backup/warcprox
1
0
Fork 0
mirror of https://github.com/internetarchive/warcprox.git synced 2025-01-18 13:22:09 +01:00
Code Issues Projects Releases Activity

Merge branch 'WT-31' into qa

Browse Source
This commit is contained in:
Barbara Miller 2021-12-09 12:20:09 -08:00
commit 6e65b5ff55
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
warcprox/dedup.py
View File

@ -435,14 +435,14 @@ class BatchTroughLoader(warcprox.BaseBatchPostfetchProcessor):
else:
if hash_plus_url in hash_plus_urls:
self.logger.debug(
'discarding duplicate {}, setting do_not_archive'.format(hash_plus_url))
'discarding duplicate and setting do_not_archive for %, hash %'.format(
recorded_url.url, warcprox.digest_str(
recorded_url.payload_digest, self.options.base32)))
recorded_url.do_not_archive = True
discards.append(
warcprox.digest_str(
recorded_url.payload_digest, self.options.base32)
if recorded_url.payload_digest else 'n/a')
self.logger.debug(
'hash_plus_urls: {}'.format(len(hash_plus_urls)))
self.logger.debug(
'len(batch)=%s len(discards)=%s buckets=%s',
len(batch), len(discards),