update sort-related docstring, and tests

open2c · Nov 15, 2021 · 4b1e193 · 4b1e193
1 parent faca970
commit 4b1e193
Show file tree

Hide file tree

Showing 3 changed files with 22 additions and 26 deletions.
diff --git a/bioframe/core/checks.py b/bioframe/core/checks.py
@@ -426,7 +426,7 @@ def is_sorted(
     df,
     view_df=None,
     reset_index=True,
-    df_view_col="view_region",
+    df_view_col=None,
     view_name_col="name",
     cols=None,
 ):
@@ -439,14 +439,19 @@ def is_sorted(
     ----------
     df : pandas.DataFrame
 
-    view_df : pandas.DataFrame or None
+    view_df : pandas.DataFrame | dict-like
         Optional view to pass to ``sort_bedframe``.
+        When it is dict-like :func:'bioframe.make_viewframe' will
+        be used to convert to viewframe. If view_df is not provided
+        df is assumed to be sorted by chrom and start.
 
     reset_index : bool
         Optional argument to pass to ``sort_bedframe``.
 
-    df_view_col: str
+    df_view_col: None | str
         Name of column from df that indicates region in view.
+        If None, :func:'bioframe.assign_view' will be used to assign view regions.
+        Default None.
 
     view_name_col: str
         Name of column from view that specifies unique region name.

diff --git a/bioframe/ops.py b/bioframe/ops.py
@@ -1577,19 +1577,21 @@ def sort_bedframe(
     df : pandas.DataFrame
         Valid bedframe.
 
-    view_df : pandas.DataFrame
-        Valid input to make a viewframe.
+    view_df : pandas.DataFrame | dict-like
+        Valid input to make a viewframe. When it is dict-like
+        :func:'bioframe.make_viewframe' will be used to convert
+        to viewframe. If view_df is not provided df is sorted by chrom and start.
 
     reset_index : bool
         Default True.
 
-    df_view_col:
+    df_view_col: None | str
         Column from 'df' used to associate intervals with view regions.
         The associated region in 'view_df' is then used for sorting.
-        If None, :func:'bioframe.ops.assign_view' will be used to assign view regions.
+        If None, :func:'bioframe.assign_view' will be used to assign view regions.
         Default None.
 
-    view_name_col:
+    view_name_col: str
         Column from view_df with names of regions.
         Default `name`.
 

diff --git a/tests/test_ops.py b/tests/test_ops.py
@@ -1649,10 +1649,10 @@ def test_sort_bedframe():
 
     df = pd.DataFrame(
         [
+            ["chr2", 5, 10, "+"],
             ["chr1", 0, 10, "+"],
             ["chrX", 5, 10, "+"],
             ["chrX", 0, 5, "+"],
-            ["chr2", 5, 10, "+"],
         ],
         columns=["chrom", "start", "end", "strand"],
     )
@@ -1674,23 +1674,12 @@ def test_sort_bedframe():
     # are placed last and view_region is returned as a categorical
     df_sorted = pd.DataFrame(
         [
-            ["chrX", 0, 5, "+", "oranges"],
-            ["chrX", 5, 10, "+", "oranges"],
-            ["chr1", 0, 10, "+", "apples"],
-            ["chr2", 5, 10, "+", pd.NA],
+            ["chrX", 0, 5, "+"],
+            ["chrX", 5, 10, "+"],
+            ["chr1", 0, 10, "+"],
+            ["chr2", 5, 10, "+"],
         ],
-        columns=["chrom", "start", "end", "strand", "view_region"],
-    )
-    df_view_cat = pd.CategoricalDtype(
-        categories=["oranges", "grapefruit", "apples"], ordered=True
-    )
-    df_sorted = df_sorted.astype(
-        {
-            "chrom": str,
-            "start": pd.Int64Dtype(),
-            "end": pd.Int64Dtype(),
-            "view_region": df_view_cat,
-        }
+        columns=["chrom", "start", "end", "strand"],
     )
 
     pd.testing.assert_frame_equal(
@@ -1737,5 +1726,5 @@ def test_sort_bedframe():
     )
     assert (
         df.dtypes
-        == bioframe.sort_bedframe(df, view_df, view_name_col="fruit").dtypes[:-1]
+        == bioframe.sort_bedframe(df, view_df, view_name_col="fruit").dtypes
     ).all()