lsprofcalltree.py @@ -30,6 +30,7 @@ WEBSITE SPEED/EFFICIENCY ================ +* cache rename detection on push (seems to be very expensive) * switch object mapping to hg->git since the many to one is that direction * don't send blobs/trees already on server (thin pack) * packfile creation benchmarking (seems to take a while sometimes) TODO.txt @@ -49,14 +49,19 @@ def gclone(ui, git_url, hg_repo_path=None): def gpush(ui, repo, remote_name='origin', branch=None): git = GitHandler(repo, ui) import cProfile, pstats + import lsprofcalltree prof = cProfile.Profile() prof = prof.runctx("git.push(remote_name)", globals(), locals()) stats = pstats.Stats(prof) + k = lsprofcalltree.KCacheGrind(prof) + data = open('/tmp/prof.kgrind', 'w+') + k.output(data) + data.close() stats.sort_stats("time") # Or cumulative stats.print_stats(80) # 80 = how many to print # The rest is optional. - # stats.print_callees() - # stats.print_callers() + #stats.print_callees() + #stats.print_callers() def gimport(ui, repo, remote_name=None): git = GitHandler(repo, ui) __init__.py @@ -187,6 +187,7 @@ class GitHandler(object): return dict(filter(is_local_head, refs.items())) def export_git_objects(self): + self.manifest_renames = {} self.ui.status(_("importing Hg objects into Git\n")) total = len(self.repo.changelog) if total: @@ -331,16 +332,24 @@ class GitHandler(object): trees = {} man = ctx.manifest() renames = [] - for filenm in man.keys(): + for filenm, nodesha in man.iteritems(): + file_id = hex(nodesha) # write blob if not in our git database - fctx = ctx.filectx(filenm) - rename = fctx.renamed() - if rename: - filerename, sha = rename + fctx = ctx.filectx(filenm) + filerename = None + if file_id in self.manifest_renames: + filerename = self.manifest_renames[file_id] + else: + rename = fctx.renamed() + if rename: + filerename, sha = rename + self.manifest_renames[file_id] = filerename + else: + self.manifest_renames[file_id] = None + if filerename: renames.append((filerename, filenm)) is_exec = 'x' in fctx.flags() is_link = 'l' in fctx.flags() - file_id = hex(fctx.filenode()) blob_sha = self.map_git_get(file_id) if not blob_sha: blob_sha = self.git.write_blob(fctx.data()) # writing new blobs to git git_handler.py 38d071d67c0d2e9a575b50cafe8b23a406a5ca23 Scott Chacon schacon@gmail.com http://github.com/schacon/hg-git/commit/18e6d096c5a0711eba644153e631700c6bfcfb5d 18e6d096c5a0711eba644153e631700c6bfcfb5d 2009-06-03T09:22:27-07:00 2009-06-03T09:22:17-07:00 fixed serious speed issue with rename detection d5a6fca748da64cb1e7c1f0aa7eca0f3ebcc324f Scott Chacon schacon@gmail.com