← Back to branch summary

~bzr-pqm/bzr/bzr.dev

« back to all changes in this revision

Viewing changes to bzrlib/fetch.py

  • Committer: Canonical.com Patch Queue Manager
  • Date: 2006-06-08 20:01:35 UTC
  • mfrom: (1753.1.2 branch)
  • Revision ID: pqm@pqm.ubuntu.com-20060608200135-842f7645e1e456b6
(robertc, j-a-meinel)Replacement xml format 5 serializer which is much more efficient - saves 0.6 seconds on add on 10K trees. (robertc, ab, jam)Fix memory usage on bzr branch.

Show diffs side-by-side

added added

removed removed

Lines of Context:
bzrlib/fetch.py
169
169
                    self.from_repository.get_transaction())
170
170
                # we fetch all the texts, because texts do
171
171
                # not reference anything, and its cheap enough
172
 
                to_weave.join(from_weave, version_ids=required_versions)
 
172
                to_weave.join(from_weave, version_ids=required_versions) 
 
173
                # we dont need *all* of this data anymore, but we dont know
 
174
                # what we do. This cache clearing will result in a new read 
 
175
                # of the knit data when we do the checkout, but probably we
 
176
                # want to emit the needed data on the fly rather than at the
 
177
                # end anyhow.
 
178
                # the from weave should know not to cache data being joined,
 
179
                # but its ok to ask it to clear.
 
180
                from_weave.clear_cache()
 
181
                to_weave.clear_cache()
173
182
        finally:
174
183
            texts_pb.finished()
175
184