Filtering "missing" tick labels

[ivan-boikov]

Description

Closes #4715
With 10 items in a CategoricalArray spacing between ticks becomes smaller than the padding and a new tick is squeezed in. Its value is not defined however and the CategoricalArrays.jl recipe gives out a missing tick label, on which a label formatter chokes.

Feels like a band-aid rather than a solution though.

Test code

using Plots, CategoricalArrays
plot(
    bar(categorical(collect(1:9)), collect(1:9)),
    bar(categorical(collect(1:10)), collect(1:10);); 
    size=(4,1).*100
)

Attribution

• I am listed in .zenodo.json (see How to get acknowledged #3503)

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.40 🎉

Comparison is base (0e0f117) 90.00% compared to head (2ec1904) 90.40%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4716      +/-   ##
==========================================
+ Coverage   90.00%   90.40%   +0.40%     
==========================================
  Files          40       40              
  Lines        8360     8727     +367     
==========================================
+ Hits         7524     7890     +366     
- Misses        836      837       +1     

Impacted Files	Coverage Δ
src/axes.jl	89.19% <100.00%> (+0.11%)	⬆️

... and 32 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[BeastyBlacksmith]

I think that the CategoricalArrays recipe should give an empty string insterad of missing in this case.

[ivan-boikov]

I think that the CategoricalArrays recipe should give an empty string insterad of missing in this case.

In JuliaData/CategoricalArrays.jl#340, they call the use of missing a hack. Perhaps they chose it and not a blank string to avoid collision -- CategoricalArrays might have empty strings as legitimate categories?