paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-08-14 00:26:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Adds better handling for files with invalid utf8 content

Browse Source
This commit is contained in:
Trenton H
2023-05-12 14:21:32 -07:00
parent 350c20d6ab
commit 111960c530
6 changed files with 47 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
src/paperless_text/parsers.py
View File

@@ -16,11 +16,7 @@ class TextDocumentParser(DocumentParser):
logging_name = "paperless.parsing.text"
def get_thumbnail(self, document_path, mime_type, file_name=None):
def read_text():
with open(document_path) as src:
lines = [line.strip() for line in src.readlines()]
text = "\n".join(lines[:50])
return text
text = self.read_file_handle_unicode_errors(document_path)
img = Image.new("RGB", (500, 700), color="white")
draw = ImageDraw.Draw(img)
@@ -29,7 +25,7 @@ class TextDocumentParser(DocumentParser):
size=20,
layout_engine=ImageFont.Layout.BASIC,
)
draw.text((5, 5), read_text(), font=font, fill="black")
draw.text((5, 5), text, font=font, fill="black")
out_path = os.path.join(self.tempdir, "thumb.webp")
img.save(out_path, format="WEBP")
@@ -37,5 +33,4 @@ class TextDocumentParser(DocumentParser):
return out_path
def parse(self, document_path, mime_type, file_name=None):
with open(document_path) as f:
self.text = f.read()
self.text = self.read_file_handle_unicode_errors(document_path)