce argument
138 constituents; the two-character combination \n doesn't act like a
139 terminating newline. The escape sequence \NNN for exactly three
140 octal digits reads as the character whose ASCII code is NNN. As
141 above, characters produced this way are argument constituents.
142 Backslash followed by other characters is not allowed.
144 * Changes to the procedure for linking libguile with your programs
146 ** Guile now builds and installs a shared guile library, if your
147 system support shared libraries. (It still builds a static library on
148 all systems.) Guile automatically detects whether your system
149 supports shared libraries. To prevent Guile from buildisg shared
150 libraries, pass the `--disable-shared' flag to the configure script.
152 Guile takes longer to compile when it builds shared libraries, because
153 it must compile every file twice --- once to produce position-
154 independent object code, and once to produce normal object code.
156 ** The libthreads library has been merged into libguile.
158 To link a program against Guile, you now need only link against
159 -lguile and -lqt; -lthreads is no longer needed. If you are using
160 autoconf to generate configuration scripts for your application, the
161 following lines should suffice to add the appropriate libraries to
164 ### Find quickthreads and libguile.
165 AC_CHECK_LIB(qt, main)
166 AC_CHECK_LIB(guile, scm_shell)
168 * Changes to Scheme functions
170 ** Guile Scheme's special syntax for keyword objects is now optional,
171 and disabled by default.
173 The syntax variation from R4RS made it difficult to port some
174 interesting packages to Guile. The routines which accepted keyword
175 arguments (mostly in the module system) have been modified to also
176 accept symbols whose names begin with `:'.
178 To change the keyword syntax, you must first import the (ice-9 debug)
180 (use-modules (ice-9 debug))
182 Then you can enable the keyword syntax as follows:
183 (read-set! keywords 'prefix)
185 To disable keyword syntax, do this:
186 (read-set! keywords #f)
188 ** Many more primitive functions accept shared substrings as
189 arguments. In the past, these functions required normal, mutable
190 strings as arguments, although they never made use of this
193 ** The uniform array functions now operate on byte vectors. These
194 functions are `array-fill!', `serial-array-copy!', `array-copy!',
195 `serial-array-map', `array-map', `array-for-each', and
198 ** The new functions `trace' and `untrace' implement simple debugging
199 support for Scheme functions.
201 The `trace' function accepts any number of procedures as arguments,
202 and tells the Guile interpreter to display each procedure's name and
203 arguments each time the procedure is invoked. When invoked with no
204 arguments, `trace' returns the list of procedures currently being
207 The `untrace' function accepts any number of procedures as arguments,
208 and tells the Guile interpreter not to trace them any more. When
209 invoked with no arguments, `untrace' untraces all curretly traced
212 The tracing in Guile has an advantage over most other systems: we
213 don't create new procedure objects, but mark the procedure objects
214 themselves. This means that anonymous and internal procedures can be
217 ** The function `assert-repl-prompt' has been renamed to
218 `set-repl-prompt!'. It takes one argument, PROMPT.
219 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
220 - If PROMPT is a string, we use it as a prompt.
221 - If PROMPT is a procedure accepting no arguments, we call it, and
222 display the result as a prompt.
223 - Otherwise, we display "> ".
225 ** The new function `eval-string' reads Scheme expressions from a
226 string and evaluates them, returning the value of the last expression
227 in the string. If the string contains no expressions, it returns an
230 ** The new function `thunk?' returns true iff its argument is a
231 procedure of zero arguments.
233 ** `defined?' is now a builtin function, instead of syntax. This
234 means that its argument should be quoted. It returns #t iff its
235 argument is bound in the current module.
237 ** The new syntax `use-modules' allows you to add new modules to your
238 environment without re-typing a complete `define-module' form. It
239 accepts any number of module names as arguments, and imports their
240 public bindings into the current module.
242 ** The new function (module-defined? NAME MODULE) returns true iff
243 NAME, a symbol, is defined in MODULE, a module object.
245 ** The new function `builtin-bindings' creates and returns a hash
246 table containing copies of all the root module's bindings.
248 ** The new function `builtin-weak-bindings' does the same as
249 `builtin-bindings', but creates a doubly-weak hash table.
251 ** The `equal?' function now considers variable objects to be
252 equivalent if they have the same name and the same value.
254 ** The new function `command-line' returns the command-line arguments
255 given to Guile, as a list of strings.
257 When using guile as a script interpreter, `command-line' returns the
258 script's arguments; those processed by the interpreter (like `-s' or
259 `-c') are omitted. (In other words, you get the normal, expected
260 behavior.) Any application that uses scm_shell to process its
261 command-line arguments gets this behavior as well.
263 ** The new function `load-user-init' looks for a file called `.guile'
264 in the user's home directory, and loads it if it exists. This is
265 mostly for use by the code generated by scm_compile_shell_switches,
266 but we thought it might also be useful in other circumstances.
268 ** The new function `log10' returns the base-10 logarithm of its
271 ** Changes to I/O functions
273 *** The functions `read', `primitive-load', `read-and-eval!', and
274 `primitive-load-path' no longer take optional arguments controlling
275 case insensitivity and a `#' parser.
277 Case sensitivity is now controlled by a read option called
278 `case-insensitive'. The user can add new `#' syntaxes with the
279 `read-hash-extend' function (see below).
281 *** The new function `read-hash-extend' allows the user to change the
282 syntax of Guile Scheme in a somewhat controlled way.
284 (read-hash-extend CHAR PROC)
285 When parsing S-expressions, if we read a `#' character followed by
286 the character CHAR, use PROC to parse an object from the stream.
287 If PROC is #f, remove any parsing procedure registered for CHAR.
289 The reader applies PROC to two arguments: CHAR and an input port.
291 *** The new functions read-delimited and read-delimited! provide a
292 general mechanism for doing delimited input on streams.
294 (read-delimited DELIMS [PORT HANDLE-DELIM])
295 Read until we encounter one of the characters in DELIMS (a string),
296 or end-of-file. PORT is the input port to read from; it defaults to
297 the current input port. The HANDLE-DELIM parameter determines how
298 the terminating character is handled; it should be one of the
301 'trim omit delimiter from result
302 'peek leave delimiter character in input stream
303 'concat append delimiter character to returned value
304 'split return a pair: (RESULT . TERMINATOR)
306 HANDLE-DELIM defaults to 'peek.
308 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
309 A side-effecting variant of `read-delimited'.
311 The data is written into the string BUF at the indices in the
312 half-open interval [START, END); the default interval is the whole
313 string: START = 0 and END = (string-length BUF). The values of
314 START and END must specify a well-defined interval in BUF, i.e.
315 0 <= START <= END <= (string-length BUF).
317 It returns NBYTES, the number of bytes read. If the buffer filled
318 up without a delimiter character being found, it returns #f. If the
319 port is at EOF when the read starts, it returns the EOF object.
321 If an integer is returned (i.e., the read is successfully terminated
322 by reading a delimiter character), then the HANDLE-DELIM parameter
323 determines how to handle the terminating character. It is described
324 above, and defaults to 'peek.
326 (The descriptions of these functions were borrowed from the SCSH
327 manual, by Olin Shivers and Brian Carlstrom.)
329 *** The `%read-delimited!' function is the primitive used to implement
330 `read-delimited' and `read-delimited!'.
332 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
334 This returns a pair of values: (TERMINATOR . NUM-READ).
335 - TERMINATOR describes why the read was terminated. If it is a
336 character or the eof object, then that is the value that terminated
337 the read. If it is #f, the function filled the buffer without finding
338 a delimiting character.
339 - NUM-READ is the number of characters read into BUF.
341 If the read is successfully terminated by reading a delimiter
342 character, then the gobble? parameter determines what to do with the
343 terminating character. If true, the character is removed from the
344 input stream; if false, the character is left in the input stream
345 where a subsequent read operation will retrieve it. In either case,
346 the character is also the first value returned by the procedure call.
348 (The descriptions of this function was borrowed from the SCSH manual,
349 by Olin Shivers and Brian Carlstrom.)
351 *** The `read-line' and `read-line!' functions have changed; they now
352 trim the terminator by default; previously they appended it to the
353 returned string. For the old behavior, use (read-line PORT 'concat).
355 *** The functions `uniform-array-read!' and `uniform-array-write!' now
356 take new optional START and END arguments, specifying the region of
357 the array to read and write.
359 *** The `ungetc-char-ready?' function has been removed.
361 ** Changes to the Unix library and system call interface
363 *** The new fcntl function provides access to the Unix `fcntl' system
366 (fcntl PORT COMMAND VALUE)
367 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
368 Values for COMMAND are:
370 F_DUPFD duplicate a file descriptor
371 F_GETFD read the descriptor's close-on-exec flag
372 F_SETFD set the descriptor's close-on-exec flag to VALUE
373 F_GETFL read the descriptor's flags, as set on open
374 F_SETFL set the descriptor's flags, as set on open to VALUE
375 F_GETOWN return the process ID of a socket's owner, for SIGIO
376 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
377 FD_CLOEXEC not sure what this is
379 For details, see the documentation for the fcntl system call.
381 *** The arguments to `select' have changed, for compatibility with
382 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
383 expected behavior. The MILLISECONDS parameter has been changed to
384 MICROSECONDS, to more closely resemble the underlying system call.
385 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
386 corresponding return set will be the same.
388 *** The arguments to the `mknod' system call have changed. They are
391 (mknod PATH TYPE PERMS DEV)
392 Create a new file (`node') in the file system. PATH is the name of
393 the file to create. TYPE is the kind of file to create; it should
394 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
395 permission bits to give the newly created file. If TYPE is
396 'block-special or 'char-special, DEV specifies which device the
397 special file refers to; its interpretation depends on the kind of
398 special file being created.
400 *** The `fork' function has been renamed to `primitive-fork', to avoid
401 clashing with various SCSH forks.
403 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
404 and `recvfrom!'. They no longer accept a size for a second argument;
405 you must pass a string to hold the received value. They no longer
406 return the buffer. Instead, `recv' returns the length of the message
407 received, and `recvfrom' returns a pair containing the packet's length
408 and originating address.
410 *** The file descriptor datatype has been removed, as have the
411 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
412 We plan to replace these functions with a SCSH-compatible interface.
414 *** The `create' function has been removed; it's just a special case
417 *** There are new functions to break down process termination status
418 values. In the descriptions below, STATUS is a value returned by
421 (status:exit-val STATUS)
422 If the child process exited normally, this function returns the exit
423 code for the child process (i.e., the value passed to exit, or
424 returned from main). If the child process did not exit normally,
425 this function returns #f.
427 (status:stop-sig STATUS)
428 If the child process was suspended by a signal, this function
429 returns the signal that suspended the child. Otherwise, it returns
432 (status:term-sig STATUS)
433 If the child process terminated abnormally, this function returns
434 the signal that terminated the child. Otherwise, this function
437 POSIX promises that exactly one of these functions will return true on
438 a valid STATUS value.
440 These functions are compatible with SCSH.
442 *** There are new accessors and setters for the broken-out time vectors
443 returned by `localtime', `gmtime', and that ilk. They are:
445 Component Accessor Setter
446 ========================= ============ ============
447 seconds tm:sec set-tm:sec
448 minutes tm:min set-tm:min
449 hours tm:hour set-tm:hour
450 day of the month tm:mday set-tm:mday
451 month tm:mon set-tm:mon
452 year tm:year set-tm:year
453 day of the week tm:wday set-tm:wday
454 day in the year tm:yday set-tm:yday
455 daylight saving time tm:isdst set-tm:isdst
456 GMT offset, seconds tm:gmtoff set-tm:gmtoff
457 name of time zone tm:zone set-tm:zone
459 *** There are new accessors for the vectors returned by `uname',
460 describing the host system:
463 ============================================== ================
464 name of the operating system implementation utsname:sysname
465 network name of this machine utsname:nodename
466 release level of the operating system utsname:release
467 version level of the operating system utsname:version
468 machine hardware platform utsname:machine
470 *** There are new accessors for the vectors returned by `getpw',
471 `getpwnam', `getpwuid', and `getpwent', describing entries from the
472 system's user database:
475 ====================== =================
476 user name passwd:name
477 user password passwd:passwd
480 real name passwd:gecos
481 home directory passwd:dir
482 shell program passwd:shell
484 *** There are new accessors for the vectors returned by `getgr',
485 `getgrnam', `getgrgid', and `getgrent', describing entries from the
486 system's group database:
489 ======================= ============
490 group name group:name
491 group password group:passwd
493 group members group:mem
495 *** There are new accessors for the vectors returned by `gethost',
496 `gethostbyaddr', `gethostbyname', and `gethostent', describing
500 ========================= ===============
501 official name of host hostent:name
502 alias list hostent:aliases
503 host address type hostent:addrtype
504 length of address hostent:length
505 list of addresses hostent:addr-list
507 *** There are new accessors for the vectors returned by `getnet',
508 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
512 ========================= ===============
513 official name of net netent:name
514 alias list netent:aliases
515 net number type netent:addrtype
516 net number netent:net
518 *** There are new accessors for the vectors returned by `getproto',
519 `getprotobyname', `getprotobynumber', and `getprotoent', describing
523 ========================= ===============
524 official protocol name protoent:name
525 alias list protoent:aliases
526 protocol number protoent:proto
528 *** There are new accessors for the vectors returned by `getserv',
529 `getservbyname', `getservbyport', and `getservent', describing
533 ========================= ===============
534 official service name servent:name
535 alias list servent:aliases
536 port number servent:port
537 protocol to use servent:proto
539 *** There are new accessors for the sockaddr structures returned by
540 `accept', `getsockname', `getpeername', `recvfrom!':
543 ======================================== ===============
544 address format (`family') sockaddr:fam
545 path, for file domain addresses sockaddr:path
546 address, for internet domain addresses sockaddr:addr
547 TCP or UDP port, for internet sockaddr:port
549 *** The `getpwent', `getgrent', `gethostent', `getnetent',
550 `getprotoent', and `getservent' functions now return #f at the end of
551 the user database. (They used to throw an exception.)
553 Note that calling MUMBLEent function is equivalent to calling the
554 corresponding MUMBLE function with no arguments.
556 *** The `setpwent', `setgrent', `sethostent', `setnetent',
557 `setprotoent', and `setservent' routines now take no arguments.
559 *** The `gethost', `getproto', `getnet', and `getserv' functions now
560 provide more useful information when they throw an exception.
562 *** The `lnaof' function has been renamed to `inet-lnaof'.
564 *** Guile now claims to have the `current-time' feature.
566 *** The `mktime' function now takes an optional second argument ZONE,
567 giving the time zone to use for the conversion. ZONE should be a
568 string, in the same format as expected for the "TZ" environment variable.
570 *** The `strptime' function now returns a pair (TIME . COUNT), where
571 TIME is the parsed time as a vector, and COUNT is the number of
572 characters from the string left unparsed. This function used to
573 return the remaining characters as a string.
575 *** The `gettimeofday' function has replaced the old `time+ticks' function.
576 The return value is now (SECONDS . MICROSECONDS); the fractional
577 component is no longer expressed in "ticks".
579 *** The `ticks/sec' constant has been removed, in light of the above change.
581 * Changes to the gh_ interface
583 ** gh_eval_str() now returns an SCM object which is the result of the
586 ** gh_scm2str() now copies the Scheme data to a caller-provided C
589 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
590 and returns the array
592 ** gh_scm2str0() is gone: there is no need to distinguish
593 null-terminated from non-null-terminated, since gh_scm2newstr() allows
594 the user to interpret the data both ways.
596 * Changes to the scm_ interface
598 ** The new function scm_symbol_value0 provides an easy way to get a
599 symbol's value from C code:
601 SCM scm_symbol_value0 (char *NAME)
602 Return the value of the symbol named by the null-terminated string
603 NAME in the current module. If the symbol named NAME is unbound in
604 the current module, return SCM_UNDEFINED.
606 ** The new function scm_sysintern0 creates new top-level variables,
607 without assigning them a value.
609 SCM scm_sysintern0 (char *NAME)
610 Create a new Scheme top-level variable named NAME. NAME is a
611 null-terminated string. Return the variable's value cell.
613 ** The function scm_internal_catch is the guts of catch. It handles
614 all the mechanics of setting up a catch target, invoking the catch
615 body, and perhaps invoking the handler if the body does a throw.
617 The function is designed to be usable from C code, but is general
618 enough to implement all the semantics Guile Scheme expects from throw.
620 TAG is the catch tag. Typically, this is a symbol, but this function
621 doesn't actually care about that.
623 BODY is a pointer to a C function which runs the body of the catch;
624 this is the code you can throw from. We call it like this:
625 BODY (BODY_DATA, JMPBUF)
627 BODY_DATA is just the BODY_DATA argument we received; we pass it
628 through to BODY as its first argument. The caller can make
629 BODY_DATA point to anything useful that BODY might need.
630 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
631 which we have just created and initialized.
633 HANDLER is a pointer to a C function to deal with a throw to TAG,
634 should one occur. We call it like this:
635 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
637 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
638 same idea as BODY_DATA above.
639 THROWN_TAG is the tag that the user threw to; usually this is
640 TAG, but it could be something else if TAG was #t (i.e., a
641 catch-all), or the user threw to a jmpbuf.
642 THROW_ARGS is the list of arguments the user passed to the THROW
645 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
646 is just a pointer we pass through to HANDLER. We don't actually
647 use either of those pointers otherwise ourselves. The idea is
648 that, if our caller wants to communicate something to BODY or
649 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
650 HANDLER can then use. Think of it as a way to make BODY and
651 HANDLER closures, not just functions; MUMBLE_DATA points to the
654 Of course, it's up to the caller to make sure that any data a
655 MUMBLE_DATA needs is protected from GC. A common way to do this is
656 to make MUMBLE_DATA a pointer to data stored in an automatic
657 structure variable; since the collector must scan the stack for
658 references anyway, this assures that any references in MUMBLE_DATA
661 ** The new function scm_internal_lazy_catch is exactly like
662 scm_internal_catch, except:
664 - It does not unwind the stack (this is the major difference).
665 - If handler returns, its value is returned from the throw.
666 - BODY always receives #f as its JMPBUF argument (since there's no
667 jmpbuf associated with a lazy catch, because we don't unwind the
670 ** scm_body_thunk is a new body function you can pass to
671 scm_internal_catch if you want the body to be like Scheme's `catch'
672 --- a thunk, or a function of one argument if the tag is #f.
674 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
675 contains the Scheme procedure to invoke as the body, and the tag
676 we're catching. If the tag is #f, then we pass JMPBUF (created by
677 scm_internal_catch) to the body procedure; otherwise, the body gets
680 ** scm_handle_by_proc is a new handler function you can pass to
681 scm_internal_catch if you want the handler to act like Scheme's catch
682 --- call a procedure with the tag and the throw arguments.
684 If the user does a throw to this catch, this function runs a handler
685 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
686 variable holding the Scheme procedure object to invoke. It ought to
687 be a pointer to an automatic variable (i.e., one living on the stack),
688 or the procedure object should be otherwise protected from GC.
690 ** scm_handle_by_message is a new handler function to use with
691 `scm_internal_catch' if you want Guile to print a message and die.
692 It's useful for dealing with throws to uncaught keys at the top level.
694 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
695 message header to print; if zero, we use "guile" instead. That
696 text is followed by a colon, then the message described by ARGS.
698 ** The return type of scm_boot_guile is now void; the function does
699 not return a value, and indeed, never returns at all.
701 ** The new function scm_shell makes it easy for user applications to
702 process command-line arguments in a way that is compatible with the
703 stand-alone guile interpreter (which is in turn compatible with SCSH,
706 To use the scm_shell function, first initialize any guile modules
707 linked into your application, and then call scm_shell with the values
708 of ARGC and ARGV your `main' function received. scm_shell will adding
709 any SCSH-style meta-arguments from the top of the script file to the
710 argument vector, and then process the command-line arguments. This
711 generally means loading a script file or starting up an interactive
712 command interpreter. For details, see "Changes to the stand-alone
715 ** The new functions scm_get_meta_args and scm_count_argv help you
716 implement the SCSH-style meta-argument, `\'.
718 char **scm_get_meta_args (int ARGC, char **ARGV)
719 If the second element of ARGV is a string consisting of a single
720 backslash character (i.e. "\\" in Scheme notation), open the file
721 named by the following argument, parse arguments from it, and return
722 the spliced command line. The returned array is terminated by a
725 For details of argument parsing, see above, under "guile now accepts
726 command-line arguments compatible with SCSH..."
728 int scm_count_argv (char **ARGV)
729 Count the arguments in ARGV, assuming it is terminated by a null
732 For an example of how these functions might be used, see the source
733 code for the function scm_shell in libguile/script.c.
735 You will usually want to use scm_shell instead of calling this
738 ** The new function scm_compile_shell_switches turns an array of
739 command-line arguments into Scheme code to carry out the actions they
740 describe. Given ARGC and ARGV, it returns a Scheme expression to
741 evaluate, and calls scm_set_program_arguments to make any remaining
742 command-line arguments available to the Scheme code. For example,
743 given the following arguments:
745 -e main -s ekko a speckled gecko
747 scm_set_program_arguments will return the following expression:
749 (begin (load "ekko") (main (command-line)) (quit))
751 You will usually want to use scm_shell instead of calling this
754 ** The function scm_shell_usage prints a usage message appropriate for
755 an interpreter that uses scm_compile_shell_switches to handle its
756 command-line arguments.
758 void scm_shell_usage (int FATAL, char *MESSAGE)
759 Print a usage message to the standard error output. If MESSAGE is
760 non-zero, write it before the usage message, followed by a newline.
761 If FATAL is non-zero, exit the process, using FATAL as the
762 termination status. (If you want to be compatible with Guile,
763 always use 1 as the exit status when terminating due to command-line
766 You will usually want to use scm_shell instead of calling this
769 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
770 expressions. It used to return SCM_EOL. Earth-shattering.
772 ** The macros for declaring scheme objects in C code have been
773 rearranged slightly. They are now:
775 SCM_SYMBOL (C_NAME, SCHEME_NAME)
776 Declare a static SCM variable named C_NAME, and initialize it to
777 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
778 be a C identifier, and SCHEME_NAME should be a C string.
780 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
781 Just like SCM_SYMBOL, but make C_NAME globally visible.
783 SCM_VCELL (C_NAME, SCHEME_NAME)
784 Create a global variable at the Scheme level named SCHEME_NAME.
785 Declare a static SCM variable named C_NAME, and initialize it to
786 point to the Scheme variable's value cell.
788 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
789 Just like SCM_VCELL, but make C_NAME globally visible.
791 The `guile-snarf' script writes initialization code for these macros
792 to its standard output, given C source code as input.
794 The SCM_GLOBAL macro is gone.
796 ** The scm_read_line and scm_read_line_x functions have been replaced
797 by Scheme code based on the %read-delimited! procedure (known to C
798 code as scm_read_delimited_x). See its description above for more
801 ** The function scm_sys_open has been renamed to scm_open. It now
802 returns a port instead of an FD object.
804 * The dynamic linking support has changed. For more information, see
805 libguile/DYNAMIC-LINKING.
810 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
813 * Changes to the 'guile' program:
815 ** Guile now loads some new files when it starts up. Guile first
816 searches the load path for init.scm, and loads it if found. Then, if
817 Guile is not being used to execute a script, and the user's home
818 directory contains a file named `.guile', Guile loads that.
820 ** You can now use Guile as a shell script interpreter.
822 To paraphrase the SCSH manual:
824 When Unix tries to execute an executable file whose first two
825 characters are the `#!', it treats the file not as machine code to
826 be directly executed by the native processor, but as source code
827 to be executed by some interpreter. The interpreter to use is
828 specified immediately after the #! sequence on the first line of
829 the source file. The kernel reads in the name of the interpreter,
830 and executes that instead. It passes the interpreter the source
831 filename as its first argument, with the original arguments
832 following. Consult the Unix man page for the `exec' system call
833 for more information.
835 Now you can use Guile as an interpreter, using a mechanism which is a
836 compatible subset of that provided by SCSH.
838 Guile now recognizes a '-s' command line switch, whose argument is the
839 name of a file of Scheme code to load. It also treats the two
840 characters `#!' as the start of a comment, terminated by `!#'. Thus,
841 to make a file of Scheme code directly executable by Unix, insert the
842 following two lines at the top of the file:
844 #!/usr/local/bin/guile -s
847 Guile treats the argument of the `-s' command-line switch as the name
848 of a file of Scheme code to load, and treats the sequence `#!' as the
849 start of a block comment, terminated by `!#'.
851 For example, here's a version of 'echo' written in Scheme:
853 #!/usr/local/bin/guile -s
855 (let loop ((args (cdr (program-arguments))))
859 (if (pair? (cdr args))
864 Why does `#!' start a block comment terminated by `!#', instead of the
865 end of the line? That is the notation SCSH uses, and although we
866 don't yet support the other SCSH features that motivate that choice,
867 we would like to be backward-compatible with any existing Guile
868 scripts once we do. Furthermore, if the path to Guile on your system
869 is too long for your kernel, you can start the script with this
873 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
876 Note that some very old Unix systems don't support the `#!' syntax.
879 ** You can now run Guile without installing it.
881 Previous versions of the interactive Guile interpreter (`guile')
882 couldn't start up unless Guile's Scheme library had been installed;
883 they used the value of the environment variable `SCHEME_LOAD_PATH'
884 later on in the startup process, but not to find the startup code
885 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
888 To run Guile without installing it, build it in the normal way, and
889 then set the environment variable `SCHEME_LOAD_PATH' to a
890 colon-separated list of directories, including the top-level directory
891 of the Guile sources. For example, if you unpacked Guile so that the
892 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
895 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
898 ** Guile's read-eval-print loop no longer prints #<unspecified>
899 results. If the user wants to see this, she can evaluate the
900 expression (assert-repl-print-unspecified #t), perhaps in her startup
903 ** Guile no longer shows backtraces by default when an error occurs;
904 however, it does display a message saying how to get one, and how to
905 request that they be displayed by default. After an error, evaluate
907 to see a backtrace, and
908 (debug-enable 'backtrace)
909 to see them by default.
913 * Changes to Guile Scheme:
915 ** Guile now distinguishes between #f and the empty list.
917 This is for compatibility with the IEEE standard, the (possibly)
918 upcoming Revised^5 Report on Scheme, and many extant Scheme
921 Guile used to have #f and '() denote the same object, to make Scheme's
922 type system more compatible with Emacs Lisp's. However, the change
923 caused too much trouble for Scheme programmers, and we found another
924 way to reconcile Emacs Lisp with Scheme that didn't require this.
927 ** Guile's delq, delv, delete functions, and their destructive
928 counterparts, delq!, delv!, and delete!, now remove all matching
929 elements from the list, not just the first. This matches the behavior
930 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
931 functions which inspired them.
933 I recognize that this change may break code in subtle ways, but it
934 seems best to make the change before the FSF's first Guile release,
938 ** The compiled-library-path function has been deleted from libguile.
940 ** The facilities for loading Scheme source files have changed.
942 *** The variable %load-path now tells Guile which directories to search
943 for Scheme code. Its value is a list of strings, each of which names
946 *** The variable %load-extensions now tells Guile which extensions to
947 try appending to a filename when searching the load path. Its value
948 is a list of strings. Its default value is ("" ".scm").
950 *** (%search-load-path FILENAME) searches the directories listed in the
951 value of the %load-path variable for a Scheme file named FILENAME,
952 with all the extensions listed in %load-extensions. If it finds a
953 match, then it returns its full filename. If FILENAME is absolute, it
954 returns it unchanged. Otherwise, it returns #f.
956 %search-load-path will not return matches that refer to directories.
958 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
959 uses %seach-load-path to find a file named FILENAME, and loads it if
960 it finds it. If it can't read FILENAME for any reason, it throws an
963 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
966 *** load uses the same searching semantics as primitive-load.
968 *** The functions %try-load, try-load-with-path, %load, load-with-path,
969 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
970 path, and load-module-with-path have been deleted. The functions
971 above should serve their purposes.
973 *** If the value of the variable %load-hook is a procedure,
974 `primitive-load' applies its value to the name of the file being
975 loaded (without the load path directory name prepended). If its value
976 is #f, it is ignored. Otherwise, an error occurs.
978 This is mostly useful for printing load notification messages.
981 ** The function `eval!' is no longer accessible from the scheme level.
982 We can't allow operations which introduce glocs into the scheme level,
983 because Guile's type system can't handle these as data. Use `eval' or
984 `read-and-eval!' (see below) as replacement.
986 ** The new function read-and-eval! reads an expression from PORT,
987 evaluates it, and returns the result. This is more efficient than
988 simply calling `read' and `eval', since it is not necessary to make a
989 copy of the expression for the evaluator to munge.
991 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
992 for the `read' function.
995 ** The function `int?' has been removed; its definition was identical
996 to that of `integer?'.
998 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
999 use the R4RS names for these functions.
1001 ** The function object-properties no longer returns the hash handle;
1002 it simply returns the object's property list.
1004 ** Many functions have been changed to throw errors, instead of
1005 returning #f on failure. The point of providing exception handling in
1006 the language is to simplify the logic of user code, but this is less
1007 useful if Guile's primitives don't throw exceptions.
1009 ** The function `fileno' has been renamed from `%fileno'.
1011 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
1014 * Changes to Guile's C interface:
1016 ** The library's initialization procedure has been simplified.
1017 scm_boot_guile now has the prototype:
1019 void scm_boot_guile (int ARGC,
1021 void (*main_func) (),
1024 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
1025 MAIN_FUNC should do all the work of the program (initializing other
1026 packages, reading user input, etc.) before returning. When MAIN_FUNC
1027 returns, call exit (0); this function never returns. If you want some
1028 other exit value, MAIN_FUNC may call exit itself.
1030 scm_boot_guile arranges for program-arguments to return the strings
1031 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
1032 scm_set_program_arguments with the final list, so Scheme code will
1033 know which arguments have been processed.
1035 scm_boot_guile establishes a catch-all catch handler which prints an
1036 error message and exits the process. This means that Guile exits in a
1037 coherent way when system errors occur and the user isn't prepared to
1038 handle it. If the user doesn't like this behavior, they can establish
1039 their own universal catcher in MAIN_FUNC to shadow this one.
1041 Why must the caller do all the real work from MAIN_FUNC? The garbage
1042 collector assumes that all local variables of type SCM will be above
1043 scm_boot_guile's stack frame on the stack. If you try to manipulate
1044 SCM values after this function returns, it's the luck of the draw
1045 whether the GC will be able to find the objects you allocate. So,
1046 scm_boot_guile function exits, rather than returning, to discourage
1047 people from making that mistake.
1049 The IN, OUT, and ERR arguments were removed; there are other
1050 convenient ways to override these when desired.
1052 The RESULT argument was deleted; this function should never return.
1054 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
1058 ** Guile's header files should no longer conflict with your system's
1061 In order to compile code which #included <libguile.h>, previous
1062 versions of Guile required you to add a directory containing all the
1063 Guile header files to your #include path. This was a problem, since
1064 Guile's header files have names which conflict with many systems'
1067 Now only <libguile.h> need appear in your #include path; you must
1068 refer to all Guile's other header files as <libguile/mumble.h>.
1069 Guile's installation procedure puts libguile.h in $(includedir), and
1070 the rest in $(includedir)/libguile.
1073 ** Two new C functions, scm_protect_object and scm_unprotect_object,
1074 have been added to the Guile library.
1076 scm_protect_object (OBJ) protects OBJ from the garbage collector.
1077 OBJ will not be freed, even if all other references are dropped,
1078 until someone does scm_unprotect_object (OBJ). Both functions
1081 Note that calls to scm_protect_object do not nest. You can call
1082 scm_protect_object any number of times on a given object, and the
1083 next call to scm_unprotect_object will unprotect it completely.
1085 Basically, scm_protect_object and scm_unprotect_object just
1086 maintain a list of references to things. Since the GC knows about
1087 this list, all objects it mentions stay alive. scm_protect_object
1088 adds its argument to the list; scm_unprotect_object remove its
1089 argument from the list.
1092 ** scm_eval_0str now returns the value of the last expression
1095 ** The new function scm_read_0str reads an s-expression from a
1096 null-terminated string, and returns it.
1098 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
1099 to a Scheme port object.
1101 ** The new function `scm_set_program_arguments' allows C code to set
1102 the value teruturned by the Scheme `program-arguments' function.
1107 * Guile no longer includes sophisticated Tcl/Tk support.
1109 The old Tcl/Tk support was unsatisfying to us, because it required the
1110 user to link against the Tcl library, as well as Tk and Guile. The
1111 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
1112 referring to widgets by names, rather than exporting widgets to Scheme
1113 code as a special datatype.
1115 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
1116 maintainers described some very interesting changes in progress to the
1117 Tcl/Tk internals, which would facilitate clean interfaces between lone
1118 Tk and other interpreters --- even for garbage-collected languages
1119 like Scheme. They expected the new Tk to be publicly available in the
1122 Since it seems that Guile might soon have a new, cleaner interface to
1123 lone Tk, and that the old Guile/Tk glue code would probably need to be
1124 completely rewritten, we (Jim Blandy and Richard Stallman) have
1125 decided not to support the old code. We'll spend the time instead on
1126 a good interface to the newer Tk, as soon as it is available.
1128 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
1131 Copyright information:
1133 Copyright (C) 1996,1997 Free Software Foundation, Inc.
1135 Permission is granted to anyone to make or distribute verbatim copies
1136 of this document as received, in any medium, provided that the
1137 copyright notice and this permission notice are preserved,
1138 thus giving the recipient permission to redistribute in turn.
1140 Permission is granted to distribute modified versions
1141 of this document, or of portions of it,
1142 under the above conditions, provided also that they
1143 carry prominent notices stating who last changed them.
1148 paragraph-separate: "[
\f]*$"