Add prepend_col_name for tags (#47)

* add prepend_col_name

* add test

* add changelog entry

docs/changelog.md

@@ -2,6 +2,12 @@
 
 This project attempts to follow [Semantic Versioning](https://semver.org) and uses [Keep-a-Changelog formatting](https://keepachangelog.com/en/1.0.0/). But I make mistakes sometimes.
 
+## [3.4.0] - 2024-08-08
+
+### Added
+
+- **SegmentProperties** New option `prepend_col_name` in `SegmentProperties.from_dataframe` will prepend the column name to the tag values.
+
 ## [3.3.7] - 2024-07-24
 
 ### Added

src/nglui/segmentprops/base.py

@@ -210,7 +210,10 @@ def _tag_descriptions(tags, tag_descriptions):
         return [tag_descriptions.get(tag, tag) for tag in tags]
 
 
-def _make_tag_map(df, value_columns, bool_columns, allow_disambiguation):
+def _make_tag_map(df, value_columns, bool_columns, allow_disambiguation, prepend_col_name):
+    if prepend_col_name:
+        for col in value_columns:
+            df[col] = df[col].apply(lambda x: f"{col}:{x}")
     unique_tags = {}
     for col in value_columns:
         col_tags = df[col].unique()
@@ -295,6 +298,7 @@ def _make_tag_property(
     tag_descriptions,
     name="tags",
     allow_disambiguation=True,
+    prepend_col_name=False,
 ):
     if value_columns is None:
         value_columns = []
@@ -305,6 +309,7 @@ def _make_tag_property(
         value_columns,
         bool_columns,
         allow_disambiguation,
+        prepend_col_name,
     )
     tag_values = _generate_tag_values(tag_df, value_columns, bool_columns, tag_map)
     return TagProperty(
@@ -412,6 +417,7 @@ def from_dataframe(
         allow_disambiguation: bool = True,
         label_separator: str = "_",
         label_format_map: Optional[str] = None,
+        prepend_col_name: bool = False,
     ):
         """Generate a segment property object from a pandas dataframe based on column
 
@@ -450,7 +456,9 @@ def from_dataframe(
             via the "format" function, replacing the column names in `{..}` with the values.
             For example, "{cell_class}: {cell_type}_{region}" would pluck values from the columns
             "cell_class", "cell_type", and "region". Label columns will be ignored and the format string is not validated.
-
+        prepend_col_name : bool, optional
+            If True, will prepend the column name to tag values, by default False.
+            
         Returns
         -------
         SegmentProperties
@@ -506,6 +514,7 @@ def from_dataframe(
                 tag_bool_cols,
                 tag_descriptions,
                 allow_disambiguation=allow_disambiguation,
+                prepend_col_name=prepend_col_name,
             )
         return cls(ids, **properties)
 

tests/test_segment_props.py

@@ -182,6 +182,22 @@ def test_categorical_props(test_categorical_df):
     assert p_dict["inline"]["properties"][2]["data_type"] == "int32"
 
 
+def test_prepend_col_name(test_categorical_df):
+    props = SegmentProperties.from_dataframe(
+        test_categorical_df,
+        id_col="seg_id",
+        label_col="seg_id",
+        number_cols=["number_int", "number_float"],
+        tag_value_cols="cell_type",
+        tag_bool_cols=["tag_a", "tag_b"],
+        prepend_col_name=True,
+    )
+
+    assert len(props) == 100
+    p_dict = props.to_dict()
+    assert "cell_type:ct_a" == p_dict["inline"]["properties"][1]["tags"][2]
+
+
 def test_segment_props_nulls(test_null_df):
     props = SegmentProperties.from_dataframe(
         test_null_df,