feat(data-process): 完善文件解析与切分存储链路

backend/tests/test_data_process_store.py

@@ -0,0 +1,54 @@
+from __future__ import annotations
+
+import pytest
+
+from app.modules.data_process.store import (
+    DataProcessStoreError,
+    _source_storage_descriptor,
+)
+
+
+def test_source_storage_descriptor_accepts_owned_local_and_legacy_db_references() -> None:
+    task_id = "dpt_task"
+    source_file_id = "dpsf_source"
+    local_reference = (
+        f"local://data-process/{task_id}/{source_file_id}/v1/source%20100%25.csv"
+    )
+
+    reference, metadata = _source_storage_descriptor(
+        {
+            "storage_object_id": local_reference,
+            "metadata": {"storage_backend": "spoofed", "content_type": "text/csv"},
+        },
+        task_id,
+        source_file_id,
+    )
+    assert reference == local_reference
+    assert metadata == {"storage_backend": "local", "content_type": "text/csv"}
+
+    legacy_reference, legacy_metadata = _source_storage_descriptor(
+        {"metadata": {"legacy": True}},
+        task_id,
+        source_file_id,
+    )
+    assert legacy_reference == f"db://data-process/{task_id}/{source_file_id}/v1"
+    assert legacy_metadata == {"storage_backend": "database", "legacy": True}
+
+
+@pytest.mark.parametrize(
+    "reference",
+    [
+        "local://data-process/dpt_other/dpsf_source/v1/source.txt",
+        "db://data-process/dpt_task/dpsf_other/v1",
+        "/var/tmp/source.txt",
+    ],
+)
+def test_source_storage_descriptor_rejects_unowned_or_unsupported_references(
+    reference: str,
+) -> None:
+    with pytest.raises(DataProcessStoreError):
+        _source_storage_descriptor(
+            {"storage_object_id": reference},
+            "dpt_task",
+            "dpsf_source",
+        )