From fc26fe0ac08faebcc0037c36fe9565e0e7b33368 Mon Sep 17 00:00:00 2001
From: Trenton Holmes <holmes.trenton@gmail.com>
Date: Sun, 22 May 2022 16:56:08 -0700
Subject: [PATCH 1/2] Updates to provide the user provided max pixel size to
ocrmypdf
---
src/paperless/settings.py | 8 ++++----
src/paperless_tesseract/parsers.py | 20 ++++++++++++++++++--
src/paperless_text/parsers.py | 4 +---
3 files changed, 23 insertions(+), 9 deletions(-)
diff --git a/src/paperless/settings.py b/src/paperless/settings.py
index cd3aafc25..7f484ad97 100644
--- a/src/paperless/settings.py
+++ b/src/paperless/settings.py
@@ -5,6 +5,7 @@ import multiprocessing
import os
import re
from typing import Final
+from typing import Optional
from typing import Set
from urllib.parse import urlparse
@@ -551,10 +552,9 @@ OCR_ROTATE_PAGES_THRESHOLD = float(
os.getenv("PAPERLESS_OCR_ROTATE_PAGES_THRESHOLD", 12.0),
)
-OCR_MAX_IMAGE_PIXELS = os.environ.get(
- "PAPERLESS_OCR_MAX_IMAGE_PIXELS",
- 256000000,
-)
+OCR_MAX_IMAGE_PIXELS: Optional[int] = None
+if os.environ.get("PAPERLESS_OCR_MAX_IMAGE_PIXELS") is not None:
+ OCR_MAX_IMAGE_PIXELS: int = int(os.environ.get("PAPERLESS_OCR_MAX_IMAGE_PIXELS"))
OCR_USER_ARGS = os.getenv("PAPERLESS_OCR_USER_ARGS", "{}")
diff --git a/src/paperless_tesseract/parsers.py b/src/paperless_tesseract/parsers.py
index 56313c5b4..f35d3a6b4 100644
--- a/src/paperless_tesseract/parsers.py
+++ b/src/paperless_tesseract/parsers.py
@@ -8,8 +8,6 @@ from documents.parsers import make_thumbnail_from_pdf
from documents.parsers import ParseError
from PIL import Image
-Image.MAX_IMAGE_PIXELS = settings.OCR_MAX_IMAGE_PIXELS
-
class NoTextFoundException(Exception):
pass
@@ -225,6 +223,24 @@ class RasterisedDocumentParser(DocumentParser):
f"they will not be used. Error: {e}",
)
+ if settings.OCR_MAX_IMAGE_PIXELS is not None:
+ # Convert pixels to mega-pixels and provide to ocrmypdf
+ max_pixels_mpixels = settings.OCR_MAX_IMAGE_PIXELS / 1_000_000.0
+ if max_pixels_mpixels > 0:
+
+ self.log(
+ "debug",
+ f"Calculated {max_pixels_mpixels} megapixels for OCR",
+ )
+
+ ocrmypdf_args["max_image_mpixels"] = max_pixels_mpixels
+ else:
+ self.log(
+ "warning",
+ "There is an issue with PAPERLESS_OCR_MAX_IMAGE_PIXELS, "
+ "this value must be at least 1 megapixel if set",
+ )
+
return ocrmypdf_args
def parse(self, document_path, mime_type, file_name=None):
diff --git a/src/paperless_text/parsers.py b/src/paperless_text/parsers.py
index 9ef5fec40..fe7e823b3 100644
--- a/src/paperless_text/parsers.py
+++ b/src/paperless_text/parsers.py
@@ -6,8 +6,6 @@ from PIL import Image
from PIL import ImageDraw
from PIL import ImageFont
-Image.MAX_IMAGE_PIXELS = settings.OCR_MAX_IMAGE_PIXELS
-
class TextDocumentParser(DocumentParser):
"""
@@ -28,7 +26,7 @@ class TextDocumentParser(DocumentParser):
font = ImageFont.truetype(
font=settings.THUMBNAIL_FONT_NAME,
size=20,
- layout_engine=ImageFont.LAYOUT_BASIC,
+ layout_engine=ImageFont.Layout.BASIC,
)
draw.text((5, 5), read_text(), font=font, fill="black")
From 985b77437808d9b468fdf1118dd6702412f46392 Mon Sep 17 00:00:00 2001
From: Trenton Holmes <holmes.trenton@gmail.com>
Date: Mon, 23 May 2022 10:44:33 -0700
Subject: [PATCH 2/2] Reworks documentation on the max pixels setting
---
docs/configuration.rst | 25 +++++++++++++++++--------
1 file changed, 17 insertions(+), 8 deletions(-)
diff --git a/docs/configuration.rst b/docs/configuration.rst
index 2068a4238..b7ab978f4 100644
--- a/docs/configuration.rst
+++ b/docs/configuration.rst
@@ -424,14 +424,23 @@ PAPERLESS_OCR_IMAGE_DPI=<num>
the produced PDF documents are A4 sized.
PAPERLESS_OCR_MAX_IMAGE_PIXELS=<num>
- Paperless will not OCR images that have more pixels than this limit.
- This is intended to prevent decompression bombs from overloading paperless.
- Increasing this limit is desired if you face a DecompressionBombError despite
- the concerning file not being malicious; this could e.g. be caused by invalidly
- recognized metadata.
- If you have enough resources or if you are certain that your uploaded files
- are not malicious you can increase this value to your needs.
- The default value is 256000000, an image with more pixels than that would not be parsed.
+ Paperless will raise a warning when OCRing images which are over this limit and
+ will not OCR images which are more than twice this limit. Note this does not
+ prevent the document from being consumed, but could result in missing text content.
+
+ If unset, will default to the value determined by
+ `Pillow <https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.MAX_IMAGE_PIXELS>`_.
+
+ .. note::
+
+ Increasing this limit could cause Paperless to consume additional resources
+ when consuming a file. Be sure you have sufficient system resources.
+
+ .. caution::
+
+ The limit is intended to prevent malicious files from consuming system resources
+ and causing crashes and other errors. Only increase this value if you are certain
+ your documents are not malicious and you need the text which was not OCRed
PAPERLESS_OCR_USER_ARGS=<json>
OCRmyPDF offers many more options. Use this parameter to specify any