CursorTool does not work with LogMapper for both axes

[spescha]

Using LogMapper for both axes

mag = LinePlot(index = ArrayDataSource(w, sort_order='ascending'),
                        value = ArrayDataSource(H),
                        color = curve_color,
                        name = curve_name,
                        index_mapper = LogMapper(range=view_mag.index_range),
                        value_mapper = LogMapper(range=view_mag.value_range))

leads to the error below, when adding a cursor with following lines

cursor1 = CursorTool(mag,
                        drag_button="left",
                        show_value_line=False,
                        show_marker=True,
                        line_width=2.0,
                        color='black')
mag.overlays.append(cursor1)

The problem is, that LogMapper.map_screen returns a 1-element array, while LinearMapper returns a scalar value, if only one data point has to be mapped.

Following subclass of LogMapper works for me as a workaround

class LogMapperCorr(LogMapper):
    """
    Corrects chaco.log_mapper.LogMapper for the usage with cursors. Cursors need
    a 1D-array iterable to unpack the x and y values and the LogMapper returns
    a 2D-array. Therefore LogMapper.map_screen has to return a scalar, if the length 
    of the data array is 1.
    """

    def __init__(self, *args, **kwargs):
        LogMapper.__init__(self, *args, **kwargs)

    def map_screen(self, data_array):
        arr = LogMapper.map_screen(self, data_array)

        if len(arr)==1:
            return arr[0]
        else:
            return arr

Traceback (most recent call last):
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\qt4\base_window.py", line 198, in paintEvent
    self.handler.paintEvent(event)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\qt4\base_window.py", line 50, in paintEvent
    self._enable_window._paint(event)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\abstract_window.py", line 467, in _paint
    self.component.draw(gc, view_bounds=(0, 0, size[0], size[1]))
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\component.py", line 424, in draw
    self._draw(gc, view_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\component.py", line 776, in _draw
    self._dispatch_draw(layer, gc, view_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\container.py", line 272, in _dispatch_draw
    component._dispatch_draw(layer, gc, new_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\container.py", line 272, in _dispatch_draw
    component._dispatch_draw(layer, gc, new_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\component.py", line 796, in _dispatch_draw
    handler(gc, view_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\enable\component.py", line 874, in _draw_overlay
    overlay.overlay(self, gc, view_bounds, mode)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\chaco\tools\line_inspector.py", line 106, in overlay
    self.draw(gc, view_bounds)
  File "C:\Users\speschda\AppData\Local\Enthought\Canopy\User\lib\site-packages\chaco\tools\cursor_tool.py", line 183, in draw
    sx, sy = plot.map_screen(self.current_position)
ValueError: need more than 1 value to unpack

[aaronayres35]

I am still able to reproduce this using the cursor_tool_demo by replacing:

chaco/examples/demo/cursor_tool_demo.py

Lines 55 to 61 in f5cd0ab

	line = create_line_plot(
	[index, value],
	add_grid=True,
	add_axis=True,
	index_sort="ascending",
	orientation="h",
	)

with

line = LinePlot(
    index=ArrayDataSource(index, sort_order='ascending'),
    value=ArrayDataSource(value),
    index_mapper=LogMapper(range=DataRange1D()),
    value_mapper=LogMapper(range=DataRange1D()),
    orientation='h',
    color='red',
    bgcolor='transparent',
    line_width=1.0,
    line_style='solid',
    border_visible=False,
)