update tests

papermerge/core/db/__init__.py

@@ -12,8 +12,8 @@
     add_document_custom_field_values,
     get_doc,
     get_doc_cfv,
+    get_docs_by_type,
     get_document_custom_field_values,
-    get_documents_by_type,
     update_doc_cfv,
     update_doc_type,
 )
@@ -85,5 +85,5 @@
     "get_document_custom_field_values",
     "get_doc_cfv",
     "update_doc_type",
-    "get_documents_by_type",
+    "get_docs_by_type",
 ]

papermerge/core/db/doc.py

@@ -374,14 +374,16 @@ def get_document_custom_field_values(
     return result
 
 
-def get_documents_by_type(
+def get_docs_by_type(
     session: Session,
     type_id: UUID,
-    parent_id: UUID,
+    ancestor_id: UUID,
     user_id: UUID,
 ):
     """
-    Returns node / cf / cfv for all documents with `document_type_id = type_id`
+    Returns list of documents + doc CFv for all documents with of given type
+
+    All fetched documents are descendants of `ancestor_id` node.
     """
     stmt = """
         SELECT node.title,
@@ -419,19 +421,28 @@ def get_documents_by_type(
             ON cfv.field_id = cf.cf_id AND cfv.document_id = doc_id
         WHERE node.parent_id = :parent_id
     """
-    str_parent_id = str(parent_id).replace("-", "")
+    str_parent_id = str(ancestor_id).replace("-", "")
     str_type_id = str(type_id).replace("-", "")
     params = {"parent_id": str_parent_id, "document_type_id": str_type_id}
     results = []
     rows = session.execute(text(stmt), params)
     for document_id, group in itertools.groupby(rows, lambda r: r.doc_id):
         items = list(group)
+        custom_fields = []
+
+        for item in items:
+            if item.cf_type == "date":
+                value = str2date(item.cf_value)
+            else:
+                value = item.cf_value
+            custom_fields.append((item.cf_name, value))
+
         results.append(
             schemas.DocumentCFV(
                 id=uuid.UUID(document_id),
                 title=items[0].title,
                 document_type_id=uuid.UUID(items[0].document_type_id),
-                custom_fields=[(i.cf_name, str(i.cf_value)) for i in items],
+                custom_fields=custom_fields,
             )
         )
 

papermerge/core/schemas/custom_fields.py

@@ -1,5 +1,6 @@
 from datetime import date
 from enum import Enum
+from typing import TypeAlias
 from uuid import UUID
 
 from pydantic import BaseModel, ConfigDict
@@ -48,6 +49,10 @@ class CustomFieldValue(CustomField):
     field_id: UUID
 
 
+CFValueType: TypeAlias = str | int | date | bool | float | None
+CFNameType: TypeAlias = str
+
+
 class CFV(BaseModel):
     # custom field value
     # `core_documents.id`
@@ -57,15 +62,15 @@ class CFV(BaseModel):
     # `custom_fields.id`
     custom_field_id: UUID
     # `custom_fields.name`
-    name: str
+    name: CFNameType
     # `custom_fields.type`
     type: CustomFieldType
     # `custom_fields.extra_data`
     extra_data: str | None
     # `custom_field_values.id`
     custom_field_value_id: UUID | None = None
     # `custom_field_values.value_text` or `custom_field_values.value_int` or ...
-    value: str | int | date | bool | float | None = None
+    value: CFValueType = None
 
 
 class DocumentCFV(BaseModel):
@@ -77,4 +82,4 @@ class DocumentCFV(BaseModel):
     # user_id: UUID
     document_type_id: UUID | None = None
     thumbnail_url: str | None = None
-    custom_fields: list[tuple[str, str]]
+    custom_fields: list[tuple[CFNameType, CFValueType]]

tests/core/models/test_document.py

@@ -483,6 +483,80 @@ def test_document_with_cfv_update_document_type_to_none(
     assert db_session.execute(stmt).scalar() == 0
 
 
+@pytest.mark.django_db(transaction=True)
+def test_get_docs_by_type_basic(db_session: Session, make_document_receipt):
+    """
+    `db.get_docs_by_type` must return all documents of specific type
+    regardless if they (documents) have or no associated custom field values.
+
+    In this scenario all returned documents must have custom fields with empty
+    values.
+    And number of returned items must be equal to the number of documents
+    of type "Grocery"
+    """
+    doc_1: Document = make_document_receipt(title="receipt_1.pdf")
+    make_document_receipt(title="receipt_2.pdf")
+    user_id = doc_1.user.id
+    parent_id = doc_1.parent.id
+    type_id = doc_1.document_type.id
+
+    items: list[schemas.DocumentCFV] = db.get_docs_by_type(
+        db_session, type_id=type_id, user_id=user_id, ancestor_id=parent_id
+    )
+
+    assert len(items) == 2
+
+    for i in range(0, 2):
+        cf = dict(items[i].custom_fields)
+        assert cf["EffectiveDate"] is None
+        assert cf["Shop"] is None
+        assert cf["Total"] is None
+
+
+@pytest.mark.django_db(transaction=True)
+def test_get_docs_by_type_one_doc_with_nonempty_cfv(
+    db_session: Session, make_document_receipt
+):
+    """
+    `db.get_docs_by_type` must return all documents of specific type
+    regardless if they (documents) have or no associated custom field values.
+
+    In this scenario one of the returned documents has all CFVs set to
+    non empty values and the other one - to all values empty
+    """
+    doc_1: Document = make_document_receipt(title="receipt_1.pdf")
+    make_document_receipt(title="receipt_2.pdf")
+    user_id = doc_1.user.id
+    parent_id = doc_1.parent.id
+    type_id = doc_1.document_type.id
+
+    # update all CFV of receipt_1.pdf to non-empty values
+    db.update_doc_cfv(
+        db_session,
+        document_id=doc_1.id,
+        custom_fields={"Shop": "rewe", "EffectiveDate": "2024-10-15", "Total": "15.63"},
+    )
+
+    items: list[schemas.DocumentCFV] = db.get_docs_by_type(
+        db_session, type_id=type_id, user_id=user_id, ancestor_id=parent_id
+    )
+
+    assert len(items) == 2
+
+    for i in range(0, 2):
+        cf = dict(items[i].custom_fields)
+        if items[i].id == doc_1.id:
+            #  receipt_1.pdf has all cf set correctly
+            assert cf["EffectiveDate"] == Date(2024, 10, 15)
+            assert cf["Shop"] == "rewe"
+            assert cf["Total"] == 15.63
+        else:
+            # receipt_2.pdf has all cf set to None
+            assert cf["EffectiveDate"] is None
+            assert cf["Shop"] is None
+            assert cf["Total"] is None
+
+
 def test_str2date():
     assert str2date("2024-10-30") == datetime(2024, 10, 30).date()
     assert str2date("2024-10-30 00:00:00") == datetime(2024, 10, 30).date()