wip

Author: nicolaskruchten

Diff for: packages/python/plotly/plotly/express/_core.py

@@ -899,20 +899,36 @@ def build_dataframe(args, attrables, array_attrables, constructor):
     if df_provided and not isinstance(args["data_frame"], pd.DataFrame):
         args["data_frame"] = pd.DataFrame(args["data_frame"])
 
-    if not args.get("x", None) and not args.get("y", None) and df_provided:
+    wide_traces = [go.Scatter, go.Bar, go.Violin, go.Box, go.Histogram]
+    has_x = args.get("x", None) is not None
+    has_y = args.get("y", None) is not None
+    if not has_x and not has_y and df_provided and constructor in wide_traces:
+        index_name = args["data_frame"].index.name or "index"
+        id_vars = [index_name]
+        # TODO multi-level index
+        # TODO multi-level columns
+        # TODO orientation
+
+        # TODO do we need to add everything to this candidate list basically? array_attrables?
+        # TODO will we need to be able to glue in non-string values here, like arrays and stuff?
+        # ...like maybe this needs to run after we've glued together the data frame?
+        for candidate in ["color", "symbol", "line_dash", "facet_row", "facet_col"] + [
+            "line_group",
+            "animation_group",
+        ]:
+            if args.get(candidate, None) not in [None, index_name, "value", "variable"]:
+                id_vars.append(args[candidate])
+        args["data_frame"] = args["data_frame"].reset_index().melt(id_vars=id_vars)
         if constructor in [go.Scatter, go.Bar]:
-            args["data_frame"] = args["data_frame"].reset_index().melt(id_vars="index")
-            args["x"] = "index"
+            args["x"] = index_name
             args["y"] = "value"
-            args["color"] = "variable"
+            args["color"] = args["color"] or "variable"
         if constructor in [go.Violin, go.Box]:
-            args["data_frame"] = args["data_frame"].reset_index().melt(id_vars="index")
             args["x"] = "variable"
             args["y"] = "value"
         if constructor in [go.Histogram]:
-            args["data_frame"] = args["data_frame"].reset_index().melt(id_vars="index")
             args["x"] = "value"
-            args["color"] = "variable"
+            args["color"] = args["color"] or "variable"
 
     df_input = args["data_frame"]
 

Diff for: packages/python/plotly/plotly/tests/test_core/test_px/test_px_input.py

@@ -225,7 +225,7 @@ def test_build_df_from_lists():
     output = {key: key for key in args}
     df = pd.DataFrame(args)
     args["data_frame"] = None
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(df.sort_index(axis=1), out["data_frame"].sort_index(axis=1))
     out.pop("data_frame")
     assert out == output
@@ -235,7 +235,7 @@ def test_build_df_from_lists():
     output = {key: key for key in args}
     df = pd.DataFrame(args)
     args["data_frame"] = None
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(df.sort_index(axis=1), out["data_frame"].sort_index(axis=1))
     out.pop("data_frame")
     assert out == output
@@ -244,7 +244,7 @@ def test_build_df_from_lists():
 def test_build_df_with_index():
     tips = px.data.tips()
     args = dict(data_frame=tips, x=tips.index, y="total_bill")
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(tips.reset_index()[out["data_frame"].columns], out["data_frame"])
 
 
@@ -254,15 +254,15 @@ def test_non_matching_index():
     expected = pd.DataFrame(dict(x=["a", "b", "c"], y=[1, 2, 3]))
 
     args = dict(data_frame=df, x=df.index, y="y")
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(expected, out["data_frame"])
 
     args = dict(data_frame=None, x=df.index, y=df.y)
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(expected, out["data_frame"])
 
     args = dict(data_frame=None, x=["a", "b", "c"], y=df.y)
-    out = build_dataframe(args, all_attrables, array_attrables)
+    out = build_dataframe(args, all_attrables, array_attrables, None)
     assert_frame_equal(expected, out["data_frame"])