Adds new setting to control color conversions (#4709)

src/paperless/settings.py

@@ -864,6 +864,11 @@ 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_COLOR_CONVERSION_STRATEGY = os.getenv(
+    "PAPERLESS_OCR_COLOR_CONVERSION_STRATEGY",
+    "RGB",
+)
+
 OCR_USER_ARGS = os.getenv("PAPERLESS_OCR_USER_ARGS", "{}")
 
 # GNUPG needs a home directory for some reason

src/paperless_tesseract/parsers.py

@@ -186,6 +186,11 @@ class RasterisedDocumentParser(DocumentParser):
             "progress_bar": False,
         }
 
+        if "pdfa" in ocrmypdf_args["output_type"]:
+            ocrmypdf_args[
+                "color_conversion_strategy"
+            ] = settings.OCR_COLOR_CONVERSION_STRATEGY
+
         if settings.OCR_MODE == "force" or safe_fallback:
             ocrmypdf_args["force_ocr"] = True
         elif settings.OCR_MODE in ["skip", "skip_noarchive"]: