mirror/borg
1
0
Fork 0
mirror of https://github.com/borgbackup/borg.git synced 2025-02-20 21:27:32 +00:00
Code Issues Releases Wiki Activity

cache sync: fix n^2 behaviour in lookup_name

Browse source
This commit is contained in:
Marian Beermann 2017-05-26 13:54:28 +02:00
parent 9f8b967a6f
commit 167875b753
1 changed files with 11 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
src/borg/cache.py
View file

@ -588,10 +588,16 @@ def fetch_and_build_idx(archive_id, repository, key, chunk_idx):
else:
os.rename(fn_tmp, fn)
def lookup_name(archive_id):
def get_archive_ids_to_names(archive_ids):
# Pass once over all archives and build a mapping from ids to names.
# The easier approach, doing a similar loop for each archive, has
# square complexity and does about a dozen million functions calls
# with 1100 archives (which takes 30s CPU seconds _alone_).
archive_names = {}
for info in self.manifest.archives.list():
if info.id == archive_id:
return info.name
if info.id in archive_ids:
archive_names[info.id] = info.name
return archive_names
def create_master_idx(chunk_idx):
logger.info('Synchronizing chunks cache...')
@ -612,8 +618,9 @@ def create_master_idx(chunk_idx):
pi = ProgressIndicatorPercent(total=len(archive_ids), step=0.1,
msg='%3.0f%% Syncing chunks cache. Processing archive %s',
msgid='cache.sync')
archive_ids_to_names = get_archive_ids_to_names(archive_ids)
for archive_id in archive_ids:
archive_name = lookup_name(archive_id)
archive_name = archive_ids_to_names.pop(archive_id)
if self.progress:
pi.show(info=[remove_surrogates(archive_name)])
if self.do_cache: