Merge pull request #434 from enthought/doc/new_scatter_inspector_example

Doc: second scatter inspector example

examples/demo/basic/scatter_inspector2.py

@@ -0,0 +1,112 @@
+""" Overlay to display the data attached to a scatter point hovered over.
+"""
+import pandas as pd
+import numpy as np
+
+from traits.api import Callable, Enum, HasTraits, Instance, on_trait_change, \
+    Str
+from traitsui.api import View, Item
+from enable.api import ComponentEditor
+from chaco.api import Plot, ArrayPlotData, ScatterInspectorOverlay, \
+    TextBoxOverlay
+from chaco.api import DataFramePlotData
+from chaco.tools.api import ScatterInspector
+
+
+class DataframeScatterInspector(ScatterInspector):
+    #: Data structure to gather all data to display neatly
+    data = Instance(pd.DataFrame)
+
+
+class DataframeScatterOverlay(TextBoxOverlay):
+    """ Overlay for displaying hovered data point information.
+    """
+    #: The inspector tool which has hover information
+    inspector = Instance(ScatterInspector)
+
+    #: Function which takes and index and returns an info string.
+    message_for_data = Callable
+
+    @on_trait_change('inspector:inspector_event')
+    def scatter_point_found(self, event):
+        data_idx = event.event_index
+        if data_idx is not None:
+            self.text = self.message_for_data(data_idx)
+        else:
+            self.text = ""
+
+        self.visible = len(self.text) > 0
+        self.component.request_redraw()
+
+    def _message_for_data_default(self):
+        def show_data(data_idx):
+            data = self.inspector.data.iloc[data_idx]
+            elements = ["idx: {}".format(data_idx)]
+            for col in data.index:
+                elements.append("{}: {}".format(col, data[col]))
+            text = "\n".join(elements)
+            return text
+
+        return show_data
+
+
+if __name__ == "__main__":
+
+    def _create_plot_component():
+        # Create a fake dataset from which 2 dimensions will be displayed in a
+        # scatter plot:
+        x = np.random.uniform(0.0, 10.0, 50)
+        y = np.random.uniform(0.0, 5.0, 50)
+        data = pd.DataFrame({"x": x, "y": y,
+                             "dataset": np.random.choice(list("abcdefg"), 50)})
+        plot_data = ArrayPlotData(x=x, y=y)
+        plot = Plot(plot_data)
+        scatter = plot.plot(("x", "y"), type="scatter")[0]
+
+        # Attach the inspector and its overlays
+        inspector = DataframeScatterInspector(
+            component=scatter, data=data
+        )
+        scatter.tools.append(inspector)
+
+        text_overlay = DataframeScatterOverlay(component=plot,
+                                               inspector=inspector,
+                                               bgcolor="black", alpha=0.6,
+                                               text_color="white",
+                                               border_color='none')
+        plot.overlays.append(text_overlay)
+
+        # Optional: add an overlay on the point to confirm what is hovered over
+        # Note that this overlay magically knows about hovered points by
+        # listening to renderer events rather than inspector events:
+        point_overlay = ScatterInspectorOverlay(component=scatter,
+                                                hover_color="red",
+                                                hover_marker_size=6)
+        scatter.overlays.append(point_overlay)
+        return plot
+
+
+# =============================================================================
+# Demo class that is used by the demo.py application.
+# =============================================================================
+
+size = (900, 500)
+title = "Tooltip demo"
+
+
+class Demo(HasTraits):
+    plot = Instance(Plot)
+
+    traits_view = View(
+        Item('plot', editor=ComponentEditor(size=size), show_label=False),
+        resizable=True, title=title
+    )
+
+    def _plot_default(self):
+        return _create_plot_component()
+
+
+demo = Demo()
+
+if __name__ == "__main__":
+    demo.configure_traits()