test(rect): reproduce unemployment rect plot from bokeh docs #85
Conversation
Tests/test_PandasBokeh.py
Outdated
| p_rect_accessor = df_unemployment.plot_bokeh.rect(**kwargs) | ||
|
|
||
| p_rect_pandas_backend = df_unemployment.plot(kind="rect", **kwargs) | ||
| p_rect_accessor_pandas_backend = df_unemployment.plot.map(**kwargs) |
There was a problem hiding this comment.
shouod be df_unemployment.plot_bokeh.rect
|
I would propose that you have a look at the code that uses the category option (e.g. scatterplot) and try to extract the corresponding code into a new function if possible. The code base is really messy, but I didn't have any time to clean it so far 👎 If the scatterplot code doesn't help you, you can also just implement the feature set on your own. Best, Patrik PS: Love it that you start your work by writting a test. Wish I would have been so smart when starting this project |
|
I tried to find the integration points for the rect plot, I think I got at least somewhere. It is still ad-hoc and sketchy, but for now I am only interested in getting the test to run - and then see it "fail" in the plot output HTML. With the latest 3d29552 I get this: where I am not sure, but suspect this has to do with Pandas upstream? |
Support for rect glyphs as proposed in #84.
The idea is to reproduce this
rectplot from the bokeh gallery.Right now it hits https://github.com/PatrikHlobil/Pandas-Bokeh/blob/master/pandas_bokeh/plot.py#L421 because the
Monthcolumn is not numeric (https://github.com/PatrikHlobil/Pandas-Bokeh/blob/master/pandas_bokeh/plot.py#L418).@PatrikHlobil what would be your general advice on how to proceed? Also, I am not yet sure how to best reuse existing code (as e.g. the
categoryargument is needed to reproduce the example plot).