Chore: remove duplicate zone OCR tests and assert on untested ones

This commit is contained in:
stumpylog committed 2026-11-18 12:48:20 -08:00
1 parent 1c6742723f
commit d706904a18
2 files changed
+43 -24

No files matched your search

@@ -2276,7 +2276,6 @@ class TestOcrTemplatePreviewPermissions:
grant_global(regular_user, "view_ocrtemplate", "add_ocrtemplate")
document = DocumentFactory(owner=None, mime_type="image/png")
document.source_path.write_bytes(b"document contents")
mock_preview.return_value = {"raw_text": "document contents", "value": None}
if action == "document-page-image":
response = user_client.get(preview_url(document))
+43 -23
View File
@@ -1448,13 +1448,6 @@ class TestApplyTransform:
"Account #123",
id="strip-punctuation",
),
pytest.param(
"2026-13-32",
"date",
"",
"2026-13-32",
id="invalid-iso-date-is-not-transformed",
),
],
)
def test_apply_transform(
@@ -1990,16 +1983,21 @@ class TestResolveDocPath:
original.touch()
assert _resolve_doc_path(document, original) == original
def test_returns_none_for_none_original_file(self, document: MagicMock) -> None:
def test_returns_none_for_missing_original_file(
self,
document: MagicMock,
tmp_path: Path,
) -> None:
"""
GIVEN:
- A document whose own files are missing and no original file
- A document whose own files are missing and an original file path
that does not exist
WHEN:
- The document path is resolved
- The document path is resolved with the original file
THEN:
- None is returned
"""
assert _resolve_doc_path(document, None) is None
assert _resolve_doc_path(document, tmp_path / "missing.pdf") is None
@pytest.mark.django_db
@@ -2081,6 +2079,7 @@ class TestRunZoneExtraction:
def test_skips_template_with_no_zones(
self,
mock_process: MagicMock,
document: Document,
pdf: Path,
) -> None:
@@ -2090,16 +2089,17 @@ class TestRunZoneExtraction:
WHEN:
- Zone extraction runs
THEN:
- No custom field instances are created
- No results are returned and the template is not processed
"""
OcrTemplateFactory(
name="Empty",
match="test",
)
run_zone_extraction(document, pdf)
results = run_zone_extraction(document, pdf)
assert CustomFieldInstance.objects.count() == 0
assert results == []
mock_process.assert_not_called()
def test_calls_process_for_enabled_template(
self,
@@ -2135,25 +2135,45 @@ class TestRunZoneExtraction:
WHEN:
- Zone extraction runs
THEN:
- The error does not propagate, so other templates can still run
- The error does not propagate, every zone of the failing template
is reported as template_failed and the template was attempted
once
"""
mock_process.side_effect = RuntimeError("test error")
template_with_zone("Failing", enabled=True)
template = template_with_zone("Failing", enabled=True)
zone = template.zones.get()
# Should not raise
run_zone_extraction(document, pdf)
results = run_zone_extraction(document, pdf)
def test_handles_none_original_file(self, document: Document) -> None:
assert [(r.template, r.zone, r.value, r.reason) for r in results] == [
("Failing", zone.name, None, "template_failed"),
]
mock_process.assert_called_once()
def test_handles_none_original_file(
self,
mock_process: MagicMock,
template_with_zone: Callable[..., OcrTemplate],
document: Document,
) -> None:
"""
GIVEN:
- A document and no original file
- A matching template, a document with no accessible file and no
original file
WHEN:
- Zone extraction runs
THEN:
- It does not crash
- Each zone is reported as missing_file and nothing is processed
"""
# No template, so it exits early, but shouldn't crash on None
run_zone_extraction(document, None)
template = template_with_zone("Matching", enabled=True)
zone = template.zones.get()
results = run_zone_extraction(document, None)
assert [(r.template, r.zone, r.reason) for r in results] == [
("Matching", zone.name, "missing_file"),
]
mock_process.assert_not_called()
def test_multiple_templates_all_process(
self,