FIX: Gracefully handle bad XML files in EGI reader (#13145)

Author: scott-huberty

doc/changes/devel/13145.bugfix.rst

@@ -0,0 +1 @@
+Improved the handling of problematic MFF files in `mne.io.read_raw_egi` by `Scott Huberty`_.

mne/io/egi/events.py

@@ -48,7 +48,9 @@ def _read_mff_events(filename, sfreq):
     orig = {}
     for xml_file in glob(join(filename, "*.xml")):
         xml_type = splitext(basename(xml_file))[0]
-        orig[xml_type] = _parse_xml(xml_file)
+        et = _parse_xml(xml_file)
+        if et is not None:
+            orig[xml_type] = et
     xml_files = orig.keys()
     xml_events = [x for x in xml_files if x[:7] == "Events_"]
     for item in orig["info"]:
@@ -80,10 +82,14 @@ def _read_mff_events(filename, sfreq):
     return events_tims, code
 
 
-def _parse_xml(xml_file):
+def _parse_xml(xml_file: str) -> list[dict[str, str]] | None:
     """Parse XML file."""
     defusedxml = _soft_import("defusedxml", "reading EGI MFF data")
-    xml = defusedxml.ElementTree.parse(xml_file)
+    try:
+        xml = defusedxml.ElementTree.parse(xml_file)
+    except defusedxml.ElementTree.ParseError as e:
+        warn(f"Could not parse the XML file {xml_file}: {e}")
+        return
     root = xml.getroot()
     return _xml2list(root)
 

mne/io/egi/tests/test_egi.py

@@ -586,3 +586,16 @@ def test_set_standard_montage_mff(fname, standard_montage):
     # Check that the reference remained
     for pick in picks:
         assert_allclose(raw.info["chs"][pick]["loc"][3:6], ref_loc)
+
+
+@requires_testing_data
+def test_egi_mff_bad_xml(tmp_path):
+    """Test that corrupt XML files are gracefully handled."""
+    pytest.importorskip("defusedxml")
+    mff_fname = shutil.copytree(egi_mff_fname, tmp_path / "test_egi_bad_xml.mff")
+    bad_xml = mff_fname / "bad.xml"
+    bad_xml.write_text("<foo>", encoding="utf-8")
+    with pytest.warns(RuntimeWarning, match="Could not parse the XML"):
+        raw = read_raw_egi(mff_fname)
+    # little check that the bad XML doesn't affect the parsing of other xml files
+    assert "DIN1" in raw.annotations.description