diff --git a/src/documents/models.py b/src/documents/models.py index a7ba0eb4db..3d989483cc 100644 --- a/src/documents/models.py +++ b/src/documents/models.py @@ -2442,18 +2442,3 @@ class OcrTemplateZone(models.Model): def __str__(self) -> str: return f"{self.template.name} -> {self.name}" - - -# Document links and select fields require object references, not OCR text. -OCR_SUPPORTED_FIELD_TYPES = frozenset( - { - CustomField.FieldDataType.STRING, - CustomField.FieldDataType.URL, - CustomField.FieldDataType.DATE, - CustomField.FieldDataType.INT, - CustomField.FieldDataType.FLOAT, - CustomField.FieldDataType.MONETARY, - CustomField.FieldDataType.LONG_TEXT, - CustomField.FieldDataType.BOOL, - }, -) diff --git a/src/documents/serialisers.py b/src/documents/serialisers.py index aeb4c2a322..9a22554ec0 100644 --- a/src/documents/serialisers.py +++ b/src/documents/serialisers.py @@ -53,7 +53,6 @@ if settings.AUDIT_LOG_ENABLED: from documents import bulk_edit from documents.data_models import DocumentSource from documents.filters import CustomFieldQueryParser -from documents.models import OCR_SUPPORTED_FIELD_TYPES from documents.models import Correspondent from documents.models import CustomField from documents.models import CustomFieldInstance @@ -99,6 +98,7 @@ from documents.validators import uri_validator from documents.validators import url_validator from documents.versioning import has_prefetched_effective_content from documents.versioning import sort_versions_newest_first +from documents.zone_ocr import OCR_SUPPORTED_FIELD_TYPES from documents.zone_ocr import ZoneOcrSkipReason from documents.zone_ocr import field_key @@ -3824,9 +3824,14 @@ class OcrTemplateZoneSerializer(serializers.ModelSerializer[OcrTemplateZone]): if value is None: return value if value.data_type not in OCR_SUPPORTED_FIELD_TYPES: + supported = ", ".join( + str(data_type.label).lower() + for data_type in CustomField.FieldDataType + if data_type in OCR_SUPPORTED_FIELD_TYPES + ) raise serializers.ValidationError( f"Custom field type '{value.data_type}' is not supported for OCR extraction. " - f"Use string, integer, float, date, monetary, boolean, URL, or long text.", + f"Use one of: {supported}.", ) return value diff --git a/src/documents/zone_ocr.py b/src/documents/zone_ocr.py index 68aa1d50b9..b12e2a1bf8 100644 --- a/src/documents/zone_ocr.py +++ b/src/documents/zone_ocr.py @@ -1325,6 +1325,21 @@ def _parse_monetary(value: str, locale: Locale) -> str | None: return result +# Document links and select fields require object references, not OCR text. +OCR_SUPPORTED_FIELD_TYPES = frozenset( + { + CustomField.FieldDataType.STRING, + CustomField.FieldDataType.URL, + CustomField.FieldDataType.DATE, + CustomField.FieldDataType.INT, + CustomField.FieldDataType.FLOAT, + CustomField.FieldDataType.MONETARY, + CustomField.FieldDataType.LONG_TEXT, + CustomField.FieldDataType.BOOL, + }, +) + + def _convert_value( value: str, data_type: CustomField.FieldDataType,