Incorporate Reverse Dependency Improvements

[Jeff-Thompson12]

Addresses #746 and incorporates #751

[codecov]

Codecov Report

Attention: Patch coverage is 85.09317% with 24 lines in your changes are missing coverage. Please review.

Project coverage is 81.27%. Comparing base (c7232ed) to head (2ba0df2).

Files	Patch %	Lines
R/mod_packageDependencies.R	60.41%	19 Missing ⚠️
R/mod_decision_automation.R	66.66%	3 Missing ⚠️
R/mod_reweightView.R	80.00%	1 Missing ⚠️
R/mod_user_roles.R	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #753      +/-   ##
==========================================
+ Coverage   79.23%   81.27%   +2.03%     
==========================================
  Files          33       33              
  Lines        5114     5169      +55     
==========================================
+ Hits         4052     4201     +149     
+ Misses       1062      968      -94     

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

[LDSamson]

@Jeff-Thompson12 I implemented suggested changes on my own fork since that's where I have write access; see #751.

[Jeff-Thompson12]

@AARON-CLARK & @Robert-Krajcik I still haven't done a deep dive into the code add by this PR. I would like your input on how the output is being produced.

For example, consider the two tidyverse packages of {dplyr} and {tibble}. {tibble} is in the imports of {dplyr}, and {dplyr} is in the suggests of {tibble}. Currently the application outputs both as reverse dependencies of each other. The table currently does not include a type for reverse dependency and we could include it but I it might be confusing. Then we would have a table where package: tibble and type: imports would tibble is in the imports of {package} but the second table it would mean {package} is in the imports of tibble. Thoughts?

[AARON-CLARK]

@Jeff-Thompson12 I implemented suggested changes on my own fork since that's where I have write access; see #751.

@LDSamson I granted you write permissions! Done ✅

[aclark02-arcus]

@Jeff-Thompson12, I would think adding a "type" column could make things slightly confusing, but maybe not. What if we moved the new table below the text output of "All reverse deps"? And we could make the "Reverse Deps" header larger. That may help folks separate the two tables in their minds.

Regardless if it's added or not, I did notice that dplyr would show up as a reverse dependency of tibble even when the Include Suggests toggle was off (or FALSE). I would think the Include Suggests toggle would apply to all output on this tab. If it's toggled on (TRUE), then dplyr should show up in the new table called "Reverse Deps available in DB". Otherwise, when FALSE, it should disappear. Thoughts?

[Jeff-Thompson12]

@aclark02-arcus It's ready for you to take another look.

@LDSamson Sorry it's taken so long to get this PR ready to review. If some desired functionality was lost, feel free to request changes.

[AARON-CLARK]

Hi @jthompson-arcus, I'm not sure if this is happening for you, but I'm getting a blank Package Dependency tab when tidyCDISC is selected on the control panel. Can you please verify?

[LDSamson]

Thanks for picking up this PR and improving it, I am happy to see that you find it useful. I will have a look this weekend.

[aclark02-arcus]

Reviewing after recent commits, everything looks good! Thanks @LDSamson & @jthompson-arcus! Merging this one. If there is anything else missing, it could be incorporated as a new PR.

stale