[Piuparts-devel] short review on daves branches
Dave Steele
dsteele at gmail.com
Sat May 26 14:54:05 UTC 2012
On Sat, May 26, 2012 at 9:56 AM, Andreas Beckmann <debian at abeckmann.de> wrote:
> * remotes/dave/purge_order
> That is wrong. We already do what you want to achieve (according to the
> changelog):
> - purge the dependencies (deps_to_purge)
> - purge the package under test afterwards (nondeps_to_purge)
> feel free to rename the variables to something understandable :-)
> (and yes, we want to keep the "insane" purge order:
> dependencies first, package-under-test last)
>
OK. My main driver in pursuing this in the short term is to drive down
the dependency-failed counts in a defensible fashion. Either #672769
or your exclude list could accomplish that.
> * remotes/dave/report-performance
> * 5e62551a8ed replace O(n^2) search in remove_old_logs with a hash
> - nice catch, although I don't see this as a big time hog, e.g.
Granted ... but there is another, far worse, one
> - the big time eater is "Writing package templates"
> that's writing 17000 files per section
That's where the far worse one is :-)
>
> * remotes/dave/rrdeps
> I'm curious, cherry-picked and tested it :)
...
> - should this be done for dependency-failed-testing,
> dependency-does-not-exist and dependency-cannot-be-tested, too?
> - the numbers should be separated by ', '
>
rrdeps was a proof-of-concept - more about infrastructure than
interface, though it sounds like the display is close to good enough.
I'll look at making the changes you describe (the
dependency-does-not-exist case is problematic).
> Oops: this happened:
> File "/org/piuparts.debian.org/share/piuparts/piuparts-report", line
> 1126, in write_state_pages
> "aside": aside,
> UnboundLocalError: local variable 'aside' referenced before assignment
>
Shouldn't be possible. 'aside' is defined in both paths of an if
statement above the error. Could this be a merge issue?
Mmm, unless there are no packages. I'll fix.
More information about the Piuparts-devel
mailing list