STY: Boolean values for bint variables

[ShaharNaveh]

• closes #xxxx
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

---

asv benchmarks: to show that performance is not changed with the stylistic changes)

All benchmarks:

       before           after         ratio
     [28e0f18a]       [6f1fd78d]
     <master>         <STY-bint-boolean-not-int-join>
         21.4±1ms       19.8±0.4ms     0.92  join_merge.Join.time_join_dataframe_index_multi(False)
       24.9±0.3ms       23.6±0.3ms     0.95  join_merge.Join.time_join_dataframe_index_multi(True)
       14.5±0.3ms       14.0±0.3ms     0.97  join_merge.Join.time_join_dataframe_index_shuffle_key_bigger_sort(False)
       17.6±0.6ms       16.7±0.5ms     0.95  join_merge.Join.time_join_dataframe_index_shuffle_key_bigger_sort(True)
       14.4±0.3ms       13.8±0.2ms     0.95  join_merge.Join.time_join_dataframe_index_single_key_bigger(False)
       17.8±0.7ms       16.4±0.5ms     0.92  join_merge.Join.time_join_dataframe_index_single_key_bigger(True)
       13.5±0.3ms       13.1±0.3ms     0.97  join_merge.Join.time_join_dataframe_index_single_key_small(False)
       14.5±0.1ms       13.9±0.5ms     0.96  join_merge.Join.time_join_dataframe_index_single_key_small(True)
       2.86±0.08s       2.94±0.05s     1.03  join_merge.JoinIndex.time_left_outer_join_index
         433±70μs         428±50μs     0.99  join_merge.JoinNonUnique.time_join_non_unique_equal

[ShaharNaveh]

This is the only place where I'm adding bint

[jorisvandenbossche]

I am not sure asv is always appropriate for changes that might only give a small change in performance. Can you maybe check the generated code to ensure that no more work is being done (unless this is documented in the cython docs to work fine?)

[ShaharNaveh]

I am not sure asv is always appropriate for changes that might only give a small change in performance. Can you maybe check the generated code to ensure that no more work is being done (unless this is documented in the cython docs to work fine?)

@jorisvandenbossche I have checked the sources of the generated C-files. there is no code change at all (except for the comments displaying the equivalent code at the .pyx file), this is purely stylistic change.

[jorisvandenbossche]

Super, thanks for checking!

[jbrockmendel]

Hah, I just made the same suggestion in the next PR

[WillAyd]

Thanks @MomIsBestFriend