Mirrors/ArchiveBox
2
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox synced 2024-11-11 07:04:14 +00:00
Code Issues Projects Releases Packages Wiki Activity

Merge pull request #446 from cdvv7788/hotfix/#445

Browse source
This commit is contained in:
Nick Sweeting 2020-08-18 02:06:32 -04:00 committed by GitHub
commit 7c16944a44
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 10 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

18
archivebox/extractors/readability.py
View file

@ -62,19 +62,21 @@ def save_readability(link: Link, out_dir: Optional[str]=None, timeout: int=TIMEO
output_folder = out_dir.absolute() / "readability" output_folder = out_dir.absolute() / "readability"
output = str(output_folder) output = str(output_folder)
document = get_html(link, out_dir)
temp_doc = NamedTemporaryFile(delete=False)
temp_doc.write(document.encode("utf-8"))
temp_doc.close()
# Readability Docs: https://github.com/mozilla/readability # Readability Docs: https://github.com/mozilla/readability
cmd = [
READABILITY_BINARY,
temp_doc.name
]
status = 'succeeded' status = 'succeeded'
timer = TimedProgress(timeout, prefix=' ') timer = TimedProgress(timeout, prefix=' ')
try: try:
document = get_html(link, out_dir)
temp_doc = NamedTemporaryFile(delete=False)
temp_doc.write(document.encode("utf-8"))
temp_doc.close()
cmd = [
READABILITY_BINARY,
temp_doc.name
]
result = run(cmd, cwd=out_dir, timeout=timeout) result = run(cmd, cwd=out_dir, timeout=timeout)
result_json = json.loads(result.stdout) result_json = json.loads(result.stdout)
output_folder.mkdir(exist_ok=True) output_folder.mkdir(exist_ok=True)