first commit

pyedf.py

@@ -0,0 +1,36 @@
+import re
+
+def read_edf_file(filename):
+    f = open(filename, 'r')
+    data = f.read()
+    f.close()
+    header = {}
+    header['version'] = data[0:7].strip() # 8
+    header['patient_id'] = data[7:88].strip() # 80
+    header['rec_id'] = data[88:168].strip() # 80
+    header['startdate'] = data[168:176].strip() # 8
+    header['starttime'] = data[176:184].strip() # 8
+    header['header_bytes'] = int(data[184:192].strip()) # 8
+    header['num_items'] = int(data[236:244].strip())
+    header['data_duration'] = float(data[244:252].strip())
+    header['num_signals'] = int(data[252:256].strip())
+    # more data 256 chars down. in header, but ignoring for now
+    print header
+    records = []
+    records_a = records.append
+    rec_pos = 512
+    rec_size = 36
+    # skip first rec
+    rec_pos += rec_size
+    print rec_pos
+    for i in range(header['num_items']-1):
+        record = {}
+        record_split = data[rec_pos:rec_pos+rec_size].split('\x14')
+        matches = re.findall(r'([\d]+)', record_split[2])
+        record['type'] = record_split[3].strip()
+        record['time'] = {'start': int(matches[0]), 'durration': int(matches[1])}
+        records_a(record)
+        rec_pos += rec_size
+    return {'header': header, 'records': records}
+data = read_edf_file('20100912_230640_EVE.edf')
+print data