From: Trenton H <797416+stumpylog@users.noreply.github.com>
Date: Mon, 4 Mar 2024 22:37:36 +0000 (-0800)
Subject: Feature: Allow a user to disable the pixel limit for OCR entirely (#5996)
X-Git-Tag: v2.6.0~9
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=6779042242a76b579bd9dfa3924900c17c04a153;p=thirdparty%2Fpaperless-ngx.git

Feature: Allow a user to disable the pixel limit for OCR entirely (#5996)
---

diff --git a/docs/advanced_usage.md b/docs/advanced_usage.md
index d4ff80f878..863be639bb 100644
--- a/docs/advanced_usage.md
+++ b/docs/advanced_usage.md
@@ -437,7 +437,7 @@ with Prometheus, as it exports metrics. For details on its capabilities,
 refer to the [Flower](https://flower.readthedocs.io/en/latest/index.html)
 documentation.
 
-Flower can be enabled with the setting [PAPERLESS_ENABLE_FLOWER](configuration/#PAPERLESS_ENABLE_FLOWER).
+Flower can be enabled with the setting [PAPERLESS_ENABLE_FLOWER](configuration.md#PAPERLESS_ENABLE_FLOWER).
 To configure Flower further, create a `flowerconfig.py` and
 place it into the `src/paperless` directory. For a Docker
 installation, you can use volumes to accomplish this:
diff --git a/docs/configuration.md b/docs/configuration.md
index 5fd14caf19..c7b710c66c 100644
--- a/docs/configuration.md
+++ b/docs/configuration.md
@@ -766,6 +766,8 @@ 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).
 
+    Setting this value to 0 will entirely disable the limit.  See the below warning.
+
     !!! note
 
         Increasing this limit could cause Paperless to consume additional
@@ -775,7 +777,7 @@ but could result in missing text content.
     !!! warning
 
         The limit is intended to prevent malicious files from consuming
-        system resources and causing crashes and other errors. Only increase
+        system resources and causing crashes and other errors. Only change
         this value if you are certain your documents are not malicious and
         you need the text which was not OCRed
 
diff --git a/src/paperless/migrations/0003_alter_applicationconfiguration_max_image_pixels.py b/src/paperless/migrations/0003_alter_applicationconfiguration_max_image_pixels.py
new file mode 100644
index 0000000000..c27feefb30
--- /dev/null
+++ b/src/paperless/migrations/0003_alter_applicationconfiguration_max_image_pixels.py
@@ -0,0 +1,24 @@
+# Generated by Django 4.2.10 on 2024-03-04 17:30
+
+import django.core.validators
+from django.db import migrations
+from django.db import models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("paperless", "0002_applicationconfiguration_app_logo_and_more"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="applicationconfiguration",
+            name="max_image_pixels",
+            field=models.FloatField(
+                null=True,
+                validators=[django.core.validators.MinValueValidator(0.0)],
+                verbose_name="Sets the maximum image size for decompression",
+            ),
+        ),
+    ]
diff --git a/src/paperless/models.py b/src/paperless/models.py
index 72805dc563..1f6cfbcedd 100644
--- a/src/paperless/models.py
+++ b/src/paperless/models.py
@@ -151,7 +151,7 @@ class ApplicationConfiguration(AbstractSingletonModel):
     max_image_pixels = models.FloatField(
         verbose_name=_("Sets the maximum image size for decompression"),
         null=True,
-        validators=[MinValueValidator(1_000_000.0)],
+        validators=[MinValueValidator(0.0)],
     )
 
     color_conversion_strategy = models.CharField(
diff --git a/src/paperless_tesseract/parsers.py b/src/paperless_tesseract/parsers.py
index 09086585ef..020922703e 100644
--- a/src/paperless_tesseract/parsers.py
+++ b/src/paperless_tesseract/parsers.py
@@ -293,20 +293,19 @@ class RasterisedDocumentParser(DocumentParser):
                     f"they will not be used. Error: {e}",
                 )
 
-        if self.settings.max_image_pixel is not None:
+        if (
+            self.settings.max_image_pixel is not None
+            and self.settings.max_image_pixel >= 0
+        ):
             # Convert pixels to mega-pixels and provide to ocrmypdf
             max_pixels_mpixels = self.settings.max_image_pixel / 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",
-                )
+            msg = (
+                "OCR pixel limit is disabled!"
+                if max_pixels_mpixels == 0
+                else f"Calculated {max_pixels_mpixels} megapixels for OCR"
+            )
+            self.log.debug(msg)
+            ocrmypdf_args["max_image_mpixels"] = max_pixels_mpixels
 
         return ocrmypdf_args