mirror/archivebox
1
0
Fork 0
mirror of https://github.com/pirate/ArchiveBox.git synced 2025-08-23 22:53:00 +02:00
Code Issues Releases Wiki Activity

log wget 404 and 403 errors

Browse Source
This commit is contained in:
Nick Sweeting
2018-06-10 21:13:07 -04:00
parent a287900345
commit c1c689cb94
1 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
archiver/archive_methods.py
View File

@@ -208,7 +208,13 @@ def fetch_wget(link_dir, link, requisites=FETCH_WGET_REQUISITES, timeout=TIMEOUT
print(' got wget response code {}:'.format(result.returncode)) print(' got wget response code {}:'.format(result.returncode))
print('\n'.join(' ' + line for line in (result.stderr or result.stdout).decode().rsplit('\n', 10)[-10:] if line.strip())) print('\n'.join(' ' + line for line in (result.stderr or result.stdout).decode().rsplit('\n', 10)[-10:] if line.strip()))
if result.returncode == 4: if result.returncode == 4:
raise Exception('Failed to wget download') raise Exception('Failed wget download')
if result.returncode > 0 and b'403: Forbidden' in result.stderr:
raise Exception('403 Forbidden (try changing WGET_USER_AGENT)')
if result.returncode > 0 and b'404: Not Found' in result.stderr:
raise Exception('404 Not Found')
except Exception as e: except Exception as e:
end() end()
print(' Run to see full output:', 'cd {}; {}'.format(link_dir, ' '.join(CMD))) print(' Run to see full output:', 'cd {}; {}'.format(link_dir, ' '.join(CMD)))