👔 Look up REDCap indexes instead of relying on +1 heuristic

Author: shnizzedy

pyproject.toml

@@ -91,6 +91,7 @@ extend-ignore = [
 ]
 
 [tool.ruff.lint.extend-per-file-ignores]
+"tests/test_redcap_format.py" = ["PLR2004"]
 "tests/**/*.py" = ["S101", "SLF001", "INP001", "D103", "ERA001"]
 "scripts/*.py" = ["T20", "D100", "N803", "B018", "ERA001", "PLR0913", "FBT003"]
 "scripts/*.ipynb" = [

src/mindlogger_data_export/outputs.py

@@ -444,20 +444,33 @@ def _parse_redcap_choice_mapping(self, field_name: str) -> dict[int, str] | None
         Parameters
         ----------
         field_name
-            The field name to look up in REDCap metadata
+            The base field name (without suffix) to look up in REDCap metadata
 
         Returns:
         -------
         dict[int, str] | None
-            Mapping from Curious index to REDCap raw value, or None if not found
+            Mapping from Mindlogger index to REDCap raw value, or None if not found
         """
         if self._redcap_metadata is None or self._redcap_metadata.is_empty():
             return None
 
-        # Get choices for this field from metadata
-        field_meta = self._redcap_metadata.filter(pl.col("field_name") == field_name)
+        # The REDCap metadata contains field names with suffixes (_response, _index, etc.)
+        # We need to try different variations to find the actual field
+        possible_field_names = [
+            field_name,  # Try base name first
+            f"{field_name}_response",  # Try with _response suffix
+            f"{field_name}_index",  # Try with _index suffix
+        ]
+
+        field_meta = None
+        for possible_name in possible_field_names:
+            field_meta = self._redcap_metadata.filter(
+                pl.col("field_name") == possible_name
+            )
+            if not field_meta.is_empty():
+                break
 
-        if field_meta.is_empty():
+        if field_meta is None or field_meta.is_empty():
             return None
 
         choices_str = field_meta["select_choices_or_calculations"].item()

tests/test_mindlogger_data.py

@@ -192,3 +192,38 @@ def test_expand_responses(report):
 def test_data_dictionary(report):
     _data = MindloggerData(report)
     assert len(list(_data.data_dictionary)) != 0
+
+
+@WITH_REPORT
+def test_redcap_format_integration_with_test_data(datafiles: Path):
+    """Integration test for RedcapImportFormat with actual test data."""
+    from mindlogger_data_export.outputs import RedcapImportFormat
+
+    mindlogger_data = MindloggerData.create(datafiles)
+
+    # Create REDCap metadata with mixed suffix patterns
+    redcap_metadata = pl.DataFrame(
+        {
+            "field_name": [
+                "mood1_1_response",  # Has _response suffix
+                "mood2_1",  # Base name
+                "sleepquality_1_index",  # Has _index suffix
+            ],
+            "select_choices_or_calculations": [
+                "0, 0-Very Cheerful | 6, 6-Very Sad",
+                "0, 0-Very Relaxed | 6, 6-Very Anxious",
+                "0, 0-Very Poor | 4, 4-Very Good",
+            ],
+        }
+    )
+
+    # Create RedcapImportFormat with metadata
+    redcap_format = RedcapImportFormat(redcap_metadata=redcap_metadata)
+
+    # Use produce() instead of non-existent to_wide()
+    outputs = redcap_format.produce(mindlogger_data)
+
+    # Should successfully process despite suffix pattern differences
+    assert outputs is not None
+    assert len(outputs) > 0
+    assert all(isinstance(output.output, pl.DataFrame) for output in outputs)