Went in to add a default species.
ended up doing a long avoided refactor to common conventions.
Particularly onerous due to headers being
numerous
disconnected from the data files (predictable out of date)
buried in hidden JS elements on an invalid html page
New behaviours to expect:
less nonsensical data due to blindly trusting column will never be inserted/deleted/rearranged
logging will scream if it detects format changes
will never be as good as (all the other) formats that integrate headers that can be checked
could be improved to at least check if the web page reported headers changed ... maybe
oh yea ... genes have a default 'in_taxon' statement now
There is also an adaption to HPOA changing its header format
( but no one seem to be making a big deal about it)
I hope to push a few more cosmetic/doc changes before this merges.
note:
unit test are failing locally because pipenv can't import psycopg2 for the mgi ingest
and claims I have a bad api key for my pypi and wants me to click on a dodgy obfuscated url
this has apparently been going on since April
Went in to add a default species. ended up doing a long avoided refactor to common conventions.
Particularly onerous due to headers being
New behaviours to expect:
oh yea ... genes have a default 'in_taxon' statement now
There is also an adaption to HPOA changing its header format
( but no one seem to be making a big deal about it)
I hope to push a few more cosmetic/doc changes before this merges.
note: unit test are failing locally because pipenv can't import psycopg2 for the mgi ingest and claims I have a bad api key for my pypi and wants me to click on a dodgy obfuscated url this has apparently been going on since April