Merge pull request cogent3#1815 from GavinHuttley/develop

MAINT: SQL where clause function now handles numpy ints for start, stop

src/cogent3/core/annotation_db.py

@@ -333,7 +333,7 @@ def _matching_conditions(
         sql.append(" AND ".join(conds))
         vals = tuple(vals)
 
-    if isinstance(start, int) and isinstance(stop, int):
+    if start is not None and stop is not None:
         if allow_partial:
             # allow matches that overlap the segment
             cond = [
@@ -347,11 +347,11 @@ def _matching_conditions(
             # only matches within bounds
             cond = f"start >= {start} AND stop <= {stop}"
         sql.append(f"({cond})")
-    elif isinstance(start, int):
+    elif start is not None:
         # if query has no stop, then any feature containing start
         cond = f"(start <= {start} AND {start} < stop)"
         sql.append(f"({cond})")
-    elif isinstance(stop, int):
+    elif stop is not None:
         # if query has no start, then any feature containing stop
         cond = f"(start <= {stop} AND {stop} < stop)"
         sql.append(f"({cond})")

tests/test_core/test_features.py

@@ -1,5 +1,6 @@
 from unittest import TestCase
 
+import numpy
 import pytest
 
 from cogent3 import ASCII, DNA, get_moltype, make_aligned_seqs
@@ -135,7 +136,9 @@ def test_feature_residue():
     db.add_feature(seqid="x", biotype="exon", name="ex1", spans=[(0, 4)])
     aln_exons = list(aln.get_features(seqid="x", biotype="exon"))
     assert "biotype='exon', name='ex1', map=[0:1, 2:5]/10" in str(aln_exons)
-    assert aln_exons[0].get_slice().to_dict() == dict(x="CCCC", y="----")
+    exon = aln_exons[0]
+    exon_seq = exon.get_slice()
+    assert exon_seq.to_dict() == dict(x="CCCC", y="----")
     # Feature.as_one_span(), is applied to the exon that
     # straddles the gap in x. The result is we preserve that feature.
     exon_full_aln = aln_exons[0].as_one_span()
@@ -621,6 +624,21 @@ def test_feature_out_range():
     assert not f
 
 
+@pytest.mark.parametrize("cast", (list, numpy.array))
+def test_search_with_ints(cast):
+    """searching for features with numpy ints should work"""
+    start, stop = cast([2, 5])
+    seq = DNA.make_seq("AAAGGGGGAACCCT", name="x")
+    db = GffAnnotationDb()
+    db.add_feature(seqid="x", biotype="exon", name="E1", spans=[(3, 8)])
+    db.add_feature(seqid="x", biotype="exon", name="E2", spans=[(10, 13)])
+    seq.annotation_db = db
+    feats = list(
+        seq.get_features(biotype="exon", allow_partial=True, start=start, stop=stop)
+    )
+    assert len(feats) == 1
+
+
 def test_roundtripped_alignment_with_slices():
     """Sliced Alignment with annotations roundtrips correctly"""
     # annotations just on member sequences
@@ -633,6 +651,8 @@ def test_roundtripped_alignment_with_slices():
     aln.annotation_db = db
     # at the alignment level
     sub_aln = aln[:-3]
+    feats = list(sub_aln.get_features(biotype="exon", allow_partial=True))
+    assert len(feats) == 2
     new = deserialise_object(sub_aln.to_json())
     gf1, gf2 = list(new.get_features(biotype="exon", allow_partial=True))
     assert gf1.get_slice().to_dict() == {"x": "GGGGG", "y": "--TTT"}