Ruth Comer

Results 454 comments of Ruth Comer

We should remove the GFI label while we decide if we still want this.

[What's new entry](https://output.circle-artifacts.com/output/job/aa92f02b-8aa8-4f6c-918f-38cd03395e1a/artifacts/0/doc/build/html/users/next_whats_new/pie_abs_labels.html) [Addition to Pie Charts example](https://output.circle-artifacts.com/output/job/aa92f02b-8aa8-4f6c-918f-38cd03395e1a/artifacts/0/doc/build/html/gallery/pie_and_polar_charts/pie_features.html#auto-label-slices) [Simplified Matplotlib Bakery](https://output.circle-artifacts.com/output/job/aa92f02b-8aa8-4f6c-918f-38cd03395e1a/artifacts/0/doc/build/html/gallery/pie_and_polar_charts/pie_and_donut_labels.html#sphx-glr-gallery-pie-and-polar-charts-pie-and-donut-labels-py)

I am not sure about `inner_labels` because there is nothing to stop you swapping the positions of these with `labels` (using the distance parameters), at which point they become outer...

Also note that `labels` is special because it will be automatically used by `legend` (and if you want them in the legend and not on the pie you set `labeldistance=None`)....

I am also wondering if it’s worth factoring out a public `pie_label` method, to rhyme with `bar_label`. The existing `labels` and `autopct` could remain as conveniences that internally call this...

OK let me see if I can summarise the current thinking: * Introduce `inner_labels` and `outer_labels` which take a list of strings or a format string to be formatted with...

I would prioritise formatting fractions over absolute values: - Since plotting fractions is basically the point of `pie` I don’t think it’s too unintuitive that those would have the convenience...

Named placeholders would give a lot more flexibility... ```python In [2]: '{abs:d}'.format(abs=42, frac=0.8) Out[2]: '42' In [3]: '{frac:.1%}'.format(abs=42, frac=0.8) Out[3]: '80.0%' In [4]: '{abs:d} ({frac:.0%})'.format(abs=42, frac=0.8) Out[4]: '42 (80%)' ```

I am in favour of making `labels` only for legend. - API consistency: almost everywhere that you pass `label(s)` in Matplotlib, you mean you want them in the legend. -...

What would be the default for `wedge_label_distance`?