paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-10-30 03:56:23 -05:00
Code Issues Packages Projects Releases Wiki Activity

Also handles confirming returned predictions are still automatic matching, in case the classifier hasn't been run since a type was changed

Browse Source
This commit is contained in:
Trenton H
2023-07-24 10:19:47 -07:00
parent 26d5730ad2
commit 802e5591ce
2 changed files with 30 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
src/documents/matching.py
View File

@@ -35,7 +35,11 @@ def match_correspondents(document, classifier, user=None):
correspondents = Correspondent.objects.all()
return list(
filter(lambda o: matches(o, document) or o.pk == pred_id, correspondents),
filter(
lambda o: matches(o, document)
or (o.pk == pred_id and o.matching_algorithm == MatchingModel.MATCH_AUTO),
correspondents,
),
)
@@ -55,7 +59,11 @@ def match_document_types(document, classifier, user=None):
document_types = DocumentType.objects.all()
return list(
filter(lambda o: matches(o, document) or o.pk == pred_id, document_types),
filter(
lambda o: matches(o, document)
or (o.pk == pred_id and o.matching_algorithm == MatchingModel.MATCH_AUTO),
document_types,
),
)
@@ -71,7 +79,14 @@ def match_tags(document, classifier, user=None):
tags = Tag.objects.all()
return list(
filter(lambda o: matches(o, document) or o.pk in predicted_tag_ids, tags),
filter(
lambda o: matches(o, document)
or (
o.matching_algorithm == MatchingModel.MATCH_AUTO
and o.pk in predicted_tag_ids
),
tags,
),
)
@@ -92,7 +107,8 @@ def match_storage_paths(document, classifier, user=None):
return list(
filter(
lambda o: matches(o, document) or o.pk == pred_id,
lambda o: matches(o, document)
or (o.pk == pred_id and o.matching_algorithm == MatchingModel.MATCH_AUTO),
storage_paths,
),
)

14
src/documents/tests/test_consumer.py
View File

@@ -561,10 +561,16 @@ class TestConsumer(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
@mock.patch("documents.consumer.load_classifier")
def testClassifyDocument(self, m):
correspondent = Correspondent.objects.create(name="test")
dtype = DocumentType.objects.create(name="test")
t1 = Tag.objects.create(name="t1")
t2 = Tag.objects.create(name="t2")
correspondent = Correspondent.objects.create(
name="test",
matching_algorithm=Correspondent.MATCH_AUTO,
)
dtype = DocumentType.objects.create(
name="test",
matching_algorithm=DocumentType.MATCH_AUTO,
)
t1 = Tag.objects.create(name="t1", matching_algorithm=Tag.MATCH_AUTO)
t2 = Tag.objects.create(name="t2", matching_algorithm=Tag.MATCH_AUTO)
m.return_value = MagicMock()
m.return_value.predict_correspondent.return_value = correspondent.pk