Merge pull request #6 from schorlton/main

Parse CS tag correctly

Author: alexomics

readpaf.py

@@ -4,7 +4,7 @@
 
 __all__ = ["parse_paf"]
 
-__version__ = "0.0.8"
+__version__ = "0.0.9"
 
 try:
     import pandas as pd
@@ -95,7 +95,7 @@ def _parse_tags(tags):
     """
     return {
         tag: SAM_TAG(tag, type_, SAM_TYPES.get(type_, lambda x: x)(val))
-        for tag, type_, val in (x.split(":") for x in tags)
+        for tag, type_, val in (x.split(":", 2) for x in tags)
     }
 
 

tests/test_readpaf.py

@@ -83,6 +83,13 @@ def test_tag_suffix_dataframe():
     assert set(df.columns) == set(cols + ["tp_tag"]), "Tag field not set correctly"
 
 
+def test_cs_tag_parsing():
+    _rec = "a7208cb4-133c-4ab9-96fe-db8630f4d9bb\t373\t15\t368\t+\tEf_genome\t2845392\t586028\t586405\t103\t377\t60\ttp:A:P\tcs:Z::6-ata:10+gtc:4*at:3\n"
+    PAF_IO = StringIO(_rec)
+    for rec in parse_paf(PAF_IO):
+        assert rec.tags["cs"].value == ":6-ata:10+gtc:4*at:3", "cs tag didn't match"
+
+
 def test_read_uncompressed():
     c = 0
     with open(PAF_FILE, "r") as fh: