Re-assign correct indices to emptyColumns at table creation

[Rdornier]

Hello @ppouchin

In some cases, when multiple columns have to be removed from a TableWrapper object, because they are considered as empty columns, wrong columns are removed instead.
I found out that because of the re-indexing performed in this method, the next indices to remove are not correct anymore and it removes the wrong columns (only the first column to remove is correct)

To correct it, I simply subtract the number of columns already removed in the table to the next indices.

Rémy.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.06%. Comparing base (6779455) to head (f45b6b8).
Report is 2 commits behind head on 5_19_0.

Additional details and impacted files

@@             Coverage Diff              @@
##             5_19_0      #89      +/-   ##
============================================
- Coverage     96.11%   96.06%   -0.05%     
- Complexity     1307     1308       +1     
============================================
  Files            46       46              
  Lines          3523     3558      +35     
  Branches        280      283       +3     
============================================
+ Hits           3386     3418      +32     
  Misses           76       76              
- Partials         61       64       +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ppouchin]

Hi!

I'm finishing some code for image analysis and I'll look into it tomorrow.
I really need to refactor the whole table code at one point, as it is quite fat but still limited (cannot have multiple ROI columns for example).

[ppouchin]

You're absolutely right!

Although if the indices are not sorted (however, they should), the current formula may not work properly either.
Alternatively, do you think that processing the columns in reverse order would work as well?

I'm wondering if a test can be added to notice this error if this creeps up again...

[Rdornier]

Alternatively, do you think that processing the columns in reverse order would work as well?

Yes, it should work and it is a nice way to handle this.

[ppouchin]

Done.