Use an appropriate NumPy dtype based on the number of unique sequences (

#118)

Add tests to check the `dtype` assignment.

src/genomicranges/GenomicRanges.py

@@ -220,9 +220,15 @@ def _sanitize_seqnames(self, seqnames, seqinfo):
             self._build_reverse_seqindex(seqinfo)
 
         if not isinstance(seqnames, np.ndarray):
-            seqnames = np.asarray(
-                [self._reverse_seqindex[x] for x in seqnames], dtype=np.int8
-            )
+            seqnames = np.asarray([self._reverse_seqindex[x] for x in seqnames])
+
+            num_uniq = len(np.unique(seqnames))
+            if num_uniq < 2**8:
+                seqnames = seqnames.astype(np.int8)
+            elif num_uniq < 2**16:
+                seqnames = seqnames.astype(np.int16)
+            elif num_uniq < 2**32:
+                seqnames = seqnames.astype(np.int32)
 
         return seqnames
 

tests/test_gr_init_seqnames.py

@@ -0,0 +1,37 @@
+import pytest
+from genomicranges import GenomicRanges
+from iranges import IRanges
+from biocframe import BiocFrame
+from random import random
+import pandas as pd
+import numpy as np
+
+__author__ = "jkanche"
+__copyright__ = "jkanche"
+__license__ = "MIT"
+
+
+def test_create_gr():
+    gr = GenomicRanges(
+        seqnames=["chr1"] * 10,
+        ranges=IRanges(start=range(100, 110), width=range(110, 120)),
+    )
+
+    assert gr is not None
+    assert gr._seqnames.dtype == np.int8
+
+    gr16 = GenomicRanges(
+        seqnames=[f"chr{i}" for i in range(500)],
+        ranges=IRanges(start=range(0, 500), width=range(10, 510)),
+    )
+
+    assert gr16 is not None
+    assert gr16._seqnames.dtype == np.int16
+
+    gr32 = GenomicRanges(
+        seqnames=[f"chr{i}" for i in range(2**16 + 1)],
+        ranges=IRanges(start=range(0, 2**16 + 1), width=range(10, 2**16 + 11)),
+    )
+
+    assert gr32 is not None
+    assert gr32._seqnames.dtype == np.int32