From 17356c79f9ed180a9e14d13a68d9561608d26a0e Mon Sep 17 00:00:00 2001 From: sophon Date: Fri, 7 Nov 2025 09:03:35 +0800 Subject: [PATCH] modify scripts --- docker/paperless/plugins/parsers.py | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/docker/paperless/plugins/parsers.py b/docker/paperless/plugins/parsers.py index ffb298d..ac4d894 100755 --- a/docker/paperless/plugins/parsers.py +++ b/docker/paperless/plugins/parsers.py @@ -335,6 +335,11 @@ class RasterisedDocumentParser(DocumentParser): os.environ["OMP_THREAD_LIMIT"] = "1" VALID_TEXT_LENGTH = 50 + # skip ocr process entirely to save time. + self.text = "defautl text" + self.log.debug("skipping reading file entirely.") + return + if mime_type == "application/pdf": text_original = self.extract_text(None, document_path) original_has_text = (