[llvm-dev] Enable Contributions Through Pull-request For LLVM
David Tellenbach via llvm-dev
llvm-dev at lists.llvm.org
Fri Nov 8 08:52:42 PST 2019
> After days of trying, I gave up trying to get arcanist to work. So it's
> all manual diff + cut and paste into a browser for me. That is
> significantly worse than simply pushing to a PR.
Agreed! But I think this is mainly due to missing documentation rather than the
tool itself. As I've already mentioned we should probably add a short example
arcanist workflow to our doc.
> They're related, right? GitHub is easier because lots of people are
> already familiar with it.
Yes, certainly related since a bad tool/platform could not get as much
acceptance as Github does. However, I think the sole fact that it has a larger
user base doesn't necessarily make Github the superior tool. What I wanted to
say: Making a Phab account is in my opinion no major problem for new
contributors. Missing documentation and a hard to find/read/understand
contribution guide is (this includes a missing Phab/arc guide).
David
> On 8. Nov 2019, at 17:23, David Greene <greened at obbligato.org> wrote:
>
> David Tellenbach via llvm-dev <llvm-dev at lists.llvm.org> writes:
>
>> My -1 holds even more since I personally find the Phabricator workflow
>> to be very good in particular as it even got easier since arcanist
>> works perfectly fine with git and the complete arc workflow can be
>> used since the migration. In contrast to e.g. Bugzilla where we
>> searched for an alternative since the current situation was bad, this
>> is not the case for Phabricator.
>
> After days of trying, I gave up trying to get arcanist to work. So it's
> all manual diff + cut and paste into a browser for me. That is
> significantly worse than simply pushing to a PR.
>
>> Regarding the hurdle for new contributors: Is the Github workflow
>> really easier or is it just the case that more people have Github
>> accounts?
>
> They're related, right? GitHub is easier because lots of people are
> already familiar with it.
>
> -David
More information about the llvm-dev
mailing list