On Mon, Feb 12, 2018 at 6:09 AM, Boris Zbarsky <bzbar...@mit.edu> wrote:
> On 2/11/18 3:57 PM, Emilio Cobos Álvarez wrote:
>> Arc wants to use something like:
> So from my point of view, having the bug# easily linked from various
> places where the short summary is all that's shown (pushlogs especially) is
> pretty useful. It saves loading a bunch of extra things when trying to go
> from regression-range pushlogs to the relevant bugs....
It's generally pretty easy to modify tooling to find linked bugs. We have a
shared Python module for parsing commit messages into useful metadata and
that's used by various tools for extracting bugs, reviewers, so they can be
rendered in various places (
And updating templating on hg.mozilla.org to render useful fields more
prominently is generally pretty turnkey. Please file bugs in the
hg.mozilla.org component if you want the display of things tweaked!
More to the point of the original question, there are several reasons why
we don't want to use Arcanist (`arc`) for Firefox development (and probably
more broadly at Mozilla). The initial comment in bug 1366401 records a lot
of them. The plan of record is to author a minimal client for submitting
reviews via Phabricator's HTTP API and to plumb that up to `mach` as
needed. This work isn't part of the Phabricator MVP, which is why the
Phabricator team hasn't worked on it.
If you use Mercurial, there is an alternative to Arcanist available today!
You can combine Mercurial's "phabricator" extension with a minimal wrapper
that Tom Prince wrote so it recognizes Mozilla's bug numbers and reviewer
annotations. Instructions are at
https://bugzilla.mozilla.org/show_bug.cgi?id=1366401#c4. This will likely
serve as the base for the eventual solution. It is probably less effort to
configure this than to install Arcanist. It is unsupported, but I think it
is better than using Arcanist.
Git users will likely have to wait until after the Phabricator MVP is
launched before there is staffing to work on a client. We kind of lucked
out that Mercurial had something that was almost drop-in ready for us to
consume and that is why there is a Mercurial alternative (i.e. this isn't
about prioritizing Mercurial over Git).
While I'm not working on either client implementation and am not part of
the Phabricator team, if someone wants to formalize the Mercurial or Git
clients in version-control-tools before the Phabricator team has time to
work on them, I'd be happy to review code or provide technical assistance.
Track things against bug 1366401 if you do any work.
dev-platform mailing list