comparison scripts/comment_parser.py @ 928:ba58cf8d0dbc

Score parsing: make csv from all XML files (one file per page ID, one column per fragment ID, one row per subject). Supports varying selections of pages and fragments across subjects. Generalised to 'page' (instead of 'song') and 'fragment' (instead of 'mix').

 for file in os.listdir("."): # You have to put this script in folder where output XML files are.
     if file.endswith(".xml"):
         tree = ET.parse(file)
         root = tree.getroot()
 
-        # get list of all songs
-        for audioholder in root.findall("./audioholder"):    # iterate over songs
-            song_name = audioholder.get('id') # get song name
+        # get list of all page names
+        for audioholder in root.findall("./audioholder"):   # iterate over pages
+            page_name = audioholder.get('id')               # get page name
 
-            # create folder [song_name] if not yet created
-            if not os.path.exists(song_name):
-                os.makedirs(song_name)
+            # create folder [page_name] if not yet created
+            if not os.path.exists(page_name):
+                os.makedirs(page_name)
 
-            # for song [song_name], print comments related to mix [id]
-            for audioelement in root.findall("*/[@id='"+song_name+"']/audioelement"):
+            # for page [page_name], print comments related to fragment [id]
+            for audioelement in root.findall("*/[@id='"+page_name+"']/audioelement"):
                 audio_id = str(audioelement.get('id'))
-                # append to file [song_name]/[song_name]-comments-[id].csv
-                with open(song_name+'/'+song_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
+                # append to file [page_name]/[page_name]-comments-[id].csv
+                with open(page_name+'/'+page_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
                     commentstr = root.find("*/[@id='"
-                                           + song_name
+                                           + page_name
                                            + "']/audioelement/[@id='"
                                            + audio_id
                                            + "']/comment/response").text
                     writer = csv.writer(csvfile, delimiter=',')
                     writer.writerow([commentstr.encode("utf-8")])