mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-14 07:04:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

fetch warc file inline with wget instead of as separate step

Browse source
This commit is contained in:
Nick Sweeting 2019-01-14 22:40:55 -05:00
parent 2d86d4dfe0
commit c42fcd42d7
2 changed files with 12 additions and 57 deletions
Download patch file Download diff file Expand all files Collapse all files

4
archivebox/config.py
View file

@ -23,7 +23,7 @@ FETCH_WGET_REQUISITES = os.getenv('FETCH_WGET_REQUISITES', 'True'
FETCH_PDF = os.getenv('FETCH_PDF', 'True' ).lower() == 'true'
FETCH_SCREENSHOT = os.getenv('FETCH_SCREENSHOT', 'True' ).lower() == 'true'
FETCH_DOM = os.getenv('FETCH_DOM', 'True' ).lower() == 'true'
FETCH_WARC = os.getenv('FETCH_WARC', 'False' ).lower() == 'true'
FETCH_WARC = os.getenv('FETCH_WARC', 'True' ).lower() == 'true'
FETCH_GIT = os.getenv('FETCH_GIT', 'True' ).lower() == 'true'
FETCH_MEDIA = os.getenv('FETCH_MEDIA', 'False' ).lower() == 'true'
FETCH_FAVICON = os.getenv('FETCH_FAVICON', 'True' ).lower() == 'true'
@ -42,7 +42,7 @@ FOOTER_INFO = os.getenv('FOOTER_INFO', 'Content is hosted
### Paths
REPO_DIR = os.path.abspath(os.path.join(os.path.dirname(os.path.abspath(__file__)), '..'))
OUTPUT_DIR = os.getenv('OUTPUT_DIR', os.path.join(REPO_DIR, 'output'))
OUTPUT_DIR = os.path.abspath(os.getenv('OUTPUT_DIR', os.path.join(REPO_DIR, 'output')))
ARCHIVE_DIR = os.path.join(OUTPUT_DIR, 'archive')
SOURCES_DIR = os.path.join(OUTPUT_DIR, 'sources')