[ENH] Scatterplot: Implement grouping of selections

[janezd]

Instead of annotating selected data with "No" and "Yes", use G0, G1, G2 ... when data is selected is subsequent groups using Shift or Cmd/Ctrl modifier.

Connect File -> Scatter plot, and then Scatter plot to a Table and also to another Scatter plot, both through signal Data (not Selection). Now select several groups in the first scatter plot by using the Shift modifier and see the results in the other scatter plot and table.

The change does not affect the other Scatter plot's output.

We can add a check box "Group selections" to enable or disable this behaviour. However, I think this should be the default behaviour and I see little reason why the user would want to disable it.

Includes

• Code changes
• Tests
• Documentation

[janezd]

@astaric, I defined a GUI for this. :)

[codecov-io]

Codecov Report

Merging #2070 into master will increase coverage by 0.05%.
The diff coverage is 95.34%.

@@            Coverage Diff             @@
##           master    #2070      +/-   ##
==========================================
+ Coverage   70.15%   70.21%   +0.05%     
==========================================
  Files         315      315              
  Lines       54096    54209     +113     
==========================================
+ Hits        37952    38061     +109     
- Misses      16144    16148       +4

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update aa30c1f...5052b08. Read the comment docs.

[markotoplak]

table.ids = data.ids

[janezd]

radon fails because of an additional case in the switch-ish select method that acts upon different possible keyboard modifiers. There's neither a reasonable way to fix this, nor is there a reason to do so as the method's flow is rather clear.

I'd rather not improve a cc score of a totally unrelated method just to make radon happy.