Re: auto07p IndexError when uses matplotlib 2.0 instead of matplotlib 1.5

[st970703]

Bug report

I am recently working on making auto07p work with matplotlib 2.0.

The source code can be found here:
https://sourceforge.net/projects/auto-07p/files/auto07p/0.9/

When using matplotlib 2.0 (or any version greater than 1.5), IndexError: invalid index to scalar variable. appears. I ran into this error using a given equation and auto file. But if you use matplotlib 2.0 and auto07p to plot most graphs in general, you are very likely to get this error if you zoom into the graph.
I would appreciate if anyone can assist me with the error, without changing the matlibplot code but only the auto code. I have read through the code but did not find anything suspicious. Since only two lines in the error stack trace involve code from auto07p. It can be more likely to be caused by matplotlib internal code.

Bug summary

IndexError: invalid index to scalar variable. in ticker.py.

Code for reproduction
Please download the files from Onedrive:
https://1drv.ms/f/s!ArJqyyCr1FQOhkITst8JdVK_mIFV

Put them in auto folder.
Run auto.
and type:

auto("couple2.auto")

It appears that there are only two lines involved. They are from the file /auto/07p/python/graphics/grapher_mpl.py.

The two methods involved are:

 def redraw(self):
        # recalculate label positions
        self.grapher.plotlabels()

        FigureCanvasTkAgg.draw(self)

And

def draw(self):
    ax = self.grapher.ax
    d = {}
    if ax is self.grapher.ax3d:
        [d["minx"], d["maxx"]] = ax.get_xlim3d()
        [d["miny"], d["maxy"]] = ax.get_ylim3d()
        [d["minz"], d["maxz"]] = ax.get_zlim3d()
        d["azimuth"] = ax.azim
        d["elevation"] = ax.elev
        d["cur_lims"] = Axes.get_xlim(ax), Axes.get_ylim(ax)
    else:
        [d["minx"], d["maxx"]] = ax.get_xlim()
        [d["miny"], d["maxy"]] = ax.get_ylim()
    for k in list(d):
        # don't adjust any unchanged settings
        if k == "cur_lims":
            if map(list, d[k]) == map(list, self.grapher._cur_lims):
                del d[k]
        elif d[k] == self.grapher.cget(k):
            del d[k]
    if d != {}:
        if "cur_lims" in d:
            del d["cur_lims"]
        if d != {}:
            self.grapher._configNoDraw(**d)

        self.redraw()
        return
    FigureCanvasTkAgg.draw(self)

# Paste your code here
#
#

Actual outcome

When using matplotlib 2.0 (or any version greater than 1.5), IndexError: invalid index to scalar variable. appears:

/home/ngb/auto/07p/python/Points.py:1086: VisibleDeprecationWarning: `rank` is deprecated; use the `ndim` attribute or function instead. To find the rank of a matrix see `numpy.linalg.matrix_rank`.
  r = rank(array_temp)
Created plot
<graphics.windowPlotter.WindowPlotter2D object at 0x7f9e42007710>
AUTO> Exception in Tkinter callback
Traceback (most recent call last):
  File "/usr/lib/python3.5/tkinter/__init__.py", line 1562, in __call__
    return self.func(*args)
  File "/usr/lib/python3.5/tkinter/__init__.py", line 608, in callit
    func(*args)
  File "/usr/lib/python3/dist-packages/matplotlib/backends/backend_tkagg.py", line 370, in idle_draw
    self.draw()
  File "/home/ngb/auto/07p/python/graphics/grapher_mpl.py", line 92, in draw
    self.redraw()
  File "/home/ngb/auto/07p/python/graphics/grapher_mpl.py", line 65, in redraw
    FigureCanvasTkAgg.draw(self)
  File "/usr/lib/python3/dist-packages/matplotlib/backends/backend_tkagg.py", line 351, in draw
    FigureCanvasAgg.draw(self)
  File "/usr/lib/python3/dist-packages/matplotlib/backends/backend_agg.py", line 464, in draw
    self.figure.draw(self.renderer)
  File "/usr/lib/python3/dist-packages/matplotlib/artist.py", line 63, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/python3/dist-packages/matplotlib/figure.py", line 1143, in draw
    renderer, self, dsu, self.suppressComposite)
  File "/usr/lib/python3/dist-packages/matplotlib/image.py", line 139, in _draw_list_compositing_images
    a.draw(renderer)
  File "/usr/lib/python3/dist-packages/matplotlib/artist.py", line 63, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/python3/dist-packages/matplotlib/axes/_base.py", line 2409, in draw
    mimage._draw_list_compositing_images(renderer, self, dsu)
  File "/usr/lib/python3/dist-packages/matplotlib/image.py", line 139, in _draw_list_compositing_images
    a.draw(renderer)
  File "/usr/lib/python3/dist-packages/matplotlib/artist.py", line 63, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/python3/dist-packages/matplotlib/axis.py", line 1136, in draw
    ticks_to_draw = self._update_ticks(renderer)
  File "/usr/lib/python3/dist-packages/matplotlib/axis.py", line 969, in _update_ticks
    tick_tups = [t for t in self.iter_ticks()]
  File "/usr/lib/python3/dist-packages/matplotlib/axis.py", line 969, in <listcomp>
    tick_tups = [t for t in self.iter_ticks()]
  File "/usr/lib/python3/dist-packages/matplotlib/axis.py", line 912, in iter_ticks
    majorLocs = self.major.locator()
  File "/usr/lib/python3/dist-packages/matplotlib/ticker.py", line 1794, in __call__
    return self.tick_values(vmin, vmax)
  File "/usr/lib/python3/dist-packages/matplotlib/ticker.py", line 1802, in tick_values
    locs = self._raw_ticks(vmin, vmax)
  File "/usr/lib/python3/dist-packages/matplotlib/ticker.py", line 1761, in _raw_ticks
    istep = np.nonzero(steps >= raw_step)[0][0]
IndexError: invalid index to scalar variable.

# If applicable, paste the console output here
#
#

Expected outcome

This program works with matplotlib 1.5 and earlier.
Auto07p could zoom in without giving errors.

Matplotlib version

• Operating System: Linux Ubuntu 16.04
• Matplotlib Version: 2.0.2
• Python Version: 3.5
• Jupyter Version (if applicable):
• Other Libraries:

[tacaswell]

Can you drop into the debugger and see what the values of steps , vmin , and vmax are?

[tacaswell]

And what version of numpy are you using?

[tacaswell]

Can you reproduce this without using auto07p ?

[st970703]

Hey tacaswell,

I don't use matplotlib apart from auto07p.
I use numpy version 1.13.1.
It appeared that the
istep = np.nonzero(steps >= raw_step)[0][0]
was trying to access a 1D array.

I replaced the line in ticker.py:
istep = np.nonzero(steps >= raw_step)[0][0]

With the following:
arr_dim = np.nonzero(steps >= raw_step).ndim
if arr_dim == 1:
istep = np.nonzero(steps >= raw_step)[0]
elif arr_dim == 2:
istep = np.nonzero(steps >= raw_step)[0][0]

It worked after the above changes.

[efiring]

From the docstring of np.nonzero in numpy 1.13.1:

Returns a tuple of arrays, one for each dimension of `a`,
containing the indices of the non-zero elements in that
dimension.

I checked, and it is true: it does return a tuple of arrays, even if its argument is a scalar, so I don't see how your replacement can work. The output of np.nonzero is not an ndarray, so it has no ndim attribute.

[tacaswell]

@st970703 Have you reported this to the auto07p developers? If you can not reproduce without auto07p it should go to them first.

[anntzer]

I cannot reproduce the issue with
auto07p 0.9.1 (built from source)
python 2.7 (auto07p does not appear to work with py3)
numpy 1.13.1
matplotlib 2.0.2 (all python deps from fresh conda env).

[efiring]

I'm going to close this, since we can't reproduce or understand it. Feel free to reopen it if new information becomes available.