pyobo 0.12.4__py3-none-any.whl → 0.12.5__py3-none-any.whl

pyobo/struct/typedef.py

@@ -15,7 +15,9 @@ __all__ = [
     "alternative_term",
     "broad_match",
     "close_match",
+    "contributes_to_condition",
     "default_typedefs",
+    "derives_from_organism",
     "editor_note",
     "enables",
     "exact_match",
@@ -24,10 +26,12 @@ __all__ = [
     "gene_product_member_of",
     "has_contributor",
     "has_dbxref",
+    "has_depiction",
     "has_end_date",
     "has_gene_product",
     "has_homepage",
     "has_inchi",
+    "has_mailbox",
     "has_mature",
     "has_member",
     "has_part",
@@ -103,12 +107,18 @@ has_component = TypeDef(
 derives_from = TypeDef(
     reference=Reference(prefix=RO_PREFIX, identifier="0001000", name="derives from"),
 )
+derives_from_organism = TypeDef(
+    reference=Reference(prefix="CLO", identifier="0037207", name="derives from organism")
+)
 molecularly_interacts_with = TypeDef(
     reference=Reference(prefix=RO_PREFIX, identifier="0002436", name="molecularly interacts with"),
 )
 located_in = TypeDef(
     reference=Reference(prefix=RO_PREFIX, identifier="0001025", name="located in"),
 )
+contributes_to_condition = TypeDef(
+    reference=Reference(prefix=RO_PREFIX, identifier="0003304", name="contributes to condition"),
+)
 exact_match = TypeDef(reference=v.exact_match, is_metadata_tag=True)
 narrow_match = TypeDef(reference=v.narrow_match, is_metadata_tag=True)
 broad_match = TypeDef(reference=v.broad_match, is_metadata_tag=True)
@@ -257,9 +267,11 @@ has_smiles = TypeDef(reference=v.has_smiles, is_metadata_tag=True).append_xref(v
 
 has_inchi = TypeDef(reference=v.has_inchi, is_metadata_tag=True).append_xref(v.debio_has_inchi)
 
-has_homepage = TypeDef(
-    reference=Reference(prefix="foaf", identifier="homepage", name="homepage"), is_metadata_tag=True
-)
+has_homepage = TypeDef(reference=v.has_homepage, is_metadata_tag=True)
+has_depiction = TypeDef(reference=v.has_depiction, is_metadata_tag=True)
+has_mailbox = TypeDef(reference=v.has_mailbox, is_metadata_tag=True)
+has_mailing_list = TypeDef(reference=v.has_mailing_list, is_metadata_tag=True)
+has_repository = TypeDef(reference=v.has_repository, is_metadata_tag=True)
 
 has_category = TypeDef(
     reference=Reference(prefix="biolink", identifier="category", name="has category"),

pyobo/struct/vocabulary.py

@@ -90,6 +90,14 @@ has_description = _c(_v.has_description)
 has_license = _c(_v.has_license)
 has_title = _c(_v.has_title)
 
+has_homepage = Reference(prefix="foaf", identifier="homepage", name="has homepage")
+has_logo = Reference(prefix="foaf", identifier="logo", name="has logo")
+has_mailbox = Reference(prefix="foaf", identifier="mbox", name="has mailbox")
+has_depiction = Reference(prefix="foaf", identifier="depicted_by", name="depicted by")
+has_repository = Reference(prefix="doap", identifier="repository", name="has repository")
+has_mailing_list = Reference(prefix="doap", identifier="mailing-list", name="has mailing list")
+has_maintainer = Reference(prefix="doap", identifier="maintainer", name="has maintainer")
+
 has_part = Reference(prefix=BFO_PREFIX, identifier="0000051", name="has part")
 part_of = Reference(prefix=BFO_PREFIX, identifier="0000050", name="part of")
 orthologous = Reference(

pyobo/utils/cache.py

@@ -12,8 +12,9 @@ from pystow.cache import CachedCollection as cached_collection  # noqa:N813
 from pystow.cache import CachedDataFrame as cached_df  # noqa:N813
 from pystow.cache import CachedJSON as cached_json  # noqa:N813
 from pystow.cache import CachedPickle as cached_pickle  # noqa:N813
+from pystow.utils import safe_open
 
-from .io import open_map_tsv, open_multimap_tsv, safe_open, write_map_tsv, write_multimap_tsv
+from .io import open_map_tsv, open_multimap_tsv, write_map_tsv, write_multimap_tsv
 
 __all__ = [
     "cached_collection",
@@ -69,13 +70,13 @@ NODE_LINK_STYLE = "links"  # TODO update to "edges"
 
 def get_gzipped_graph(path: str | Path) -> nx.MultiDiGraph:
     """Read a graph that's gzipped nodelink."""
-    with safe_open(path, read=True) as file:
+    with safe_open(path, operation="read") as file:
         return nx.node_link_graph(json.load(file), edges=NODE_LINK_STYLE)
 
 
 def write_gzipped_graph(graph: nx.MultiDiGraph, path: str | Path) -> None:
     """Write a graph as gzipped nodelink."""
-    with safe_open(path, read=False) as file:
+    with safe_open(path, operation="write") as file:
         json.dump(nx.node_link_data(graph, edges=NODE_LINK_STYLE), file)
 
 

pyobo/utils/io.py

@@ -1,27 +1,24 @@
 """I/O utilities."""
 
 import collections.abc
-import contextlib
-import csv
 import gzip
 import logging
 from collections import defaultdict
 from collections.abc import Generator, Iterable, Mapping
 from contextlib import contextmanager
 from pathlib import Path
-from typing import Literal, TextIO, TypeVar
+from typing import TypeVar, cast
 
 import pandas as pd
+import pystow.utils
+from pystow.utils import safe_open_reader, safe_open_writer
 from tqdm.auto import tqdm
 
 __all__ = [
-    "get_reader",
     "multidict",
     "multisetdict",
     "open_map_tsv",
     "open_multimap_tsv",
-    "open_reader",
-    "safe_open",
     "safe_open_writer",
     "write_iterable_tsv",
     "write_map_tsv",
@@ -34,35 +31,22 @@ X = TypeVar("X")
 Y = TypeVar("Y")
 
 
-@contextmanager
-def open_reader(path: str | Path, sep: str = "\t"):
-    """Open a file and get a reader for it."""
-    path = Path(path)
-    with safe_open(path, read=True) as file:
-        yield get_reader(file, sep=sep)
-
-
-def get_reader(x, sep: str = "\t"):
-    """Get a :func:`csv.reader` with PyOBO default settings."""
-    return csv.reader(x, delimiter=sep, quoting=csv.QUOTE_MINIMAL)
-
-
 def open_map_tsv(
     path: str | Path, *, use_tqdm: bool = False, has_header: bool = True
 ) -> Mapping[str, str]:
     """Load a mapping TSV file into a dictionary."""
-    with safe_open(path, read=True) as file:
+    rv = {}
+    with pystow.utils.safe_open_reader(path) as reader:
         if has_header:
-            next(file)  # throw away header
+            next(reader)  # throw away header
         if use_tqdm:
-            file = tqdm(file, desc=f"loading TSV from {path}")
-        rv = {}
-        for row in get_reader(file):
+            reader = tqdm(reader, desc=f"loading TSV from {path}")
+        for row in reader:
             if len(row) != 2:
                 logger.warning("[%s] malformed row can not be put in dict: %s", path, row)
                 continue
             rv[row[0]] = row[1]
-        return rv
+    return rv
 
 
 def open_multimap_tsv(
@@ -72,24 +56,27 @@ def open_multimap_tsv(
     has_header: bool = True,
 ) -> Mapping[str, list[str]]:
     """Load a mapping TSV file that has multiple mappings for each."""
-    return multidict(_help_multimap_tsv(path=path, use_tqdm=use_tqdm, has_header=has_header))
+    with _help_multimap_tsv(path=path, use_tqdm=use_tqdm, has_header=has_header) as file:
+        return multidict(file)
 
 
+@contextmanager
 def _help_multimap_tsv(
     path: str | Path,
     *,
     use_tqdm: bool = False,
     has_header: bool = True,
-) -> Iterable[tuple[str, str]]:
-    with safe_open(path, read=True) as file:
+) -> Generator[Iterable[tuple[str, str]], None, None]:
+    with safe_open_reader(path) as reader:
         if has_header:
             try:
-                next(file)  # throw away header
+                next(reader)  # throw away header
             except gzip.BadGzipFile as e:
                 raise ValueError(f"could not open file {path}") from e
         if use_tqdm:
-            file = tqdm(file, desc=f"loading TSV from {path}")
-        yield from get_reader(file)
+            yield tqdm(reader, desc=f"loading TSV from {path}")
+        else:
+            yield cast(Iterable[tuple[str, str]], reader)
 
 
 def multidict(pairs: Iterable[tuple[X, Y]]) -> Mapping[X, list[Y]]:
@@ -149,28 +136,3 @@ def write_iterable_tsv(
         if header is not None:
             writer.writerow(header)
         writer.writerows(it)
-
-
-@contextlib.contextmanager
-def safe_open(
-    path: str | Path, read: bool, encoding: str | None = None
-) -> Generator[TextIO, None, None]:
-    """Safely open a file for reading or writing text."""
-    path = Path(path).expanduser().resolve()
-    mode: Literal["rt", "wt"] = "rt" if read else "wt"
-    if path.suffix.endswith(".gz"):
-        with gzip.open(path, mode=mode, encoding=encoding) as file:
-            yield file
-    else:
-        with open(path, mode=mode) as file:
-            yield file
-
-
-@contextlib.contextmanager
-def safe_open_writer(f: str | Path | TextIO, *, delimiter: str = "\t"):  # type:ignore
-    """Open a CSV writer, wrapping :func:`csv.writer`."""
-    if isinstance(f, str | Path):
-        with safe_open(f, read=False) as file:
-            yield csv.writer(file, delimiter=delimiter)
-    else:
-        yield csv.writer(f, delimiter=delimiter)

pyobo/utils/misc.py

@@ -1,9 +1,17 @@
 """Miscellaneous utilities."""
 
+from __future__ import annotations
+
 import logging
+from collections.abc import Callable
 from datetime import datetime
 
+import bioversions.utils
+
+from pyobo.constants import ONTOLOGY_GETTERS, OntologyFormat
+
 __all__ = [
+    "VERSION_GETTERS",
     "cleanup_version",
 ]
 
@@ -15,8 +23,11 @@ BIZARRE_LOGGED = set()
 VERSION_REWRITES = {
     "$Date: 2009/11/15 10:54:12 $": "2009-11-15",  # for owl
     "http://www.w3.org/2006/time#2016": "2016",  # for time
+    "https://purl.org/ontology/modalia#1.0.0": "1.0.0",  # for dalia
+}
+STATIC_VERSION_REWRITES = {
+    "orth": "2",
 }
-STATIC_VERSION_REWRITES = {"orth": "2"}
 VERSION_PREFIXES = [
     "http://www.orpha.net/version",
     "https://www.orphadata.com/data/ontologies/ordo/last_version/ORDO_en_",
@@ -27,17 +38,34 @@ VERSION_PREFIXES = [
     "http://purl.dataone.org/odo/SASAP/",  # like in http://purl.dataone.org/odo/SASAP/0.3.1
     "http://purl.dataone.org/odo/SENSO/",  # like in http://purl.dataone.org/odo/SENSO/0.1.0
     "https://purl.dataone.org/odo/ADCAD/",
+    "http://identifiers.org/combine.specifications/teddy.rel-",
+    "https://nfdi.fiz-karlsruhe.de/ontology/",
+    "http://www.w3.org/ns/prov-",
+    "https://raw.githubusercontent.com/enpadasi/Ontology-for-Nutritional-Studies/releases/download/v",
+    "http://purl.jp/bio/4/ontology/iobc/",  # like http://purl.jp/bio/4/ontology/iobc/1.6.0
+    "http://w3id.org/nfdi4ing/metadata4ing/",  # like http://w3id.org/nfdi4ing/metadata4ing/1.3.1
+    "http://www.semanticweb.com/OntoRxn/",  # like http://www.semanticweb.com/OntoRxn/0.2.5
+    "https://w3id.org/lehrplan/ontology/",  # like in https://w3id.org/lehrplan/ontology/1.0.0-4
 ]
 VERSION_PREFIX_SPLITS = [
     "http://www.ebi.ac.uk/efo/releases/v",
     "http://www.ebi.ac.uk/swo/swo.owl/",
     "http://semanticscience.org/ontology/sio/v",
     "http://ontology.neuinfo.org/NIF/ttl/nif/version/",
+    "http://nmrml.org/cv/v",  # as in http://nmrml.org/cv/v1.1.0/nmrCV
+    "http://enanomapper.github.io/ontologies/releases/",  # as in http://enanomapper.github.io/ontologies/releases/10.0/enanomapper
 ]
+BAD = {
+    "http://purl.obolibrary.org/obo",
+    "http://www.bioassayontology.org/bao/bao_complete",
+}
 
 
 def cleanup_version(data_version: str, prefix: str) -> str:
     """Clean the version information."""
+    # in case a literal string that wasn't parsed properly gets put in
+    data_version = data_version.strip('"')
+
     if data_version in VERSION_REWRITES:
         return VERSION_REWRITES[data_version]
 
@@ -74,3 +102,109 @@ def cleanup_version(data_version: str, prefix: str) -> str:
         logger.debug("[%s] bizarre version: %s", prefix, data_version)
         BIZARRE_LOGGED.add((prefix, data_version))
     return data_version
+
+
+def _get_obo_version(prefix: str, url: str, *, max_lines: int = 200) -> str | None:
+    rv = bioversions.utils.get_obo_version(url, max_lines=max_lines)
+    if rv is None:
+        return None
+    return cleanup_version(rv, prefix)
+
+
+def _get_owl_version(prefix: str, url: str, *, max_lines: int = 200) -> str | None:
+    rv = bioversions.utils.get_owl_xml_version(url, max_lines=max_lines)
+    if rv is None:
+        return None
+    return cleanup_version(rv, prefix)
+
+
+def _get_obograph_json_version(prefix: str, url: str) -> str | None:
+    rv = bioversions.utils.get_obograph_json_version(url)
+    if rv is None:
+        return None
+    return cleanup_version(rv, prefix)
+
+
+#: A mapping from data type to gersion getter function
+VERSION_GETTERS: dict[OntologyFormat, Callable[[str, str], str | None]] = {
+    "obo": _get_obo_version,
+    "owl": _get_owl_version,
+    "json": _get_obograph_json_version,
+}
+
+
+def _prioritize_version(
+    data_version: str | None,
+    ontology_prefix: str,
+    version: str | None,
+    date: datetime | None,
+) -> str | None:
+    """Process version information coming from several sources and normalize them."""
+    if ontology_prefix in STATIC_VERSION_REWRITES:
+        return STATIC_VERSION_REWRITES[ontology_prefix]
+
+    if version:
+        if version in BAD:
+            logger.debug("[%s] had known bad version, returning None: ", ontology_prefix, version)
+            return None
+
+        clean_injected_version = cleanup_version(version, prefix=ontology_prefix)
+        if not data_version:
+            logger.debug(
+                "[%s] did not have a version, overriding with %s",
+                ontology_prefix,
+                clean_injected_version,
+            )
+            return clean_injected_version
+
+        clean_data_version = cleanup_version(data_version, prefix=ontology_prefix)
+        if clean_data_version != clean_injected_version:
+            # in this case, we're going to trust the one that's passed
+            # through explicitly more than the graph's content
+            logger.debug(
+                "[%s] had version %s, overriding with %s",
+                ontology_prefix,
+                data_version,
+                version,
+            )
+        return clean_injected_version
+
+    if data_version:
+        if data_version in BAD:
+            logger.debug(
+                "[%s] had known bad version, returning None: ", ontology_prefix, data_version
+            )
+            return None
+
+        clean_data_version = cleanup_version(data_version, prefix=ontology_prefix)
+        logger.debug("[%s] using version %s", ontology_prefix, clean_data_version)
+        return clean_data_version
+
+    if date is not None:
+        derived_date_version = date.strftime("%Y-%m-%d")
+        logger.debug(
+            "[%s] does not report a version. falling back to date: %s",
+            ontology_prefix,
+            derived_date_version,
+        )
+        return derived_date_version
+
+    logger.debug("[%s] does not report a version nor a date", ontology_prefix)
+    return None
+
+
+def _get_version_from_artifact(prefix: str) -> str | None:
+    # assume that all possible files that can be downloaded
+    # are in sync and have the same version
+    for ontology_format, func in ONTOLOGY_GETTERS:
+        url = func(prefix)
+        if url is None:
+            continue
+        # Try to peak into the file to get the version without fully downloading
+        version_func = VERSION_GETTERS.get(ontology_format)
+        if version_func is None:
+            continue
+        version = version_func(prefix, url)
+        if version:
+            return cleanup_version(version, prefix=prefix)
+    return None

pyobo/utils/path.py

@@ -1,6 +1,7 @@
 """Utilities for building paths."""
 
 import enum
+import json
 import logging
 from pathlib import Path
 from typing import Any, Literal
@@ -14,6 +15,7 @@ from ..constants import CACHE_SUBDIRECTORY_NAME, RAW_MODULE, RELATION_SUBDIRECTO
 __all__ = [
     "CacheArtifact",
     "ensure_df",
+    "ensure_json",
     "ensure_path",
     "get_cache_path",
     "get_relation_cache_path",
@@ -46,11 +48,13 @@ def ensure_path(
     version: VersionHint = None,
     name: str | None = None,
     force: bool = False,
-    backend: Literal["requests", "urllib"] = "urllib",
+    backend: Literal["requests", "urllib"] | None = None,
     verify: bool = True,
     **download_kwargs: Any,
 ) -> Path:
     """Download a file if it doesn't exist."""
+    if backend is None:
+        backend = "urllib"
     if verify:
         download_kwargs = {"backend": backend}
     else:
@@ -79,7 +83,7 @@ def ensure_df(
     sep: str = "\t",
     dtype=str,
     verify: bool = True,
-    backend: Literal["requests", "urllib"] = "urllib",
+    backend: Literal["requests", "urllib"] | None = None,
     **kwargs,
 ) -> pd.DataFrame:
     """Download a file and open as a dataframe."""
@@ -96,6 +100,34 @@ def ensure_df(
     return pd.read_csv(_path, sep=sep, dtype=dtype, **kwargs)
 
 
+def ensure_json(
+    prefix: str,
+    *parts: str,
+    url: str,
+    version: VersionHint = None,
+    name: str | None = None,
+    force: bool = False,
+    sep: str = "\t",
+    dtype=str,
+    verify: bool = True,
+    backend: Literal["requests", "urllib"] | None = None,
+    **kwargs,
+) -> pd.DataFrame:
+    """Download a file and open as JSON."""
+    _path = ensure_path(
+        prefix,
+        *parts,
+        url=url,
+        version=version,
+        name=name,
+        force=force,
+        verify=verify,
+        backend=backend,
+    )
+    with _path.open() as file:
+        return json.load(file)
+
+
 class CacheArtifact(enum.Enum):
     """An enumeration for."""
 

pyobo/version.py

@@ -12,7 +12,7 @@ __all__ = [
     "get_version",
 ]
 
-VERSION = "0.12.4"
+VERSION = "0.12.5"
 
 
 def get_git_hash() -> str:

{pyobo-0.12.4.dist-info → pyobo-0.12.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobo
-Version: 0.12.4
+Version: 0.12.5
 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
@@ -36,9 +36,9 @@ Requires-Dist: more-click>=0.0.2
 Requires-Dist: humanize
 Requires-Dist: tabulate
 Requires-Dist: cachier
-Requires-Dist: pystow>=0.7.0
-Requires-Dist: bioversions>=0.8.0
-Requires-Dist: bioregistry>=0.12.7
+Requires-Dist: pystow>=0.7.5
+Requires-Dist: bioversions>=0.8.101
+Requires-Dist: bioregistry>=0.12.30
 Requires-Dist: bioontologies>=0.7.2
 Requires-Dist: ssslm>=0.0.13
 Requires-Dist: zenodo-client>=0.3.6
@@ -46,7 +46,7 @@ Requires-Dist: class-resolver>=0.6.0
 Requires-Dist: psycopg2-binary
 Requires-Dist: pydantic>=2.0
 Requires-Dist: curies>=0.10.17
-Requires-Dist: curies-processing>=0.1.0
+Requires-Dist: curies-processing>=0.1.2
 Requires-Dist: python-dateutil
 Requires-Dist: networkx>=3.4
 Requires-Dist: drugbank-downloader