[mp1] 1.28 release candidate #1

["Philippe M. Chiasson" <go...@cpan.org>]

Finally, the first mod_perl 1.28 release candidate #1 has arrived. It
can be downloaded here:

http://www.apache.org/~gozer/mp1/mod_perl-1.27_01-dev-rc1.tar.gz

MD5 : ad64dfdb4f5b056b00d87288ef31bd56
SHA1: 03409c6f44c408acae44a258fe4e59b408c0040f

The summary of what has changed since 1.27 are (from STATUS):

Add Win32 support to Apache::SizeLimit [Matt Phillips
<mp...@virage.com> and Mohamed Hendawi <mh...@virage.com>]

Change Apache::SizeLimit to not push a cleanup handler if already in
the cleanup handler phase, and adjust docs to show that cleanup
handler is the preferred phase to use [Perrin Harkins
<pe...@elem.com>]

Rename Apache::test to Apache::testold because Apache::test on 
case-insensitive systems will collide with Apache::Test which 
supercedes Apache::test. So if you want to keep on using Apache::test,
either bundle it with your project (putting it under inc/ or t/ so it
won't be installed) or require mod_perl 1.28 and use Apache::testold
instead. Of course the best route is to port your test suite to use
a much better Apache::Test which work with mod_perl 1.0 and 2.0.
[Philippe M. Chiasson, Stas Bekman]

Tweak mod_perl.h to defined _INCLUDE_APACHE_FIRST only after apache
headers were included [Stas Bekman]

avoid various warnings under src/modules/perl/:
- declare bufsiz to be STRLEN in Apache.xs, and add
  STRLEN to Apache/typemap
- add I32 typecast in Constants.xs
- avoid use of unregistered local variables for Win32
  in mod_perl.c and perl_config.c
- s/I32/U8/ in mod_perl.h, perl_config.c, and perl_util.c
- declare i and http_code to be STRLEN in perl_util.c
[Stas Bekman, Randy Kobes]

don't use $r variable in Apache::PerlRun::compile(), so the script
won't use use inherited $r by mistake [Stas Bekman]

define PERL_DIRECTIVE_HANDLERS so that ModuleConfig.c gets
generated when building on Win32 within Visual Studio
[John Petrakis <jp...@rcn.com>]

enable PERL_SECTIONS for Win32 [Dirk Maetens <di...@chello.be>]

use touch() from ExtUtils::Command, rather than a system touch(),
for the benefit of platforms without touch(). [Randy Kobes
<ra...@theoryx5.uwinnipeg.ca>]
 
can't let the default typemap rule to convert sv into char* in
unescape_url, since it doesn't handle correctly undefs (returns an
unallocated "" string, which then causes a segfault in
ap_unescape_url. use SvPV_force, instead, which does the right
thing. [Stas Bekman]

Make sure to start perl, if it's not running, before processing Perl*
directives, with threaded perl and PERL_STACKED_HANDLERS=0 [Stas
Bekman]

Add Apache::Module to Bundle::Apache [Stas Bekman]

use $Config{'installstyle'} instead of hardcoded 'lib', to handle
Makefile.PL's PREFIX option correctly [Philippe M. Chiasson
<go...@cpan.org>]

prevent segfaults in mod_perl_mark_where() when a sub can't get
resolved [Gerald Richter <ri...@ecos.de>]

Apache::Status: Need to load B::Terse/TerseSize if it wasn't loaded
yet in that child before using it.  [Dan Sully <da...@electricrain.com>]

document the server_root_relative() method [Stas Bekman <st...@stason.org>]

eliminate warnings when flushing functions with empty () prototypes in
Apache::PerlRun::flush_namespace [Yair Lenga <ya...@citigroup.com>]

fix Apache::Status to not use :: in filenames, which is not allowed on
certain OSs [DH <cr...@yahoo.com>]

various cygwin fixes [Per Einar Ellefsen <pe...@skynet.be>]

fix to compile with 5.8.0 on win32 [Randy Kobes <ra...@theoryx5.uwinnipeg.ca>]

Document the possible misuses of the Apache::Constant constants
[Per Einar Ellefsen <pe...@skynet.be>]

--

A more detailled review of each patch included in this release candidate can be found here:
http://www.apache.org/~gozer/mp1/1.27-dev/

Please give this release a spin and report back any problems or failed tests to:
dev@perl.apache.org as soon as possible. The more platforms & configurations, the
merrier!

Thank you!

-- 
-- -----------------------------------------------------------------------------
Philippe M. Chiasson /gozer\@(cpan|ectoplasm)\.org/ 88C3A5A5 (122FF51B/C634E37B)
http://gozer.ectoplasm.org/    F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 88C3 A5A5
Q: It is impossible to make anything foolproof because fools are so ingenious.
perl -e'$$=\${gozer};{$_=unpack(P7,pack(L,$$));/^JAm_pH\n$/&&print||$$++&&redo}'