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

add do_not_archive check to should_archive

Browse Source
This commit is contained in:
Barbara Miller 2018-02-14 15:48:21 -08:00
parent a6fa04bcae
commit 46dd01de89
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
warcprox/writerthread.py
View File

@ -81,8 +81,12 @@ class WarcWriterProcessor(warcprox.BaseStandardPostfetchProcessor):
if recorded_url.warcprox_meta if recorded_url.warcprox_meta
and 'warc-prefix' in recorded_url.warcprox_meta and 'warc-prefix' in recorded_url.warcprox_meta
else self.options.prefix) else self.options.prefix)
do_not_archive = (recorded_url.do_not_archive
if recorded_url.do_not_archive
else False)
# special warc name prefix '-' means "don't archive" # special warc name prefix '-' means "don't archive"
return prefix != '-' and self._filter_accepts(recorded_url) return prefix != '-' and (not do_not_archive) and
self._filter_accepts(recorded_url)
def _log(self, recorded_url, records): def _log(self, recorded_url, records):
try: try: