bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-12 13:18:28 +03:00
Code Issues Packages Projects Releases Wiki Activity

Tidy ups + unit tests:

Browse Source 
1. Allow loading modules from --module_paths=/extra/path/here
2. Improved unit tests for module loading
3. Further small tidy ups/clean ups
This commit is contained in:
Patrick Robertson
2025-01-29 18:42:12 +01:00
parent dcd5576f29
commit 3d37c494aa
13 changed files with 216 additions and 81 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/auto_archiver/modules/html_formatter/html_formatter.py
View File

@@ -12,7 +12,7 @@ from auto_archiver.core import Metadata, Media, ArchivingContext
from auto_archiver.core import Formatter
from auto_archiver.modules.hash_enricher import HashEnricher
from auto_archiver.utils.misc import random_str
from auto_archiver.core.module import get_module
@dataclass
class HtmlFormatter(Formatter):
@@ -53,7 +53,7 @@ class HtmlFormatter(Formatter):
outf.write(content)
final_media = Media(filename=html_path, _mimetype="text/html")
he = HashEnricher({"hash_enricher": {"algorithm": ArchivingContext.get("hash_enricher.algorithm"), "chunksize": 1.6e7}})
he = get_module('hash_enricher', self.config)
if len(hd := he.calculate_hash(final_media.filename)):
final_media.set("hash", f"{he.algorithm}:{hd}")