#acl PaulHowarth:read,write,admin,revert,delete All:read

=== Tuesday 21st February 2012 ===

==== Fedora Project ====

 * Became co-maintainer of `perl-ExtUtils-ParseXS` in EPEL-5
 * Updated `perl-Compress-Raw-Zlib` to 2.050 in Rawhide:
  * Fix build failure on Irix and Solaris ([[CPAN:75151|CPAN RT#75151]])
 * Built `perl-Params-Classify` (0.013) in EPEL-5 as `perl-ExtUtils-ParseXS` became available

==== Local Packages ====

 * New package `perl-Perl-Critic-Lax` (0.008)
 * Updated `perl-Compress-Raw-Zlib` to 2.050 as per the Fedora version
 * Updated `perl-Filter` to 1.42:
  * Improve `t/tee.t` test 5 on Windows, which allows read access for all `Administrator` group members
 * Updated `perl-Module-CoreList` to 2.61:
  * Updated for v5.15.8
 * Updated `perl-Net-FTPSSL` to 0.21:
  * Fixed bad call to `_croak_or_return ()` in `_get_data_channel()` when dealing with failed certificates for data connections
  * New option to execute `PRET` (for distributed FTP data channel transfers) before all `PASV` calls ([[CPAN:74547|CPAN RT#74547]])
  * Updated `t/10-complex.t` to dynamically check if `PRET` is needed
  * Added checks to avoid warnings with `uc(undef)` ([[CPAN:74035|CPAN RT#74035]])
  * Fixed both test cases to no longer send `Port` as undefined to get rid of warnings in log file (passing `undef` wasn't realistic, was just convenient)
  * `t/10-complex.t` now explicitly checks if any warnings are generated and does a request for feadback instead of as an error if any are found, which should help considerably with debugging any code changes in the future
  * Added new internal method `_feat()` to get additional commands to be recognized by `supported()`
  * Enhanced `_help()` in parsing its response from the server and other significant enhancements (major rewrite of this one method)
  * Above rewrite uncovered bug in `t/10-complex.t` to fix in a `supported()` test
  * Did some tweaking to the workings of the `OverrideHELP` option due to changes in `_help()` and `supported()`
  * Updated POD for `supported()` to describe changes to functionality
 * Updated `perl-Params-Validate` to 1.06:
  * Shut up warnings when `XS` is loaded on older 5.8.x `Perl` versions ([[CPAN:74742|CPAN RT#74742]])
 . Now that this problem is fixed, I've dropped the workaround of only building the Pure-Perl version on affected distributions
 * Updated `perl-Sub-Uplevel` to 0.24:
  * Added: the `$Sub::Uplevel::CHECK_FRAMES` global constant to enable the frame depth checking and compile it out if not needed
  * Fixed: `Perl` 5.15 series `Carp` changes output format of messages (adds a period); this fixes the test regular expressions
 . This release shipped with a host of new author/release tests, so I added a bunch of build-requirements for them (some excluded when bootstrapping) and ran the additional tests too
 * Updated `perl-XML-LibXML` to 1.92:
  * Overload hash dereferencing on `XML::LibXML::Elements`, to provide access to the element's attributes (see `XML::LibXML::AttributeHash` for details)
  * Add more convenient methods to `XML::LibXML::NodeList` such as `sort`, `map`, `grep`, etc.
  * Printed some warnings regardless of if `DEBUG` is on
  * Fix for test failure of 1.91 on `perl`s < 5.10 ([[CPAN:75195|CPAN RT#75195]])
 . There's a curious failure of the `t/72destruction.t` test on RHEL-3 x86_64 (not on i386), after all of the subtests passed:
 . {{{
t/72destruction.....................dubious
        Test returned status 0 (wstat 139, 0x8b)
Scalar found where operator expected at (eval 153) line 1, near "'int' $__val"
        (Missing operator before $__val?)
        after all the subtests completed successfully
t/80registryleak....................ok
t/90threads.........................skipped
        all skipped: optional (set THREAD_TEST=1 to run these tests)
t/pod...............................ok
Failed 1/54 test scripts, 98.15% okay. 0/2379 subtests failed, 100.00% okay.
Failed Test       Stat Wstat Total Fail  Failed  List of Failed -------------------------------------------------------------------------------
t/72destruction.t    0   139     1    0   0.00%  ??
3 tests skipped.
}}}
 . As everything pretty well worked, I put this down to a weird bug in that ancient `perl` and just skipped that test for that particular build
 * Updated `php-Smarty` to 3.1.8:
  * Bugfix: template inheritance: `{$smarty.block.child}` in nested child `{block}` tags did not return content after `{$smarty.block.child}` (Forum Topic 20564)
  * Bugfix: the new lexer failed if `mbstring.func_overload != 0` (Issue 70, Forum Topic 20680)
  * Bugfix: template inheritance failed if `mbstring.func_overload != 0` (Issue 70, Forum Topic 20680)
  * Bugfix: `Smarty`'s internal header code must be excluded from postfilters (Issue 71)
  * Bugfix: enable more entropy in `Smarty_Internal_Write_File` for "more uniqueness" and Cygwin compatibility (Forum Topic 20724)
  * Bugfix: embedded quotes in single quoted strings did not compile correctly in `{nocache}` sections (Forum Topic 20730)
  * Bugfix: `{block foo nocache}` did not load plugins within child `{block}` in `nocache` mode (Forum Topic 20753)
  * Bugfix: typo in `Smarty_Internal_Get_IncludePath` caused runtime overhead (Issue 74)
  * Improvement: remove unneeded assigments (Issue 75 and 76)
  * Fixed typo in template parser
  * Bugfix: output filter must not run before writing cache when template contains `nocache` code (Issue 71)
  * Bugfix: `Smarty` did not handle a template name of "`0`" (Forum Topic 20895)
  * Bugfix: `Smarty_Security` internal `$_resource_dir` cache wasn't properly propagated
  * Improvement: `stream_resolve_include_path()` added to `Smarty_Internal_Get_Include_Path` (Forum Topic 20980)
  * Bugfix: `fetch('extends:foo.tpl')` always yielded `$source->exists == true` (Forum Topic 20980)
  * Added modifier `unescape:"url"`, fix (Forum Topic 20980)
  * Improvement: replaced some calls of `preg_replace` with `str_replace` (Issue 73)
  * Bugfix: increase entropy of internal function names in compiled and cached template files (Forum Topic 20996)
  * Enhancement: `cacheable` parameter added to default plugin handler, with the same functionality as `registerPlugin`
  * Bugfix: `{include}` could result in a fatal error if used in appended or prepended nested `{block}` tags (Issue 83)
  * Enhancement: added `Smarty` special variable `$smarty.template_object` to return the current template object (Forum Topic 20289)

----