| 1 |
On 5/22/11 10:13 AM, Thomas Kahle wrote: |
| 2 |
> I like it. Funnily It digged up some stable bugs from the stone-age |
| 3 |
> that have been processed, but x86 was still on the CC-list. |
| 4 |
|
| 5 |
Thanks. Those old bugs being displayed was a bug, I fixed it. |
| 6 |
|
| 7 |
> Have you seen app-portage/tatt ? I think there is a huge overlap between |
| 8 |
> your project and tatt. |
| 9 |
|
| 10 |
Yeah, the project was in fact inspired by tatt and the earlier gatt. |
| 11 |
|
| 12 |
> tatt can already fetch bugs from bugzilla, |
| 13 |
> create USE-flags and reverse dependency, as well as cvs commit scripts. |
| 14 |
> You can use your own templates for the scripts, but tatt also provides |
| 15 |
> some. |
| 16 |
|
| 17 |
One of the main things I wanted was being able to see all bug comments, |
| 18 |
bug dependencies, related bugs, and repoman output in the bugzilla |
| 19 |
viewer. Can tatt do that? |
| 20 |
|
| 21 |
I think I'm going to either re-use or copy tatt's parts responsible for |
| 22 |
compiling and committing the changes. For now I just focused on the |
| 23 |
first part of the workflow (identifying bugs to work on, and identifying |
| 24 |
possible problems as early as possible). |
| 25 |
|
| 26 |
> Regarding the work-order I intentionally try to be agnostic. I do |
| 27 |
> security bugs first, and then work on the oldest bug. |
| 28 |
|
| 29 |
Right, but sometimes there are bugs with no input from the maintainer. |
| 30 |
Another thing I'd like to do is to mark such a bug as ignored, or even |
| 31 |
"ignored until the next update" (e.g. when someone replies on that bug). |
Archives