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

Fix: Allows pre-consume scripts to modify the working path again (#5260)

Browse Source 
* Allows pre-consume scripts to modify the working path again and generally cleans up some confusion about working copy vs original
This commit is contained in:
Trenton H
2024-01-05 21:01:57 -08:00
committed by GitHub
parent 3115106dc1
commit a82e3771ae
5 changed files with 37 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
src/paperless_tesseract/parsers.py
View File

@@ -90,16 +90,18 @@ class RasterisedDocumentParser(DocumentParser):
with Image.open(image) as im:
return im.mode in ("RGBA", "LA")
def remove_alpha(self, image_path: str):
def remove_alpha(self, image_path: str) -> Path:
no_alpha_image = Path(self.tempdir) / "image-no-alpha"
subprocess.run(
[
settings.CONVERT_BINARY,
"-alpha",
"off",
image_path,
image_path,
no_alpha_image,
],
)
return no_alpha_image
def get_dpi(self, image) -> Optional[int]:
try:
@@ -251,7 +253,8 @@ class RasterisedDocumentParser(DocumentParser):
f"Removing alpha layer from {input_file} "
"for compatibility with img2pdf",
)
self.remove_alpha(input_file)
# Replace the input file with the non-alpha
ocrmypdf_args["input_file"] = self.remove_alpha(input_file)
if dpi:
self.log.debug(f"Detected DPI for image {input_file}: {dpi}")