mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-15 07:34:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix broken wget user agents

Browse source
This commit is contained in:
Nick Sweeting 2019-02-04 20:02:48 -08:00
parent d8c7c98c01
commit 4e5b466bb2
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
archivebox/archive_methods.py
View file

@ -222,7 +222,7 @@ def fetch_wget(link_dir, link, requisites=FETCH_WGET_REQUISITES, warc=FETCH_WARC
*(() if warc else ('--timestamping',)), *(() if warc else ('--timestamping',)),
*(('--warc-file={}'.format(warc_path),) if warc else ()), *(('--warc-file={}'.format(warc_path),) if warc else ()),
*(('--page-requisites',) if FETCH_WGET_REQUISITES else ()), *(('--page-requisites',) if FETCH_WGET_REQUISITES else ()),
*(('--user-agent="{}"'.format(WGET_USER_AGENT),) if WGET_USER_AGENT else ()), *(('--user-agent={}'.format(WGET_USER_AGENT),) if WGET_USER_AGENT else ()),
*((() if CHECK_SSL_VALIDITY else ('--no-check-certificate',))), *((() if CHECK_SSL_VALIDITY else ('--no-check-certificate',))),
link['url'], link['url'],
] ]
@ -257,7 +257,7 @@ def fetch_wget(link_dir, link, requisites=FETCH_WGET_REQUISITES, warc=FETCH_WARC
print(' {}Some resources were skipped: {}{}'.format(ANSI['lightyellow'], e, ANSI['reset'])) print(' {}Some resources were skipped: {}{}'.format(ANSI['lightyellow'], e, ANSI['reset']))
print(' Run to see full output:') print(' Run to see full output:')
print(' cd {};'.format(link_dir)) print(' cd {};'.format(link_dir))
print(' {}'.format(' '.join(CMD))) print(' {}'.format(' '.join(CMD).replace(WGET_USER_AGENT, '"{}"'.format(WGET_USER_AGENT))))
output = e output = e
return { return {