paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-07-28 18:24:38 -05:00
Code Issues Packages Projects Releases Wiki Activity

Chore: Backend bulk updates (#4509)

Browse Source
This commit is contained in:
Trenton H
2023-11-13 09:09:56 -08:00
committed by GitHub
parent b671f54cb7
commit facb7226fe
15 changed files with 1764 additions and 1596 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/documents/classifier.py
View File

@@ -248,7 +248,7 @@ class DocumentClassifier:
data_vectorized = self.data_vectorizer.fit_transform(content_generator())
# See the notes here:
# https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html # noqa: E501
# https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html
# This attribute isn't needed to function and can be large
self.data_vectorizer.stop_words_ = None

10
src/documents/parsers.py
View File

@@ -36,13 +36,13 @@ from documents.utils import copy_file_with_basic_stats
# TODO: isnt there a date parsing library for this?
DATE_REGEX = re.compile(
r"(\b|(?!=([_-])))([0-9]{1,2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{4}|[0-9]{2})(\b|(?=([_-])))|" # noqa: E501
r"(\b|(?!=([_-])))([0-9]{4}|[0-9]{2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{1,2})(\b|(?=([_-])))|" # noqa: E501
r"(\b|(?!=([_-])))([0-9]{1,2}[\. ]+[a-zA-Z]{3,9} ([0-9]{4}|[0-9]{2}))(\b|(?=([_-])))|" # noqa: E501
r"(\b|(?!=([_-])))([0-9]{1,2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{4}|[0-9]{2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{4}|[0-9]{2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{1,2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{1,2}[\. ]+[a-zA-Z]{3,9} ([0-9]{4}|[0-9]{2}))(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{1,2}, ([0-9]{4}))(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{1,2}[^ ]{2}[\. ]+[^ ]{3,9}[ \.\/-][0-9]{4})(\b|(?=([_-])))|" # noqa: E501
r"(\b|(?!=([_-])))(\b[0-9]{1,2}[ \.\/-][a-zA-Z]{3}[ \.\/-][0-9]{4})(\b|(?=([_-])))", # noqa: E501
r"(\b|(?!=([_-])))([0-9]{1,2}[^ ]{2}[\. ]+[^ ]{3,9}[ \.\/-][0-9]{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\b[0-9]{1,2}[ \.\/-][a-zA-Z]{3}[ \.\/-][0-9]{4})(\b|(?=([_-])))",
)

1
src/documents/serialisers.py
View File

@@ -1023,7 +1023,6 @@ class AcknowledgeTasksViewSerializer(serializers.Serializer):
)
def _validate_task_id_list(self, tasks, name="tasks"):
pass
if not isinstance(tasks, list):
raise serializers.ValidationError(f"{name} must be a list")
if not all(isinstance(i, int) for i in tasks):

4
src/documents/views.py
View File

@@ -158,10 +158,10 @@ class IndexView(TemplateView):
context["main_js"] = f"frontend/{self.get_frontend_language()}/main.js"
context[
"webmanifest"
] = f"frontend/{self.get_frontend_language()}/manifest.webmanifest" # noqa: E501
] = f"frontend/{self.get_frontend_language()}/manifest.webmanifest"
context[
"apple_touch_icon"
] = f"frontend/{self.get_frontend_language()}/apple-touch-icon.png" # noqa: E501
] = f"frontend/{self.get_frontend_language()}/apple-touch-icon.png"
return context