[WIP] [FIX] New signals: default output

[jerneju]

Issue

Selected Data should be default instead of Features.

Data should be default instead of Features.

Description of changes

Includes

• Code changes
• Tests
• Documentation

[codecov-io]

Codecov Report

Merging #2390 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master   #2390      +/-   ##
=========================================
+ Coverage   73.88%   73.9%   +0.01%     
=========================================
  Files         322     322              
  Lines       55843   55843              
=========================================
+ Hits        41262   41270       +8     
+ Misses      14581   14573       -8

[janezd]

As far as I recall, we said that the problem here is that Data became a dynamic signal, and static signals have precedence over dynamic.

@ales-erjavec, I had an idea that we could change the signal connection policy so that when the signal is dynamic, it still has the same priority as static if the types actually match. That is, if Table is connected to Table (not to Corpus) it remains the default signal although it's dynamic (because it would match even if it was static). I'm not sure I still like it; it's messy. Do you have a better idea?

Another option is to eliminate non-dynamic signals. There are currently almost no static signals. @astaric?

At any rate, while this modification fixes the problem, I don't think it fixes it correctly. If we accept it, we should at least add a comment why the change was necessary.

[ales-erjavec]

@ales-erjavec, I had an idea that we could change the signal connection policy so that when the signal is dynamic, it still has the same priority as static if the types actually match. That is, if Table is connected to Table (not to Corpus) it remains the default signal although it's dynamic (because it would match even if it was static). I'm not sure I still like it; it's messy. Do you have a better idea?

I think this is the best solution. Please see #2431

[janezd]

Closed in favor of #2431.