feat: forward delimiter parameters from copy_umi_from_read_name

Expose read_name_delimiter and umi_delimiter on copy_umi_from_read_name
and forward them to extract_umis_from_read_name, so callers working
with non-Illumina read-name conventions can use the higher-level helper.
The remove_umi branch now uses the passed-in read_name_delimiter rather
than the hardcoded Illumina constant.

Defaults match the existing Illumina constants, so all existing callers
are unaffected.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: msto

fgpyo/platform/illumina.py

@@ -87,17 +87,32 @@ def extract_umis_from_read_name(
 
 
 def copy_umi_from_read_name(
-    rec: AlignedSegment, strict: bool = False, remove_umi: bool = False
+    rec: AlignedSegment,
+    strict: bool = False,
+    remove_umi: bool = False,
+    read_name_delimiter: str = _ILLUMINA_READ_NAME_DELIMITER,
+    umi_delimiter: str = _ILLUMINA_UMI_DELIMITER,
 ) -> bool:
     """
     Copy a UMI from an alignment's read name to its `RX` SAM tag.
 
     The UMI will not be copied to RX tag if it is invalid.
 
+    `strict`, `read_name_delimiter`, and `umi_delimiter` are forwarded to
+    [`extract_umis_from_read_name`][fgpyo.platform.illumina.extract_umis_from_read_name] — see
+    that function for their semantics. If you change the signature of either function, update
+    the other to keep them in sync.
+
     Args:
         rec: The alignment record to update.
-        strict: If `True` and UMI invalid, will throw an exception
+        strict: If `True` and UMI invalid, will throw an exception.
+            Forwarded to `extract_umis_from_read_name`.
         remove_umi: If `True`, the UMI will be removed from the read name after copying.
+        read_name_delimiter: The delimiter separating the components of the read name.
+            Forwarded to `extract_umis_from_read_name`, and also used to strip the UMI
+            segment when `remove_umi` is `True`.
+        umi_delimiter: The delimiter separating multiple UMIs.
+            Forwarded to `extract_umis_from_read_name`.
 
     Returns:
         `True` if the UMI was successfully extracted, False if otherwise.
@@ -111,13 +126,15 @@ def copy_umi_from_read_name(
     umi = extract_umis_from_read_name(
         read_name=rec.query_name,
         strict=strict,
+        read_name_delimiter=read_name_delimiter,
+        umi_delimiter=umi_delimiter,
     )
     if umi is not None:
         if rec.has_tag("RX"):
             raise ValueError(f"Record {rec.query_name} already has a populated RX tag")
         rec.set_tag(tag="RX", value=umi)
         if remove_umi:
-            last_index = rec.query_name.rfind(_ILLUMINA_READ_NAME_DELIMITER)
+            last_index = rec.query_name.rfind(read_name_delimiter)
             rec.query_name = rec.query_name[:last_index] if last_index != -1 else rec.query_name
         return True
     elif strict:

tests/fgpyo/platform/test_illumina.py

@@ -149,6 +149,25 @@ def test_copy_invalid_multi_umi_non_strict_returns_false() -> None:
     assert read.has_tag("RX") is False
 
 
+def test_copy_umi_from_read_name_custom_read_name_delimiter() -> None:
+    """Test that a custom read_name_delimiter is used for both extraction and UMI removal."""
+    builder = SamBuilder()
+    read = builder.add_single(name="abc_def_ghi_AAAA+TTTT")
+    assert (
+        copy_umi_from_read_name(read, remove_umi=True, read_name_delimiter="_") is True
+    )
+    assert read.get_tag("RX") == "AAAA-TTTT"
+    assert read.query_name == "abc_def_ghi"
+
+
+def test_copy_umi_from_read_name_custom_umi_delimiter() -> None:
+    """Test that a custom umi_delimiter is forwarded when parsing multi-UMI segments."""
+    builder = SamBuilder()
+    read = builder.add_single(name="abc:def:ghi:AAAA|TTTT")
+    assert copy_umi_from_read_name(read, umi_delimiter="|") is True
+    assert read.get_tag("RX") == "AAAA-TTTT"
+
+
 def test_populated_rx_tag_raises() -> None:
     """Test that we raise a ValueError when a record already has a populated RX tag."""
     builder = SamBuilder()