From 138fcdb977f2b6c8faab2fd0f30b59b8b93343b9 Mon Sep 17 00:00:00 2001
From: Jeff Stout <stoutjd@nih.gov>
Date: Tue, 2 Jul 2024 13:25:31 -0400
Subject: [PATCH] Create a standardized ID input for BIDS

---
 nih2mne/utilities/standardize_eventID_list.py | 69 +++++++++++++++++++
 pyproject.toml                                |  1 +
 2 files changed, 70 insertions(+)
 create mode 100755 nih2mne/utilities/standardize_eventID_list.py

diff --git a/nih2mne/utilities/standardize_eventID_list.py b/nih2mne/utilities/standardize_eventID_list.py
new file mode 100755
index 0000000..9678f21
--- /dev/null
+++ b/nih2mne/utilities/standardize_eventID_list.py
@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Created on Tue Jul  2 11:06:33 2024
+
+@author: jstout
+"""
+
+import pandas as pd
+import mne
+import os, os.path as op
+import glob
+
+
+def write_standardized_event_list(meg_dir, task=None):
+    '''
+    Read out the annotations from all runs of a task a make a standardized
+    list. This function will write out a task_TASK_standardized_EventId_list.csv 
+    file for each task set in the folder.
+
+    Parameters
+    ----------
+    meg_dir : str 
+        CTF meg dataset directory.
+    task : str
+        Specify task or leave empty and it will iterate over all tasks
+
+    Returns
+    -------
+    None.
+
+    '''
+    if task==None:
+        tmp_ = glob.glob(op.join(meg_dir, '*.ds'))
+        task_list = [op.basename(i).split('_')[1] for i in tmp_]
+    else:
+        task_list = [task]
+        
+    for task in task_list:
+        print(f'Generating unique entry ID list for {task}')
+        task_dsets = glob.glob(op.join(meg_dir, f'*_{task}_*.ds'))
+        sorted(task_dsets)
+        
+        id_list = []
+        for dset in task_dsets:
+            raw = mne.io.read_raw_ctf(dset, system_clock='ignore')
+            id_list.extend(list(raw.annotations.description)[:])
+        
+        id_list = sorted(set(id_list)) #Sort unique entries
+        id_list = pd.DataFrame(zip(id_list,range(1, 1+len(id_list))), 
+                               columns=['ID_names', 'ID_vals'])    
+        id_list.to_csv(op.join(meg_dir, f'task_{task}_standardized_EventId_list.csv')) 
+        
+def main_parser():
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-megdir', 
+                        help='CTF acquisition folder - generally a date')
+    parser.add_argument('-task',
+                        help='''Specify a task to write the standardized IDs.
+                        Leave blank to process all tasks''',
+                        default=None)
+    args = parser.parse_args()
+    write_standardized_event_list(args.megdir, task=args.task)
+                        
+                        
+    
+    
+    
diff --git a/pyproject.toml b/pyproject.toml
index ae22362..a36b9f8 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,6 +37,7 @@ testing = ['datalad','pytest','pygit2']
 "fix_dsname.py"='nih2mne.utilities.fix_dsname:main'
 "megcore_prep_mri_bids.py"='nih2mne.megcore_prep_mri_bids:main'
 "qa_fids.py"='nih2mne.utilities.qa_fids:main'
+"standardize_eventID_list.py"='nih2mne.utilities.standardize_eventID_list:main_parser'
 #'nih2mne/eyetracking_preprocessing.py',
 #'cmdline/make_bids_fs_swarm.sh',