3 Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001 Free software Foundation, Inc.
5 Permission is granted to anyone to make or distribute verbatim copies
6 of this document as received, in any medium, provided that the
7 copyright notice and permission notice are preserved,
8 and that the distributor grants the recipient permission
9 for further redistribution as permitted by this notice.
11 Permission is granted to distribute modified versions
12 of this document, or of portions of it,
13 under the above conditions, provided also that they
14 carry prominent notices stating who last changed them,
15 and that any new or changed statements about the activities
16 of the Free Software Foundation are approved by the Foundation.
19 What to Hack =========================================================
21 You can hack whatever you want, thank GNU.
23 However, to see what others have indicated as their interest (and avoid
24 potential wasteful duplication of effort), see file TODO. Note that
25 the version you find may be out of date; a CVS checkout is recommended
26 (see also file SNAPSHOTS).
28 It's also a good idea to join the guile-devel@gnu.org mailing list.
29 See http://www.gnu.org/software/guile/mail/mail.html for more info.
32 Hacking It Yourself ==================================================
34 As distributed, Guile needs only an ANSI C compiler and a Unix system
35 to compile. However, Guile's makefiles, configuration scripts, and a
36 few other files are automatically generated, not written by hand. If
37 you want to make changes to the system (which we encourage!) you will
38 find it helpful to have the tools we use to develop Guile. They
41 Autoconf 2.50 --- a system for automatically generating `configure'
42 scripts from templates which list the non-portable features a
43 program would like to use. Available in
44 "ftp://ftp.gnu.org/pub/gnu/autoconf"
46 Automake 1.4-p4 --- a system for automatically generating Makefiles that
47 conform to the (rather Byzantine) GNU coding standards. The
48 nice thing is that it takes care of hairy targets like 'make
49 dist' and 'make distclean', and automatically generates
50 Makefile dependencies. Automake is available in
51 "ftp://ftp.gnu.org/pub/gnu/automake"
53 Before using automake, you may need to copy `threads.m4' and
54 `guile.m4' from the top directory of the Guile core disty to
55 `/usr/local/share/aclocal'.
57 libtool 1.4 --- a system for managing the zillion hairy options needed
58 on various systems to produce shared libraries. Available in
59 "ftp://ftp.gnu.org/pub/gnu/libtool"
61 To avoid reported tricky errors during the Guile build: After
62 unpacking the libtool distribution, use autoconf (2.50) to
63 regenerate its ./configure script. Then build and install as
66 flex 2.5.4 (or newer) --- a scanner generator. earlier versions will
67 most probably work too.
69 You are lost in a little maze of automatically generated files, all
73 Sample GDB Initialization File=========================================
75 Here is a sample .gdbinit posted by Bill Schottstaedt (modified to
76 use `set' instead of `call' in some places):
83 Executes (object->string arg)
88 call gdb_eval(gdb_result)
89 set gdb_print(gdb_result)
93 Executes (print (eval (read arg))): ge "(+ 1 2)" => 3
97 call g_help(scm_str2symbol($arg0), 20)
102 Prints help string for arg: gh "enved-target"
107 so in gdb if you see something useless like:
109 #32 0x081ae8f4 in scm_primitive_load (filename=1112137128) at load.c:129
111 You can get the file name with gp:
114 $1 = 0x40853fac "\"/home/bil/test/share/guile/1.5.0/ice-9/session.scm\""
117 Contributing Your Changes ============================================
119 - If you have put together a change that meets the coding standards
120 described below, we encourage you to submit it to Guile. The best
121 place to post it is guile-devel@gnu.org. Please don't send it
122 directly to me; I often don't have time to look things over. If you
123 have tested your change, then you don't need to be shy.
125 - Please submit patches using either context or unified diffs (diff -c
126 or diff -u). Don't include a patch for ChangeLog; such patches don't
127 apply cleanly, since we've probably changed the top of ChangeLog too.
128 Instead, provide the unaltered text at the top of your patch.
130 - For proper credit, also make sure you update the AUTHORS file
131 (for new files for which you've assigned copyright to the FSF), or
132 the THANKS file (for everything else).
134 Please don't include patches for generated files like configure,
135 aclocal.m4, or any Makefile.in. Such patches are often large, and
136 we're just going to regenerate those files anyway.
139 CVS conventions ======================================================
141 - We use CVS to manage the Guile sources. The repository lives on
142 subversions.gnu.org, in /cvs; you will need an
143 account on that machine to access the repository. Also, for security
144 reasons, subversions presently only supports CVS connections via the SSH
145 protocol, so you must first install the SSH client. Then, you should
146 set your CVS_RSH environment variable to ssh, and use the following as
149 :ext:USER@subversions.gnu.org:/cvs
151 Either set your CVSROOT environment variable to that, or give it as
152 the value of the global -d option to CVS when you check out a working
155 For more information on SSH, see http://www.cs.hut.fi/ssh.
157 The Guile sources live in several modules:
159 - guile-core --- the interpreter, QuickThreads, and ice-9
160 - guile-tcltk --- the Guile/Tk interface
161 - guile-tk --- the new Guile/Tk interface, based on STk's modified Tk
162 - guile-rgx-ctax --- the Guile/Rx interface, and the ctax implementation
163 - guile-scsh --- the port of SCSH to guile, talk to Gary Houston
164 - guile-www --- A Guile module for making HTTP requests.
165 - guile-statprof --- an experimental statistical profiler.
167 There is a mailing list for CVS commit messages; see README for details.
169 - The guile-core tree is now versioned similarly to the Linux kernel.
170 Guile now always uses three numbers to represent the version,
171 i.e. "1.6.5". The first number, 1, is the major version number, the
172 second number, 6, is the minor version number, and the third number,
173 5, is the micro version number. Changes in major version number
174 indicate major changes in Guile.
176 Minor version numbers that are even denote stable releases, and odd
177 minor version numbers denote development versions (which may be
178 unstable). The micro version number indicates a minor sub-revision of
179 a given MAJOR.MINOR release.
181 - A default CVS checkout will get the current unstable development
182 tree. However, for each stable release, a CVS branch is created so
183 that release (and ongoing maintenance) of the stable version can
184 proceed independent of the development of the next unstable version.
185 To check out a particular stable branch, you just need to specify "-r
186 branch_release-X-Y" to your CVS checkout command (or to any update).
187 For example, if you wanted to check out the 1.6 stable branch, you
188 would specify "-r branch_release-1-6".
190 So, for example, during a normal development cycle, work will proceed
191 on an unstable version, say 1.5.X, until it is decided that it's time
192 for a stable release. At that point, a branch named
193 branch_release-1-6 will be created, and the version numbers on the
194 HEAD of the CVS tree (the trunk, i.e. what you get by default), will
195 be changed to reflect the new unstable version 1.7.X. Then unstable
196 development will proceed on the unstable version, while the stable
197 1.5.X branch is fixed up for the eventual 1.6.0 release.
199 Anytime you want to yank an existing checked out tree to the stable
200 branch, you can run a command like this:
202 cvs -z3 update -r branch_release-1-6 -Pd
204 This will yank the working directory over on to the stable release
205 branch. Note that this directory will track that branch from then on
206 unless you do something to yank it back to the main (unstable) trunk.
208 To go back to the unstable branch, you can use
210 cvs -z3 update -A -Pd
212 Note that in either case, you should probably make sure you've
213 commited or removed all local changes before running the commands or
214 you're likely to have some unexpected results.
216 Finally note that one approach, should you need to work on both
217 branches, is to keep two trees checked out, one stable, the other
218 unstable and you can work in whichever is appropriate.
220 To save some initial bandwidth, you can check out either the stable
221 tree or the unstable tree, and then do something like this:
223 cp -a core-unstable core-1.5
225 cvs -z3 update -r branch_release-1-6 -Pd
227 - The stable and unstable CVS trees are distinct, and no changes will
228 automatically propagate between them. If you make changes that need
229 to show up both places, you'll need to apply the changes both places.
230 You *might* be able to do this with a cvs command, but often you'll
231 probably need to apply the changes by hand or risk migrating
232 superfluous modifications between the two versions. This is
233 particularly important when moving a change from the unstable branch
234 to the stable branch.
236 - In general, please don't be adventurous with the stable branch. We
237 mostly want bugfixes, documentation improvements, build improvements,
238 etc., though exceptions will doubtless exist.
240 - There are a few CVS tagging conventions which follow the Scheme
241 convention that dashes are used to separate words within a single
242 symbol, and so dashes bind more tightly than underscores. This means
243 that foo-bar_baz-bax indicates that foo-bar is somehow separate from
244 baz-bax. The conventions are as follows:
248 anytime just before you create a branch it's a good
249 idea to create a normal tag so that you can refer to the branch point
250 on the main trunk as well as on the branch. So please use a tag of
253 branch-root-release-1-X
255 or more generally, for other non-release branches:
261 for the branch tag itself please use
265 or more generally, for other non-release branches:
271 Whenever you're merging a branch back into the trunk (or into another
272 branch repeatedly) you need to tag the branch each time you merge. If
273 you don't do that, you won't be able to merge repeatedly without
274 possibly tedious conflicts. For those tags, we suggest:
276 branch-merge_SOME-FOO_to_SOME-BAR_1
277 branch-merge_SOME-FOO_to_SOME-BAR_2
280 As an example, SOME-BAR might be trunk, or even perhaps another branch
281 like branch-mvo-super-fixes :>
283 More mundanely, you might have
285 branch-merge_release-1-6_to_trunk_1
287 (Merging the stable branch to the trunk like this
288 will probably be much more common, when it happens, than the
289 reverse for the reasons mentioned above.
293 When releasing a new version of guile, please use:
301 - If you hack on a stable branch, please apply any relevant patches or
302 fixes to the current unstable version (the main CVS trunk) as well.
303 Similarly, please back-port any important fixes to the unstable CVS
304 tree to the current stable branch.
306 - We check Makefile.am and configure.in files into CVS, but the
307 "autogen.sh" script must be run from the top-level to generate the
308 actual "configure" script that then must be run to create the various
309 Makefile-s to build guile. The general rule is that you should be able
310 to check out a working directory of Guile from CVS, and then type
311 "./autogen.sh", then "configure", and finally "make". No
312 automatically generated files should be checked into the CVS
315 - The .cvsignore file is contained in the repository, to provide a
316 reasonable list of auto-generated files that should not be checked in.
317 This, however, prohibits one from having local additions to the
318 .cvsignore file (yes, you can modify it and never check it in, but
319 that doesn't seem to be a good solution to me). To get around this
320 problem, you might want to patch your cvs program so that it uses a
321 .cvsignore-local file (say) instead of the one from the repository. A
322 patch for this can be found at the very end of this file.
324 - (Automake 1.4 only) Be sure to run automake at the top of the tree
325 with no arguments. Do not use `automake Makefile' to regenerate
326 specific Makefile.in files, and do not trust the Makefile rules to
327 rebuild them when they are out of date. Automake 1.4 will add
328 extraneous rules to the top-level Makefile if you specify specific
329 Makefiles to rebuild on the command line. Running the command
330 `autoreconf --force' should take care of everything correctly.
332 - Make sure your changes compile and work, at least on your own
333 machine, before checking them into the main branch of the Guile
334 repository. If you really need to check in untested changes, make a
337 - Include each log entry in both the ChangeLog and in the CVS logs.
338 If you're using Emacs, the pcl-cvs interface to CVS has features to
339 make this easier; it checks the ChangeLog, and generates good default
340 CVS log entries from that.
343 Coding standards =====================================================
345 - Before contributing larger amounts of code to Guile, please read the
346 documents in `guile-core/devel/policy' in the CVS source tree.
348 - As for any part of Project GNU, changes to Guile should follow the
349 GNU coding standards. The standards are available via anonymous FTP
350 from prep.ai.mit.edu, as /pub/gnu/standards/standards.texi and
353 - The Guile tree should compile without warnings under the following
354 GCC switches, which are the default in the current configure script:
356 -O2 -Wall -Wpointer-arith -Wmissing-prototypes
358 To make sure of this, you can use the --enable-error-on-warning option
359 to configure. This option will make GCC fail if it hits a warning.
361 Note that the warnings generated vary from one version of GCC to the
362 next, and from one architecture to the next (apparently). To provide
363 a concrete common standard, Guile should compile without warnings from
364 GCC 2.7.2.3 in a Red Hat 5.2 i386 Linux machine. Furthermore, each
365 developer should pursue any additional warnings noted by on their
366 compiler. This means that people using more stringent compilers will
367 have more work to do, and assures that everyone won't switch to the
368 most lenient compiler they can find. :)
370 Note also that EGCS (as of November 3 1998) doesn't handle the
371 `noreturn' attribute properly, so it doesn't understand that functions
372 like scm_error won't return. This may lead to some silly warnings
373 about uninitialized variables. You should look into these warnings to
374 make sure they are indeed spurious, but you needn't correct warnings
375 caused by this EGCS bug.
377 - If you add code which uses functions or other features that are not
378 entirely portable, please make sure the rest of Guile will still
379 function properly on systems where they are missing. This usually
380 entails adding a test to configure.in, and then adding #ifdefs to your
381 code to disable it if the system's features are missing.
383 - The normal way of removing a function, macro or variable is to mark
384 it as "deprecated", keep it for a while, and remove it in a later
385 release. If a function or macro is marked as "deprecated" it
386 indicates that people shouldn't use it in new programs, and should try
387 to remove it in old. Make sure that an alternative exists unless it
388 is our purpose to remove functionality. Don't deprecate definitions
389 if it is unclear when they will be removed. (This is to ensure that a
390 valid way of implementing some functionality always exists.)
392 When deprecating a definition, always follow this procedure:
394 1. Mark the definition using
396 #if (SCM_DEBUG_DEPRECATED == 0)
400 or, for Scheme code, wrap it using
405 2. Make the deprecated code issue a warning when it is used, by using
406 scm_c_issue_deprecation_warning (in C) or issue-deprecation-warning
409 3. Write a comment at the definition explaining how a programmer can
410 manage without the deprecated definition.
412 4. Add an entry that the definition has been deprecated in NEWS and
413 explain what do do instead.
415 5. In file TODO, there is a list of releases with reminders about what
416 to do at each release. Add a reminder about the removal of the
417 deprecated defintion at the appropriate release.
419 - When you make a user-visible change (i.e. one that should be
420 documented, and appear in NEWS, put an asterisk in column zero of the
421 start of the ChangeLog entry, like so:
423 Sat Aug 3 01:27:14 1996 Gary Houston <ghouston@actrix.gen.nz>
425 * * fports.c (scm_open_file): don't return #f, throw error.
427 When you've written a NEWS entry and updated the documentation, go
428 ahead and remove the asterisk. The asterisks are used to find and
429 document changes that haven't been dealt with before a release.
431 - Please write log entries for functions written in C under the
432 functions' C names, and write log entries for functions written in
433 Scheme under the functions' Scheme names. Please don't do this:
435 * procs.c, procs.h (procedure-documentation): Moved from eval.c.
437 Entries like this make it harder to search the ChangeLogs, because you
438 can never tell which name the entry will refer to. Instead, write this:
440 * procs.c, procs.h (scm_procedure_documentation): Moved from eval.c.
442 Changes like adding this line are special:
444 SCM_PROC (s_map_in_order, "map-in-order", 2, 0, 1, scm_map);
446 Since the change here is about the name itself --- we're adding a new
447 alias for scm_map that guarantees the order in which we process list
448 elements, but we're not changing scm_map at all --- it's appropriate
449 to use the Scheme name in the log entry.
451 - There's no need to keep a change log for a ChangeLog file. For any
452 other kind of file (including documentation, since our documentation
453 is indeed precisely engineered -- we surpass GNU standards here), add
454 an appropriate ChangeLog entry when you change it. Simple!
456 - Make sure you have papers from people before integrating their
457 changes or contributions. This is very frustrating, but very
458 important to do right. From maintain.texi, "Information for
459 Maintainers of GNU Software":
461 When incorporating changes from other people, make sure to follow the
462 correct procedures. Doing this ensures that the FSF has the legal
463 right to distribute and defend GNU software.
465 For the sake of registering the copyright on later versions ofthe
466 software you need to keep track of each person who makes significant
467 changes. A change of ten lines or so, or a few such changes, in a
468 large program is not significant.
470 *Before* incorporating significant changes, make sure that the person
471 has signed copyright papers, and that the Free Software Foundation has
474 If you receive contributions you want to use from someone, let me know
475 and I'll take care of the administrivia. Put the contributions aside
476 until we have the necessary papers.
478 Once you accept a contribution, be sure to keep the files AUTHORS and
481 - When you make substantial changes to a file, add the current year to
482 the list of years in the copyright notice at the top of the file.
484 - When you get bug reports or patches from people, be sure to list
488 Naming conventions =================================================
490 We use certain naming conventions to structure the considerable number
491 of global identifiers. All identifiers should be either all lower
492 case or all upper case. Syllables are separated by underscores `_'.
493 All non-static identifiers should start with scm_ or SCM_. Then might
494 follow zero or more syllables giving the category of the identifier.
495 The currently used category identifiers are
499 c,C - something with a interface suited for C use. This is used
500 to name functions that behave like Scheme primitives but
501 have a more C friendly calling convention.
503 i,I - internal to libguile. It is global, but not considered part
506 f - a SCM variable pointing to a Scheme function object.
508 F - a bit mask for a flag.
510 m - a macro transformer procedure
512 n,N - a count of something
514 s - a constant C string
516 k - a SCM variable pointing to a keyword.
518 sym - a SCM variable pointing to a symbol.
520 var - a SCM variable pointing to a variable object.
522 The follwing syllables also have a technical meaning:
524 str - this denotes a zero terminated C string
526 mem - a C string with an explicit count
529 See also the file `devel/names.text'.
532 Helpful hints ========================================================
534 - [From Mikael Djurfeldt] When working on the Guile internals, it is
535 quite often practical to implement a scheme-level procedure which
536 helps you examine the feature you're working on.
538 Examples of such procedures are: pt-size, debug-hand and
541 I've now put #ifdef GUILE_DEBUG around all such procedures, so that
542 they are not compiled into the "normal" Guile library. Please do the
543 same when you add new procedures/C functions for debugging purpose.
545 You can define the GUILE_DEBUG flag by passing --enable-guile-debug to
546 the configure script.
548 - You'll see uses of the macro SCM_P scattered throughout the code;
549 those are vestiges of a time when Guile was meant to compile on
550 pre-ANSI compilers. Guile now requires ANSI C, so when you write new
551 functions, feel free to use ANSI declarations, and please provide
552 prototypes for everything. You don't need to use SCM_P in new code.
555 Jim Blandy, and others
558 Patches ===========================================================
560 This one makes cvs-1.10 consider the file $CVSDOTIGNORE instead of
561 .cvsignore when that environment variable is set.
564 diff -r -u cvs-1.10/src/cvs.h cvs-1.10.ignore-hack/src/cvs.h
565 --- cvs-1.10/src/cvs.h Mon Jul 27 04:54:11 1998
566 +++ cvs-1.10.ignore-hack/src/cvs.h Sun Jan 23 12:58:09 2000
569 extern int ign_name PROTO ((char *name));
570 void ign_add PROTO((char *ign, int hold));
571 -void ign_add_file PROTO((char *file, int hold));
572 +int ign_add_file PROTO((char *file, int hold));
573 void ign_setup PROTO((void));
574 void ign_dir_add PROTO((char *name));
575 int ignore_directory PROTO((char *name));
576 diff -r -u cvs-1.10/src/ignore.c cvs-1.10.ignore-hack/src/ignore.c
577 --- cvs-1.10/src/ignore.c Mon Sep 8 01:04:15 1997
578 +++ cvs-1.10.ignore-hack/src/ignore.c Sun Jan 23 12:57:50 2000
581 * Open a file and read lines, feeding each line to a line parser. Arrange
582 * for keeping a temporary list of wildcards at the end, if the "hold"
584 + * argument is set. Return true when the file exists and has been handled.
588 ign_add_file (file, hold)
594 if (! existence_error (errno))
595 - error (0, errno, "cannot open %s", file);
597 + error (0, errno, "cannot open %s", file);
600 while (getline (&line, &line_allocated, fp) >= 0)
601 ign_add (line, hold);
604 error (0, errno, "cannot close %s", file);
609 /* Parse a line of space-separated wildcards and add them to the list. */
614 + char *cvsdotignore;
616 /* Set SUBDIRS if we have subdirectory information in ENTRIES. */
622 - ign_add_file (CVSDOTIGNORE, 1);
623 + cvsdotignore = getenv("CVSDOTIGNORE");
624 + if (cvsdotignore == NULL || !ign_add_file (cvsdotignore, 1))
625 + ign_add_file (CVSDOTIGNORE, 1);
627 wrap_add_file (CVSDOTWRAPPER, 1);
629 while ((dp = readdir (dirp)) != NULL)
632 This one is for pcl-cvs-2.9.2, so that `i' adds to the local
636 --- pcl-cvs.el~ Mon Nov 1 12:33:46 1999
637 +++ pcl-cvs.el Tue Jan 25 21:46:27 2000
638 @@ -1177,7 +1177,10 @@
639 "Append the file in FILEINFO to the .cvsignore file.
640 Can only be used in the *cvs* buffer."
641 (save-window-excursion
642 - (set-buffer (find-file-noselect (expand-file-name ".cvsignore" dir)))
643 + (set-buffer (find-file-noselect
644 + (expand-file-name (or (getenv "CVSDOTIGNORE")
647 (goto-char (point-max))
648 (unless (zerop (current-column)) (insert "\n"))