fastplotlib · kushalkolar · Sep 17, 2024 · Jun 18, 2024 · Jun 18, 2024 · Jun 18, 2024
@@ -18,7 +18,7 @@ def parent(self) -> Graphic | None:
         return self._parent
 
     @property
-    def selection(self) -> Sequence[float] | List[Sequence[float]]:
+    def selection(self) -> np.ndarray[float]:
         """
         (xmin, xmax, ymin, ymax) of the rectangle selection
         """
@@ -319,10 +319,11 @@ def get_selected_data(
         # do not need to check for mode for images, because the selector is bounded by the image shape
         # will always be `full`
         if "Image" in source.__class__.__name__:
-            col_ixs = slice(ixs[0][0], ixs[0][-1] + 1)
-            row_ixs = slice(ixs[1][0], ixs[1][-1] + 1)
+            row_ixs, col_ixs = ixs
+            row_slice = slice(row_ixs[0], row_ixs[-1] + 1)
+            col_slice = slice(col_ixs[0], col_ixs[-1] + 1)
 
-            return source.data[row_ixs, col_ixs]
+            return source.data[row_slice, col_slice]
 
         if mode not in ["full", "partial", "ignore"]:
             raise ValueError(
@@ -414,7 +415,7 @@ def get_selected_data(
 
     def get_selected_indices(
         self, graphic: Graphic = None
-    ) -> Union[np.ndarray, List[np.ndarray]]:
+    ) -> np.ndarray | tuple[np.ndarray]:
         """
         Returns the indices of the ``Graphic`` data bounded by the current selection.
 
@@ -429,43 +430,43 @@ def get_selected_indices(
         -------
         Union[np.ndarray, List[np.ndarray]]
             data indicies of the selection
-            | list of [x_indices_array, y_indices_array] if the graphic is an image
+            | tuple of [row_indices, y_indices_array] if the graphic is an image
             | list of indices along the x-dimension for each line if graphic is a line collection
             | array of indices along the x-dimension if graphic is a line
         """
         # get indices from source
         source = self._get_source(graphic)
 
         # selector (xmin, xmax, ymin, ymax) values
-        bounds = self.selection
+        xmin, xmax, ymin, ymax = self.selection
 
         # image data does not need to check for mode because the selector is always bounded
         # to the image
         if "Image" in source.__class__.__name__:
-            xs = np.arange(bounds[0], bounds[1], dtype=int)
-            ys = np.arange(bounds[2], bounds[3], dtype=int)
-            return [xs, ys]
+            col_ixs = np.arange(xmin, xmax, dtype=int)
+            row_ixs = np.arange(ymin, ymax, dtype=int)
+            return row_ixs, col_ixs
 
         if "Line" in source.__class__.__name__:
             if isinstance(source, GraphicCollection):
                 ixs = list()
                 for g in source.graphics:
                     data = g.data.value
                     g_ixs = np.where(
-                        (data[:, 0] >= bounds[0] - g.offset[0])
-                        & (data[:, 0] <= bounds[1] - g.offset[0])
-                        & (data[:, 1] >= bounds[2] - g.offset[1])
-                        & (data[:, 1] <= bounds[3] - g.offset[1])
+                        (data[:, 0] >= xmin - g.offset[0])
+                        & (data[:, 0] <= xmax - g.offset[0])
+                        & (data[:, 1] >= ymin - g.offset[1])
+                        & (data[:, 1] <= ymax - g.offset[1])
                     )[0]
                     ixs.append(g_ixs)
             else:
                 # map only this graphic
                 data = source.data.value
                 ixs = np.where(
-                    (data[:, 0] >= bounds[0])
-                    & (data[:, 0] <= bounds[1])
-                    & (data[:, 1] >= bounds[2])
-                    & (data[:, 1] <= bounds[3])
+                    (data[:, 0] >= xmin)
+                    & (data[:, 0] <= xmax)
+                    & (data[:, 1] >= ymin)
+                    & (data[:, 1] <= ymax)
                 )[0]
 
             return ixs