Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
RE: [] IP Logs for Ganymede

Thanks Bjorn.
The basic idea looks good, and I'll happily have our manually generated
tm-log.csv file cross checked against your automatically generated
version. Our bugzilla input data should be good, since we've been
using bugzilla queries with the "contributed" keyword already to
cross-check our manually maintained log against bugzilla.
I think I'd like to stay in bucket 3 ("Stay the Course") for now since
I'm very confident that our manually maintained and cross-checked
logs are good. But please do verify them with your generator.
FYI, here's another corner case that we have had to deal with:
where a contributor submits some patches, becomes a committer
afterwards, and then applies his own patches. In my understanding,
these contributions must be logged because they were made before
the contributor became committer, so they are under different
legal terms.
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member

From: [] On Behalf Of Bjorn Freeman-Benson
Sent: Freitag, 16. Mai 2008 02:47
Cc: cross-project-issues-dev-bounces@xxxxxxxxxxx
Subject: Re: [] IP Logs for Ganymede

Martin (and everyone),
Can you share a little bit of technical detail, e.g. based on what information
you're planning to have the automatic IP Log generated? Are you going to
use Bugzilla's "attachment isPatch" field together with a check whether the
attachment has been created by a non-committer, and the bugzilla FIXED state?
Are you going to take the bugzilla "contributed" keyword into account?
Based on the feedback on bug 220977, yes, I will :-)
Note that we have bugs where contributors attached patches, but the final
fix was created by a committer without using the contribution. I'm really
curios how you're planning to sort these things out -- personally, as well
as on behalf of my project, since I want to know whether we'll be in
bucket 2 or bucket 3.
As with any automated system, the results will only be as good as the data quality going in. Or something like that. I'm open to all suggestions about how to handle the various corner cases, but for your specific question, my idea was that the unused patch would be marked obsolete and thus be ignored by the automatic IP Log generator.

- Bjorn

[end of message]

Back to the top