TPF Devel::Cover grant report Week 9
In accordance with the terms of my grant from TPF this is the report for week 9 of my work on improving Devel::Cover. This report covers 14.07 to 20.07. This week I released Devel::Cover 0.91 and 0.92. I made a change I've been considering for quite some time, which is to get the @INC directories from perl itself. At build time we put the @INC directories in @Devel::Cover::Inc::Inc so that we will be able to ignore them by default at runtime. That works well normally, but sometimes the build environment differs from the run environment which can result in default coverage of core modules. The solution is to shell out to perl and ask it what its @INC is. Of course it's not quite that easy. And shelling out isn't very nice, but when the new V module for introspection is in place we can use that instead. I knew this was a risky commit, but cpantesters didn't show any problems. Nevertheless, I had broken things for people using local::lib. Thankfully, Olivier Mengué noticed and patched it up for me. I also fixed up a change I made last week which made all the tests fail on Windows. Devel::Cover itself still worked, but almost all of the tests didn't. Christian Walde's cpantesters smoker detected the problem quickly and we diagnosed the problem together on IRC. So I eventually got around to installing Perl on Windows (which turned out to be pretty simple) and then it was just a matter of getting the quoting such that both *nix and Windows were happy, and hoping that other platforms work too. (On which note, does anyone use Devel::Cover on anything other than *nix and Windows?) As usual, I made other various fixes and updates. Closed RT tickets: 68991 t/e2e epic fail when prefix for perl is not the physical location 76528 Devel/Cover/Inc.pm contains temporary install destination DESTDIR 66881 Symlinked directories can cause ignore functionality to fail 37349 /usr/local/* ommitted in Inc 32345 Devel::Cover 0.63 fails 1 test under Perl 5.8.7, Ubuntu 6.06 Closed Github tickets: 25 Fix inc directories when working with local::lib Merged pull requests: 25 Fix inc directories when working with local::lib Fixed cpantesters reports: http://www.cpantesters.org/cpan/report/251a6589-6ef1-1014-92c1-140f3373d4b5 http://www.cpantesters.org/cpan/report/d50bb259-6bf5-1014-86c6-f8fbafede0e8 http://www.cpantesters.org/cpan/report/675f887f-6fa2-1014-9b4e-081b3373d4b5 You can see the commits at https://github.com/pjcj/Devel--Cover/commits/master Hours worked: 14.07 03:20 16.07 04:25 20.07 01:15 Total 09:00 -- Paul Johnson - p...@pjcj.net http://www.pjcj.net
TPF Devel::Cover grant report Week 10
In accordance with the terms of my grant from TPF this is the report for week 10 of my work on improving Devel::Cover. This report covers 21.07 to 27.07. I wasn't able to do any grant related work this week. -- Paul Johnson - p...@pjcj.net http://www.pjcj.net
TPF Devel::Cover grant report July 2012
In accordance with the terms of my grant from TPF this is the monthly report for my work on improving Devel::Cover covering July 2012. This month I released Devel::Cover 0.90, 0.91 and 0.92. I wasn't able to spend as much time working on Devel::Cover this month as I had hoped to. The time I did spend was primarily on working through RT and github tickets, resolving those I could and merging pull requests. RT 62423 concerns Devel::Cover failing its tests when it is built in a directory containing spaces, which appears to be the default when building with cpan on OSX. I couldn't build on OSX, but I did fix things up so that Devel::Cover should now work in directories which contain spaces. (But if you prefer your life to be easier, i suggest you don't do that.) I made a change I've been considering for quite some time, which is to get the @INC directories from perl itself. At build time we put the @INC directories in @Devel::Cover::Inc::Inc so that we will be able to ignore them by default at runtime. That works well normally, but sometimes the build environment differs from the run environment which can result in default coverage of core modules. The solution is to shell out to perl and ask it what its @INC is. Of course it's not quite that easy. And shelling out isn't very nice, but when the new V module for introspection is in place we can use that instead. I knew this was a risky commit, but cpantesters didn't show any problems. Nevertheless, I had broken things for people using local::lib. Thankfully, Olivier Mengué noticed and patched it up for me. I also fixed up a n earlier change I had made which made all the tests fail on Windows. Devel::Cover itself still worked, but almost all of the tests didn't. Christian Walde's cpantesters smoker detected the problem quickly and we diagnosed the problem together on IRC. So I eventually got around to installing Perl on Windows (which turned out to be pretty simple) and then it was just a matter of getting the quoting such that both *nix and Windows were happy, and hoping that other platforms work too. (On which note, does anyone use Devel::Cover on anything other than *nix and Windows?) As usual, I made other various fixes and updates. The work I have completed in the time covered by this report is: Closed RT tickets: 75565 cover -test fails Writing HTML output on Windows 62423 Devel::Cover 0.73 tests fail due to unescaped path 57173 Devel::Cover 0.66 is causing Moose constraint failures 70115 malformed JSON string errors when gcov2perl is ran for XS distributions 39162 Devel::Cover test 't/at2.t' failure 38765 1 test fails (acond_branch.t) on Cygwin 1.5.25 / Perl 5.10 38453 Devel::Cover - Use of uninitialized value in sprintf 68991 t/e2e epic fail when prefix for perl is not the physical location 76528 Devel/Cover/Inc.pm contains temporary install destination DESTDIR 66881 Symlinked directories can cause ignore functionality to fail 37349 /usr/local/* ommitted in Inc 32345 Devel::Cover 0.63 fails 1 test under Perl 5.8.7, Ubuntu 6.06 Closed Github tickets: 23 RE list with spaces doesn't work 24 protect slashes on windows 25 Fix inc directories when working with local::lib Merged pull requests: 24 protect slashes on windows 25 Fix inc directories when working with local::lib Fixed cpantesters reports: http://www.cpantesters.org/cpan/report/251a6589-6ef1-1014-92c1-140f3373d4b5 http://www.cpantesters.org/cpan/report/d50bb259-6bf5-1014-86c6-f8fbafede0e8 http://www.cpantesters.org/cpan/report/675f887f-6fa2-1014-9b4e-081b3373d4b5 You can see the commits at https://github.com/pjcj/Devel--Cover/commits/master Hours worked: 25:20 Total hours works on grant: 88:50 -- Paul Johnson - p...@pjcj.net http://www.pjcj.net
Re: version.pm and developer version numbers
On 2012.7.27 12:48 PM, Jeffrey Thalhammer wrote: I just discovered that version.pm always treats version numbers with an underscore as less than the equivalent version number without the underscore. So 6.63_02 is less than 6.6302. Is it it just me, or does that seem crazy? Dealing with $VERSION numbers in Perl is hard enough as it is. This doesn't help. IMO it makes sense. 6.6302 and 6.63_02 are not the same thing. One is stable, one is alpha. One is version 6.6302, one is 6.63 alpha 2. The alpha version should always be less than an equivalent stable version. What's really going on is this: 6.63_02 = 6.63_2 6.6302 = 6.6302.0 And 6.63 6.6302. I know this gets in the way because people like to collapse alpha version numbers to avoid a warning when they're compared as numbers. Ironically the very thing people do which trips up version.pm is unnecessary with version.pm. -- Life is like a sewer - what you get out of it depends on what you put into it. - Tom Lehrer
Re: TPF Devel::Cover grant report July 2012
* Paul Johnson p...@pjcj.net [2012-08-01T11:48:43] In accordance with the terms of my grant from TPF this is the monthly report for my work on improving Devel::Cover covering July 2012. Thanks Paul, +1. -- rjbs signature.asc Description: Digital signature
