Closes 2661 - ignore_seconds_flag now defaults to TRUE.

[jimrothstein]

… Need to review vignettes.

Thank you for your Pull Request! We have developed this task checklist from the Development Process Guide to help with the final steps of the process. Completing the below tasks helps to ensure our reviewers can maximize their time on your code as well as making sure the admiral codebase remains robust and consistent.

Please check off each taskbox as an acknowledgment that you completed the task or check off that it is not relevant to your Pull Request. This checklist is part of the Github Action workflows and the Pull Request will not be merged into the main branch until you have checked off each task.

• Place Closes #<insert_issue_number> into the beginning of your Pull Request Title (Use Edit button in top-right if you need to update)
• Code is formatted according to the tidyverse style guide. Run styler::style_file() to style R and Rmd files
• Updated relevant unit tests or have written new unit tests, which should consider realistic data scenarios and edge cases, e.g. empty datasets, errors, boundary cases etc. - See Unit Test Guide
• If you removed/replaced any function and/or function parameters, did you fully follow the deprecation guidance?
• Review the Cheat Sheet. Make any required updates to it by editing the file inst/cheatsheet/admiral_cheatsheet.pptx and re-upload a PDF and a PNG version of it to the same folder. (The PNG version can be created by taking a screenshot of the PDF version.)
• Update to all relevant roxygen headers and examples, including keywords and families. Refer to the categorization of functions to tag appropriate keyword/family.
• Run devtools::document() so all .Rd files in the man folder and the NAMESPACE file in the project root are updated appropriately
• Address any updates needed for vignettes and/or templates
• Update NEWS.md under the header # admiral (development version) if the changes pertain to a user-facing function (i.e. it has an @export tag) or documentation aimed at users (rather than developers). A Developer Notes section is available in NEWS.md for tracking developer-facing issues.
• Build admiral site pkgdown::build_site() and check that all affected examples are displayed correctly and that all new functions occur on the "Reference" page.
• Address or fix all lintr warnings and errors - lintr::lint_package()
• Run R CMD check locally and address all errors and warnings - devtools::check()
• Link the issue in the Development Section on the right hand side.
• Address all merge conflicts and resolve appropriately
• Pat yourself on the back for a job well done! Much love to your accomplishment!

[jimrothstein]

@bms63 @zdz2101

• changed default ignore_seconds_flag to TRUE in derive_vars_dtm()
• there are problems: first code example (approx line # 59) has non-zero seconds;
• so error thrown at line # 752 in compute_tmf()

 if (any(!is.na(partial[["second"]]))) {

This is draft, not PR. The discussion has details. Any suggestions what I should do next?

[bms63]

Hi @jimrothstein as far as I know @zdz2101 is not involved in admiral anymore so no need to ping him or assign him as a reviewer.

[bms63]

Hi @jimrothstein I started to investigate each issue that was causing errors in the checks. Most are related to what the argument is checking. You can see how I adjusted things as I went in the files changed. I would start with understanding what the argument is doing originally and how I have changed it and what that is impacting on the examples, vignettes and tests. I left you the test to update.

We might need to re-visit how we present the examples and vignettes later, but for now I think you should focus on what has been updated in this one argument and how this minor changes percolates throughout admiral. Once the checks are all pass we can revisit if the examples and vignettes are coherent.

Also note we might want to alert the users that we have changed this default behavior...but not sure on that yet

[jimrothstein]

@jeffreyad

Changes made:

• default value for ignore_seconds_flag in derive_vars_dtm is NOW TRUE
• several tests in testthat were removed because they contained seconds, which threw an error.

NEXT? If this much is ok, then will go through vignettes/examples etc.

[bundfussr]

• The default of ignore_seconds_flag in compute_tmf() should be updated as well to be consistent.
• The change should added to the "Breaking Changes" section in NEWS.md.

[bundfussr]

The first and the second record are the same. One of them should be removed.

Further updates of the examples are required as the first and third example are the same.

• The third example should be updated such that it shows the behaviour when ignore_seconds_flag = FALSE is used.
• An example should be added to show that an error is issued if seconds are present in the DTC variable and ignore_seconds_flag is not set to FALSE. This should also be mentioned in the description of the ignore_seconds_flag argument.

[jimrothstein]

@bundfussr
Working on the changes ... thanks

An example should be added to show that an error is issued if seconds are present in the DTC variable and >ignore_seconds_flag is not set to FALSE.

Meaning: "2019-07-18T15:25:40" will throw error when ignore_seconds_flag is DEFAULT (ie TRUE)?

• By hand, I will work out a chart of what happens, given --DTF and various parameters.

[bundfussr]

Meaning: "2019-07-18T15:25:40" will throw error when ignore_seconds_flag is DEFAULT (ie TRUE)?

Yes

[bundfussr]

This unit test should test that an error is issued. I.e., the input data needs to be updated such that the error is thrown.

[bundfussr]

Should be removed as it is the default now.

[bundfussr]

The note in the "Imputation Flags" section needs to be updated.

[bms63]

@jimrothstein we discussed in today's core meeting. We want to try something new in admiral where we are first going to alert the users to this upcoming change in 1.3 and in 1.4 implement the change.

so i'll make a new issue messaging alert and then this change can be done much later

[github-actions]

Package	Line Rate	Health
admiral	98%	✔
Summary	98% (5316 / 5421)	✔

[jimrothstein]

@bundfussr

• not done, but made several changes
• please take a quick look/feedback
• next: vignette imputation
• thx !

[bms63]

@jimrothstein we discussed in today's core meeting. We want to try something new in admiral where we are first going to alert the users to this upcoming change in 1.3 and in 1.4 implement the change.

so i'll make a new issue messaging alert and then this change can be done much later

Hi @jimrothstein just want to make sure you understand that we are not going to implement this change for 1.3 (June release). We are just going to alert users to the coming change which will be implemented in 1.4 (EOY release).

[jimrothstein]

@bms63
Please check slack - thanks.

[jeffreyad]

Hi @jimrothstein and @bundfussr for derive_vars_dtm() when time_imputation is set to a time value, the seconds are still required or there is a warning that it fails to parse. Do you think this should be updated to accept either values like "00:00" or "00:00:00"?

Warning message:
All formats failed to parse. No formats found. 

[jimrothstein]

@jeffreyad
I defer to @bundfussr (I will only be guessing)

[bundfussr]

Hi @jimrothstein and @bundfussr for derive_vars_dtm() when time_imputation is set to a time value, the seconds are still required or there is a warning that it fails to parse. Do you think this should be updated to accept either values like "00:00" or "00:00:00"?

Warning message:
All formats failed to parse. No formats found. 

I think we should check if the value specified for time_imputation is valid and issue an error if it is invalid. The current message is not helpful.

I'm not sure if we should accept "00:00". Even if ignore_seconds_flag = TRUE is specified, seconds are imputed. I.e., if we accept "00:00" or "23:59", we would need to decide which value for the seconds should be used. I would tend to keep it simple and require that the user provides this information.

@bms63 , @manciniedoardo , what do you think?