Re: Revision tracking in phpdoc and peardoc
| From: | Philip Olson | Date: | Wed, 12 Nov 2008 02:09:38 +0000 |
| Subject: | Re: Revision tracking in phpdoc and peardoc | ||
| References: | 1 2 | Groups: | php.doc |
| Request: | Send a blank email to phpdoc+get-969379669@lists.php.net to get a copy of this message | ||
On 10 Nov 2008, at 12:32, Hannes Magnusson wrote:
On Sun, Nov 9, 2008 at 00:05, Christian Weiske <cweiske@cweiske.de> wrote:If the committer does nothing with a revision tag, then it should be assumed that the commit is translation worthy. However, if the committer feels it should not be translated (like simply typo, WS, ...), then said committer should be able to easily say so thus not outdate any translations with the commit. In other words, I prefer we lean towards the side of caution here by not requiring a manual bump. How? I'm not sure. But once we all agree on a general route then I'm guessing we could figure out how to implement. Also, I'm guessing the move to SVN will open up a few more options here too. Regards, PhilipHello all, Now that peardoc also uses PhD, we need a good method to track revisions for the english version and translations. We need a way to determine if translations are outdated, and if those outdatedness is only minor or a major outdatance. There has been a discussion about this [1], but that was over a year ago and nothing has been done yet. I personally favor the docbook revision attribute for the file's top level tag, with a major and minor rev number. Incrementing the major one determines that big/major changes have been done, while a minor number enhancement tells you that only smaller things have changed. The numbers would be updated manually instead of automatically - which has the advantage that typos can be fixed without bumping the number. When a translation differs in the major number, a big fat warning should be displayed at top of the page. In comparison to the currently used Revcheck comment in the files, the revision attributes can be read programmatically using plain XML libraries, without own parsing scripts. So what do people think now, one year after the last discussion?I still think manual bumping is the way-to-go, but I can live with automaticbumping if someone writes such a svn hook.