New function merge_camtrapdp()

[sannegovaert]

fix #75

Remarks

• Assume that locationID's and individualID's must not be unique as the same location or individual can be used in different data packages.
• Assume that duplicatesd ID's are between packages, not within.
• If duplicates are present, all values of that identifier get a prefix.

Helper functions
I created 6 helper functions. Maybe they can be simplified or reduced in number!
They now live at utils.R.

function	functionality in merge_camtrapdp()	suggestion
check_duplicate_ids()	add prefix to identifiers with duplicates	helper function @ utils.R
add_prefix()	add prefix to identifiers with duplicates	helper function @ utils.R
normalize_list()	remove duplicated elements of lists	helper function @ utils.R or separate file.R
is_subset()	remove duplicated elements of lists	helper function @ utils.R or separate file.R
update_unique()	remove duplicated elements of lists	helper function @ utils.R or separate file.R
remove_duplicates()	remove duplicated elements of lists	helper function @ utils.R or separate file.R

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.91%. Comparing base (4e52a04) to head (722b71f).
Report is 149 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #112      +/-   ##
==========================================
+ Coverage   99.89%   99.91%   +0.01%     
==========================================
  Files          23       25       +2     
  Lines         983     1151     +168     
==========================================
+ Hits          982     1150     +168     
  Misses          1        1              

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

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests

[PietrH]

@peterdesmet : 0f9a28a, why remove the subdir in tempdir()? This guarantees that there are no filename collisions in tempdir() which is shared over the whole R session.

• Avoids accidental collisions (usage of same filename) between tests
• Avoids collisions with developer writing to tempdir()

Much less an issue for CI, but still, I'm interested to see what negatives you see

[peterdesmet]

@PietrH regarding removing subdir

• I assumed that on.exit(unlink(temp_dir, recursive = TRUE)) clears everything from the tempdir after a test anyway? So collisions would only happen if a developer did not copy that line
• The same subdir was used between tests, so collisions could happen anyway?

My reasoning was mainly: only create subdirs if these is necessary within a test.

[PietrH]

@PietrH regarding removing subdir

* I assumed that `on.exit(unlink(temp_dir, recursive = TRUE))` clears everything from the tempdir after a test anyway? So collisions would only happen if a developer did not copy that line

* The same subdir was used between tests, so collisions could happen anyway?

My reasoning was mainly: only create subdirs if these is necessary within a test.

Fair enough, I prefer leaving the clearing of the tempdir() up to the OS, and using withr if I really need to reset state. Since you could still leave tempdir() intact when a function fails before unlink() is called. withr really shines for things like environmental variables and (re)setting random seeds. I don't think it's use is warranted here. My curiosity is satisfied!

[PietrH]

I only checked my previous remarks. Well done Sanne! This wasn't an easy development 👍