TestWrite.test_missing_entries()
#533
Conversation
joeflack4
added
enhancement
joeflack4
force-pushed
the
missing-entries
branch
2 times, most recently
from
ef955af
to
97d9ab1
Compare
| # TODO: Determine assertions that should pass | ||
| # TODO: Fix failing assertions that should pass | ||
| # TODO: before merging, remove pass/fail comments at end of assertion lines | ||
| def test_missing_entries(self): |
There was a problem hiding this comment.
Get assertions to pass
- 1. Determine assertions that should pass @matentzn
- 2. Fix failing assertions that should pass
- 3. before merging, remove pass/fail comments at end of assertion lines
| # TODO: Determine assertions that should pass | ||
| # TODO: Fix failing assertions that should pass | ||
| # TODO: before merging, remove pass/fail comments at end of assertion lines | ||
| def test_missing_entries(self): |
There was a problem hiding this comment.
Which assertions are appropriate?
- @matentzn review
Can you look at the assertions I've made and determine which ones you think should be expected to pass, and which should not?
I've left temporary comments next to each assertion showing their current pass/fail status.
| "mapping_provider": "https://www.orpha.net/", | ||
| } | ||
|
|
||
| # When passing Converter |
There was a problem hiding this comment.
FYI: 2 different setups
I have a set of assertions for 2 cases of MappingSetDataFrame instantiation:
- with inclusion of a
Converter - with no inclusion of a
Converter
joeflack4
force-pushed
the
missing-entries
branch
3 times, most recently
from
001557e
to
43fa8f5
Compare
| with open(tmp_file, "w") as f: | ||
| write_table(msdf, f) | ||
| msdf2 = parse_sssom_table(tmp_file) | ||
| self.assertIn("curie_map", msdf2.metadata) # fail |
There was a problem hiding this comment.
This behaviour is expected for now, because curie_map sort if lives outside the metadata. however, it will soon be solved: mapping-commons/sssom#349
There was a problem hiding this comment.
Wow, I'm surprised it wasn't part of the data model.
From an implementation standpoint, I was thinking that maybe msdf.prefix_map and msdf.metadata['curie_map'] could point to the same Python dictionary object. Right now they are different objects and, as these assertions show, can get out of sync.
joeflack4
force-pushed
the
missing-entries
branch
from
43fa8f5
to
c82302d
Compare
addresses #534
Changes