New issue 612: Reloading an athena dataset leads to corruption. https://bitbucket.org/yt_analysis/yt/issue/612/reloading-an-athena-dataset-l... Nathan Goldbaum: I'm using a 2D orszag tang dataset to test this. The failing script is here: ``` #!python from yt.mods import * fns = sorted(glob.glob('OrszagTang.????.vtk')) fn = fns[0] #fn = '/home/goldbaum/tar_files/IsolatedGalaxy/galaxy0030/galaxy0030' pf = load(fn) slc = SlicePlot(pf, 2, 'Density') slc.save() pf = load(fn) slc = SlicePlot(pf, 2, 'Density') slc.save() ``` This produces the following traceback: ``` #!python Traceback (most recent call last): File "test.py", line 15, in <module> slc.save() File "/home/goldbaum/yt-x86_64/src/yt-hg/yt/visualization/plot_window.py", line 1091, in save names.append(v.save(n,mpl_kwargs)) File "/home/goldbaum/yt-x86_64/src/yt-hg/yt/visualization/base_plot_types.py", line 70, in save canvas.print_figure(name, **mpl_kwargs) File "/home/goldbaum/yt-x86_64/lib/python2.7/site-packages/matplotlib/backend_bases.py", line 2096, in print_figure **kwargs) File "/home/goldbaum/yt-x86_64/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", line 492, in print_png FigureCanvasAgg.draw(self) File "/home/goldbaum/yt-x86_64/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", line 435, in draw self.renderer = self.get_renderer() File "/home/goldbaum/yt-x86_64/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", line 454, in get_renderer self.renderer = RendererAgg(w, h, self.figure.dpi) File "/home/goldbaum/yt-x86_64/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", line 85, in __init__ self._renderer = _RendererAgg(int(width), int(height), dpi, debug=False) ValueError: cannot convert float NaN to integer ``` When I switch out the athena dataset for the enzo IsolatedGalaxy dataset, I get no crash, so I think this has to be an issue with the athena frontend. I will try to dive in and fix this tonight, but if @samskillman, @jzuhone, or @MatthewTurk have any idea where the error might be, I'd love to hear your thoughts.