Re: PEAR issues vs Github issues - open question
| From: | Alexey Borzov | Date: | Tue, 04 Mar 2014 15:01:22 +0000 |
| Subject: | Re: PEAR issues vs Github issues - open question | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-55326@lists.php.net to get a copy of this message | ||
Hi Daniel,
On 04.03.2014 18:22, Daniel O'Connor wrote:
So recently, a decent conversation picked up around PEAR core; getting it stable; etc. One comment stood out to meThe problem is not that our tracker is bad, it is that no one checks it (I'm talking about tracker for PEAR-the-package specifically). There is a lot of noise, obvious pilot errors, duplicate issues. We are also highly unlikely to migrate 100+ open issues to github, so we'll end up with two trackers that no one checks. Hardly an improvement...It was also pointed out that there was a large number of bugs that are resolved via github, but no obvious connection back to the original bug report exists. How would the community feel about increasing the use of github issue tracking, and on some or maybe even all packages; going read only for new pearweb based bug reports?I'd suggest at least documenting that PEAR lives @ github and list people having push access to the repository.
Given the fact we're fairly quiet/short handed, and the bug tracking code is a touch on the older side, this is a maintenance problem we needn't necessarily have (there is occasional spam, odd issues with html escaping bugs, etc).I'm currently scratching my head about a much bigger maintenance problem: I see 48 of 720 PEAR unit tests failing under Linux and 66 under Windows. Looks like no one bothers to run them even when merging pull requests...