Here are some guidelines for members of the Guile developers team. Contributing Your Changes ============================================ - If you have put together a change that meets the coding standards described below, we encourage you to submit it to Guile. The best place to post it is guile@sourceware.cygnus.com. Please don't send it directly to me; I often don't have time to look things over. If you have tested your change, then you don't need to be shy. - Please submit patches using either context or unified diffs (diff -c or diff -u). Don't include a patch for ChangeLog; such patches don't apply cleanly, since we've probably changed the top of ChangeLog too. Instead, provide the unaltered text at the top of your patch. Please don't include patches for generated files like configure, aclocal.m4, or any Makefile.in. Such patches are often large, and we're just going to regenerate those files anyway. CVS conventions ====================================================== - We use CVS to manage the Guile sources. The repository lives on egcs.cygnus.com, in /cvs/guile; you will need an account on that machine to access the repository. Also, for security reasons, egcs presently only supports CVS connections via the SSH protocol, so you must first install the SSH client. Then, you should set your CVS_RSH environment variable to ssh, and use the following as your CVS root: :ext:USER@egcs.cygnus.com:/cvs/guile Either set your CVSROOT environment variable to that, or give it as the value of the global -d option to CVS when you check out a working directory. For more information on SSH, see http://www.cs.hut.fi/ssh. The Guile sources live in several modules: - guile-core --- the interpreter, QuickThreads, and ice-9 - guile-doc --- documentation in progress. When complete, this will be incorporated into guile-core. - guile-oops --- The Guile Object-Oriented Programming System (talk to mdj) - guile-tcltk --- the Guile/Tk interface - guile-tk --- the new Guile/Tk interface, based on STk's modified Tk - guile-rgx-ctax --- the Guile/Rx interface, and the ctax implementation - guile-scsh --- the port of SCSH to guile, talk to Gary Houston - guile-www --- A Guile module for making HTTP requests. There is a mailing list for CVS commit messages; see README for details. - We check Makefile.in and configure files into CVS, as well as the files they are built from (Makefile.am, configure.in); we do not check in Makefiles or header files generated by configuration scripts. The general rule is that you should be able to check out a working directory of Guile from CVS, and then type "configure" and "make", without running any other tools. - (Automake 1.4 only) Be sure to run automake at the top of the tree with no arguments. Do not use `automake Makefile' to regenerate specific Makefile.in files, and do not trust the Makefile rules to rebuild them when they are out of date. Automake 1.4 will add extraneous rules to the top-level Makefile if you specify specific Makefiles to rebuild on the command line. Running the command `autoreconf --force' should take care of everything correctly. - Make sure your changes compile and work, at least on your own machine, before checking them into the main branch of the Guile repository. If you really need to check in untested changes, make a branch. - Include each log entry in both the ChangeLog and in the CVS logs. If you're using Emacs, the pcl-cvs interface to CVS has features to make this easier; it checks the ChangeLog, and generates good default CVS log entries from that. Coding standards ===================================================== - As for any part of Project GNU, changes to Guile should follow the GNU coding standards. The standards are available via anonymous FTP from prep.ai.mit.edu, as /pub/gnu/standards/standards.texi and make-stds.texi. - The Guile tree should compile without warnings under the following GCC switches, which are the default in the current configure script: -O2 -Wall -Wpointer-arith -Wmissing-prototypes The only exceptions are the warnings about variables being clobbered by longjmp/vfork in eval.c. (Tho' if you can figure out how to get rid of those, too, I'd be happy.) Note that the warnings generated vary from one version of GCC to the next, and from one architecture to the next (apparently). To provide a concrete common standard, Guile should compile without warnings from GCC 2.7.2.3 in a Red Hat 5.0 i386 Linux machine. Furthermore, each developer should pursue any additional warnings noted by on their compiler. This means that people using more stringent compilers will have more work to do, and assures that everyone won't switch to the most lenient compiler they can find. :) Note also that EGCS (as of November 3 1998) doesn't handle the `noreturn' attribute properly, so it doesn't understand that functions like scm_error won't return. This may lead to some silly warnings about uninitialized variables. You should look into these warnings to make sure they are indeed spurious, but you needn't correct warnings caused by this EGCS bug. - If you add code which uses functions or other features that are not entirely portable, please make sure the rest of Guile will still function properly on systems where they are missing. This usually entails adding a test to configure.in, and then adding #ifdefs to your code to disable it if the system's features are missing. - When you make a user-visible change (i.e. one that should be documented, and appear in NEWS, put an asterisk in column zero of the start of the ChangeLog entry, like so: Sat Aug 3 01:27:14 1996 Gary Houston * * fports.c (scm_open_file): don't return #f, throw error. When you've written a NEWS entry and updated the documentation, go ahead and remove the asterisk. I will use the asterisks to find and document changes that haven't been dealt with before a release. - Please write log entries for functions written in C under the functions' C names, and write log entries for functions written in Scheme under the functions' Scheme names. Please don't do this: * procs.c, procs.h (procedure-documentation): Moved from eval.c. Entries like this make it harder to search the ChangeLogs, because you can never tell which name the entry will refer to. Instead, write this: * procs.c, procs.h (scm_procedure_documentation): Moved from eval.c. Changes like adding this line are special: SCM_PROC (s_serial_map, "serial-map", 2, 0, 1, scm_map); Since the change here is about the name itself --- we're adding a new alias for scm_map that guarantees the order in which we process list elements, but we're not changing scm_map at all --- it's appropriate to use the Scheme name in the log entry. - There's no need to keep a change log for documentation files. This is because documentation is not susceptible to bugs that are hard to fix. Documentation does not consist of parts that must interact in a precisely engineered fashion; to correct an error, you need not know the history of the erroneous passage. (This is copied from the GNU coding standards.) - Make sure you have papers from people before integrating their changes or contributions. This is very frustrating, but very important to do right. From maintain.texi, "Information for Maintainers of GNU Software": When incorporating changes from other people, make sure to follow the correct procedures. Doing this ensures that the FSF has the legal right to distribute and defend GNU software. For the sake of registering the copyright on later versions ofthe software you need to keep track of each person who makes significant changes. A change of ten lines or so, or a few such changes, in a large program is not significant. *Before* incorporating significant changes, make sure that the person has signed copyright papers, and that the Free Software Foundation has received them. If you receive contributions you want to use from someone, let me know and I'll take care of the administrivia. Put the contributions aside until we have the necessary papers. - When you make substantial changes to a file, add the current year to the list of years in the copyright notice at the top of the file. Helpful hints ======================================================== - [From Mikael Djurfeldt] When working on the Guile internals, it is quite often practical to implement a scheme-level procedure which helps you examine the feature you're working on. Examples of such procedures are: pt-size, debug-hand and current-pstate. I've now put #ifdef GUILE_DEBUG around all such procedures, so that they are not compiled into the "normal" Guile library. Please do the same when you add new procedures/C functions for debugging purpose. You can define the GUILE_DEBUG flag by passing --enable-guile-debug to the configure script. - You'll see uses of the macro SCM_P scattered throughout the code; those are vestiges of a time when Guile was meant to compile on pre-ANSI compilers. Guile now requires ANSI C, so when you write new functions, feel free to use ANSI declarations, and please provide prototypes for everything. You don't need to use SCM_P in new code. Jim Blandy