filter prefixes - option to require all prefixes (#379)

* nothing commit

* committing so i can install in koza

* add to filter_out_prefix as well

* add extra tests for new options

* revert to two methods, but keep require_all_prefixes option

* bump version

* uncomment tests (#2)

* match filter out and filter return

* fix lint issues i think

* bumped version too far

* Remove unnecessary Args: from filter_prefixes

Author: Glass

.gitignore

@@ -1,4 +1,5 @@
-venv/
+.venv
+venv
 sssom/__pycache__/
 .idea
 sssom.egg-info
@@ -24,4 +25,4 @@ schema/sssom.schema.json
 schema/sssom.yaml
 schema/sssom_datamodel.py
 sssom/internal_context.py
-sssom/sssom_datamodel.py
+sssom/sssom_datamodel.py

README.md

@@ -23,7 +23,9 @@
 SSSOM (Simple Standard for Sharing Ontology Mappings) is a TSV and RDF/OWL standard for ontology mappings
 
 ```
-WARNING: the export formats (json, rdf) of sssom-py are not yet finalised! Please expect changes in future releases!
+WARNING: 
+    The export formats (json, rdf) of sssom-py are not yet finalised! 
+    Please expect changes in future releases!
 ```
 
 See https://github.com/OBOFoundry/SSSOM

setup.cfg

@@ -3,7 +3,7 @@
 ##########################
 [metadata]
 name = sssom
-version = 0.3.30
+version = 0.3.31
 description = Operations on SSSOM mapping tables
 long_description = file: README.md
 long_description_content_type = text/markdown

sssom/util.py

@@ -1170,55 +1170,57 @@ def get_prefixes_used_in_metadata(meta: MetadataType) -> List[str]:
 
 
 def filter_out_prefixes(
-    df: pd.DataFrame, filter_prefixes: List[str], features: list = KEY_FEATURES
+    df: pd.DataFrame,
+    filter_prefixes: List[str],
+    features: list = KEY_FEATURES,
+    require_all_prefixes: bool = False,
 ) -> pd.DataFrame:
-    """Filter any row where a CURIE in one of the key column uses one of the given prefixes.
+    """Filter out rows which contains a CURIE with a prefix in the filter_prefixes list.
 
-    :param df: Pandas DataFrame
+    :param df: Pandas DataFrame of SSSOM Mapping
     :param filter_prefixes: List of prefixes
     :param features: List of dataframe column names dataframe to consider
+    :param require_all_prefixes: If True, all prefixes must be present in a row to be filtered out
     :return: Pandas Dataframe
     """
     filter_prefix_set = set(filter_prefixes)
     rows = []
+    selection = all if require_all_prefixes else any
 
     for _, row in df.iterrows():
         prefixes = {get_prefix_from_curie(curie) for curie in row[features]}
-        # Confirm if none of the CURIEs in the list above appear in the filter_prefixes list.
-        # If TRUE, append row.
-        if not any(prefix in prefixes for prefix in filter_prefix_set):
+        if not selection(prefix in prefixes for prefix in filter_prefix_set):
             rows.append(row)
-    if rows:
-        return pd.DataFrame(rows)
-    else:
-        return pd.DataFrame(columns=features)
+
+    return pd.DataFrame(rows) if rows else pd.DataFrame(columns=features)
 
 
 def filter_prefixes(
-    df: pd.DataFrame, filter_prefixes: List[str], features: list = KEY_FEATURES
+    df: pd.DataFrame,
+    filter_prefixes: List[str],
+    features: list = KEY_FEATURES,
+    require_all_prefixes: bool = True,
 ) -> pd.DataFrame:
-    """Filter any row where a CURIE in one of the key column uses one of the given prefixes.
+    """Filter out rows which do NOT contain a CURIE with a prefix in the filter_prefixes list.
 
-    :param df: Pandas DataFrame
+    :param df: Pandas DataFrame of SSSOM Mapping
     :param filter_prefixes: List of prefixes
     :param features: List of dataframe column names dataframe to consider
+    :param require_all_prefixes: If True, all prefixes must be present in a row to be filtered out
     :return: Pandas Dataframe
     """
     filter_prefix_set = set(filter_prefixes)
     rows = []
+    selection = all if require_all_prefixes else any
 
     for _, row in df.iterrows():
         prefixes = {
             get_prefix_from_curie(curie) for curie in row[features] if curie is not None
         }
-        # Confirm if all of the CURIEs in the list above appear in the filter_prefixes list.
-        # If TRUE, append row.
-        if all(prefix in filter_prefix_set for prefix in prefixes):
+        if selection(prefix in filter_prefix_set for prefix in prefixes):
             rows.append(row)
-    if rows:
-        return pd.DataFrame(rows)
-    else:
-        return pd.DataFrame(columns=features)
+
+    return pd.DataFrame(rows) if rows else pd.DataFrame(columns=features)
 
 
 # TODO this is not used anywhere

tests/test_utils.py

@@ -34,24 +34,54 @@ def test_broken_predicate_list(self):
                 iri_list.extend(p_iri)
         self.assertEqual(3, len(iri_list))
 
-    def test_filter_prefixes(self):
+    def test_filter_prefixes_any(self):
         """Test filtering MSDF.df by prefixes provided."""
         prefix_filter_list = ["x", "y"]
         original_msdf = self.msdf
         filtered_df = filter_prefixes(
-            original_msdf.df, prefix_filter_list, self.features
+            original_msdf.df,
+            prefix_filter_list,
+            self.features,
+            require_all_prefixes=False,
+        )
+        self.assertEqual(len(filtered_df), 136)
+
+    def test_filter_prefixes_all(self):
+        """Test filtering MSDF.df by prefixes provided."""
+        prefix_filter_list = ["x", "y"]
+        original_msdf = self.msdf
+        filtered_df = filter_prefixes(
+            original_msdf.df,
+            prefix_filter_list,
+            self.features,
+            require_all_prefixes=True,
         )
         self.assertEqual(len(filtered_df), 40)
 
-    def test_filter_out_prefixes(self):
+    def test_filter_out_prefixes_any(self):
         """Test filtering MSDF.df by prefixes provided."""
         prefix_filter_list = ["x", "y"]
         original_msdf = self.msdf
         filtered_df = filter_out_prefixes(
-            original_msdf.df, prefix_filter_list, self.features
+            original_msdf.df,
+            prefix_filter_list,
+            self.features,
+            require_all_prefixes=False,
         )
         self.assertEqual(len(filtered_df), 5)
 
+    def test_filter_out_prefixes_all(self):
+        """Test filtering MSDF.df by prefixes provided."""
+        prefix_filter_list = ["x", "y"]
+        original_msdf = self.msdf
+        filtered_df = filter_out_prefixes(
+            original_msdf.df,
+            prefix_filter_list,
+            self.features,
+            require_all_prefixes=True,
+        )
+        self.assertEqual(len(filtered_df), 101)
+
     def test_remove_mappings(self):
         """Test remove mappings."""
         prefix_filter_list = ["x", "y"]