paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-09-14 21:45:37 -05:00
Code Issues Packages Projects Releases Wiki Activity

Enhancement: Limit excessively long content length when computing suggestions (#10656)

Browse Source 
This helps prevent excessive processing times on very large documents
by limiting the text analyzed during date parsing, tag prediction,
and correspondent matching.

If the document exceeds 1.2M chars, crop to 1M char.
This commit is contained in:
Antoine Mérino
2025-09-09 22:02:16 +02:00
committed by GitHub
parent 84d85d7a23
commit 8adc26e09d
3 changed files with 65 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
src/documents/matching.py
View File

@@ -41,7 +41,11 @@ def log_reason(
def match_correspondents(document: Document, classifier: DocumentClassifier, user=None):
pred_id = classifier.predict_correspondent(document.content) if classifier else None
pred_id = (
classifier.predict_correspondent(document.suggestion_content)
if classifier
else None
)
if user is None and document.owner is not None:
user = document.owner
@@ -65,8 +69,11 @@ def match_correspondents(document: Document, classifier: DocumentClassifier, use
def match_document_types(document: Document, classifier: DocumentClassifier, user=None):
pred_id = classifier.predict_document_type(document.content) if classifier else None
pred_id = (
classifier.predict_document_type(document.suggestion_content)
if classifier
else None
)
if user is None and document.owner is not None:
user = document.owner
@@ -89,7 +96,9 @@ def match_document_types(document: Document, classifier: DocumentClassifier, use
def match_tags(document: Document, classifier: DocumentClassifier, user=None):
predicted_tag_ids = classifier.predict_tags(document.content) if classifier else []
predicted_tag_ids = (
classifier.predict_tags(document.suggestion_content) if classifier else []
)
if user is None and document.owner is not None:
user = document.owner
@@ -112,7 +121,11 @@ def match_tags(document: Document, classifier: DocumentClassifier, user=None):
def match_storage_paths(document: Document, classifier: DocumentClassifier, user=None):
pred_id = classifier.predict_storage_path(document.content) if classifier else None
pred_id = (
classifier.predict_storage_path(document.suggestion_content)
if classifier
else None
)
if user is None and document.owner is not None:
user = document.owner