3 This file is for people who are interested in making modifications to
4 INN. Normal users can safely skip reading it. It is intended primarily
5 as a guide, resource, and accumulation of tips for maintainers and
6 contributors, and secondarily as documentation of some of INN's
9 This is $Revision: 7736 $ dated $Date: 2006-04-15 04:52:06 +0200 (Sat,
12 First of all, if you plan on working on INN source, please start from
13 the current development tree. There may be significant changes from the
14 previous full release, so starting from development sources will make it
15 considerably easier to integrate your work. You can get nightly
16 snapshots of the current development source from ftp.isc.org in
17 /isc/inn/snapshots (the snapshots named inn-CURRENT-*.tar.gz), or you
18 can get the current CVS tree by using CVSup (see "Using CVSup").
20 Configuring and Portability
22 All INN code should be written expecting ANSI C and POSIX. There is no
23 need to attempt to support pre-ANSI compilers, and ANSI-only features
24 such as <stdarg.h>, string concatenation, #elif, and token pasting may
25 be used freely. So far as possible, INN is written to attempt to be
26 portable to any system new enough that someone is likely to want to run
27 a news server on it, but whenever possible this portability should be
28 provided by checking for standard behavior in configure and supplying
29 replacements for standard functions that are missing.
31 When there is a conflict between ANSI C and C99, INN code should be
32 written expecting C99 and autoconf used to patch up the differences.
34 Try to avoid using #ifdef and the like in the middle of code as much as
35 possible. Instead, try to isolate the necessary portability bits and
36 include them in libinn or at least in conditional macros separate from
37 the code. Trying to read code littered with conditional compilation
38 directives is much more difficult.
40 The shell script configure at the top level of the source tree is
41 generated by autoconf from configure.in, and include/config.h.in is
42 generated by autoheader from configure.in and include/acconfig.h. At
43 configure time, configure generates include/config.h and several other
44 files based on options it was given and what it discovers about the
47 All modifications to configure should instead be made to configure.in.
48 Similarly, modifications to include/config.h.in should instead be made
49 to include/acconfig.h. The autoconf manual (available using info
50 autoconf if you have autoconf and the GNU info utilities installed on
51 your system) is a valuable reference when making any modifications.
53 To regenerate configure, just run "autoconf". To regenerate
54 include/config.h.in, run:
58 to tell it where to find acconfig.h. Please don't include patches to
59 either configure or include/config.h.in when sending patches to INN;
60 instead, note in your patch that those files must be regenerated.
62 The generated files are checked into the CVS repository so that people
63 working on INN don't have to have autoconf on their system, and to make
66 At the time of this writing, autoconf 2.13 is required.
68 The supporting files for autoconf are in the support subdirectory,
69 including the files config.guess and config.sub to determine the system
70 name and and ltmain.sh for libtool support. The latter file comes from
71 the libtool distribution; the canonical version of the former two are
72 available from ftp.gnu.org in /gnu/config. In addition, m4/libtool.m4
73 is just a copy of libtool.m4 from the libtool distribution. (Using
74 libtool without using automake requires a few odd hacks.) These files
75 used to be on a separate vendor branch so that we could make local
76 modifications, but local modifications have not been necessary for some
77 time. Now, new versions can just be checked in like any other file
80 INN should not compile with libtool by default, only when requested,
81 since otherwise normal compilations are quite slow. (Using libtool is
82 not without some cost.) Basic compilation with libtool works fine as of
83 this writing, with both static and shared compiles, but the dependencies
84 aren't quite right for make -j using libtool.
88 INN's documentation is currently somewhat in a state of flux. The vast
89 majority is still in the form of man pages written directly in nroff.
90 Some parts of the documentation have been rewritten in POD; that
91 documentation can be found in doc/pod. The canonical source for README,
92 INSTALL, NEWS, doc/hook-perl, doc/hook-python, and this file are also in
95 If you're modifying some part of INN's documentation and see that it has
96 a POD version in doc/pod, it's preferred if you can make the
97 modifications to the POD source and then regenerate the derived files.
98 For a quick introduction to POD, see the perlpod(1) man page on your
99 system (it should be installed if you have Perl installed).
101 When writing new documentation, write in whatever format you care to; if
102 necessary, we can always convert it to POD or whatever else we want to
103 use. Having the documentation exist in *some* form is more important
104 than what language you write it in. If you really don't have any
105 particular preference, there's a slight preference currently for POD.
107 If you use POD or regenerate POD documentation, please install something
108 close to the latest versions of the POD processing utilities to avoid
109 changes to the documentation depending on who generated it last. You
110 can find the latest version on CPAN (ftp.perl.org or another mirror) in
111 modules/by-module/Pod. You'll need PodParser (for versions of Perl
112 before 5.6.1; 5.6.1 and later come with a recent enough version) and the
113 latest version of podlators. For versions of Perl earlier than 5.005,
114 you'll also need File::Spec in modules/by-module/File.
116 podlators 1.25 or later will build INN's documentation without
117 significant changes from the versions that are checked into the
120 There are Makefile rules in doc/pod/Makefile to build all of the
121 documentation whose master form is POD; if you add additional
122 documentation, please add a rule there as well. Documentation should be
123 generated by cd'ing to doc/pod and typing "make file" where "file" is
124 the relative path to the documentation file. This will get all of the
125 various flags right for pod2text or pod2man.
129 INN has a set of generic error handling routines that should be used as
130 much as possible so that the same syntax can be used for reporting
131 errors everywhere in INN. The four basic functions are warn, syswarn,
132 die, and sysdie; warn prints or logs a warning, and die does the same
133 and then exits the current program. The sys* versions add a colon, a
134 space, and the value of strerror(errno) to the end of the message, and
135 should be used to report failing system calls.
137 All of the actual error reporting is done via error handlers, and a
138 program can register its own handlers in addition to or instead of the
139 default one. The default error handler (error_log_stderr) prints to
140 stderr, prepending the value of error_program_name if it's set to
141 something other than NULL. Three other error handlers are available,
142 error_log_syslog_crit, error_log_syslog_err, and
143 error_log_syslog_warning, which log the message to syslog at LOG_CRIT,
144 LOG_ERR, or LOG_WARNING priority, respectively.
146 There is a different set of error handlers for warn/syswarn and
147 die/sysdie. To set them, make calls like:
149 warn_set_handlers(2, error_log_stderr, error_log_syslog_warning);
150 die_set_handlers(2, error_log_stderr, error_log_syslog_err);
152 The first argument is the number of handlers, and the remaining
153 arguments are pointers to functions taking an int (the length of the
154 formatted message), a const char * (the format), a va_list (the
155 arguments), and an int that's 0 if warn or die was called and equal to
156 the value of errno if syswarn or sysdie was called. The length of the
157 formatted message is obtained by calling vsnprintf with the provided
158 format and arguments, and therefore is reliable to use as the size of a
159 buffer to malloc to hold the result of formatting the message provided
160 that vsnprintf is used to format it (warning: the system vsprintf may
161 produce more output under some circumstances, so always use vsnprintf).
163 The error handler can do anything it wishes; each error handler is
164 called in the sequence given. Error handlers shouldn't call warn or die
165 unless great caution is taken to prevent infinite recursion. Also be
166 aware that sysdie is called if malloc fails in xmalloc, so if the error
167 handler needs to allocate memory, it must not use xmalloc or a related
168 function to do so and it shouldn't call die to report failure. The
169 default syslog handlers report memory allocation failure to stderr and
172 Finally, die and sysdie support an additional handler that's called
173 immediate before exiting, takes no arguments, and returns an int which
174 is used as the argument for exit. It can do any necessary global
175 cleanup, call abort instead to generate a core dump or the like.
177 The advantage of using this system everywhere in INN is that library
178 code can use warn and die to report errors and each calling program can
179 set up the error handlers as appropriate to make sure the errors go to
180 the right place. The default handler is fine for interactive programs;
181 for programs that run from interactive scripts, adding something like:
183 error_program_name = "program";
185 to the beginning of main (where program is the name of the program) will
186 make it easier to figure out which program the script calls is failing.
187 For programs that may also be called non-interactively, like inndstart,
188 one may want to set up handlers like:
190 warn_set_handlers(2, error_log_stderr, error_log_syslog_warning);
191 die_set_handlers(2, error_log_stderr, error_log_syslog_err);
193 Finally, for daemons and other non-interactive programs, one may want to
196 warn_set_handlers(1, error_log_syslog_warning);
197 die_set_handlers(1, error_log_syslog_err);
199 to report errors only via syslog. (Note that if you use syslog error
200 handlers, the program should call openlog first thing to make sure they
201 are logged with the right facility.)
203 For historical reasons, error messages that are fatal to the news
204 subsystem are logged at the LOG_CRIT priority, and therefore die in innd
205 should use error_log_syslog_crit.
209 The test suite for INN is located in the tests directory and is just
210 getting started. The test suite consists of a set of programs listed in
211 tests/TESTS and the scaffolding in the runtests program.
213 Adding new tests is very straightforward and very flexible. Just write
214 a program that tests some part of INN, put it in a directory under tests
215 named after the part of INN it's testing (all the tests so far are in
216 lib because they're testing libinn routines), and have it output first a
217 line containing the count of test cases in that file, and then for each
218 test a line saying "ok n" or "not ok n" where n is the test case number.
219 (If a test is skipped for some reason, such as a test of an optional
220 feature that wasn't compiled into INN, the test program should output
221 "ok n # skip".) Add any rules necessary to build the test to
222 tests/Makefile (note that for simplicity it doesn't recurse into
223 subdirectories) and make sure it creates an executable ending in .t.
224 Then add the name of the test to tests/TESTS, without the .t ending.
226 One naming convention: to distinguish more easily between e.g.
227 lib/error.c (the implementation) and tests/lib/error-t.c (the test
228 suite), we add -t to the end of the test file names. So
229 tests/lib/error-t.c is the source that compiles into an executable
230 tests/lib/error.t which is run by putting a line in tests/TESTS of just
233 Note that tests don't have to be written in C; in fact, lib/xmalloc.t is
234 just a shell script (that calls a supporting C program). Tests can be
235 written in shell or Perl (but other languages should be avoided because
236 someone who wants to run the test suite may not have it) and just have
237 to follow the above output conventions.
239 Additions to the test suite, no matter how simple, are very welcome.
243 All INN makefiles include Makefile.global at the top level, and only
244 that makefile is a configure substitution target. This has the
245 disadvantage that configure's normal support for building in a tree
246 outside of the source tree doesn't work, but it has the significant
247 advantage of making configure run much faster and allowing one to run
248 make in any subdirectory and pick up all the definitions and settings
249 from the top level configuration.
251 All INN makefiles should also set $(top) to be the path to the top of
252 the build directory (usually relative). This path is used to find
253 various programs like fixscript and libtool so that the same macros (set
254 in Makefile.global) can be used all over INN.
256 The format of INN's makefiles is mostly standardized; the best examples
257 of the format are probably frontends/Makefile and backends/Makefile, at
258 least for directories with lots of separate programs. The ALL variable
259 holds all the files that should be generated, EXTRA those additional
260 files that were generated by configure, and SOURCES the C source files
261 for generating tag information.
263 There are a set of standard installation commands defined in make
264 variables by Makefile.global, and these should be used for all file
265 installations. See the comment blocks in Makefile.global.in for
266 information on what commands are available and when they should be used.
267 There are also variables set for each of the installation directories
268 that INN uses, for use in building the list of installed paths to files.
270 Each subdirectory makefile should have the targets all (the default),
271 clean, clobber, install, tags, and profiled. The tags target generates
272 vi tags files, and the profiled target generates a profiling version of
273 the programs (although this hasn't been tested much recently). These
274 rules should be present and empty in those directories where they don't
277 Be sure to test compiling with both static and dynamic libraries and
278 make sure that all the libtool support works correctly. All linking
279 steps, and the compile steps for all library source, should be done
280 through $(LIBTOOL) (which will be set to empty in Makefile.global if
281 libtool support isn't desired).
285 INN comes with and installs a large number of different scripts, both
286 Bourne shell and Perl, and also comes with support for Tcl scripts
287 (although it doesn't come with any). Shell variables containing both
288 configure-time information and configuration information from inn.conf
289 are set by the innshellvars support libraries, so the only
290 system-specific configuration that should have to be done is fixing the
291 right path to the interpretor and adding a line to load the appropriate
294 support/fixscript, built by configure, does this. It takes a .in file
295 and generates the final script (removing the .in) by fixing the path to
296 the interpretor on the first line and replacing the second line,
297 whatever it is, with code to load the innshellvars appropriate for that
298 interpretor. (If invoked with -i, it just fixes the interpretor path.)
300 Scripts should use innshellvars (via fixscript) to get the right path
301 and the right variables whenever possible, rather than having configure
302 substitute values in them. Any values needed at run-time should instead
303 be available from all of the different innshellvars.
305 See the existing scripts for examples of how this is done.
309 Include files relevant to all of INN, or relevant to the two libraries
310 built as part of INN (the utility libinn library and the libstorage
311 library that contains all storage and overview functions) are found in
312 the include directory; other include files relevant only to a portion of
313 INN are found in the relevant directory.
315 Practically all INN source files will start with:
318 #include "clibrary.h"
320 The first picks up all defines generated by autoconf and is necessary
321 for types that may not be present on all systems (uid_t, pid_t, size_t,
322 int32_t, and the like). It therefore should be included before any
323 other headers that use those types, as well as to get general
324 configuration information.
326 The second is portably equivalent to:
328 #include <sys/types.h>
337 except that it doesn't include headers that are missing on a given
338 system, replaces functions not found on the system with the INN
339 equivalents, provides macros that INN assumes are available but which
340 weren't found, and defines some additional portability things. Even if
341 this is more headers than the source file actually needs, it's generally
342 better to just include clibrary.h rather than trying to duplicate the
343 autoconf-driven hackery that it does to do things portably. The primary
344 exception is for source files in lib that only define a single function
345 and are used for portability; those may want to include only config.h so
346 that they can be easily used in other projects that use autoconf.
347 config.h is a fairly standard header name for this purpose.
349 clibrary.h does also include config.h, but it's somewhat poor form to
350 rely on this; it's better to explicitly list the header dependencies for
351 the benefit of someone else reading the code.
353 There are portable wrappers around several header files that have known
354 portability traps or that need some fixing up on some platforms. Look
355 in include/portable and familiarize yourself with them and use them
358 Another frequently included header file is libinn.h, which among other
359 things defines xmalloc(), xrealloc(), xstrdup(), and xcalloc(), which
360 are checked versions of the standard memory allocation routines that
361 terminate the program if the memory allocation fails. These should
362 generally always be used instead of the regular C versions. libinn.h
363 also provides various other utility functions that are frequently used.
365 paths.h includes a wide variety of paths determined at configure time,
366 both default paths to various parts of INN and paths to programs. Don't
367 just use the default paths, though, if they're also configurable in
368 inn.conf; instead, call ReadInnConf() and use the global innconf
371 Other files in include are interfaces to particular bits of INN library
372 functionality or are used for other purposes; see the comments in each
375 Eventually, the header files will be separated into installed header
376 files and uninstalled header files; the latter are those headers that
377 are used only for compiling INN and aren't useful for users of INN's
378 libraries (such as clibrary.h). All of the installed headers will live
379 in include/inn and be installed in a subdirectory named inn in the
380 configured include directory. This conversion is still in progress.
382 When writing header files, remember that C reserves all identifiers
383 beginning with two underscores and all identifiers beginning with an
384 underscore and a capital letter for the use of the implementation; don't
385 use any identifiers with names like that. Additionally, any identifier
386 beginning with an underscore and a lower-case letter is reserved in file
387 scope, which means that such identifiers can only be used by INN for the
388 name of structure members or function arguments in function prototypes.
390 Try to pay attention to the impact of a header file on the program
391 namespace, particularly for installed header files in include/inn. All
392 symbols defined by a header file should ideally begin with INN_, inn_,
393 or some other unique prefix indicating the subsystem that symbol is part
394 of, to avoid accidental conflicts with symbols defined by the program
395 that uses that header file.
399 INN has quite a variety of coding styles intermixed. As with all
400 programs, it's preferrable when making minor modifications to keep the
401 coding style of the code you're modifying. In INN, that will vary by
402 file. (Over time we're trying to standardize on one coding style, so
403 changing the region you worked on to fit the general coding style is
406 If you're writing a substantial new piece of code, the prevailing
407 "standard" INN coding style appears to be something like the following:
409 * Write in regular ANSI C whenever possible. Use the normal ANSI and
410 POSIX constructs and use autoconf or portability wrappers to fix
411 things up beforehand so that the code itself can read like regular
412 ANSI or POSIX code. Code should be written so that it works as
413 expected on a modern platform and is fixed up with portability tricks
414 for older platforms, not the other way around. You may assume an
417 Try to use const wherever appropriate. Don't use register; modern
418 compilers will do as good of a job as you will in choosing what to
419 put into a register. Don't bother with restrict (at least yet).
421 * Use string handling functions that take counts for the size of the
422 buffer whenever possible. This means using snprintf in preference to
423 sprintf and using strlcpy and strlcat in preference to strcpy and
424 strcat. Also, use strlcpy and strlcat instead of strncpy and strncat
425 unless the behavior of the latter is specifically required, as it is
426 much easier to audit uses of the former than the latter. (strlcpy is
427 like strncpy except that it always nul-terminates and doesn't fill
428 the rest of the buffer with nuls, making it more efficient. strlcat
429 is like strncat except that it always nul-terminates and it takes the
430 total size of the buffer as its third argument rather than just the
431 amount of space left.) All of these functions are guaranteed to be
432 available; there are replacements in lib for systems that don't have
435 * Avoid #ifdef and friends whenever possible. Particularly avoid using
436 them in the middle of code blocks. Try to hide all portability
437 preprocessor magic in header files or in portability code in lib.
438 When something just has to be done two completely different ways
439 depending on the platform or compile options or the like, try to
440 abstract that functionality out into a generic function and provide
441 two separate implementations using #ifdef; then the main code can
442 just call that function.
444 If you do have to use preprocessor defines, note that if you always
445 define them to either 0 or 1 (never use #define without a second
446 argument), you can use the preprocessor define in a regular if
447 statement rather than using #if or #ifdef. Make use of this instead
448 of #ifdef when possible, since that way the compiler will still
449 syntax-check the other branch for you and it makes it far easier to
450 convert the code to use a run-time check if necessary.
451 (Unfortunately, this trick can't be used if one branch may call
452 functions unavailable on a particular platform.)
454 * Avoid uses of fixed-width buffers except in performance-critical
455 code, as it's harder to be sure that such code is correct and it
456 tends to be less flexible later on. If you need a reusable,
457 resizable memory buffer, one is provided in lib/buffer.c.
459 * Avoid uses of static variables whenever possible, particularly in
460 libraries, because it interferes with making the code re-entrant down
461 the road and makes it harder to follow what's going on. Similarly,
462 avoid using global variables whenever possible, and if they are
463 required, try to wrap them into structures that could later be
464 changed into arguments to the affected functions.
466 * Roughly BSD style but with four-space indents. This means no space
467 before the parens around function arguments, open brace on the same
468 line as if/while/for, and close and open brace on the same line as
471 * Introductory comments for functions or files are generally written
475 ** Introductory comment.
478 Other multiline comments in the source are generally written as:
481 multiline comment. */
483 Comments before functions saying what they do are nice to have. In
484 general, the RCS/CVS Id tag is on the first line of each source file
485 since it's useful to know when a file was last modified.
487 * Checks for NULL pointers are preferrably written out explicitly; in
496 to make it clearer what the code is assuming.
498 * It's better to always put the body of an if statement on a separate
499 line, even if it's only a single line. In other words, write:
506 if (p != NULL) return p;
508 This is in part for a practical reason: some code coverage analysis
509 tools like purecov will count the second example above as a single
510 line and won't notice if the condition always evaluates the same way.
512 * Plain structs make perfectly reasonable abstract data types; it's not
513 necessary to typedef the struct to something else. Structs are
514 actually very useful for opaque data structures, since you can
515 predeclare them and then manipulate pointers to them without ever
516 having to know what the contents look like. Please try to avoid
517 typedefs except for function pointers or other extremely confusing
518 data types, or for data types where we really gain some significant
519 data abstraction from hiding the underlying data type. Also avoid
520 using the _t suffix for any type; all types ending in _t are reserved
521 by POSIX. For typedefs of function pointer types, a suffix of _func
524 This style point is currently widely violated inside of INN itself;
525 INN originally made extensive use of typedefs.
527 * When noting something that should be improved later, add a comment
528 containing "FIXME:" so that one can easily grep for such comments.
530 INN's indentation style roughly corresponds to that produced by GNU
531 indent 2.2.6 with the following options:
533 -bad -bap -nsob -fca -lc78 -cd41 -cp1 -br -ce -cdw -cli0 -ss -npcs
534 -ncs -di1 -nbc -psl -brs -i4 -ci4 -lp -ts8 -nut -ip5 -lps -l78 -bbo
537 Unfortunately, indent currently doesn't get everything right (it has
538 problems with spacing around struct pointer arguments in functions,
539 wants to put in a space between a dereference of a function pointer and
540 the arguments to the called function, misidentifies some macro calls as
541 being type declarations, and fouls up long but simple case statements).
542 It would be excellent if someday we could just run all of INN's code
543 through indent routinely to enforce a consistant coding style, but
544 indent isn't quite ready for that.
546 For users of emacs cc-mode, use the "bsd" style but with:
548 (setq c-basic-offset 4)
550 Finally, if possible, please don't use tabs in source files, since they
551 can expand differently in different environments. In particular, please
552 try not to use the mix of tabs and spaces that is the default in emacs.
553 If you use emacs to edit INN code, you may want to put:
555 ; Use only spaces when indenting or centering, no tabs.
556 (setq-default indent-tabs-mode nil)
558 in your ~/.emacs file.
560 Note that this is only a rough guideline and the maintainers aren't
561 style nazis; we're more interested in your code contribution than in how
566 If you want to get updated INN source more easily or more quickly than
567 by downloading nightly snapshots, or if you want to see the full CVS
568 history, you may want to use CVSup to download the source. CVSup is a
569 client and server designed for replicating CVS repositories between
572 Unfortunately, CVSup is written in Modula-3, so getting a working binary
573 can be somewhat difficult. Binaries are available in the *BSD ports
574 collection or (for a wide variety of different platforms) available from
575 <ftp://ftp.freebsd.org/pub/FreeBSD/CVSup/binaries/> and its mirrors.
576 Alternately, you can get a compiler from <http://m3.polymtl.ca/m3/>
577 (this is more actively maintained than the DEC Modula-3 compiler) and
578 the source from <ftp://ftp.freebsd.org/pub/FreeBSD/CVSup/>.
580 After you have the CVSup client, you need to have space to download the
581 INN repository and space for CVSup to store its data files. You also
582 need to write a configuration file (a supfile) for CVSup. The following
583 supfile will download the latest versions from the mainline source:
585 *default host=inn-cvs.isc.org
586 *default base=<data-location>
587 *default prefix=<download-location>
590 *default delete use-rel-suffix
593 where <data-location> should be a directory where CVSup can put its data
594 files and <download-location> is where the downloaded source will go (it
595 will be put into a subdirectory named inn). If you want to pull down
596 the entire CVS repository instead (warning: this is much larger than
597 just the latest versions of the source), delete the "*default tag=."
598 line. The best way to download the CVS repository is to download it
599 into a portion of a locally-created CVS repository, so that then you can
600 perform standard CVS operations (like cvs log) against the downloaded
601 repository. Creating your own local CVS repository is outside the scope
604 Note that only multiplexed mode is supported (this mode should be the
607 For more general information on using CVSup, see the FreeBSD page on it
608 at <http://www.freebsd.org/handbook/mirrors-cvsup.html>.
612 This is a checklist that INN maintainers should go through when
613 preparing a new release of INN.
615 1. If making a major release, branch the source tree and create a new
616 STABLE branch tag. This branch will be used for minor releases
617 based on that major release and can be done a little while before
618 the .0 release of that major release. At the same time as the
619 branch is cut, tag the trunk with a STABLE-<version>-branch marker
620 tag so that it's easy to refer to the trunk at the time of the
623 2. Update doc/pod/news.pod and regenerate NEWS. Be more detailed for a
624 minor release than for a major release. For a major release, also
625 add information on how to upgrade from the last major release,
626 including anything special to be aware of. (Minor releases
627 shouldn't require any special care when upgrading.)
629 3. Make sure that support/config.sub and support/config.guess are the
630 latest versions (from <ftp://ftp.gnu.org/gnu/config/>). See the
631 instructions in "Configuring and Portability" for details on how to
634 4. Make sure that samples/control.ctl is in sync with the master
635 version at <ftp://ftp.isc.org/pub/usenet/CONFIG/control.ctl>.
637 5. Check out a copy of the release branch. It's currently necessary to
638 run configure to generate Makefile.global. Then run "make
639 check-manifest". The only differences should be files that are
640 generated by configure; if there are any other differences, fix the
643 6. Run "make release". Note that you need to have a copy of svn2cl
644 from <http://ch.tudelft.nl/~arthur/svn2cl/> to do this; at least
645 version 0.7 is required. Start the ChangeLog at the time of the
646 previous release. (Eventually, the script will be smart enough to
649 7. Make the resulting tar file available for testing in a non-listable
650 directory on ftp.isc.org and announce its availability on
651 inn-workers. Install it on at least one system and make sure that
652 system runs fine for at least a few days. This is also a good time
653 to send out a draft of the release announcement to inn-workers for
656 8. Generate a diff between this release and the previous release if
657 feasible (always for minor releases, possibly not a good idea due to
658 the length of the diff for major releases).
660 9. Move the release into the public area of the ftp site and update the
661 inn.tar.gz link. Make an MD5 checksum of the release tarball and
662 put it on the ftp site as well, and update the inn.tar.gz.md5 link.
663 Put the diff up on the ftp site as well. Contact the ISC folks to
664 get the release PGP-signed. Possibly move older releases off into
667 10. Announce the new release on inn-announce and in news.software.nntp.
669 11. Tag the checked-out tree that was used for generating the release
670 with a release tag (INN-<version>).
672 12. Bump the revision number in Makefile.global.in.
676 Some additional references that may be hard to find and may be of use to
677 people working on INN:
679 <http://www.eyrie.org/~eagle/nntp/>
680 The home page for the IETF NNTP standardization effort, including
681 links to the IETF NNTP working group archives and copies of the
682 latest drafts of the new NNTP standard. The old archived mailing
683 list traffic contains a lot of interesting discussion of why NNTP is
686 <http://www.imc.org/ietf-usefor/>
687 The archives for the USEFOR IETF working group, the working group
688 for the RFC 1036 replacement (the format of Usenet articles). Also
689 contains a lot of references to other relevant work, such as the RFC
690 822 replacement work.
692 <http://www.mibsoftware.com/userkt/inn/dev/>
693 Forrest Cavalier provides several tools for following INN
694 development at this page and elsewhere in the Usenet RKT. Under
695 here is a web-accessible checked-out copy of the current INN source
696 tree and pointers to how to use CVSup.
698 <http://www.sas.com/standards/large.file/>
699 The standards for large file support on Unix that are being
700 generally implemented by vendors. INN sort of partially uses these,
701 but a good full audit of the code to check them should really be
702 done and there are occasional problems.
704 <http://v6web.litech.org/ipv6primer/>
705 A primer on IPv6 with pointers to the appropriate places for more
706 technical details as needed, useful when working on IPv6 support in