Custom-cohort creation script patch

[MattWellie]

Interface changes

Needed

• dry-run doesn't work. The boolean is captured appropriately, but when sent to create_cohort_from_criteria it's ignored. This PR has not resolved that
• project, name, description made mandatory (?)

QOL

• projects changed from type=str, nargs='*' to a nargs='*', default=[] - the type here was misleading as we want a list?
• multiple args changed from required=False, type=list[str] to nargs='*', default=[]
  • these are all lists of strings with none-to-many elements (though typically 1)
  • if the flag is not used, the value becomes an empty list
  • argparse's type doesn't work with nested types (afaik, the passed type is used as a callable), so list[str] is just list
  • argparse's type also casts values to the given type, so --sg_platform illumina becomes ['i', 'l', 'l', 'u', 'm', 'i', 'n', 'a'], instead of ['illumina'] (the result of list(input) instead of [input]

Strict Checking

• where there's a strict enumeration of valid values in metamist I've applied these using choices - could bin this off, or query metamist for the available choices at runtime.

Challenge:

python scripts/create_custom_cohort.py --project matt --name MATT --description "Matt desc" --projects matt not_matt --sg_technology short-read --sg_platform illumina --sg_type genome --sg_ids_internal CPGA --sample_type blood

Current Result: (I couldn't find any way to use multiple CPG IDs, so just using a list of one in this toy example)

{'project': 'matt', 'name': 'MATT', 'description': 'Matt desc', 'template_id': None, 'projects': ['matt', 'not_matt'], 'sg_ids_internal': ['C', 'P', 'G', 'A'], 'excluded_sgs_internal': None, 'sg_technology': ['s', 'h', 'o', 'r', 't', '-', 'r', 'e', 'a', 'd'], 'sg_platform': ['i', 'l', 'l', 'u', 'm', 'i', 'n', 'a'], 'sg_type': ['g', 'e', 'n', 'o', 'm', 'e'], 'sample_type': ['b', 'l', 'o', 'o', 'd'], 'dry_run': False}

New Result:

{'project': 'matt', 'name': 'MATT', 'description': 'Matt desc', 'template_id': None, 'projects': ['matt', 'not_matt'], 'sg_ids_internal': ['CPGA'], 'excluded_sgs_internal': [], 'sg_technology': ['short-read'], 'sg_platform': ['illumina'], 'sg_type': ['genome'], 'sample_type': ['blood'], 'dry_run': False}

[vivbak]

Huge thanks for this @MattWellie!
I'm just trying to think of a simple way to make codecov happy, presumably by testing that the parser is working correctly (?)

Not a super clear part of the code to test, to me.

[MattWellie]

I'm just trying to think of a simple way to make codecov happy, presumably by testing that the parser is working correctly (?)

Yeah, there's a couple of ways to solve that. I'll have a tinker this morning!

[codecov]

Codecov Report

Attention: Patch coverage is 98.38710% with 1 line in your changes missing coverage. Please review.

Project coverage is 80.83%. Comparing base (e3a7f39) to head (c6272c4).
Report is 5 commits behind head on dev.

Files	Patch %	Lines
scripts/create_custom_cohort.py	90.90%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #875      +/-   ##
==========================================
+ Coverage   80.67%   80.83%   +0.15%     
==========================================
  Files         172      172              
  Lines       14551    14603      +52     
==========================================
+ Hits        11739    11804      +65     
+ Misses       2812     2799      -13     

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