This is the mail archive of the
mailing list for the glibc project.
Re: Update on commit and release workflow discussions
- From: Siddhesh Poyarekar <siddhesh at redhat dot com>
- To: Joseph Myers <joseph at codesourcery dot com>
- Cc: libc-alpha at sourceware dot org, carlos at redhat dot com, roland at hack dot frob dot com
- Date: Tue, 18 Aug 2015 01:05:36 +0530
- Subject: Re: Update on commit and release workflow discussions
- Authentication-results: sourceware.org; auth=none
- References: <20150817115747 dot GC2415 at spoyarek dot pnq dot redhat dot com> <alpine dot DEB dot 2 dot 10 dot 1508171441030 dot 29836 at digraph dot polyomino dot org dot uk> <20150817171102 dot GD2415 at spoyarek dot pnq dot redhat dot com> <alpine dot DEB dot 2 dot 10 dot 1508171720590 dot 29836 at digraph dot polyomino dot org dot uk>
On Mon, Aug 17, 2015 at 05:24:40PM +0000, Joseph Myers wrote:
> Yes, that would do the job. You still need a way to avoid a bug getting
> in the list of fixed bugs if the bug number in the Resolves line was wrong
> (this includes it being a bug that was fixed long ago, as well as a bug
> that the Resolves line caused to be wrongly closed and was then reopened),
> or if the fix had to be reverted or turned out to be incomplete (without
> the bug getting fully fixed again before the release), and a way to get a
> bug in that list if someone omitted the Resolves line when committing the
> fix (or only realized later that there was a bug open for the issue).
We could do this using empty commits of the form:
Reopen bug #bznumber
or even to fix up ChangeLog entries like so:
Fix up commit id <commit id>
<the fixed up changelog>
similarly to add missing attributions.
This kind of thing may get overused though, so if this is acceptable,
then we need to be careful that it's not extended so much that these
empty commits become significantly visible.