Simplify typing (#213)

Author: cthoyt

pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["uv_build>=0.6.6,<0.7"]
+requires = ["uv_build>=0.6.6,<1.0"]
 build-backend = "uv_build"
 
 [project]
@@ -60,8 +60,8 @@ dependencies = [
     "click",
     "pyyaml",
     "tqdm",
-    "pystow>=0.2.7",
-    "bioregistry>=0.10.43",
+    "pystow>=0.7.11",
+    "bioregistry>=0.12.42",
 ]
 
 [project.optional-dependencies]
@@ -91,10 +91,10 @@ web = [
     "flask-wtf",
 ]
 gilda = [
-    "pyobo[gilda]>=0.12.0",
+    "pyobo[gilda]>=0.12.8",
 ]
 gilda-slim = [
-    "pyobo[gilda-slim]>=0.12.0",
+    "pyobo[gilda-slim]>=0.12.8",
 ]
 ndex = [
     "ndex2",

src/biomappings/cli.py

@@ -60,12 +60,12 @@ def curate(
         resolver_base: str | None,
     ) -> None:
         """Run a target curation web app."""
-        from curies import NamableReference
+        from curies import Reference
 
         from .resources import _load_table
         from .wsgi import get_app
 
-        target_references: list[NamableReference] = []
+        target_references: list[Reference] = []
         for mapping in _load_table(path, standardize=True):
             target_references.append(mapping.subject)
             target_references.append(mapping.object)

src/biomappings/graph.py

@@ -87,7 +87,7 @@ def _graph_from_mappings(
             mapping.subject,
             mapping.object,
             relation=mapping.predicate.curie,
-            provenance=mapping.author.name if mapping.author else None,
+            provenance=mapping.author.curie if mapping.author else None,
             type=mapping.mapping_justification.curie,
             strata=strata,
         )

src/biomappings/resources/__init__.py

@@ -56,6 +56,18 @@
 
 logger = logging.getLogger(__name__)
 
+COLUMNS = [
+    "subject_id",
+    "subject_label",
+    "predicate_id",
+    "object_id",
+    "object_label",
+    "mapping_justification",
+    "author_id",
+    "mapping_tool",
+    "predicate_modifier",
+]
+
 
 class _CuratedTuple(NamedTuple):
     """A tuple for writing manual curations to SSSOM TSV."""
@@ -283,8 +295,8 @@ def append_true_mapping_tuples(mappings: Iterable[SemanticMapping]) -> None:
 
 
 def write_true_mappings(mappings: Iterable[SemanticMapping], *, path: Path | None = None) -> None:
-    """Write mappigns to the true mappings file."""
-    _write_helper(mappings, path=path or POSITIVES_SSSOM_PATH, mode="w", t="curated")
+    """Write mappings to the true mappings file."""
+    _write_helper(mappings, path or POSITIVES_SSSOM_PATH, mode="w", t="curated")
 
 
 def lint_true_mappings(*, path: Path | None = None, standardize: bool) -> None:
@@ -316,7 +328,7 @@ def append_false_mappings(
     """Append new lines to the false mappings table."""
     if path is None:
         path = NEGATIVES_SSSOM_PATH
-    _write_helper(mappings=mappings, path=path, mode="a", t="curated")
+    _write_helper(mappings, path=path, mode="a", t="curated")
     if sort:
         lint_false_mappings(path=path, standardize=standardize)
 
@@ -412,7 +424,7 @@ def append_predictions(
 
     if path is None:
         path = PREDICTIONS_SSSOM_PATH
-    _write_helper(mappings, path, mode="a", t="predicted")
+    _write_helper(mappings, path=path, mode="a", t="predicted")
     if sort:
         lint_predictions(path=path, standardize=standardize)
 
@@ -441,8 +453,7 @@ def lint_predictions(
             additional_curated_mappings or [],
         ),
     )
-    mappings = _remove_redundant(mappings)
-    mappings = sorted(mappings)
+    mappings = _clean_mappings(mappings)
     write_predictions(mappings, path=path)
 
 
@@ -454,6 +465,11 @@ def remove_mappings(
     return (mapping for mapping in mappings if get_canonical_tuple(mapping) not in skip_tuples)
 
 
+def _clean_mappings(mappings: Iterable[SemanticMapping]):
+    m = sorted(mappings)
+    return _remove_redundant(m)
+
+
 def _remove_redundant(mappings: Iterable[SemanticMapping]) -> Iterable[SemanticMapping]:
     dd = defaultdict(list)
     for mapping in mappings:

src/biomappings/resources/export_sssom.py

@@ -70,9 +70,9 @@ def get_sssom_df(*, use_tqdm: bool = False) -> SSSOMReturnTuple:
     prefixes: set[str] = {"semapv"}
 
     # NEW WAY: load all DFs, concat them, reorder columns
-    a = pd.read_csv(POSITIVES_SSSOM_PATH, sep="\t")
-    b = pd.read_csv(NEGATIVES_SSSOM_PATH, sep="\t")
-    c = pd.read_csv(PREDICTIONS_SSSOM_PATH, sep="\t")
+    a = pd.read_csv(POSITIVES_SSSOM_PATH, sep="\t", comment="#")
+    b = pd.read_csv(NEGATIVES_SSSOM_PATH, sep="\t", comment="#")
+    c = pd.read_csv(PREDICTIONS_SSSOM_PATH, sep="\t", comment="#")
     df = pd.concat([a, b, c])
     df = df[columns]
 

src/biomappings/testing.py

@@ -12,13 +12,11 @@
 
 import bioregistry
 from bioregistry import NormalizedNamableReference
-from curies import NamableReference
+from curies import Reference
 
 from biomappings.resources import (
     CURATORS_PATH,
     SemanticMapping,
-    _CuratedTuple,
-    _PredictedTuple,
     load_mappings,
     load_predictions,
 )
@@ -38,7 +36,6 @@
 
 SEMAPV_ID_TO_NAME = get_semapv_id_to_name()
 
-TPL = TypeVar("TPL", _CuratedTuple, _PredictedTuple)
 X = TypeVar("X")
 Y = TypeVar("Y")
 
@@ -98,7 +95,7 @@ def test_valid_curies(self) -> None:
             if mapping.author is not None:
                 self.assert_valid(label, line, mapping.author)
 
-    def assert_valid(self, label: str, line: int, reference: NamableReference) -> None:
+    def assert_valid(self, label: str, line: int, reference: Reference) -> None:
         """Assert a reference is valid and normalized to the Bioregistry."""
         norm_prefix = bioregistry.normalize_prefix(reference.prefix)
         self.assertIsNotNone(
@@ -167,9 +164,7 @@ def test_cross_redundancy(self) -> None:
 
     def assert_no_internal_redundancies(self, mappings: list[SemanticMapping]) -> None:
         """Assert that the list of mappings doesn't have any redundancies."""
-        counter: defaultdict[tuple[NamableReference, NamableReference], list[int]] = defaultdict(
-            list
-        )
+        counter: defaultdict[tuple[Reference, Reference], list[int]] = defaultdict(list)
         for line_number, mapping in enumerate(mappings, start=1):
             counter[mapping.subject, mapping.object].append(line_number)
         redundant = _extract_redundant(counter)

src/biomappings/wsgi.py

@@ -15,7 +15,7 @@
 import pydantic
 import werkzeug
 from bioregistry import NormalizedNamableReference
-from curies import NamableReference
+from curies import NamableReference, Reference
 from flask import current_app
 from flask_wtf import FlaskForm
 from pydantic import BaseModel
@@ -104,7 +104,7 @@ def url_for_state(endpoint, state: State, **kwargs: Any) -> str:
 
 def get_app(
     *,
-    target_references: Iterable[NamableReference] | None = None,
+    target_references: Iterable[Reference] | None = None,
     predictions_path: Path | None = None,
     positives_path: Path | None = None,
     negatives_path: Path | None = None,
@@ -153,7 +153,7 @@ class Controller:
     def __init__(
         self,
         *,
-        target_references: Iterable[NamableReference] | None = None,
+        target_references: Iterable[Reference] | None = None,
         predictions_path: Path | None = None,
         positives_path: Path | None = None,
         negatives_path: Path | None = None,

tests/test_wsgi.py

@@ -7,7 +7,7 @@
 from bioregistry import NormalizedNamableReference as Reference
 
 from biomappings import SemanticMapping
-from biomappings.resources import _write_helper, write_predictions
+from biomappings.resources import COLUMNS, write_predictions
 from biomappings.wsgi import Controller, State, get_app
 
 TEST_USER = Reference(prefix="orcid", identifier="0000-0000-0000-0000", name="Max Mustermann")
@@ -69,9 +69,9 @@ def setUp(self) -> None:
         unsure_path = directory.joinpath("unsure.tsv")
 
         write_predictions(predictions, path=predictions_path)
-        _write_helper([], path=positives_path, mode="w", t="curated")
-        _write_helper([], path=negatives_path, mode="w", t="curated")
-        _write_helper([], path=unsure_path, mode="w", t="curated")
+        for path in [positives_path, negatives_path, unsure_path]:
+            with path.open("w") as file:
+                print(*COLUMNS, sep="\t", file=file)
 
         self.controller = Controller(
             predictions_path=predictions_path,