Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 102456 - HTML-Mason-1.30 update
Summary: HTML-Mason-1.30 update
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: New packages (show other bugs)
Hardware: All All
: High normal (vote)
Assignee: Gentoo Perl team
URL: http://www.masonhq.com/code/history.h...
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-08-13 22:53 UTC by kfm
Modified: 2005-08-14 08:32 UTC (History)
0 users

See Also:
Package list:
Runtime testing required: ---


Attachments
HTML-Mason-1.29_beta02 -> HTML-Mason-1.30 (ebuild diff) (mason,1.60 KB, patch)
2005-08-13 22:55 UTC, kfm
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description kfm 2005-08-13 22:53:48 UTC
HTML-Mason-1.30 is now available upstream. I'm attaching a diff against the
previous HTML-Mason-1.29_beta02 ebuild which is in portage. Please review and
consider for inclusion. The changes are as follows:

(1) Now requires >=dev-perl/Params-Validate-0.70 (according to Build.PL)
(2) Now requires >=dev-perl/Exception-Class-1.15 (according to Build.PL)
(3) Absence of the "modperl" USE flag now *explicitly* introduces CGI dep
(4) Presence of the "modperl" USE flag now *explicitly* introduces
libapreq/libapreq2 dep (and consequently mod_perl)
(5) Changed ">=www-apache/libapreq-1.0-r2" to ">=www-apache/libapreq-1.0-r3"
(cosmetic change, the former is no longer in portage)
(6) Changed "|| ( perl-core/File-Spec >=dev-lang/perl-5.8.0-r12 )" to
">=dev-lang/perl-5.8.6-r5"

The rationale behind 3, 4 and 6 is that the ebuild was previously
non-deterministic in terms of how it responds to USE flags when determining its
effective dependencies. This is plain wrong in my view. Here's one example of why:

  # emerge -pv HTML-Mason

  These are the packages that I would merge, in order:

  Calculating dependencies ...done!
  [ebuild  N    ] dev-perl/yaml-0.36  58 kB
  [ebuild  N    ] perl-core/Test-Harness-2.42  0 kB
  [ebuild  N    ] dev-perl/Compress-Zlib-1.35  0 kB
  [ebuild  N    ] dev-perl/IO-Zlib-1.01  0 kB
  [ebuild  N    ] dev-perl/IO-String-1.05  0 kB
  [ebuild  N    ] dev-perl/Archive-Tar-1.23-r1  0 kB
  [ebuild  N    ] dev-perl/module-build-0.25  102 kB
  [ebuild  N    ] perl-core/File-Spec-3.06  107 kB
  [ebuild  N    ] perl-core/CGI-3.10  220 kB
  [ebuild  N    ] dev-perl/Scalar-List-Utils-1.14  15 kB
  [ebuild  N    ] dev-perl/Class-Data-Inheritable-0.02  3 kB
  [ebuild  N    ] perl-core/Test-Simple-0.54  57 kB
  [ebuild  N    ] dev-perl/Devel-StackTrace-1.11  8 kB
  [ebuild  N    ] dev-perl/Exception-Class-1.21  15 kB
  [ebuild  N    ] dev-perl/Params-Validate-0.76  49 kB
  [ebuild  N    ] dev-perl/Class-Container-0.11  18 kB
  [ebuild  N    ] perl-core/MIME-Base64-3.05  0 kB
  [ebuild  N    ] perl-core/digest-base-1.10  0 kB
  [ebuild  N    ] dev-perl/Digest-SHA1-2.10  0 kB
  [ebuild  N    ] perl-core/Storable-2.13  0 kB
  [ebuild  N    ] dev-perl/IPC-ShareLite-0.09  35 kB
  [ebuild  N    ] dev-perl/Error-0.15-r2  7 kB
  [ebuild  N    ] dev-perl/Cache-Cache-1.03  33 kB
  [ebuild  N    ] dev-perl/HTML-Mason-1.30  +apache2 +modperl 0 kB [1]

Note that the list includes CGI (which I don't really need or want in this
case), and fails to include both apache-2* and mod_perl! My changes correct it
so that all combinations of the apache2/modperl flags cause it to behave as one
would expect.

I'm less certain about the File-Spec change but I think that perl itself
provides the required functionality. Not heavily tested (nor do I intend to try
it with apache-1*) but I'll post back if I have anything noteworthy to add.
Comment 1 kfm 2005-08-13 22:55:34 UTC
Created attachment 65898 [details, diff]
HTML-Mason-1.29_beta02 -> HTML-Mason-1.30 (ebuild diff)

diff -u /usr/portage/dev-perl/HTML-Mason/HTML-Mason-1.29_beta02.ebuild
HTML-Mason-1.30.ebuild (UTF-8 encoding)
Comment 2 Michael Cummings (RETIRED) gentoo-dev 2005-08-14 08:32:43 UTC
yeah, better use of the modperl use flag, thanks :)

you know, i'd have caught this in a day or so - we have a list of cpan vs ebuild
versions, honest (though at least i know someone's looking at this and its not
just me pushing ebuilds into the ether)