Auto translate strings

src-ui/messages.xlf

@@ -2544,11 +2544,11 @@
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1017</context>
+          <context context-type="linenumber">1018</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1382</context>
+          <context context-type="linenumber">1383</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-list/bulk-editor/bulk-editor.component.ts</context>
@@ -3156,7 +3156,7 @@
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">970</context>
+          <context context-type="linenumber">971</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-list/bulk-editor/bulk-editor.component.ts</context>
@@ -6579,7 +6579,7 @@
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1381</context>
+          <context context-type="linenumber">1382</context>
         </context-group>
       </trans-unit>
       <trans-unit id="6490688569532630280" datatype="html">
@@ -6904,21 +6904,21 @@
         <source>Next document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">573</context>
+          <context context-type="linenumber">574</context>
         </context-group>
       </trans-unit>
       <trans-unit id="651985345816518480" datatype="html">
         <source>Previous document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">583</context>
+          <context context-type="linenumber">584</context>
         </context-group>
       </trans-unit>
       <trans-unit id="2885986061416655600" datatype="html">
         <source>Close document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">591</context>
+          <context context-type="linenumber">592</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/services/open-documents.service.ts</context>
@@ -6929,67 +6929,67 @@
         <source>Save document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">598</context>
+          <context context-type="linenumber">599</context>
         </context-group>
       </trans-unit>
       <trans-unit id="1784543155727940353" datatype="html">
         <source>Save and close / next</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">607</context>
+          <context context-type="linenumber">608</context>
         </context-group>
       </trans-unit>
       <trans-unit id="5758784066858623886" datatype="html">
         <source>Error retrieving metadata</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">659</context>
+          <context context-type="linenumber">660</context>
         </context-group>
       </trans-unit>
       <trans-unit id="3456881259945295697" datatype="html">
         <source>Error retrieving suggestions.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">688</context>
+          <context context-type="linenumber">689</context>
         </context-group>
       </trans-unit>
       <trans-unit id="2194092841814123758" datatype="html">
         <source>Document &quot;<x id="PH" equiv-text="newValues.title"/>&quot; saved successfully.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">860</context>
+          <context context-type="linenumber">861</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">884</context>
+          <context context-type="linenumber">885</context>
         </context-group>
       </trans-unit>
       <trans-unit id="6626387786259219838" datatype="html">
         <source>Error saving document &quot;<x id="PH" equiv-text="this.document.title"/>&quot;</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">890</context>
+          <context context-type="linenumber">891</context>
         </context-group>
       </trans-unit>
       <trans-unit id="448882439049417053" datatype="html">
         <source>Error saving document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">939</context>
+          <context context-type="linenumber">940</context>
         </context-group>
       </trans-unit>
       <trans-unit id="8410796510716511826" datatype="html">
         <source>Do you really want to move the document &quot;<x id="PH" equiv-text="this.document.title"/>&quot; to the trash?</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">971</context>
+          <context context-type="linenumber">972</context>
         </context-group>
       </trans-unit>
       <trans-unit id="282586936710748252" datatype="html">
         <source>Documents can be restored prior to permanent deletion.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">972</context>
+          <context context-type="linenumber">973</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-list/bulk-editor/bulk-editor.component.ts</context>
@@ -7000,7 +7000,7 @@
         <source>Move to trash</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">974</context>
+          <context context-type="linenumber">975</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-list/bulk-editor/bulk-editor.component.ts</context>
@@ -7011,14 +7011,14 @@
         <source>Error deleting document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">993</context>
+          <context context-type="linenumber">994</context>
         </context-group>
       </trans-unit>
       <trans-unit id="619486176823357521" datatype="html">
         <source>Reprocess confirm</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1013</context>
+          <context context-type="linenumber">1014</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-list/bulk-editor/bulk-editor.component.ts</context>
@@ -7029,67 +7029,67 @@
         <source>This operation will permanently recreate the archive file for this document.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1014</context>
+          <context context-type="linenumber">1015</context>
         </context-group>
       </trans-unit>
       <trans-unit id="302054111564709516" datatype="html">
         <source>The archive file will be re-generated with the current settings.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1015</context>
+          <context context-type="linenumber">1016</context>
         </context-group>
       </trans-unit>
       <trans-unit id="8251197608401006898" datatype="html">
         <source>Reprocess operation for &quot;<x id="PH" equiv-text="this.document.title"/>&quot; will begin in the background. Close and re-open or reload this document after the operation has completed to see new content.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1025</context>
+          <context context-type="linenumber">1026</context>
         </context-group>
       </trans-unit>
       <trans-unit id="4409560272830824468" datatype="html">
         <source>Error executing operation</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1036</context>
+          <context context-type="linenumber">1037</context>
         </context-group>
       </trans-unit>
       <trans-unit id="6030453331794586802" datatype="html">
         <source>Error downloading document</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1085</context>
+          <context context-type="linenumber">1086</context>
         </context-group>
       </trans-unit>
       <trans-unit id="4458954481601077369" datatype="html">
         <source>Page Fit</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1162</context>
+          <context context-type="linenumber">1163</context>
         </context-group>
       </trans-unit>
       <trans-unit id="4663705961777238777" datatype="html">
         <source>PDF edit operation for &quot;<x id="PH" equiv-text="this.document.title"/>&quot; will begin in the background.</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1400</context>
+          <context context-type="linenumber">1401</context>
         </context-group>
       </trans-unit>
       <trans-unit id="9043972994040261999" datatype="html">
         <source>Error executing PDF edit operation</source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1412</context>
+          <context context-type="linenumber">1413</context>
         </context-group>
       </trans-unit>
       <trans-unit id="6085793215710522488" datatype="html">
         <source>An error occurred loading tiff: <x id="PH" equiv-text="err.toString()"/></source>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1479</context>
+          <context context-type="linenumber">1480</context>
         </context-group>
         <context-group purpose="location">
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
-          <context context-type="linenumber">1483</context>
+          <context context-type="linenumber">1484</context>
         </context-group>
       </trans-unit>
       <trans-unit id="4958946940233632319" datatype="html">

src-ui/src/app/components/document-detail/document-detail.component.ts

@@ -472,6 +472,7 @@ export class DocumentDetailComponent
               if (titleValue !== this.titleInput.value) return
               this.title = titleValue
               this.documentForm.patchValue({ title: titleValue })
+              this.documentForm.get('title').markAsDirty()
             })
           this.setupDirtyTracking(useDoc, doc)
         },

src/documents/parsers.py

@@ -19,6 +19,8 @@ from documents.loggers import LoggingMixin
 from documents.signals import document_consumer_declaration
 from documents.utils import copy_file_with_basic_stats
 from documents.utils import run_subprocess
+from paperless.config import OcrConfig
+from paperless.utils import ocr_to_dateparser_languages
 
 if TYPE_CHECKING:
     import datetime
@@ -272,6 +274,11 @@ def parse_date_generator(filename, text) -> Iterator[datetime.datetime]:
         """
         import dateparser
 
+        ocr_config = OcrConfig()
+        languages = settings.DATE_PARSER_LANGUAGES or ocr_to_dateparser_languages(
+            ocr_config.language,
+        )
+
         return dateparser.parse(
             ds,
             settings={
@@ -280,7 +287,7 @@ def parse_date_generator(filename, text) -> Iterator[datetime.datetime]:
                 "RETURN_AS_TIMEZONE_AWARE": True,
                 "TIMEZONE": settings.TIME_ZONE,
             },
-            locales=settings.DATE_PARSER_LANGUAGES,
+            locales=languages,
         )
 
     def __filter(date: datetime.datetime) -> datetime.datetime | None:

src/documents/tests/test_date_parsing.py

@@ -1,12 +1,14 @@
 import datetime
 from zoneinfo import ZoneInfo
 
+import pytest
 from pytest_django.fixtures import SettingsWrapper
 
 from documents.parsers import parse_date
 from documents.parsers import parse_date_generator
 
 
+@pytest.mark.django_db()
 class TestDate:
     def test_date_format_1(self):
         text = "lorem ipsum 130218 lorem ipsum"
@@ -49,7 +51,7 @@ class TestDate:
         settings: SettingsWrapper,
         settings_timezone: ZoneInfo,
     ):
-        settings.DATE_PARSER_LANGUAGES = []
+        settings.DATE_PARSER_LANGUAGES = ["de"]
         text = "lorem ipsum\nMärz 2019\nlorem ipsum"
         date = parse_date("", text)
         assert date == datetime.datetime(2019, 3, 1, 0, 0, tzinfo=settings_timezone)

src/locale/en_US/LC_MESSAGES/django.po

@@ -2,7 +2,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: paperless-ngx\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2025-08-16 14:34+0000\n"
+"POT-Creation-Date: 2025-08-31 22:24+0000\n"
 "PO-Revision-Date: 2022-02-17 04:17\n"
 "Last-Translator: \n"
 "Language-Team: English\n"
@@ -1645,147 +1645,147 @@ msgstr ""
 msgid "paperless application settings"
 msgstr ""
 
-#: paperless/settings.py:774
+#: paperless/settings.py:772
 msgid "English (US)"
 msgstr ""
 
-#: paperless/settings.py:775
+#: paperless/settings.py:773
 msgid "Arabic"
 msgstr ""
 
-#: paperless/settings.py:776
+#: paperless/settings.py:774
 msgid "Afrikaans"
 msgstr ""
 
-#: paperless/settings.py:777
+#: paperless/settings.py:775
 msgid "Belarusian"
 msgstr ""
 
-#: paperless/settings.py:778
+#: paperless/settings.py:776
 msgid "Bulgarian"
 msgstr ""
 
-#: paperless/settings.py:779
+#: paperless/settings.py:777
 msgid "Catalan"
 msgstr ""
 
-#: paperless/settings.py:780
+#: paperless/settings.py:778
 msgid "Czech"
 msgstr ""
 
-#: paperless/settings.py:781
+#: paperless/settings.py:779
 msgid "Danish"
 msgstr ""
 
-#: paperless/settings.py:782
+#: paperless/settings.py:780
 msgid "German"
 msgstr ""
 
-#: paperless/settings.py:783
+#: paperless/settings.py:781
 msgid "Greek"
 msgstr ""
 
-#: paperless/settings.py:784
+#: paperless/settings.py:782
 msgid "English (GB)"
 msgstr ""
 
-#: paperless/settings.py:785
+#: paperless/settings.py:783
 msgid "Spanish"
 msgstr ""
 
-#: paperless/settings.py:786
+#: paperless/settings.py:784
 msgid "Persian"
 msgstr ""
 
-#: paperless/settings.py:787
+#: paperless/settings.py:785
 msgid "Finnish"
 msgstr ""
 
-#: paperless/settings.py:788
+#: paperless/settings.py:786
 msgid "French"
 msgstr ""
 
-#: paperless/settings.py:789
+#: paperless/settings.py:787
 msgid "Hungarian"
 msgstr ""
 
-#: paperless/settings.py:790
+#: paperless/settings.py:788
 msgid "Italian"
 msgstr ""
 
-#: paperless/settings.py:791
+#: paperless/settings.py:789
 msgid "Japanese"
 msgstr ""
 
-#: paperless/settings.py:792
+#: paperless/settings.py:790
 msgid "Korean"
 msgstr ""
 
-#: paperless/settings.py:793
+#: paperless/settings.py:791
 msgid "Luxembourgish"
 msgstr ""
 
-#: paperless/settings.py:794
+#: paperless/settings.py:792
 msgid "Norwegian"
 msgstr ""
 
-#: paperless/settings.py:795
+#: paperless/settings.py:793
 msgid "Dutch"
 msgstr ""
 
-#: paperless/settings.py:796
+#: paperless/settings.py:794
 msgid "Polish"
 msgstr ""
 
-#: paperless/settings.py:797
+#: paperless/settings.py:795
 msgid "Portuguese (Brazil)"
 msgstr ""
 
-#: paperless/settings.py:798
+#: paperless/settings.py:796
 msgid "Portuguese"
 msgstr ""
 
-#: paperless/settings.py:799
+#: paperless/settings.py:797
 msgid "Romanian"
 msgstr ""
 
-#: paperless/settings.py:800
+#: paperless/settings.py:798
 msgid "Russian"
 msgstr ""
 
-#: paperless/settings.py:801
+#: paperless/settings.py:799
 msgid "Slovak"
 msgstr ""
 
-#: paperless/settings.py:802
+#: paperless/settings.py:800
 msgid "Slovenian"
 msgstr ""
 
-#: paperless/settings.py:803
+#: paperless/settings.py:801
 msgid "Serbian"
 msgstr ""
 
-#: paperless/settings.py:804
+#: paperless/settings.py:802
 msgid "Swedish"
 msgstr ""
 
-#: paperless/settings.py:805
+#: paperless/settings.py:803
 msgid "Turkish"
 msgstr ""
 
-#: paperless/settings.py:806
+#: paperless/settings.py:804
 msgid "Ukrainian"
 msgstr ""
 
-#: paperless/settings.py:807
+#: paperless/settings.py:805
 msgid "Vietnamese"
 msgstr ""
 
-#: paperless/settings.py:808
+#: paperless/settings.py:806
 msgid "Chinese Simplified"
 msgstr ""
 
-#: paperless/settings.py:809
+#: paperless/settings.py:807
 msgid "Chinese Traditional"
 msgstr ""
 

src/paperless/settings.py

@@ -17,8 +17,6 @@ from dateparser.languages.loader import LocaleDataLoader
 from django.utils.translation import gettext_lazy as _
 from dotenv import load_dotenv
 
-from paperless.utils import ocr_to_dateparser_languages
-
 logger = logging.getLogger("paperless.settings")
 
 # Tap paperless.conf if it's available
@@ -1184,61 +1182,6 @@ DATE_ORDER = os.getenv("PAPERLESS_DATE_ORDER", "DMY")
 FILENAME_DATE_ORDER = os.getenv("PAPERLESS_FILENAME_DATE_ORDER")
 
 
-def _ocr_to_dateparser_languages(ocr_languages: str) -> list[str]:
-    """
-    Convert Tesseract OCR_LANGUAGE codes (ISO 639-2, e.g. "eng+fra", with optional scripts like "aze_Cyrl")
-    into a list of locales compatible with the `dateparser` library.
-
-    - If a script is provided (e.g., "aze_Cyrl"), attempts to use the full locale (e.g., "az-Cyrl").
-    Falls back to the base language (e.g., "az") if needed.
-    - If a language cannot be mapped or validated, it is skipped with a warning.
-    - Returns a list of valid locales, or an empty list if none could be converted.
-    """
-    ocr_to_dateparser = ocr_to_dateparser_languages()
-    loader = LocaleDataLoader()
-    result = []
-    try:
-        for ocr_language in ocr_languages.split("+"):
-            # Split into language and optional script
-            ocr_lang_part, *script = ocr_language.split("_")
-            ocr_script_part = script[0] if script else None
-
-            language_part = ocr_to_dateparser.get(ocr_lang_part)
-            if language_part is None:
-                logger.debug(
-                    f'Unable to map OCR language "{ocr_lang_part}" to dateparser locale. ',
-                )
-                continue
-
-            # Ensure base language is supported by dateparser
-            loader.get_locale_map(locales=[language_part])
-
-            # Try to add the script part if it's supported by dateparser
-            if ocr_script_part:
-                dateparser_language = f"{language_part}-{ocr_script_part.title()}"
-                try:
-                    loader.get_locale_map(locales=[dateparser_language])
-                except Exception:
-                    logger.info(
-                        f"Language variant '{dateparser_language}' not supported by dateparser; falling back to base language '{language_part}'. You can manually set PAPERLESS_DATE_PARSER_LANGUAGES if needed.",
-                    )
-                    dateparser_language = language_part
-            else:
-                dateparser_language = language_part
-            if dateparser_language not in result:
-                result.append(dateparser_language)
-    except Exception as e:
-        logger.warning(
-            f"Error auto-configuring dateparser languages. Set PAPERLESS_DATE_PARSER_LANGUAGES parameter to avoid this. Detail: {e}",
-        )
-        return []
-    if not result:
-        logger.info(
-            "Unable to automatically determine dateparser languages from OCR_LANGUAGE, falling back to multi-language support.",
-        )
-    return result
-
-
 def _parse_dateparser_languages(languages: str | None):
     language_list = languages.split("+") if languages else []
     # There is an unfixed issue in zh-Hant and zh-Hans locales in the dateparser lib.
@@ -1253,12 +1196,14 @@ def _parse_dateparser_languages(languages: str | None):
     return list(LocaleDataLoader().get_locale_map(locales=language_list))
 
 
-if os.getenv("PAPERLESS_DATE_PARSER_LANGUAGES"):
-    DATE_PARSER_LANGUAGES = _parse_dateparser_languages(
+# If not set, we will infer it at runtime
+DATE_PARSER_LANGUAGES = (
+    _parse_dateparser_languages(
         os.getenv("PAPERLESS_DATE_PARSER_LANGUAGES"),
     )
-else:
-    DATE_PARSER_LANGUAGES = _ocr_to_dateparser_languages(OCR_LANGUAGE)
+    if os.getenv("PAPERLESS_DATE_PARSER_LANGUAGES")
+    else None
+)
 
 
 # Maximum number of dates taken from document start to end to show as suggestions for

src/paperless/tests/test_settings.py

@@ -6,7 +6,6 @@ from unittest import mock
 import pytest
 from celery.schedules import crontab
 
-from paperless.settings import _ocr_to_dateparser_languages
 from paperless.settings import _parse_base_paths
 from paperless.settings import _parse_beat_schedule
 from paperless.settings import _parse_dateparser_languages
@@ -476,33 +475,6 @@ class TestPathSettings(TestCase):
         self.assertEqual("/foobar/", base_paths[4])  # LOGOUT_REDIRECT_URL
 
 
-@pytest.mark.parametrize(
-    ("ocr_language", "expected"),
-    [
-        # One language
-        ("eng", ["en"]),
-        # Multiple languages
-        ("fra+ita+lao", ["fr", "it", "lo"]),
-        # Languages that don't have a two-letter equivalent
-        ("fil", ["fil"]),
-        # Languages with a script part supported by dateparser
-        ("aze_cyrl+srp_latn", ["az-Cyrl", "sr-Latn"]),
-        # Languages with a script part not supported by dateparser
-        # In this case, default to the language without script
-        ("deu_frak", ["de"]),
-        # Traditional and simplified chinese don't have the same name in dateparser,
-        # so they're converted to the general chinese language
-        ("chi_tra+chi_sim", ["zh"]),
-        # If a language is not supported by dateparser, fallback to the supported ones
-        ("eng+unsupported_language+por", ["en", "pt"]),
-        # If no language is supported, fallback to default
-        ("unsupported1+unsupported2", []),
-    ],
-)
-def test_ocr_to_dateparser_languages(ocr_language, expected):
-    assert sorted(_ocr_to_dateparser_languages(ocr_language)) == sorted(expected)
-
-
 @pytest.mark.parametrize(
     ("languages", "expected"),
     [

src/paperless/tests/test_utils.py

@@ -0,0 +1,52 @@
+import logging
+
+import pytest
+
+from paperless import utils
+from paperless.utils import ocr_to_dateparser_languages
+
+
+@pytest.mark.parametrize(
+    ("ocr_language", "expected"),
+    [
+        # One language
+        ("eng", ["en"]),
+        # Multiple languages
+        ("fra+ita+lao", ["fr", "it", "lo"]),
+        # Languages that don't have a two-letter equivalent
+        ("fil", ["fil"]),
+        # Languages with a script part supported by dateparser
+        ("aze_cyrl+srp_latn", ["az-Cyrl", "sr-Latn"]),
+        # Languages with a script part not supported by dateparser
+        # In this case, default to the language without script
+        ("deu_frak", ["de"]),
+        # Traditional and simplified chinese don't have the same name in dateparser,
+        # so they're converted to the general chinese language
+        ("chi_tra+chi_sim", ["zh"]),
+        # If a language is not supported by dateparser, fallback to the supported ones
+        ("eng+unsupported_language+por", ["en", "pt"]),
+        # If no language is supported, fallback to default
+        ("unsupported1+unsupported2", []),
+        # Duplicate languages, should not duplicate in result
+        ("eng+eng", ["en"]),
+        # Language with script, but script is not mapped
+        ("ita_unknownscript", ["it"]),
+    ],
+)
+def test_ocr_to_dateparser_languages(ocr_language, expected):
+    assert sorted(ocr_to_dateparser_languages(ocr_language)) == sorted(expected)
+
+
+def test_ocr_to_dateparser_languages_exception(monkeypatch, caplog):
+    # Patch LocaleDataLoader.get_locale_map to raise an exception
+    class DummyLoader:
+        def get_locale_map(self, locales=None):
+            raise RuntimeError("Simulated error")
+
+    with caplog.at_level(logging.WARNING):
+        monkeypatch.setattr(utils, "LocaleDataLoader", lambda: DummyLoader())
+        result = utils.ocr_to_dateparser_languages("eng+fra")
+        assert result == []
+        assert (
+            "Set PAPERLESS_DATE_PARSER_LANGUAGES parameter to avoid this" in caplog.text
+        )

src/paperless/utils.py

@@ -1,4 +1,10 @@
-def ocr_to_dateparser_languages() -> dict[str, str]:
+import logging
+
+from dateparser.languages.loader import LocaleDataLoader
+
+logger = logging.getLogger("paperless.utils")
+
+OCR_TO_DATEPARSER_LANGUAGES = {
     """
     Translation map from languages supported by Tesseract OCR
     to languages supported by dateparser.
@@ -14,7 +20,6 @@ def ocr_to_dateparser_languages() -> dict[str, str]:
     # agq, asa, bez, brx, cgg, ckb, dav, dje, dyo, ebu, guz, jgo, jmc, kde, kea, khq, kln,
     # ksb, ksf, ksh, lag, lkt, lrc, luy, mer, mfe, mgh, mgo, mua, mzn, naq, nmg, nnh, nus,
     # rof, rwk, saq, sbp, she, ses, shi, teo, twq, tzm, vun, wae, xog, yav, yue
-    return {
     "afr": "af",
     "amh": "am",
     "ara": "ar",
@@ -108,3 +113,57 @@ def ocr_to_dateparser_languages() -> dict[str, str]:
     "yor": "yo",
     "chi": "zh",
 }
+
+
+def ocr_to_dateparser_languages(ocr_languages: str) -> list[str]:
+    """
+    Convert Tesseract OCR_LANGUAGE codes (ISO 639-2, e.g. "eng+fra", with optional scripts like "aze_Cyrl")
+    into a list of locales compatible with the `dateparser` library.
+
+    - If a script is provided (e.g., "aze_Cyrl"), attempts to use the full locale (e.g., "az-Cyrl").
+    Falls back to the base language (e.g., "az") if needed.
+    - If a language cannot be mapped or validated, it is skipped with a warning.
+    - Returns a list of valid locales, or an empty list if none could be converted.
+    """
+    loader = LocaleDataLoader()
+    result = []
+    try:
+        for ocr_language in ocr_languages.split("+"):
+            # Split into language and optional script
+            ocr_lang_part, *script = ocr_language.split("_")
+            ocr_script_part = script[0] if script else None
+
+            language_part = OCR_TO_DATEPARSER_LANGUAGES.get(ocr_lang_part)
+            if language_part is None:
+                logger.debug(
+                    f'Unable to map OCR language "{ocr_lang_part}" to dateparser locale. ',
+                )
+                continue
+
+            # Ensure base language is supported by dateparser
+            loader.get_locale_map(locales=[language_part])
+
+            # Try to add the script part if it's supported by dateparser
+            if ocr_script_part:
+                dateparser_language = f"{language_part}-{ocr_script_part.title()}"
+                try:
+                    loader.get_locale_map(locales=[dateparser_language])
+                except Exception:
+                    logger.info(
+                        f"Language variant '{dateparser_language}' not supported by dateparser; falling back to base language '{language_part}'. You can manually set PAPERLESS_DATE_PARSER_LANGUAGES if needed.",
+                    )
+                    dateparser_language = language_part
+            else:
+                dateparser_language = language_part
+            if dateparser_language not in result:
+                result.append(dateparser_language)
+    except Exception as e:
+        logger.warning(
+            f"Error auto-configuring dateparser languages. Set PAPERLESS_DATE_PARSER_LANGUAGES parameter to avoid this. Detail: {e}",
+        )
+        return []
+    if not result:
+        logger.info(
+            "Unable to automatically determine dateparser languages from OCR_LANGUAGE, falling back to multi-language support.",
+        )
+    return result