pyobo 0.12.1__py3-none-any.whl → 0.12.3__py3-none-any.whl

pyobo/struct/struct.py

@@ -22,7 +22,7 @@ import curies
 import networkx as nx
 import pandas as pd
 import ssslm
-from curies import ReferenceTuple
+from curies import Converter, ReferenceTuple
 from curies import vocabulary as _cv
 from more_click import force_option, verbose_option
 from tqdm.auto import tqdm
@@ -84,6 +84,7 @@ __all__ = [
     "Synonym",
     "SynonymTypeDef",
     "Term",
+    "TypeDef",
     "abbreviation",
     "acronym",
     "make_ad_hoc_ontology",
@@ -91,9 +92,6 @@ __all__ = [
 
 logger = logging.getLogger(__name__)
 
-#: This is what happens if no specificity is given
-DEFAULT_SPECIFICITY: _cv.SynonymScope = "RELATED"
-
 #: Columns in the SSSOM dataframe
 SSSOM_DF_COLUMNS = [
     "subject_id",
@@ -104,7 +102,6 @@ SSSOM_DF_COLUMNS = [
     "confidence",
     "contributor",
 ]
-UNSPECIFIED_MATCHING_CURIE = "sempav:UnspecifiedMatching"
 FORMAT_VERSION = "1.4"
 
 
@@ -153,14 +150,14 @@ class Synonym(HasReferencesMixin):
     def _sort_key(self) -> tuple[str, _cv.SynonymScope, str]:
         return (
             self.name,
-            self.specificity or DEFAULT_SPECIFICITY,
+            self.specificity or _cv.DEFAULT_SYNONYM_SCOPE,
             self.type.curie if self.type else "",
         )
 
     @property
     def predicate(self) -> curies.NamedReference:
         """Get the specificity reference."""
-        return _cv.synonym_scopes[self.specificity or DEFAULT_SPECIFICITY]
+        return _cv.synonym_scopes[self.specificity or _cv.DEFAULT_SYNONYM_SCOPE]
 
     def to_obo(
         self,
@@ -189,7 +186,7 @@ class Synonym(HasReferencesMixin):
         elif self.type is not None:
             # it's not valid to have a synonym type without a specificity,
             # so automatically assign one if we'll need it
-            x = f"{x} {DEFAULT_SPECIFICITY}"
+            x = f"{x} {_cv.DEFAULT_SYNONYM_SCOPE}"
 
         # Add on the synonym type, if exists
         if self.type is not None:
@@ -429,9 +426,8 @@ class Term(Stanza):
         if self.definition:
             yield f"def: {self._definition_fp()}"
         # 7
-        for x in self.get_property_values(v.comment):
-            if isinstance(x, OBOLiteral):
-                yield f'comment: "{x.value}"'
+        for comment in self.get_comments():
+            yield f'comment: "{comment}"'
         # 8
         yield from _reference_list_tag("subset", self.subsets, ontology_prefix)
         # 9
@@ -723,17 +719,11 @@ class Obo:
         """Iterate over terms in this ontology."""
         raise NotImplementedError
 
-    def get_graph(self):
-        """Get an OBO Graph object."""
-        from ..obographs import graph_from_obo
-
-        return graph_from_obo(self)
-
-    def write_obograph(self, path: str | Path) -> None:
+    def write_obograph(self, path: str | Path, *, converter: Converter | None = None) -> None:
         """Write OBO Graph json."""
-        graph = self.get_graph()
-        with safe_open(path, read=False) as file:
-            file.write(graph.model_dump_json(indent=2, exclude_none=True, exclude_unset=True))
+        from . import obograph
+
+        obograph.write_obograph(self, path, converter=converter)
 
     @classmethod
     def cli(cls, *args, default_rewrite: bool = False) -> Any:
@@ -1642,13 +1632,13 @@ class Obo:
     #############
 
     def iterate_edges(
-        self, *, use_tqdm: bool = False
+        self, *, use_tqdm: bool = False, include_xrefs: bool = True
     ) -> Iterable[tuple[Stanza, TypeDef, Reference]]:
         """Iterate over triples of terms, relations, and their targets."""
         _warned: set[ReferenceTuple] = set()
         typedefs = self._index_typedefs()
         for stanza in self._iter_stanzas(use_tqdm=use_tqdm, desc=f"[{self.ontology}] edge"):
-            for predicate, reference in stanza._iter_edges():
+            for predicate, reference in stanza._iter_edges(include_xrefs=include_xrefs):
                 if td := self._get_typedef(stanza, predicate, _warned, typedefs):
                     yield stanza, td, reference
 

pyobo/struct/struct_utils.py

@@ -231,6 +231,8 @@ class Stanza(Referenced, HasReferencesMixin):
     ) -> None:
         self._axioms[_property_resolve(p, o)].append(annotation)
 
+    # TODO check different usages of this
+
     def append_equivalent(
         self,
         reference: ReferenceHint,
@@ -242,6 +244,15 @@ class Stanza(Referenced, HasReferencesMixin):
             stanza_type_to_eq_prop[self.type], reference, annotations=annotations
         )
 
+    def append_equivalent_to(
+        self, reference: ReferenceHint, *, annotations: Iterable[Annotation] | None = None
+    ) -> Self:
+        """Append to the "equivalent to" list."""
+        reference = _ensure_ref(reference)
+        self.equivalent_to.append(reference)
+        self._extend_annotations(stanza_type_to_eq_prop[self.type], reference, annotations)
+        return self
+
     def append_xref(
         self,
         reference: ReferenceHint,
@@ -317,15 +328,6 @@ class Stanza(Referenced, HasReferencesMixin):
         self.union_of.append(_ensure_ref(reference))
         return self
 
-    def append_equivalent_to(
-        self, reference: ReferenceHint, *, annotations: Iterable[Annotation] | None = None
-    ) -> Self:
-        """Append to the "equivalent to" list."""
-        reference = _ensure_ref(reference)
-        self.equivalent_to.append(reference)
-        self._extend_annotations(stanza_type_to_eq_prop[self.type], reference, annotations)
-        return self
-
     def _iterate_intersection_of_obo(self, *, ontology_prefix: str) -> Iterable[str]:
         for element in sorted(self.intersection_of, key=self._intersection_of_key):
             match element:
@@ -679,14 +681,18 @@ class Stanza(Referenced, HasReferencesMixin):
         """Add a comment property."""
         return self.annotate_string(v.comment, value, annotations=annotations, language=language)
 
+    def get_comments(self) -> list[str]:
+        """Get all comment strings."""
+        return [x.value for x in self.get_property_values(v.comment) if isinstance(x, OBOLiteral)]
+
     @property
     def alt_ids(self) -> Sequence[Reference]:
         """Get alternative terms."""
         return tuple(self.get_property_objects(v.alternative_term))
 
-    def get_edges(self) -> list[tuple[Reference, Reference]]:
+    def get_edges(self, *, include_xrefs: bool = True) -> list[tuple[Reference, Reference]]:
         """Get edges."""
-        return list(self._iter_edges())
+        return list(self._iter_edges(include_xrefs=include_xrefs))
 
     def _iter_parents(self) -> Iterable[tuple[Reference, Reference]]:
         parent_prop = stanza_type_to_prop[self.type]
@@ -702,7 +708,7 @@ class Stanza(Referenced, HasReferencesMixin):
                 case (predicate, target):
                     yield predicate, target
 
-    def _iter_edges(self) -> Iterable[tuple[Reference, Reference]]:
+    def _iter_edges(self, *, include_xrefs: bool = True) -> Iterable[tuple[Reference, Reference]]:
         # The following are "object" properties, meaning
         # they're part of the definition of the object
         yield from self.iterate_relations()
@@ -715,8 +721,10 @@ class Stanza(Referenced, HasReferencesMixin):
         for subset in self.subsets:
             yield v.in_subset, subset
         yield from self.iterate_object_properties()
-        for xref_reference in self.xrefs:
-            yield v.has_dbxref, xref_reference
+
+        if include_xrefs:
+            for xref_reference in self.xrefs:
+                yield v.has_dbxref, xref_reference
 
         # TODO disjoint_from
 

pyobo/struct/typedef.py

@@ -294,6 +294,10 @@ has_end_date = TypeDef(
 has_title = TypeDef(reference=v.has_title, is_metadata_tag=True)
 has_license = TypeDef(reference=v.has_license, is_metadata_tag=True)
 has_description = TypeDef(reference=v.has_description, is_metadata_tag=True)
+obo_autogenerated_by = TypeDef(reference=v.obo_autogenerated_by, is_metadata_tag=True)
+obo_has_format_version = TypeDef(reference=v.obo_has_format_version, is_metadata_tag=True)
+obo_is_metadata_tag = TypeDef(reference=v.obo_is_metadata_tag, is_metadata_tag=True)
+obo_has_id = TypeDef(reference=v.obo_has_id, is_metadata_tag=True)
 
 in_subset = TypeDef(reference=v.in_subset, is_metadata_tag=True)
 has_term_editor = TypeDef(reference=v.has_term_editor, is_metadata_tag=True)

pyobo/struct/vocabulary.py

@@ -47,6 +47,13 @@ has_dbxref = _c(_v.has_dbxref)
 
 in_subset = _c(_v.obo_in_subset)
 has_obo_namespace = _c(_v.obo_has_namespace)
+obo_is_metadata_tag = Reference(
+    prefix="oboinowl", identifier="is_metadata_tag", name="is metadata tag"
+)
+obo_has_id = Reference(prefix="oboinowl", identifier="id", name="has ID")
+obo_has_format_version = Reference(
+    prefix="oboinowl", identifier="hasOBOFormatVersion", name="has OBO format version"
+)
 obo_autogenerated_by = _c(_v.obo_autogenerated_by)
 obo_creation_date = _c(_v.obo_creation_date)
 

pyobo/version.py

@@ -12,7 +12,7 @@ __all__ = [
     "get_version",
 ]
 
-VERSION = "0.12.1"
+VERSION = "0.12.3"
 
 
 def get_git_hash() -> str:

{pyobo-0.12.1.dist-info → pyobo-0.12.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobo
-Version: 0.12.1
+Version: 0.12.3
 Summary: A python package for handling and generating OBO
 Keywords: snekpack,cookiecutter,ontologies,biomedical ontologies,life sciences,natural sciences,bioinformatics,cheminformatics,Open Biomedical Ontologies,OBO
 Author: Charles Tapley Hoyt
@@ -39,13 +39,14 @@ Requires-Dist: cachier
 Requires-Dist: pystow>=0.7.0
 Requires-Dist: bioversions>=0.8.0
 Requires-Dist: bioregistry>=0.12.7
-Requires-Dist: bioontologies>=0.7.1
+Requires-Dist: bioontologies>=0.7.2
 Requires-Dist: ssslm>=0.0.13
 Requires-Dist: zenodo-client>=0.3.6
 Requires-Dist: class-resolver>=0.6.0
 Requires-Dist: psycopg2-binary
 Requires-Dist: pydantic>=2.0
-Requires-Dist: curies>=0.10.13
+Requires-Dist: curies>=0.10.17
+Requires-Dist: curies-processing>=0.1.0
 Requires-Dist: python-dateutil
 Requires-Dist: networkx>=3.4
 Requires-Dist: drugbank-downloader
@@ -55,6 +56,7 @@ Requires-Dist: clinicaltrials-downloader>=0.0.2
 Requires-Dist: nih-reporter-downloader>=0.0.1
 Requires-Dist: typing-extensions
 Requires-Dist: rdflib
+Requires-Dist: obographs>=0.0.8
 Requires-Dist: ssslm[gilda] ; extra == 'gilda'
 Requires-Dist: ssslm[gilda-slim] ; extra == 'gilda-slim'
 Maintainer: Charles Tapley Hoyt
@@ -395,19 +397,6 @@ class MyTestCase(unittest.TestCase):
           pyobo.get_name("chebi", "1234")
 ```
 
-## Preprocessing CURIEs, URIs, and unqualified identifiers
-
-In order to normalize references and identify resources, PyOBO uses the
-[Bioregistry](https://github.com/bioregistry/bioregistry). It used to be a part
-of PyOBO, but has since been externalized for more general reuse.
-
-At
-[src/pyobo/identifier_utils/preprocessing.json](https://github.com/pyobo/pyobo/blob/master/src/pyobo/src/pyobo/identifier_utils/preprocessing.json)
-is the curated set of pre-processing rules. These are used in combination with
-the `curies` package to do pre-processing steps on CURIEs, URIs, and unqualified
-identifiers beyond what is possible with the Bioregistry. See
-https://curies.readthedocs.io/en/latest/preprocessing.html.
-
 ## Troubleshooting
 
 The OBO Foundry seems to be pretty unstable with respect to the URLs to OBO