Making prefix mapping less strict, fixes #702 #761
Conversation
There was a problem hiding this comment.
Nice suite! Cant see anything particularly wrong, left two comments.
Probably too late now but this is a robot jar we could run after this is merged to check all the changes in obographs-json and obo format against the suite:
https://incenp.org/files/softs/robot/1.10/robot-1.10.0-SNAPSHOT.jar
| # Class: <http://purl.obolibrary.org/obo/X_1> (<http://purl.obolibrary.org/obo/X_1>) | ||
|
|
||
| AnnotationAssertion(<http://www.geneontology.org/formats/oboInOwl#id> <http://purl.obolibrary.org/obo/X_1> "X:1") | ||
| SubClassOf(Annotation(<http://www.geneontology.org/formats/oboInOwl#gci_predicate> "R:2") <http://purl.obolibrary.org/obo/X_1> <http://purl.obolibrary.org/obo/X_2>) |
There was a problem hiding this comment.
Is this expected? Should it be R:2 some X:2 subClassOf X:1?
There was a problem hiding this comment.
I think this is because the input file is using the wrong qualifier (see my other comment below).
There was a problem hiding this comment.
Good catch.
The main issue with this suite (as I mentioned here #751) is that it's self-bootstrapping there is a one-off TestByGuru run where both inputs and outputs need carefully checked against the spec.
| @@ -0,0 +1,4 @@ | |||
| name: prefixes-conflict-main-idspace | |||
| name: prefixes-conflict-main-idspace | |||
There was a problem hiding this comment.
Why do the metadata yaml files have the same line twice? All of them.
There was a problem hiding this comment.
I believe this is a bug in the split_compiled_obo function (src/tests/test_converters/test_obo_format.py), lines 113–119. The name: ... line is added to the yamls[ontology_id] dictionary twice:
if k == "name":
ontology_id = v
if ontology_id in blocks:
raise ValueError(f"Duplicate ontology id: {ontology_id}")
metadata[ontology_id] = {}
yamls[ontology_id] = [yaml.rstrip()] # First addition of `name: ...`
blocks[ontology_id] = [
"format-version: 1.4\n",
f"ontology: {ontology_id}\n",
]
metadata[ontology_id][k.strip()] = v.strip()
yamls[ontology_id].append(yaml.strip()) # Second addition of `name: ...`
tests/input/obo-compliance.obo
Outdated
|
|
||
| [Term] | ||
| id: X:1 | ||
| is_a: X:2 {gci_predicate="R:2", gci_filler="X:3"} |
There was a problem hiding this comment.
Shouldn’t that be gci_relation instead of gci_predicate? I don’t think I have ever seen that gci_predicate qualifier, and the OBO spec does not mention it at all.
|
I totally lost track of this, and this is now breaking some OAK pipelines we have (especially the multilingual stuff). Oops. |
Adds test that currently codifies ambiguous behavior.
See also #760 for broader issue.
This PR also extends the obo test suite to include this,
and adds some derived files previously missing.