add get_bad_bpms

[awegsche]

New Feature

This script collects a list of all BPMs declared as bad by either SVD cleaning or isolation forest and compiles a list with number of occurances for each BPM that has been declared at least 50% (25%) of the time.

Purpose

Every now and then BI asks for a list of BPMs that cause trouble for our measurements so that they can take a closer look during the next (E?YET|L)S . This script is intended to facilitate the process of retrieving such a list.

Notes

This script is just a template, if specific measurements / globbing sounds better than just taking unchecked all the measurements in one (or several) output folder, feel free to adapt if necessary

I am not going to write tests for this. It's a script that's run once every 2-3 years.

[fsoubelet]

Details here and there. I'm open to contributing about them if you would like.

I'm wondering if this should be included in omc3 itself. Maybe this is a good fit for pylhc? Would like @JoschD's opinion.

Also, in hindsight to my review in #pylhc/py, I like the little purpose paragraph you have in this PR. Maybe it could be copy-pasted or adapted there?

[JoschD]

Please don't make me write tests :-/

[JoschD]

Docstring for the args missing

[fsoubelet]

Cool, mostly would like to see the module docstring include script parameters and potentially an example.

Worth noting: with the amount of tfs.concat we're doing here, without providing any setup for the headers handling, the results will always be headers-less. Not sure if it's something we care, or it we want to write some basic info in the final output's headers.

[codeclimate]

Code Climate has analyzed commit 09b5ff6 and detected 0 issues on this pull request.

The test coverage on the diff in this pull request is 91.3% (50% is the threshold).

This pull request will bring the total coverage in the repository to 85.9% (0.0% change).

View more on Code Climate.