skip blanks in add_plate_info

src/pmotools/pmo_builder/metatable_to_pmo.py

@@ -487,6 +487,14 @@ def check_columns_exist(df, columns):
         )
 
 
+def _is_blank(value):
+    if value is None:
+        return True
+    if isinstance(value, str):
+        return not value.strip()
+    return bool(pd.isna(value))
+
+
 def add_plate_info(
     plate_col_col,
     plate_name_col,
@@ -516,38 +524,39 @@ def add_plate_info(
             plate_row_col = "plate_row"
             plate_col_col = "plate_col"
 
-            try:
-                df[plate_row_col] = (
-                    df[plate_position_col].str.extract(r"(?i)^([A-H])")[0].str.upper()
-                )
-                df[plate_col_col] = (
-                    df[plate_position_col]
-                    .str.extract(r"(?i)^[A-H]0*([1-9]|1[0-2])$")[0]
-                    .astype(int)
-                )
-            except (AttributeError, ValueError, IndexError, KeyError) as e:
+            positions = df[plate_position_col]
+            blank = positions.map(_is_blank)
+            parsed = (
+                positions[~blank]
+                .astype(str)
+                .str.strip()
+                .str.extract(r"(?i)^([A-H])0*([1-9]|1[0-2])$")
+            )
+            if parsed.isna().any().any():
                 raise ValueError(
                     f"Values in '{plate_position_col}' must start with a single letter A-H/a-h followed by number 1-12."
-                ) from e
+                )
+            df[plate_row_col] = None
+            df.loc[~blank, plate_row_col] = parsed[0].str.upper()
+            df[plate_col_col] = pd.Series(pd.NA, index=df.index, dtype="Int64")
+            df.loc[~blank, plate_col_col] = parsed[1].astype(int)
 
     for row in meta_json:
         content_row = df[df[match_col] == row[match_col]]
         plate_name_val = content_row[plate_name_col].iloc[0] if plate_name_col else None
-        plate_row_val = (
-            content_row[plate_row_col].iloc[0].upper() if plate_row_col else None
-        )
+        plate_row_val = content_row[plate_row_col].iloc[0] if plate_row_col else None
         plate_col_val = content_row[plate_col_col].iloc[0] if plate_col_col else None
-        if plate_col_val is not None and not pd.isna(plate_col_val):
+        if not _is_blank(plate_col_val):
             try:
                 plate_col_val = int(plate_col_val)
             except (TypeError, ValueError):
                 plate_col_val = plate_col_val
         plate_info = {}
-        if plate_name_val:
+        if not _is_blank(plate_name_val):
             plate_info["plate_name"] = plate_name_val
-        if plate_row_val:
-            plate_info["plate_row"] = plate_row_val
-        if plate_col_val is not None and not pd.isna(plate_col_val):
+        if not _is_blank(plate_row_val):
+            plate_info["plate_row"] = str(plate_row_val).strip().upper()
+        if not _is_blank(plate_col_val):
             plate_info["plate_col"] = plate_col_val
 
         if plate_info:

tests/test_pmo_builder/test_metatable_to_pmo.py

@@ -219,6 +219,81 @@ def test_add_plate_info_adds_nothing(self):
             result, [{"specimen_name": "sample1"}, {"specimen_name": "sample2"}]
         )
 
+    def test_add_plate_info_row_col_skips_blank_cells(self):
+        df = pd.DataFrame(
+            {
+                "specimen_name": ["sample1", "sample2"],
+                "plate_row": ["A", None],
+                "plate_col": [1, None],
+                "plate_name": ["Plate1", None],
+            }
+        )
+        result = add_plate_info(
+            "plate_col",
+            "plate_name",
+            "plate_row",
+            None,
+            self.small_json_example,
+            df,
+            "specimen_name",
+        )
+        self.assertEqual(
+            result[0]["plate_info"],
+            {"plate_name": "Plate1", "plate_row": "A", "plate_col": 1},
+        )
+        self.assertNotIn("plate_info", result[1])
+
+    def test_add_plate_info_position_skips_blank_cells(self):
+        df = pd.DataFrame(
+            {
+                "specimen_name": ["sample1", "sample2", "sample3"],
+                "plate_position": ["A01", None, "  "],
+                "plate_name": ["Plate1", float("nan"), ""],
+            }
+        )
+        json_example = [
+            {"specimen_name": "sample1"},
+            {"specimen_name": "sample2"},
+            {"specimen_name": "sample3"},
+        ]
+        result = add_plate_info(
+            None,
+            "plate_name",
+            None,
+            "plate_position",
+            json_example,
+            df,
+            "specimen_name",
+        )
+        self.assertEqual(
+            result[0]["plate_info"],
+            {"plate_name": "Plate1", "plate_row": "A", "plate_col": 1},
+        )
+        self.assertNotIn("plate_info", result[1])
+        self.assertNotIn("plate_info", result[2])
+
+    def test_add_plate_info_position_fails_with_invalid_value_among_blanks(self):
+        df = pd.DataFrame(
+            {
+                "specimen_name": ["sample1", "sample2"],
+                "plate_position": [None, "K10"],
+            }
+        )
+        with self.assertRaises(ValueError) as context:
+            add_plate_info(
+                None,
+                None,
+                None,
+                "plate_position",
+                self.small_json_example,
+                df,
+                "specimen_name",
+            )
+        self.assertEqual(
+            "Values in 'plate_position' must start with a single letter A-H/a-h followed by number 1-12.",
+            str(context.exception),
+        )
+
     def test_add_parasite_density_info_single_value(self):
         df = pd.DataFrame(
             {