pyobo 0.10.11__py3-none-any.whl → 0.10.12__py3-none-any.whl

pyobo/api/names.py

@@ -2,11 +2,15 @@
 
 """High-level API for nomenclature."""
 
+from __future__ import annotations
+
 import logging
 import subprocess
 from functools import lru_cache
 from typing import Callable, List, Mapping, Optional, Set, TypeVar
 
+from curies import Reference, ReferenceTuple
+
 from .alts import get_primary_identifier
 from .utils import get_version
 from ..getters import NoBuild, get_ontology
@@ -32,6 +36,8 @@ logger = logging.getLogger(__name__)
 
 def get_name_by_curie(curie: str, *, version: Optional[str] = None) -> Optional[str]:
     """Get the name for a CURIE, if possible."""
+    if version is None:
+        version = get_version(curie.split(":")[0])
     prefix, identifier = normalize_curie(curie)
     if prefix and identifier:
         return get_name(prefix, identifier, version=version)
@@ -40,7 +46,8 @@ def get_name_by_curie(curie: str, *, version: Optional[str] = None) -> Optional[
 
 X = TypeVar("X")
 
-NO_BUILD_PREFIXES = set()
+NO_BUILD_PREFIXES: Set[str] = set()
+NO_BUILD_LOGGED: Set = set()
 
 
 def _help_get(
@@ -59,8 +66,10 @@ def _help_get(
             logger.warning("[%s] unable to look up results with %s", prefix, f)
             NO_BUILD_PREFIXES.add(prefix)
         return None
-    except ValueError:
-        logger.warning("[%s] unable to look up results with %s", prefix, f)
+    except ValueError as e:
+        if prefix not in NO_BUILD_PREFIXES:
+            logger.warning("[%s] value error while looking up results with %s: %s", prefix, f, e)
+            NO_BUILD_PREFIXES.add(prefix)
         return None
 
     if not mapping:
@@ -74,9 +83,17 @@ def _help_get(
 
 
 @wrap_norm_prefix
-def get_name(prefix: str, identifier: str, *, version: Optional[str] = None) -> Optional[str]:
+def get_name(
+    prefix: str | Reference | ReferenceTuple,
+    identifier: Optional[str] = None,
+    /,
+    *,
+    version: Optional[str] = None,
+) -> Optional[str]:
     """Get the name for an entity."""
-    return _help_get(get_id_name_mapping, prefix, identifier, version=version)
+    if isinstance(prefix, (ReferenceTuple, Reference)):
+        prefix, identifier = prefix.prefix, prefix.identifier
+    return _help_get(get_id_name_mapping, prefix, identifier, version=version)  # type:ignore
 
 
 @lru_cache()
@@ -159,8 +176,12 @@ def get_name_id_mapping(
 
 
 @wrap_norm_prefix
-def get_definition(prefix: str, identifier: str, *, version: Optional[str] = None) -> Optional[str]:
+def get_definition(
+    prefix: str, identifier: str | None = None, *, version: Optional[str] = None
+) -> Optional[str]:
     """Get the definition for an entity."""
+    if identifier is None:
+        prefix, _, identifier = prefix.rpartition(":")
     return _help_get(get_id_definition_mapping, prefix, identifier, version=version)
 
 

pyobo/api/utils.py

@@ -7,6 +7,7 @@ from typing import Optional
 
 import bioversions
 
+from ..constants import VERSION_PINS
 from ..utils.path import prefix_directory_join
 
 __all__ = [
@@ -25,6 +26,10 @@ def get_version(prefix: str) -> Optional[str]:
     :param prefix: the resource name
     :return: The version if available else None
     """
+    # Prioritize loaded environmental variable VERSION_PINS dictionary
+    version = VERSION_PINS.get(prefix)
+    if version:
+        return version
     try:
         version = bioversions.get_version(prefix)
     except KeyError:

pyobo/cli/lookup.py

@@ -282,7 +282,7 @@ def ancestors(prefix: str, identifier: str, force: bool, version: Optional[str])
     """Look up ancestors."""
     curies = get_ancestors(prefix=prefix, identifier=identifier, force=force, version=version)
     for curie in sorted(curies or []):
-        click.echo(f"{curie}\t{get_name_by_curie(curie)}")
+        click.echo(f"{curie}\t{get_name_by_curie(curie, version=version)}")
 
 
 @lookup.command()
@@ -295,7 +295,7 @@ def descendants(prefix: str, identifier: str, force: bool, version: Optional[str
     """Look up descendants."""
     curies = get_descendants(prefix=prefix, identifier=identifier, force=force, version=version)
     for curie in sorted(curies or []):
-        click.echo(f"{curie}\t{get_name_by_curie(curie)}")
+        click.echo(f"{curie}\t{get_name_by_curie(curie, version=version)}")
 
 
 @lookup.command()

pyobo/constants.py

@@ -2,7 +2,9 @@
 
 """Constants for PyOBO."""
 
+import json
 import logging
+import os
 import re
 
 import pystow
@@ -11,6 +13,7 @@ __all__ = [
     "RAW_DIRECTORY",
     "DATABASE_DIRECTORY",
     "SPECIES_REMAPPING",
+    "VERSION_PINS",
 ]
 
 logger = logging.getLogger(__name__)
@@ -80,7 +83,6 @@ TYPEDEFS_FILE = "typedefs.tsv.gz"
 SPECIES_RECORD = "5334738"
 SPECIES_FILE = "species.tsv.gz"
 
-
 NCBITAXON_PREFIX = "NCBITaxon"
 DATE_FORMAT = "%d:%m:%Y %H:%M"
 PROVENANCE_PREFIXES = {
@@ -99,3 +101,31 @@ PROVENANCE_PREFIXES = {
     "isbn",
     "issn",
 }
+
+# Load version pin dictionary from the environmental variable VERSION_PINS
+try:
+    VERSION_PINS_STR = os.getenv("VERSION_PINS")
+    if not VERSION_PINS_STR:
+        VERSION_PINS = {}
+    else:
+        VERSION_PINS = json.loads(VERSION_PINS_STR)
+        for k, v in VERSION_PINS.items():
+            if not isinstance(k, str) or not isinstance(v, str):
+                logger.error("The prefix and version name must both be " "strings")
+            VERSION_PINS = {}
+            break
+except ValueError as e:
+    logger.error(
+        "The value for the environment variable VERSION_PINS must be a valid JSON string: %s" % e
+    )
+    VERSION_PINS = {}
+
+if VERSION_PINS:
+    logger.debug(
+        f"These are the resource versions that are pinned.\n{VERSION_PINS}. "
+        f"\nPyobo will download the latest version of a resource if it's "
+        f"not pinned.\nIf you want to use a specific version of a "
+        f"resource, edit your VERSION_PINS environmental "
+        f"variable which is a JSON string to include a prefix and version "
+        f"name."
+    )

pyobo/gilda_utils.py

@@ -15,6 +15,7 @@ from gilda.term import filter_out_duplicates
 from tqdm.auto import tqdm
 
 from pyobo import (
+    get_descendants,
     get_id_name_mapping,
     get_id_species_mapping,
     get_id_synonyms_mapping,
@@ -247,3 +248,23 @@ def get_gilda_terms(
             )
             if term is not None:
                 yield term
+
+
+def get_gilda_term_subset(
+    source: str, ancestors: Union[str, List[str]], **kwargs
+) -> Iterable[gilda.term.Term]:
+    """Get a subset of terms."""
+    subset = {
+        descendant
+        for parent_curie in _ensure_list(ancestors)
+        for descendant in get_descendants(*parent_curie.split(":")) or []
+    }
+    for term in get_gilda_terms(source, **kwargs):
+        if bioregistry.curie_to_str(term.db, term.id) in subset:
+            yield term
+
+
+def _ensure_list(s: Union[str, List[str]]) -> List[str]:
+    if isinstance(s, str):
+        return [s]
+    return s

pyobo/identifier_utils.py

@@ -2,11 +2,14 @@
 
 """Utilities for handling prefixes."""
 
+from __future__ import annotations
+
 import logging
 from functools import wraps
 from typing import Optional, Tuple, Union
 
 import bioregistry
+from curies import Reference, ReferenceTuple
 
 from .registries import (
     curie_has_blacklisted_prefix,
@@ -108,11 +111,25 @@ def wrap_norm_prefix(f):
     """Decorate a function that take in a prefix to auto-normalize, or return None if it can't be normalized."""
 
     @wraps(f)
-    def _wrapped(prefix, *args, **kwargs):
-        norm_prefix = bioregistry.normalize_prefix(prefix)
-        if norm_prefix is None:
-            raise ValueError(f"Invalid prefix: {prefix}")
-        return f(norm_prefix, *args, **kwargs)
+    def _wrapped(prefix: str | Reference | ReferenceTuple, *args, **kwargs):
+        if isinstance(prefix, str):
+            norm_prefix = bioregistry.normalize_prefix(prefix)
+            if norm_prefix is None:
+                raise ValueError(f"Invalid prefix: {prefix}")
+            prefix = norm_prefix
+        elif isinstance(prefix, Reference):
+            norm_prefix = bioregistry.normalize_prefix(prefix.prefix)
+            if norm_prefix is None:
+                raise ValueError(f"Invalid prefix: {prefix.prefix}")
+            prefix = Reference(prefix=norm_prefix, identifier=prefix.identifier)
+        elif isinstance(prefix, ReferenceTuple):
+            norm_prefix = bioregistry.normalize_prefix(prefix.prefix)
+            if norm_prefix is None:
+                raise ValueError(f"Invalid prefix: {prefix.prefix}")
+            prefix = ReferenceTuple(norm_prefix, prefix.identifier)
+        else:
+            raise TypeError
+        return f(prefix, *args, **kwargs)
 
     return _wrapped
 

pyobo/reader.py

@@ -417,7 +417,7 @@ def _clean_definition(s: str) -> str:
     # if '\t' in s:
     #     logger.warning('has tab')
     return (
-        s.replace('\\"', '"').replace("\n", " ").replace("\t", " ").replace("\d", "")  # noqa:W605
+        s.replace('\\"', '"').replace("\n", " ").replace("\t", " ").replace(r"\d", "")  # noqa:W605
     )
 
 

pyobo/sources/__init__.py

@@ -12,6 +12,7 @@ from .civic_gene import CIVICGeneGetter
 from .complexportal import ComplexPortalGetter
 from .conso import CONSOGetter
 from .cpt import CPTGetter
+from .credit import CreditGetter
 from .cvx import CVXGetter
 from .depmap import DepMapGetter
 from .dictybase_gene import DictybaseGetter
@@ -69,6 +70,7 @@ __all__ = [
     "CVXGetter",
     "ChEMBLCompoundGetter",
     "ComplexPortalGetter",
+    "CreditGetter",
     "DepMapGetter",
     "DictybaseGetter",
     "DrugBankGetter",

pyobo/sources/antibodyregistry.py

@@ -5,12 +5,12 @@
 import logging
 from typing import Iterable, Mapping, Optional
 
-import bioversions
 import pandas as pd
 from bioregistry.utils import removeprefix
 from tqdm.auto import tqdm
 
 from pyobo import Obo, Term
+from pyobo.api.utils import get_version
 from pyobo.utils.path import ensure_df
 
 __all__ = [
@@ -27,7 +27,7 @@ CHUNKSIZE = 20_000
 def get_chunks(*, force: bool = False, version: Optional[str] = None) -> pd.DataFrame:
     """Get the BioGRID identifiers mapping dataframe."""
     if version is None:
-        version = bioversions.get_version(PREFIX)
+        version = get_version(PREFIX)
     df = ensure_df(
         PREFIX,
         url=URL,

pyobo/sources/biogrid.py

@@ -5,9 +5,9 @@
 from functools import partial
 from typing import Mapping, Optional
 
-import bioversions
 import pandas as pd
 
+from pyobo.api.utils import get_version
 from pyobo.resources.ncbitaxon import get_ncbitaxon_id
 from pyobo.utils.cache import cached_mapping
 from pyobo.utils.path import ensure_df, prefix_directory_join
@@ -52,7 +52,7 @@ def _lookup(name: str) -> Optional[str]:
 
 def get_df() -> pd.DataFrame:
     """Get the BioGRID identifiers mapping dataframe."""
-    version = bioversions.get_version("biogrid")
+    version = get_version("biogrid")
     url = f"{BASE_URL}/BIOGRID-{version}/BIOGRID-IDENTIFIERS-{version}.tab.zip"
     df = ensure_df(PREFIX, url=url, skiprows=28, dtype=str, version=version)
     df["taxonomy_id"] = df["ORGANISM_OFFICIAL_NAME"].map(_lookup)
@@ -65,7 +65,7 @@ def get_df() -> pd.DataFrame:
         "cache",
         "xrefs",
         name="ncbigene.tsv",
-        version=partial(bioversions.get_version, PREFIX),
+        version=partial(get_version, PREFIX),
     ),
     header=["biogrid_id", "ncbigene_id"],
 )

pyobo/sources/credit.py

@@ -0,0 +1,68 @@
+"""Converter for the Contributor Roles Taxonomy."""
+
+from __future__ import annotations
+
+import json
+from typing import Iterable
+
+from more_itertools import chunked
+
+from pyobo.struct import Obo, Term
+from pyobo.utils.path import ensure_path
+
+__all__ = [
+    "CreditGetter",
+]
+
+url = "https://api.github.com/repos/CASRAI-CRedIT/Dictionary/contents/Picklists/Contributor%20Roles"
+PREFIX = "credit"
+
+
+class CreditGetter(Obo):
+    """An ontology representation of the Contributor Roles Taxonomy."""
+
+    ontology = PREFIX
+    static_version = "2022"
+    idspaces = {
+        PREFIX: "https://credit.niso.org/contributor-roles/",
+    }
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return get_terms(force=force)
+
+
+def get_obo(force: bool = False) -> Obo:
+    """Get RGD as OBO."""
+    return CreditGetter(force=force)
+
+
+def get_terms(force: bool = False) -> list[Term]:
+    """Get terms from the Contributor Roles Taxonomy via GitHub."""
+    path = ensure_path(PREFIX, url=url, name="picklist-api.json", force=force)
+    with open(path) as f:
+        data = json.load(f)
+    terms = []
+    for x in data:
+        name = x["name"].removesuffix(".md").lower()
+
+        pp = ensure_path(PREFIX, "picklist", url=x["download_url"], backend="requests")
+        with open(pp) as f:
+            header, *rest = f.read().splitlines()
+            name = header = header.removeprefix("# Contributor Roles/")
+            dd = {k.removeprefix("## "): v for k, v in chunked(rest, 2)}
+            identifier = (
+                dd["Canonical URL"]
+                .removeprefix("https://credit.niso.org/contributor-roles/")
+                .rstrip("/")
+            )
+            desc = dd["Short definition"]
+            terms.append(
+                Term.from_triple(prefix=PREFIX, identifier=identifier, name=name, definition=desc)
+            )
+
+    return terms
+
+
+if __name__ == "__main__":
+    get_obo(force=True).write_default(write_obo=True)

pyobo/sources/geonames.py

@@ -1,5 +1,7 @@
 """Get terms from geonames."""
 
+from __future__ import annotations
+
 import logging
 from typing import Collection, Iterable, Mapping
 
@@ -146,15 +148,7 @@ def get_code_to_admin2(
     return code_to_admin2
 
 
-def get_cities(
-    code_to_country,
-    code_to_admin1,
-    code_to_admin2,
-    *,
-    minimum_population: int = 100_000,
-    force: bool = False,
-) -> Mapping[str, Term]:
-    """Get a mapping from city code to term."""
+def _get_cities_df(force: bool = False) -> pd.DataFrame:
     columns = [
         "geonames_id",
         "name",
@@ -184,7 +178,19 @@ def get_cities(
         names=columns,
         dtype=str,
     )
+    return cities_df
+
 
+def get_cities(
+    code_to_country,
+    code_to_admin1,
+    code_to_admin2,
+    *,
+    minimum_population: int = 100_000,
+    force: bool = False,
+) -> Mapping[str, Term]:
+    """Get a mapping from city code to term."""
+    cities_df = _get_cities_df(force=force)
     cities_df = cities_df[cities_df.population.astype(int) > minimum_population]
     cities_df.synonyms = cities_df.synonyms.str.split(",")
 
@@ -235,5 +241,17 @@ def get_cities(
     return terms
 
 
+def get_city_to_country() -> dict[str, str]:
+    """Get a mapping from city GeoNames to country GeoNames id."""
+    rv = {}
+    code_to_country = get_code_to_country()
+    cities_df = _get_cities_df()
+    for city_geonames_id, country_code in cities_df[["geonames_id", "country_code"]].values:
+        if pd.isna(city_geonames_id) or pd.isna(country_code):
+            continue
+        rv[city_geonames_id] = code_to_country[country_code].identifier
+    return rv
+
+
 if __name__ == "__main__":
     GeonamesGetter().write_default(write_obo=True, force=True)

pyobo/sources/hgnc.py

@@ -10,10 +10,10 @@ from collections import Counter, defaultdict
 from operator import attrgetter
 from typing import DefaultDict, Dict, Iterable, Optional
 
-import bioversions
 from tabulate import tabulate
 from tqdm.auto import tqdm
 
+from pyobo.api.utils import get_version
 from pyobo.struct import (
     Obo,
     Reference,
@@ -241,7 +241,7 @@ def get_obo(*, force: bool = False) -> Obo:
 def get_terms(version: Optional[str] = None, force: bool = False) -> Iterable[Term]:  # noqa:C901
     """Get HGNC terms."""
     if version is None:
-        version = bioversions.get_version("hgnc")
+        version = get_version("hgnc")
     unhandled_entry_keys: typing.Counter[str] = Counter()
     unhandle_locus_types: DefaultDict[str, Dict[str, Term]] = defaultdict(dict)
     path = ensure_path(

pyobo/sources/mesh.py

@@ -11,6 +11,7 @@ from xml.etree.ElementTree import Element
 
 from tqdm.auto import tqdm
 
+from pyobo.api.utils import get_version
 from pyobo.identifier_utils import standardize_ec
 from pyobo.struct import Obo, Reference, Synonym, Term
 from pyobo.utils.cache import cached_json, cached_mapping
@@ -331,9 +332,8 @@ def get_mesh_category_curies(
     .. seealso:: https://meshb.nlm.nih.gov/treeView
     """
     if version is None:
-        import bioversions
-
-        version = bioversions.get_version("mesh")
+        version = get_version("mesh")
+        assert version is not None
     tree_to_mesh = get_tree_to_mesh_id(version=version)
     rv = []
     for i in range(1, 100):

pyobo/sources/msigdb.py

@@ -137,7 +137,7 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
 def _get_definition(attrib) -> Optional[str]:
     rv = attrib["DESCRIPTION_FULL"].strip() or attrib["DESCRIPTION_BRIEF"].strip() or None
     if rv is not None:
-        return rv.replace("\d", "").replace("\s", "")  # noqa: W605
+        return rv.replace(r"\d", "").replace(r"\s", "")  # noqa: W605
     return None
 
 

pyobo/sources/npass.py

@@ -64,7 +64,7 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
         )
 
         for xref_prefix, xref_id in [
-            ("chembl", chembl_id),
+            ("chembl.compound", chembl_id),
             # ("zinc", zinc_id),
         ]:
             if pd.notna(xref_id):

pyobo/sources/pubchem.py

@@ -5,12 +5,12 @@
 import logging
 from typing import Iterable, Mapping, Optional
 
-import bioversions
 import pandas as pd
 from bioregistry.utils import removeprefix
 from tqdm.auto import tqdm
 
 from ..api import get_name_id_mapping
+from ..api.utils import get_version
 from ..struct import Obo, Reference, Synonym, Term
 from ..utils.iter import iterate_gzips_together
 from ..utils.path import ensure_df, ensure_path
@@ -26,7 +26,7 @@ PREFIX = "pubchem.compound"
 
 def _get_pubchem_extras_url(version: Optional[str], end: str) -> str:
     if version is None:
-        version = bioversions.get_version("pubchem")
+        version = get_version("pubchem")
     return f"ftp://ftp.ncbi.nlm.nih.gov/pubchem/Compound/Monthly/{version}/Extras/{end}"
 
 
@@ -100,7 +100,7 @@ def get_pubchem_id_to_mesh_id(version: str) -> Mapping[str, str]:
 
 def _ensure_cid_name_path(*, version: Optional[str] = None, force: bool = False) -> str:
     if version is None:
-        version = bioversions.get_version("pubchem")
+        version = get_version("pubchem")
     # 2 tab-separated columns: compound_id, name
     cid_name_url = _get_pubchem_extras_url(version, "CID-Title.gz")
     cid_name_path = ensure_path(PREFIX, url=cid_name_url, version=version, force=force)

pyobo/sources/rgd.py

@@ -28,7 +28,7 @@ old_name_type = SynonymTypeDef.from_text("old_name")
 
 # NOTE unigene id was discontinue in January 18th, 2021 dump
 
-GENES_URL = "https://download.rgd.mcw.edu/data_release/GENES.RAT.txt"
+GENES_URL = "https://download.rgd.mcw.edu/data_release/GENES_RAT.txt"
 GENES_HEADER = [
     "GENE_RGD_ID",
     "SYMBOL",

pyobo/sources/rhea.py

@@ -5,9 +5,9 @@
 import logging
 from typing import TYPE_CHECKING, Dict, Iterable, Optional
 
-import bioversions
 import pystow
 
+from pyobo.api.utils import get_version
 from pyobo.struct import Obo, Reference, Term
 from pyobo.struct.typedef import (
     TypeDef,
@@ -63,7 +63,7 @@ def ensure_rhea_rdf(version: Optional[str] = None, force: bool = False) -> "rdfl
     """Get the Rhea RDF graph."""
     # see docs: https://ftp.expasy.org/databases/rhea/rdf/rhea_rdf_documentation.pdf
     if version is None:
-        version = bioversions.get_version(PREFIX)
+        version = get_version(PREFIX)
     return pystow.ensure_rdf(
         "pyobo",
         "raw",