|
And, SearchCVS uses the format [123456] for bug numbers in
CVS commit comments, although it can understand many
other
formats too including Mylyn's default
format:
Cheers,
--
Martin Oberhuber, Senior Member of Technical
Staff, Wind River
Target Management Project
Lead, DSDP PMC Member
Jan,
I’m
not best qualified to answer your question since it addresses administration
of bugs.eclipse.org, but I’ve been involved with similar studies and am
guessing that you will find it easiest to make an exception for how you handle
the missing bugs. They’re a tiny part of the sample, and they were
probably deleted for a reason (e.g., private information was accidentally
posted).
Fyi,
for any users of Mylyn, and commits made on the Mylyn component in Bugzilla
you will see automatically-formatted comments that explicitly associate
commits with bugs. They have the following form:
235889:
edit working sets dialog misbehaves
https://bugs.eclipse.org/bugs/show_bug.cgi?id=235889
Mik
From:
cross-project-issues-dev-bounces@xxxxxxxxxxx
[mailto:cross-project-issues-dev-bounces@xxxxxxxxxxx] On Behalf Of Jan
Vlegels
Sent: Wednesday, June 04, 2008 7:43 AM
To:
cross-project-issues-dev@xxxxxxxxxxx
Subject:
[cross-project-issues-dev] non-existing bugs that have existed
before...
My name is Jan
Vlegels and i am a researcher at the University of Antwerp (Belgium).
Currently i'm investigating existing links between the version control
system (CVS) and the bug-reporting system (bugzilla) for several projects (one
of them is eclipse)
Oftentimes
researchers assume that developers refer in the commit messages of the CVS to
the bug they fixed by mentioning its bug-report id number (f.e. "fixed bug
4512"). We all know that this is not always the case, due to the fact
that the CVS does not "enforce" this behaviour and so the reliability of this
heuristic depends on the project's internal directives and the discipline of
the developers.
What i'm doing
is trying to estimate the reliability of this heuristic, based on a case study
in several projects. Therefore I need to have the CVS and Bugzilla data
from the projects in my case study.
Concerning
Eclipse, I have a problem with the bugzilla data. At a Software
Engineering Conference (see http://msr.uwaterloo.ca/msr2007/challenge/) I
got the bug report information for bugs 1 through 162655. Because i also
need the Bug Activity tables I downloaded those from https://bugs.eclipse.org/bugs/ .
However, when I checked the completeness of the gathered data, i
encountered a problem: There are some bugs (not that much) about which https://bugs.eclipse.org/bugs/
says that they do not exist (f.e. bug 4208). However, I got their bug
reports at the Software Engineering Conference...
At the bottom of
this mail, you find the bug report id's of the "problematic" bugs. True,
it concerns of only a few bugs so it's not fundamental to proceed with my
research, but it would be nice to have a complete data-set. So, can
anyone provide me the activity logs of the bugs that "don't exist" anymore,
but clearly have existed before?
Thank you very
much for your help!
Dept.
of Mathematics and Computer Science
Office
location: CMI - G.3.04
Office
phone: +32 3 265 37 81
|
