Adding tests for import behavior. Adding import_depth to OntologyReso…

…urce and CLI, to determine how deep in import hierarchy to traverse (#249) * Adding tests for import behavior. Adding import_depth to OntologyResource and CLI, to determine how deep in import hierarchy to traverse * black * Skipping pronto import test; awaiting althonos/pronto#186
INCATools · Aug 30, 2022 · 7e9e8b4 · 7e9e8b4
1 parent 1ac1ac4
commit 7e9e8b4
Show file tree

Hide file tree

Showing 12 changed files with 254 additions and 23 deletions.
diff --git a/src/oaklib/cli.py b/src/oaklib/cli.py
@@ -560,11 +560,23 @@ def chain_it(v):
     "--autosave/--no-autosave",
     help="For commands that mutate the ontology, this determines if these are automatically saved in place",
 )
+@click.option(
+    "--import-depth",
+    type=click.INT,
+    help="Maximum depth in the import tree to traverse",
+)
 @input_option
 @input_type_option
 @add_option
 def main(
-    verbose: int, quiet: bool, input: str, input_type: str, add: List, save_as: str, autosave: bool
+    verbose: int,
+    quiet: bool,
+    input: str,
+    input_type: str,
+    add: List,
+    save_as: str,
+    autosave: bool,
+    import_depth: Optional[int],
 ):
     """Run the oaklib Command Line.
 
@@ -593,7 +605,7 @@ def main(
 
     if input:
         impl_class: Type[OntologyInterface]
-        resource = get_resource_from_shorthand(input, format=input_type)
+        resource = get_resource_from_shorthand(input, format=input_type, import_depth=import_depth)
         impl_class = resource.implementation_class
         logging.info(f"RESOURCE={resource}")
         settings.impl = impl_class(resource)

diff --git a/src/oaklib/implementations/pronto/pronto_implementation.py b/src/oaklib/implementations/pronto/pronto_implementation.py
@@ -108,12 +108,15 @@ def __post_init__(self):
         if self.wrapped_ontology is None:
             resource = self.resource
             logging.info(f"Pronto using resource: {resource}")
+            kwargs = {}
+            if resource and resource.import_depth is not None:
+                kwargs["import_depth"] = resource.import_depth
             if resource is None:
                 ontology = Ontology()
             elif resource.local:
-                ontology = Ontology(str(resource.local_path))
+                ontology = Ontology(str(resource.local_path), **kwargs)
             else:
-                ontology = Ontology.from_obo_library(resource.slug)
+                ontology = Ontology.from_obo_library(resource.slug, **kwargs)
             self.wrapped_ontology = ontology
 
     @classmethod

diff --git a/src/oaklib/interfaces/search_interface.py b/src/oaklib/interfaces/search_interface.py
@@ -5,9 +5,7 @@
 from oaklib.interfaces.basic_ontology_interface import BasicOntologyInterface
 from oaklib.types import CURIE
 
-__all__ = [
-    "SearchConfiguration",
-]
+__all__ = ["SearchConfiguration", "SearchInterface"]
 
 
 class SearchInterface(BasicOntologyInterface, ABC):

diff --git a/src/oaklib/resource.py b/src/oaklib/resource.py
@@ -1,6 +1,6 @@
 from dataclasses import dataclass
 from pathlib import Path
-from typing import TYPE_CHECKING, Type, Union
+from typing import TYPE_CHECKING, Optional, Type, Union
 
 from class_resolver import HintOrType
 
@@ -21,16 +21,36 @@ class OntologyResource:
     """
 
     slug: str = None
-    directory: str = None
-    scheme: str = None
-    format: str = None
-    url: str = None
-    readonly: bool = False
-    provider: str = None
-    local: bool = False
-    in_memory: bool = False
-    data: str = None
-    implementation_class: Union[Type] = None
+    """Name or path of ontology resource"""
+
+    directory: Optional[str] = None
+    """For local resources, the directory where the serialization is found"""
+
+    scheme: Optional[str] = None
+    """The scheme of the resource, e.g pronto, sqlite"""
+
+    format: Optional[str] = None
+    """For serialized resources, the serialization format"""
+
+    url: Optional[str] = None
+    """For remote resources, the URL from which it can be obtained"""
+
+    readonly: Optional[bool] = False
+    """Typically true for remote resources"""
+
+    provider: Optional[str] = None
+
+    local: Optional[bool] = True
+    """Is the resource locally on disk or remote?"""
+
+    in_memory: Optional[bool] = False
+
+    data: Optional[str] = None
+
+    implementation_class: Optional[Union[Type]] = None
+
+    import_depth: Optional[int] = None
+    """If set, this determines the maximum depth in the import tree to follow"""
 
     @property
     def local_path(self) -> Path:

diff --git a/src/oaklib/selector.py b/src/oaklib/selector.py
@@ -122,12 +122,15 @@ def get_resource_imp_class_from_suffix_descriptor(
     return impl_class, resource
 
 
-def get_resource_from_shorthand(descriptor: str, format: str = None) -> OntologyResource:
+def get_resource_from_shorthand(
+    descriptor: str, format: str = None, import_depth: Optional[int] = None
+) -> OntologyResource:
     """
     Maps from a shorthand descriptor to an OntologyResource.
 
     :param descriptor:
-    :param format:
+    :param format: file format/syntax, e.g obo, turtle
+    :param import_depth: maximum import depth to traverse
     :return:
     """
     from oaklib.implementations import (
@@ -137,6 +140,7 @@ def get_resource_from_shorthand(descriptor: str, format: str = None) -> Ontology
     )
 
     resource = OntologyResource(format=format)
+    resource.import_depth = import_depth
     resource.slug = descriptor
     impl_class: Optional[Type[OntologyInterface]] = None
     if descriptor:

diff --git a/tests/input/catalog-v001.xml b/tests/input/catalog-v001.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<catalog prefer="public" xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog">
+
+    <!-- ./ -->
+    <uri id="User Entered Import Resolution" name="http://purl.obolibrary.org/obo/test_imported_ontology.owl" uri="./test_imported_ontology.owl"/>
+</catalog>
diff --git a/tests/input/test_import_root.obo b/tests/input/test_import_root.obo
@@ -0,0 +1,7 @@
+format-version: 1.2
+ontology: test_import_root
+import: http://purl.obolibrary.org/obo/test_imported_ontology.owl
+
+[Term]
+id: PATO:0000001
+name: term in root
diff --git a/tests/input/test_import_root.owl b/tests/input/test_import_root.owl
@@ -0,0 +1,61 @@
+<?xml version="1.0"?>
+<rdf:RDF xmlns="http://purl.obolibrary.org/obo/test_import_root.owl#"
+     xml:base="http://purl.obolibrary.org/obo/test_import_root.owl"
+     xmlns:owl="http://www.w3.org/2002/07/owl#"
+     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+     xmlns:xml="http://www.w3.org/XML/1998/namespace"
+     xmlns:xsd="http://www.w3.org/2001/XMLSchema#"
+     xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
+     xmlns:oboInOwl="http://www.geneontology.org/formats/oboInOwl#">
+    <owl:Ontology rdf:about="http://purl.obolibrary.org/obo/test_import_root.owl">
+        <owl:imports rdf:resource="http://purl.obolibrary.org/obo/test_imported_ontology.owl"/>
+    </owl:Ontology>
+
+
+
+    <!-- 
+    ///////////////////////////////////////////////////////////////////////////////////////
+    //
+    // Annotation properties
+    //
+    ///////////////////////////////////////////////////////////////////////////////////////
+     -->
+
+
+
+
+    <!-- http://www.geneontology.org/formats/oboInOwl#id -->
+
+    <owl:AnnotationProperty rdf:about="http://www.geneontology.org/formats/oboInOwl#id"/>
+
+
+
+    <!-- http://www.w3.org/2000/01/rdf-schema#label -->
+
+    <owl:AnnotationProperty rdf:about="http://www.w3.org/2000/01/rdf-schema#label"/>
+
+
+
+    <!-- 
+    ///////////////////////////////////////////////////////////////////////////////////////
+    //
+    // Classes
+    //
+    ///////////////////////////////////////////////////////////////////////////////////////
+     -->
+
+
+
+
+    <!-- http://purl.obolibrary.org/obo/PATO_1 -->
+
+    <owl:Class rdf:about="http://purl.obolibrary.org/obo/PATO_1">
+        <oboInOwl:id rdf:datatype="http://www.w3.org/2001/XMLSchema#string">PATO:1</oboInOwl:id>
+        <rdfs:label rdf:datatype="http://www.w3.org/2001/XMLSchema#string">term in root</rdfs:label>
+    </owl:Class>
+</rdf:RDF>
+
+
+
+<!-- Generated by the OWL API (version 4.5.6) https://github.com/owlcs/owlapi -->
+
diff --git a/tests/input/test_imported_ontology.obo b/tests/input/test_imported_ontology.obo
@@ -0,0 +1,6 @@
+format-version: 1.2
+ontology: test_imported_ontology
+
+[Term]
+id: PATO:2
+name: term in imported
diff --git a/tests/input/test_imported_ontology.owl b/tests/input/test_imported_ontology.owl
@@ -0,0 +1,69 @@
+<?xml version="1.0"?>
+<rdf:RDF xmlns="http://purl.obolibrary.org/obo/test_imported_ontology.owl#"
+     xml:base="http://purl.obolibrary.org/obo/test_imported_ontology.owl"
+     xmlns:owl="http://www.w3.org/2002/07/owl#"
+     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+     xmlns:xml="http://www.w3.org/XML/1998/namespace"
+     xmlns:xsd="http://www.w3.org/2001/XMLSchema#"
+     xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
+     xmlns:oboInOwl="http://www.geneontology.org/formats/oboInOwl#">
+    <owl:Ontology rdf:about="http://purl.obolibrary.org/obo/test_imported_ontology.owl">
+        <oboInOwl:hasOBOFormatVersion rdf:datatype="http://www.w3.org/2001/XMLSchema#string">1.2</oboInOwl:hasOBOFormatVersion>
+    </owl:Ontology>
+
+
+
+    <!-- 
+    ///////////////////////////////////////////////////////////////////////////////////////
+    //
+    // Annotation properties
+    //
+    ///////////////////////////////////////////////////////////////////////////////////////
+     -->
+
+
+
+
+    <!-- http://www.geneontology.org/formats/oboInOwl#hasOBOFormatVersion -->
+
+    <owl:AnnotationProperty rdf:about="http://www.geneontology.org/formats/oboInOwl#hasOBOFormatVersion">
+        <rdfs:label rdf:datatype="http://www.w3.org/2001/XMLSchema#string">has_obo_format_version</rdfs:label>
+    </owl:AnnotationProperty>
+
+
+
+    <!-- http://www.geneontology.org/formats/oboInOwl#id -->
+
+    <owl:AnnotationProperty rdf:about="http://www.geneontology.org/formats/oboInOwl#id"/>
+
+
+
+    <!-- http://www.w3.org/2000/01/rdf-schema#label -->
+
+    <owl:AnnotationProperty rdf:about="http://www.w3.org/2000/01/rdf-schema#label"/>
+
+
+
+    <!-- 
+    ///////////////////////////////////////////////////////////////////////////////////////
+    //
+    // Classes
+    //
+    ///////////////////////////////////////////////////////////////////////////////////////
+     -->
+
+
+
+
+    <!-- http://purl.obolibrary.org/obo/PATO_2 -->
+
+    <owl:Class rdf:about="http://purl.obolibrary.org/obo/PATO_2">
+        <oboInOwl:id rdf:datatype="http://www.w3.org/2001/XMLSchema#string">PATO:2</oboInOwl:id>
+        <rdfs:label rdf:datatype="http://www.w3.org/2001/XMLSchema#string">term in imported</rdfs:label>
+    </owl:Class>
+</rdf:RDF>
+
+
+
+<!-- Generated by the OWL API (version 4.5.6) https://github.com/owlcs/owlapi -->
+
diff --git a/tests/test_implementations/test_pronto.py b/tests/test_implementations/test_pronto.py
@@ -1,6 +1,7 @@
 import logging
 import unittest
 
+import pronto
 from kgcl_schema.datamodel import kgcl
 
 from oaklib.datamodels import obograph
@@ -103,7 +104,6 @@ def test_metadata(self):
     def test_labels(self):
         """
         Tests labels can be retrieved, and no label is retrieved when a term does not exist
-        :return:
         """
         oi = self.oi
         label = oi.label("GO:0005773")
@@ -143,8 +143,6 @@ def test_synonyms(self):
     def test_mappings(self):
         oi = self.oi
         mappings = list(oi.get_sssom_mappings_by_curie(NUCLEUS))
-        # for m in mappings:
-        #    logging.info(yaml_dumper.dumps(m))
         assert any(m for m in mappings if m.object_id == "Wikipedia:Cell_nucleus")
         self.assertEqual(len(mappings), 2)
         for m in mappings:
@@ -177,6 +175,38 @@ def test_from_obo_library(self):
         curies = oi.curies_by_label("shape")
         self.assertEqual(["PATO:0000052"], curies)
 
+    @unittest.skip("https://github.com/althonos/pronto/issues/186")
+    def test_import_behavior(self):
+        """
+        Tests behavior of owl:imports
+
+        by default, imports should be followed
+
+        See: https://github.com/INCATools/ontology-access-kit/issues/248
+        """
+        for slug in ["test_import_root.obo", "test_import_root.obo"]:
+            resource = OntologyResource(slug=slug, directory=INPUT_DIR, local=True)
+            # print(resource.local_path)
+            # currently throws exception
+            pronto.Ontology(resource.local_path)
+            oi = ProntoImplementation.create(resource)
+            terms = list(oi.entities(owl_type="owl:Class"))
+            self.assertEqual(2, len(terms))
+
+    def test_no_import_depth(self):
+        """
+        Tests behavior of owl:imports
+
+        do not follow imports if depth is set to zero
+
+        See: https://github.com/INCATools/ontology-access-kit/issues/248
+        """
+        for slug in ["test_import_root.obo", "test_import_root.obo"]:
+            resource = OntologyResource(slug=slug, directory=INPUT_DIR, local=True, import_depth=0)
+            oi = ProntoImplementation(resource)
+            terms = list(oi.entities(owl_type="owl:Class"))
+            self.assertEqual(1, len(terms))
+
     @unittest.skip("Hide warnings")
     def test_from_owl(self):
         r = OntologyResource(local=True, slug="go-nucleus.owl", directory=INPUT_DIR)

diff --git a/tests/test_implementations/test_sparql.py b/tests/test_implementations/test_sparql.py
@@ -34,6 +34,7 @@
 TEST_RDF = INPUT_DIR / "go-nucleus.owl.ttl"
 TEST_INST_RDF = INPUT_DIR / "inst.owl.ttl"
 TEST_MUTABLE_RDF = OUTPUT_DIR / "go-nucleus.owl.ttl"
+TEST_IMPORTER = INPUT_DIR / "test_import_root.owl"
 
 
 class TestSparqlImplementation(unittest.TestCase):
@@ -249,3 +250,17 @@ def test_mutable(self):
         )
         self.assertCountEqual([], list(oi.descendants(NUCLEUS, predicates=preds, reflexive=False)))
         oi.save()
+
+    @unittest.skip("not yet implemented")
+    def test_import_behavior(self):
+        """
+        Tests behavior of owl:imports
+
+        by default, imports should be followed
+
+        See: https://github.com/INCATools/ontology-access-kit/issues/248
+        """
+        resource = OntologyResource(slug=str(TEST_IMPORTER))
+        oi = SparqlImplementation(resource)
+        terms = list(oi.entities(owl_type="owl:Class"))
+        self.assertCountEqual(["PATO:1", "PATO:2"], terms)