> I end up doing this every time. I'll put a branch in ec2 that calls
> flush_database_updates() in StormStatementRecorder.__enter__().
Good idea ;)
> [2]
>
> + build_farm_jobs.extend([build.build_farm_job for build in builds2])
>
> No need for the intermediate list, though I guess it does little harm.
Okay.
> [3]
>
> + # rvb 2011-11-11: Each build issues 2 new queries.
> + # This is because of the way SourcePackageRecipe._recipe_data
> + # fetches SourcePackageRecipeData. I've no idea how to prefetch
> + # this.
>
> I guess this is work in progress, but should probably be XXX'd
> anyway. What do you think?
I'll keep it as is but I'll work on a branch to fix this right away.
Thanks for the review.
> I end up doing this every time. I'll put a branch in ec2 that calls updates( ) in StormStatementR ecorder. __enter_ _().
> flush_database_
Good idea ;)
> [2] jobs.extend( [build. build_farm_ job for build in builds2])
>
> + build_farm_
>
> No need for the intermediate list, though I guess it does little harm.
Okay.
> [3] cipe._recipe_ data cipeData. I've no idea how to prefetch
>
> + # rvb 2011-11-11: Each build issues 2 new queries.
> + # This is because of the way SourcePackageRe
> + # fetches SourcePackageRe
> + # this.
>
> I guess this is work in progress, but should probably be XXX'd
> anyway. What do you think?
I'll keep it as is but I'll work on a branch to fix this right away.