new files types inferred from file names 🟥

backend/pashub_fetcher/tests/test_core_files.py

@@ -0,0 +1,64 @@
+import pytest
+
+from backend.pashub_fetcher.core_files import infer_file_type
+
+
+# --- GREEN: pre-existing file types (startswith match) ---
+
+
+def test_infer_photopack():
+    assert infer_file_type("Photopack_123456_V1.pdf") == "photo_pack"
+
+
+def test_infer_sitenote():
+    assert infer_file_type("SiteNote_123456_V1.pdf") == "site_note"
+
+
+def test_infer_rdsap_sitenote():
+    assert infer_file_type("RdSAP_SiteNote_9510890_V1_Assessmet.pdf") == "rd_sap_site_note"
+
+
+def test_infer_pas2023_ventilation():
+    assert infer_file_type("PAS 2023 Ventilation Assessment Report_123456.pdf") == "pas_2023_ventilation"
+
+
+def test_infer_pas2023_condition():
+    assert infer_file_type("PAS 2023 Condition Report_123456.pdf") == "pas_2023_condition"
+
+
+def test_infer_pas_significance():
+    assert infer_file_type("PAS Significance_123456.pdf") == "pas_significance"
+
+
+def test_infer_par_photopack():
+    assert infer_file_type("PAR Photo Pack_95101890_V2_Assessment.pdf") == "par_photo_pack"
+
+
+def test_infer_pas2023_property():
+    assert infer_file_type("PAS 2023 Property Assessment Report_123456.pdf") == "pas_2023_property"
+
+
+def test_infer_pas2023_occupancy():
+    assert infer_file_type("PAS 2023 Occupancy Assessment Report_123456.pdf") == "pas_2023_occupancy"
+
+
+def test_infer_unknown_returns_none():
+    assert infer_file_type("unknown_document_123.pdf") is None
+
+
+# --- RED: new file types (suffix match not yet implemented) ---
+
+
+def test_infer_improvement_option_evaluation():
+    # filename: "{job_id} - {postcode} - Improvement Option Evaluation.pdf"
+    assert infer_file_type("6000802 - NG4 4HD - Improvement Option Evaluation.pdf") == "improvement_option_evaluation"
+
+
+def test_infer_medium_term_improvement_plan():
+    # filename: "{job_id} - {postcode} - Medium Term Improvement Plan IOE.pdf"
+    assert infer_file_type("60800802 - NG4 4HD - Medium Term Improvement Plan IOE.pdf") == "medium_term_improvement_plan"
+
+
+@pytest.mark.skip(reason="Retrofit Design Doc filename pattern not yet known")
+def test_infer_retrofit_design_doc():
+    assert infer_file_type("2512-OSM-H56M900-XX-DR-N-A_Radford Road 408.pdf") == "retrofit_design_doc"

pytest.ini

@@ -3,6 +3,27 @@ pythonpath = .
 log_cli = true
 log_cli_level = INFO
 addopts = --cov-report term-missing --cov=etl/epc --cov=recommendations --cov=backend --cov=etl/epc_clean --cov=etl/spatial
-testpaths = recommendations/tests backend/tests etl/epc/tests etl/epc_clean/tests etl/spatial/tests backend/condition/tests backend/address2UPRN/tests backend/onboarders/tests backend/categorisation/tests backend/export/tests etl/hubspot/tests datatypes/epc/schema/tests datatypes/epc/surveys/tests datatypes/epc/domain/tests backend/ecmk_fetcher/tests/ backend/pashub_fetcher/tests backend/documents_parser/tests backend/magic_plan/tests datatypes/magicplan/api/tests datatypes/magicplan/domain/tests backend/app/db/functions/tests
+testpaths =
+    recommendations/tests
+    backend/tests
+    backend/address2UPRN/tests
+    backend/app/db/functions/tests
+    backend/categorisation/tests
+    backend/condition/tests
+    backend/documents_parser/tests
+    backend/ecmk_fetcher/tests
+    backend/export/tests
+    backend/magic_plan/tests
+    backend/onboarders/tests
+    backend/pashub_fetcher/tests
+    datatypes/epc/domain/tests
+    datatypes/epc/schema/tests
+    datatypes/epc/surveys/tests
+    datatypes/magicplan/api/tests
+    datatypes/magicplan/domain/tests
+    etl/epc/tests
+    etl/epc_clean/tests
+    etl/hubspot/tests
+    etl/spatial/tests
 markers =
     integration: mark a test as an integration test