Facilitate research on comparing self-reported nap/nonwear with acc-based daytime rest detection

[vincentvanhees]

This PR expands part 5 to help us get a better understanding of the meaning of daytime SIBs compared with self-reported naps and self-reported nonwear by looking at their overlap.

• Re-uses the already existing argument possible_nap_duration to optionally filter out all SIBs with a duration shorter than this value (default = 0).
• New argument possible_nap_edge_acc to specify the maximum acceleration value to consider before and/or after the sustained inactivity bout (default = Inf).

If the user finds an optimal setting for these arguments then that could possibly inform a new nap classification algorithm.

Note: The nap detection itself is still limited to the model trained in 3-year olds and I am not changing anything to that. This functionality should be able to run without using the current nap detection option.

Fixes #687

Checklist before merging:

• Existing tests still work (check by running the test suite, e.g. from RStudio).
• Added tests (if you added functionality) or fixed existing test (if you fixed a bug).
• Updated or expanded the documentation.
• Updated release notes in inst/NEWS.Rd with a user-readable summary. Please, include references to relevant issues or PR discussions.
• Added your name to the contributors lists in the DESCRIPTION and CITATION.cff files.

[jhmigueles]

I was not getting the expected output when testing this, and it is because time here is introduced as ISO6801, but sibreport$start and sibreport$end are stored as POSIX objects, so the function was not able to find any srep_tmp.
I fixed this locally by introducing these lines at the begining of this function:

# transform time to POSIX
  if (is.ISO8601(time[1])) {
    time = iso8601chartime2POSIX(time, tz = tz)
  }

@vincentvanhees Should I commit this change or do you want to investigate why do you get the expected output and I don't?

[vincentvanhees]

Yes, please. I have been working so far with the csv variant of ms5.outraw. Maybe if you use the RData variant time is also stored different. It is good to facilitate both.

[vincentvanhees]

Looks like the problem is that is.ISO8601 assumes the input x is a character.
I will now fix is.ISO8601 such that it always first checks that input is a character.

[jhmigueles]

Looks like the problem is that is.ISO8601 assumes the input x is a character. I will now fix is.ISO8601 such that it always first checks that input is a character.

Great, I have already merged the changes from the master branch, I will not do any more commits.

[codecov-commenter]

Codecov Report

Merging #900 (f7b53e3) into master (c7bfa8c) will increase coverage by 0.21%.
The diff coverage is 86.91%.

❗ Current head f7b53e3 differs from pull request most recent head 3c36304. Consider uploading reports for the commit 3c36304 to get more accurate results

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

@@            Coverage Diff             @@
##           master     #900      +/-   ##
==========================================
+ Coverage   78.01%   78.23%   +0.21%     
==========================================
  Files          91       92       +1     
  Lines       12138    12302     +164     
==========================================
+ Hits         9469     9624     +155     
- Misses       2669     2678       +9     

Files Changed	Coverage Δ
R/g.part5.R	80.28% <50.00%> (-0.80%)	⬇️
R/g.sibreport.R	91.66% <89.47%> (+0.64%)	⬆️
R/g.part5.analyseRest.R	97.79% <97.79%> (ø)
R/check_params.R	97.61% <100.00%> (ø)
R/g.part5_analyseSegment.R	93.39% <100.00%> (+0.19%)	⬆️
R/load_params.R	99.13% <100.00%> (+<0.01%)	⬆️