fixed string subID bug (#1291)

* fixed string subID bug

* added alphanumeric subject to participants testing

* added name to authors list

* add author name to whatsnew

* make test more lean

* add whatsnew entry

* add to CITATION.cff

---------

Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org>

CITATION.cff

@@ -189,8 +189,11 @@ authors:
       orcid: 'https://orcid.org/0000-0001-6364-7272'
     - given-names: Thomas
       family-names: Hartmann
-      affiliation: 'Paris-Lodron-University Salzburg, Centre for Cogntitive Neuroscience, Department of Psychology, Salzburg, Austria'
+      affiliation: 'Paris-Lodron-University Salzburg, Centre for Cognitive Neuroscience, Department of Psychology, Salzburg, Austria'
       orcid: 'https://orcid.org/0000-0002-8298-8125'
+    - given-names: Aaron
+      family-names: Earle-Richardson
+      affiliation: 'Duke University School of Medicine, Department of Neurology'
     - given-names: Alexandre
       family-names: Gramfort
       affiliation: 'Université Paris-Saclay, Inria, CEA, Palaiseau, France'

doc/authors.rst

@@ -1,3 +1,4 @@
+.. _Aaron Earle-Richardson: https://github.com/Aaronearlerichardson
 .. _Mainak Jas: https://jasmainak.github.io/
 .. _Teon Brooks: https://teonbrooks.com
 .. _Chris Holdgraf: https://bids.berkeley.edu/people/chris-holdgraf

doc/whats_new.rst

@@ -17,8 +17,9 @@ Version 0.16 (unreleased)
 
 The following authors contributed for the first time. Thank you so much! 🤩
 
-* `Kaare Mikkelsen`_
+* `Aaron Earle-Richardson`_
 * `Amaia Benitez`_
+* `Kaare Mikkelsen`_
 * `Thomas Hartmann`_
 
 The following authors had contributed before. Thank you for sticking around! 🤘
@@ -53,6 +54,7 @@ Detailed list of changes
 ^^^^^^^^^^^^
 
 - When anonymizing the date of a recording, MNE-BIDS will no longer error during `~mne_bids.write_raw_bids` if passing a `~mne.io.Raw` instance to ``empty_room``, by `Daniel McCloy`_ (:gh:`1270`)
+- Dealing with alphanumeric ``sub`` entity labels is now fixed for :func:`~mne_bids.write_raw_bids`, by `Aaron Earle-Richardson`_ (:gh:`1291`)
 
 ⚕️ Code health
 ^^^^^^^^^^^^^^

mne_bids/tests/test_write.py

@@ -254,15 +254,16 @@ def test_write_participants(_bids_validate, tmp_path):
     _to_tsv(data, participants_tsv)
 
     # write in now another subject
-    bids_path.update(subject="03")
+    # (use alphanumeric label to test for GH-1291)
+    bids_path.update(subject="D03")
     write_raw_bids(raw, bids_path, verbose=False)
     data = _from_tsv(participants_tsv)
 
     # hand should have been written properly with now 'n/a' for sub-01 and
-    # sub-03, but 'L' for sub-03
+    # sub-02, but 'L' for sub-D03
     assert data["hand"][data["participant_id"].index("sub-01")] == "n/a"
     assert data["hand"][data["participant_id"].index("sub-02")] == "n/a"
-    assert data["hand"][data["participant_id"].index("sub-03")] == "L"
+    assert data["hand"][data["participant_id"].index("sub-D03")] == "L"
 
     # check to make sure participant data is overwritten, but keeps the fields
     # if there are extra fields that were user defined

mne_bids/tsv_handler.py

@@ -82,8 +82,8 @@ def _contains_row(data, row_data):
         # Cast row_value to the same dtype as data_value to avoid a NumPy
         # FutureWarning, see
         # https://github.com/mne-tools/mne-bids/pull/372
-        row_value = np.array(row_value, dtype=data_value.dtype)
-
+        if data_value.size > 0:
+            row_value = np.array(row_value, dtype=data_value.dtype)
         column_mask = np.isin(data_value, row_value)
         mask = column_mask if mask is None else (mask & column_mask)
     return np.any(mask)