ENH: use LineCollection for wide DataFrame.line plots

Author: EvMossan

pandas/plotting/_matplotlib/core.py

@@ -1538,6 +1538,7 @@ def _make_legend(self) -> None:
 
 class LinePlot(MPLPlot):
     _default_rot = 0
+    _wide_line_threshold: int = 200
 
     @property
     def orientation(self) -> PlottingOrientation:
@@ -1547,93 +1548,100 @@ def orientation(self) -> PlottingOrientation:
     def _kind(self) -> Literal["line", "area", "hist", "kde", "box"]:
         return "line"
 
-    def __init__(self, data, **kwargs) -> None:
-        MPLPlot.__init__(self, data, **kwargs)
+    def __init__(self, data, **kwargs):
+        super().__init__(data, **kwargs)
         if self.stacked:
             self.data = self.data.fillna(value=0)
 
-    def _make_plot(self, fig: Figure) -> None:
-        threshold = 200  # switch when DataFrame has more than this many columns
-        can_use_lc = (
-            not self._is_ts_plot()  # not a TS plot
-            and not self.stacked  # stacking not requested
-            and not com.any_not_none(*self.errors.values())  # no error bars
-            and len(self.data.columns) > threshold
+    def _make_plot(self, fig):
+        is_ts = self._is_ts_plot()
+
+        use_lc = (
+            not is_ts
+            and not self.stacked
+            and not com.any_not_none(*self.errors.values())
+            and len(self.data.columns) > self._wide_line_threshold
         )
-        if can_use_lc:
-            ax = self._get_ax(0)
-            x = self._get_xticks()
-            segments = [
-                np.column_stack((x, self.data[col].values)) for col in self.data.columns
-            ]
-            base_colors = mpl.rcParams["axes.prop_cycle"].by_key()["color"]
-            colors = list(itertools.islice(itertools.cycle(base_colors), len(segments)))
-            lc = LineCollection(
-                segments,
-                colors=colors,
-                linewidths=self.kwds.get("linewidth", mpl.rcParams["lines.linewidth"]),
-            )
-            ax.add_collection(lc)
-            ax.margins(0.05)
-            return  # skip the per-column Line2D loop
 
-        if self._is_ts_plot():
+        if is_ts:
             data = maybe_convert_index(self._get_ax(0), self.data)
-
-            x = data.index  # dummy, not used
+            x_vals = data.index  # (ignored by _ts_plot)
             plotf = self._ts_plot
-            it = data.items()
+            iterator: Iterable = data.items()
+        else:
+            x_vals = self._get_xticks()
+            plotf = self._plot
+            iterator = self._iter_data(self.data)
+
+        # drawing step
+        if use_lc:
+            self._draw_with_linecollection(x_vals, iterator)
         else:
-            x = self._get_xticks()
-            # error: Incompatible types in assignment (expression has type
-            # "Callable[[Any, Any, Any, Any, Any, Any, KwArg(Any)], Any]", variable has
-            # type "Callable[[Any, Any, Any, Any, KwArg(Any)], Any]")
-            plotf = self._plot  # type: ignore[assignment]
-            # error: Incompatible types in assignment (expression has type
-            # "Iterator[tuple[Hashable, ndarray[Any, Any]]]", variable has
-            # type "Iterable[tuple[Hashable, Series]]")
-            it = self._iter_data(data=self.data)  # type: ignore[assignment]
+            self._draw_iteratively(x_vals, plotf, iterator, is_ts=is_ts)
+
+        self._post_plot_logic(self._get_ax(0), self.data)
+
+    # fast path: one LineCollection
+    def _draw_with_linecollection(self, x_vals, iterator):
+        ax: Axes = self._get_ax(0)
+        n_rows = len(x_vals)
+        n_cols = len(self.data.columns)
+
+        # vertices: vectorised (n_cols, n_rows, 2)
+        x_vec = np.asarray(x_vals, dtype=float)
+        x2d = np.broadcast_to(x_vec[:, None], (n_rows, n_cols))
+        seg_array = np.dstack((x2d, self.data.values)).transpose(1, 0, 2)
+
+        # colours
+        base = mpl.rcParams["axes.prop_cycle"].by_key()["color"]
+        colours = list(itertools.islice(itertools.cycle(base), n_cols))
 
+        # legend handling
+        if self.legend:
+            # Build proxy Line2D handles once (keeps API parity and tests).
+            proxy_handles = [
+                mpl.lines.Line2D([], [], color=colours[i], label=pprint_thing(lbl))
+                for i, (lbl, _) in enumerate(iterator)
+            ]
+            for h in proxy_handles:
+                self._append_legend_handles_labels(h, h.get_label())
+
+        lc = LineCollection(
+            seg_array,
+            colors=colours,
+            linewidths=self.kwds.get("linewidth", mpl.rcParams["lines.linewidth"]),
+        )
+        ax.add_collection(lc)
+        ax.margins(0.05)
+
+    def _draw_iteratively(self, x_vals, plotf, iterator, *, is_ts: bool):
         stacking_id = self._get_stacking_id()
         is_errorbar = com.any_not_none(*self.errors.values())
+        colours = self._get_colors()
 
-        colors = self._get_colors()
-        for i, (label, y) in enumerate(it):
-            ax = self._get_ax(i)
+        for i, (label, y) in enumerate(iterator):
+            ax: Axes = self._get_ax(i)
             kwds = self.kwds.copy()
             if self.color is not None:
                 kwds["color"] = self.color
-            style, kwds = self._apply_style_colors(
-                colors,
-                kwds,
-                i,
-                # error: Argument 4 to "_apply_style_colors" of "MPLPlot" has
-                # incompatible type "Hashable"; expected "str"
-                label,  # type: ignore[arg-type]
-            )
 
-            errors = self._get_errorbars(label=label, index=i)
-            kwds = dict(kwds, **errors)
-
-            label = pprint_thing(label)
-            label = self._mark_right_label(label, index=i)
-            kwds["label"] = label
+            style, kwds = self._apply_style_colors(colours, kwds, i, label)
+            kwds.update(self._get_errorbars(label, i))
+            kwds["label"] = self._mark_right_label(pprint_thing(label), index=i)
 
             newlines = plotf(
                 ax,
-                x,
+                x_vals,
                 y,
                 style=style,
                 column_num=i,
                 stacking_id=stacking_id,
                 is_errorbar=is_errorbar,
                 **kwds,
             )
-            self._append_legend_handles_labels(newlines[0], label)
+            self._append_legend_handles_labels(newlines[0], kwds["label"])
 
-            if self._is_ts_plot():
-                # reset of xlim should be used for ts data
-                # TODO: GH28021, should find a way to change view limit on xaxis
+            if is_ts:
                 lines = get_all_lines(ax)
                 left, right = get_xlim(lines)
                 ax.set_xlim(left, right)