New reporting approach

[milechin]

So I discovered a nextflow function called collectFile, which opens up new opportunities for reporting. Essentially if each process creates a file, this function will collect the individual files and merge them into one file. More info here:
https://nextflow-io.github.io/patterns/collect-into-file/

So in pkgtest.nf, I updated runTests process so that it generates a csv file called test_metrics.csv with the following columns:

• results - PASSED/FAILED value - A test will receive FAILED result if one of the following is true:

  1. If in the results.txt file contains words other than "Passed"
  2. If the exit code return by the test.qsub is not 0.
  3. If the log.txt file contains the word(s) "error". (Not sure if this will work for all tests, as some tests may have the word "error" in their summary but not indicate anything wrong.)

• module - The name of the module

• tests_passed - The number of "Passed" words found in the results.txt

• tests_failed - The number of "Error" words found in the results.txt

• log_error_count - The number of "Error" words found in the log.txt

• exit_code - The exit code for the test.qsub execution

• installer - The username of the installer who installed the module.

• category - The module category

• install_date - The install date of the module.

• workdir - The nextflow working directory that contains the logs for the module test.

In the Workflow function, I included the collectFile function that combines the test_metrics.csv results from each process and generates a csv report in the current directory. The name of the report is the input csv file name with "report_" as a prefix.

In this commit I am including an example output report called report_example.csv for review.

[aramp10]

I like the report_example.csv table, it's simple and easy to sort by the columns. This would be a good way to see the results of the many tests that we plan to run.

[bu-bgregor]

reviewed during meeting, approved :)