> Your changes look good, but I have a few comments:
>
> * Have we updated to the new bzr-hg that can take advantage of the tarball
> cache?
That's what this branch was doing (see 11347), but that change was reverted later when I remerged devel. I've now reintroduced it.
> * You've removed the global db_file variable, but you have the magic string
> 'hg-cache.tar.gz' in the code twice.
It's not really a magic filename like 'hg-v2.db', hg-cache.tar.gz is just the name of the temporary filename that's used to generate the tarball of the hg cache that's uploaded - similar to git-cache.tar.gz in the bzr-git worker class.
> * From my reading of the code, it looks like it is going to migrate new pushed
> branches to the tarball cache, is that the real intent?
Yes, that's correct.
Hi Steve,
Thanks for the review. :-)
> Your changes look good, but I have a few comments:
>
> * Have we updated to the new bzr-hg that can take advantage of the tarball
> cache?
That's what this branch was doing (see 11347), but that change was reverted later when I remerged devel. I've now reintroduced it.
> * You've removed the global db_file variable, but you have the magic string
> 'hg-cache.tar.gz' in the code twice.
It's not really a magic filename like 'hg-v2.db', hg-cache.tar.gz is just the name of the temporary filename that's used to generate the tarball of the hg cache that's uploaded - similar to git-cache.tar.gz in the bzr-git worker class.
> * From my reading of the code, it looks like it is going to migrate new pushed
> branches to the tarball cache, is that the real intent?
Yes, that's correct.