lgcosta/paperless-ngx
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: update ocr

Browse Source
This commit is contained in:
otxtan@gmail.com 2024-05-19 18:01:56 +07:00
parent 58a0bd55c0
commit 619534239b
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
src/paperless_ocr_custom/parsers.py
View File

@ -257,6 +257,7 @@ class RasterisedDocumentParser(DocumentParser):
rolate_width = width_api_img /page_width rolate_width = width_api_img /page_width
for block in data["pages"][page_num]["blocks"]: for block in data["pages"][page_num]["blocks"]:
for line in block.get("lines", []): for line in block.get("lines", []):
text_line = ''
for word in line.get("words", []): for word in line.get("words", []):
x1 = word["bbox"][0][0] / float(rolate_width) x1 = word["bbox"][0][0] / float(rolate_width)
y1 = word["bbox"][0][1] / float(rolate_height) y1 = word["bbox"][0][1] / float(rolate_height)