show_harmo_error shows success rate <100% when there are no errors

[twey2]

show_harmo_error shows success rate <100% when there are warnings but no errors, which is confusing. I think it would be better for the success rate to reflect only the errors, and I think the warnings and errors should be split up. But the way the warnings are highlighted in this version are useful!

[GuiFabre]

This feature (specificatlly separte errors from warnings) will be updated in the future. This has multiple implications backstage (the reason is because some warnings in R are conceptually errors, but cast as warnings)

[rwissa]

20/10/2023 - To change before CRAN. Harmo team vote

[GuiFabre]

Here is the proposition :

in addtion, the errors and warnings are now shown only if there are some

[a-trottier]

Works well.
but why the alignment of the middle columns changed?
rule_category and success are aligned left while the others are aligned right.
It makes the column next to success harder to read.

Should all aligned left or the middle columns could be centered

[GuiFabre]

done

[twey2]

Layout looks nice overall!
Why is there a 'case_whe' and 'case_when' in this example? (This shouldn't happen in a real scenario?)
Minor comment that we might eventually want 'Total number of algorithms' to match terminology in the glossary.

[GuiFabre]

Hi @twey2 and thank you for your contribution.

For the 'case_whe' rule category, this example is generated with the DPE - with errors in DEMO_files_harmo object. In this scenario, we tried to mimic some errors, including typo in the rule category. Let's say the user writes 'direct_maping' or, like here, 'case_whe' instead of 'case_when'

For the terminology, good point, I'll make the change right away.

[GuiFabre]

done

[twey2]

Great!