llvm dev - Nov 2021 - IMPORTANT: LLVM Bugzilla migration

On Tue, Nov 23, 2021 at 11:39 AM Anton Korobeynikov
<anton at korobeynikov.info> wrote:
>
> > 1) GitHub's search is pretty abysmal, but I was not expecting it
to be
> > this level of problematic. Doing a search for something like
> > "attribute" to see what kind of attribute-related bugs are
open is
> > trivial in Bugzilla and utterly useless in GitHub. I'm guessing
this
> > wasn't expected, but I'm wondering if there's anything to
be done
> > about it?
> We can try to report this issue to github folks and see what could be
> done. Maybe it's a matter of some text index to be rebuilt.
Thanks! I'm hopeful that we can find a solution because this is almost
a show-stopper for me. I've spent several weeks going through Bugzilla
to triage C++20-specific issues and as best I can tell, GitHub would
be unable to accomplish this task *at all* (because many bugs were
reported before we had a tag for users to apply and not all bugs are
things that we had bugzilla tags for anyway -- we need a searchable
bug database). So hopefully it's a matter of rebuilding some text
index or something straightforward like that.
> > 2) It seems that some user accounts are being created as mannequins
> > rather than being hooked up to the actual user. For example,
> > https://github.com/llvm/test8/issues/37237#issuecomment-973565691 is a
> > comment by zygoloid and it is marked as "mannequin" rather
than
> > linking to https://github.com/zygoloid. Is that expected behavior or
> > does this indicate some of the data wasn't migrated properly?
> "mannequins" are those who are not members of LLVM organization
(they
> will be treated separately after the migration).
> It was also a surprise to me that Richard does not have LLVM GH repo
> commit access :)
Oh, interesting. Thanks for the explanation

~Aaron
>
> --
> With best regards, Anton Korobeynikov
> Department of Statistical Modelling, Saint Petersburg State University

On Tue, Nov 23, 2021 at 11:50 AM Aaron Ballman via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
> On Tue, Nov 23, 2021 at 11:39 AM Anton Korobeynikov
> <anton at korobeynikov.info> wrote:
> >
> > > 1) GitHub's search is pretty abysmal, but I was not expecting
it to be
> > > this level of problematic. Doing a search for something like
> > > "attribute" to see what kind of attribute-related bugs
are open is
> > > trivial in Bugzilla and utterly useless in GitHub. I'm
guessing this
> > > wasn't expected, but I'm wondering if there's
anything to be done
> > > about it?
> > We can try to report this issue to github folks and see what could be
> > done. Maybe it's a matter of some text index to be rebuilt.
>
> Thanks! I'm hopeful that we can find a solution because this is almost
> a show-stopper for me. I've spent several weeks going through Bugzilla
> to triage C++20-specific issues and as best I can tell, GitHub would
> be unable to accomplish this task *at all* (because many bugs were
> reported before we had a tag for users to apply and not all bugs are
> things that we had bugzilla tags for anyway -- we need a searchable
> bug database). So hopefully it's a matter of rebuilding some text
> index or something straightforward like that.
>
> > > 2) It seems that some user accounts are being created as
mannequins
> > > rather than being hooked up to the actual user. For example,
> > > https://github.com/llvm/test8/issues/37237#issuecomment-973565691
is a
> > > comment by zygoloid and it is marked as "mannequin"
rather than
> > > linking to https://github.com/zygoloid. Is that expected behavior
or
> > > does this indicate some of the data wasn't migrated properly?
> > "mannequins" are those who are not members of LLVM
organization (they
> > will be treated separately after the migration).
> > It was also a surprise to me that Richard does not have LLVM GH repo
> > commit access :)
>
> Oh, interesting. Thanks for the explanation
>
Richard is making ongoing commits to LLVM Github which are properly
attributed to his GH account, though:
https://github.com/llvm/llvm-project/commits?author=zygoloid
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20211123/18deeea6/attachment.html>

On Tue, Nov 23, 2021 at 8:50 AM Aaron Ballman via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
> On Tue, Nov 23, 2021 at 11:39 AM Anton Korobeynikov
> <anton at korobeynikov.info> wrote:
> >
> > > 1) GitHub's search is pretty abysmal, but I was not expecting
it to be
> > > this level of problematic. Doing a search for something like
> > > "attribute" to see what kind of attribute-related bugs
are open is
> > > trivial in Bugzilla and utterly useless in GitHub. I'm
guessing this
> > > wasn't expected, but I'm wondering if there's
anything to be done
> > > about it?
> > We can try to report this issue to github folks and see what could be
> > done. Maybe it's a matter of some text index to be rebuilt.
>
> Thanks! I'm hopeful that we can find a solution because this is almost
> a show-stopper for me. I've spent several weeks going through Bugzilla
> to triage C++20-specific issues and as best I can tell, GitHub would
> be unable to accomplish this task *at all* (because many bugs were
> reported before we had a tag for users to apply and not all bugs are
> things that we had bugzilla tags for anyway -- we need a searchable
> bug database). So hopefully it's a matter of rebuilding some text
> index or something straightforward like that.
>
> > > 2) It seems that some user accounts are being created as
mannequins
> > > rather than being hooked up to the actual user. For example,
> > > https://github.com/llvm/test8/issues/37237#issuecomment-973565691
is a
> > > comment by zygoloid and it is marked as "mannequin"
rather than
> > > linking to https://github.com/zygoloid. Is that expected behavior
or
> > > does this indicate some of the data wasn't migrated properly?
> > "mannequins" are those who are not members of LLVM
organization (they
> > will be treated separately after the migration).
> > It was also a surprise to me that Richard does not have LLVM GH repo
> > commit access :)
>
> Oh, interesting. Thanks for the explanation
>
It looks like Richard Smith, under zygoloid, does have LLVM GH repo commit
access, though? eg:
https://github.com/llvm/llvm-project/commits?author=zygoloid

So it seems maybe there are people with commit access but are not part of
the "LLVM organization" on github?

>
> ~Aaron
>
> >
> > --
> > With best regards, Anton Korobeynikov
> > Department of Statistical Modelling, Saint Petersburg State University
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20211123/ca3404bf/attachment.html>

The issue was reported to GitHub as well.

On Tue, Nov 23, 2021 at 7:50 PM Aaron Ballman <aaron at aaronballman.com>
wrote:
>
> On Tue, Nov 23, 2021 at 11:39 AM Anton Korobeynikov
> <anton at korobeynikov.info> wrote:
> >
> > > 1) GitHub's search is pretty abysmal, but I was not expecting
it to be
> > > this level of problematic. Doing a search for something like
> > > "attribute" to see what kind of attribute-related bugs
are open is
> > > trivial in Bugzilla and utterly useless in GitHub. I'm
guessing this
> > > wasn't expected, but I'm wondering if there's
anything to be done
> > > about it?
> > We can try to report this issue to github folks and see what could be
> > done. Maybe it's a matter of some text index to be rebuilt.
>
> Thanks! I'm hopeful that we can find a solution because this is almost
> a show-stopper for me. I've spent several weeks going through Bugzilla
> to triage C++20-specific issues and as best I can tell, GitHub would
> be unable to accomplish this task *at all* (because many bugs were
> reported before we had a tag for users to apply and not all bugs are
> things that we had bugzilla tags for anyway -- we need a searchable
> bug database). So hopefully it's a matter of rebuilding some text
> index or something straightforward like that.
>
> > > 2) It seems that some user accounts are being created as
mannequins
> > > rather than being hooked up to the actual user. For example,
> > > https://github.com/llvm/test8/issues/37237#issuecomment-973565691
is a
> > > comment by zygoloid and it is marked as "mannequin"
rather than
> > > linking to https://github.com/zygoloid. Is that expected behavior
or
> > > does this indicate some of the data wasn't migrated properly?
> > "mannequins" are those who are not members of LLVM
organization (they
> > will be treated separately after the migration).
> > It was also a surprise to me that Richard does not have LLVM GH repo
> > commit access :)
>
> Oh, interesting. Thanks for the explanation
>
> ~Aaron
>
> >
> > --
> > With best regards, Anton Korobeynikov
> > Department of Statistical Modelling, Saint Petersburg State University


-- 
With best regards, Anton Korobeynikov
Department of Statistical Modelling, Saint Petersburg State University