nicholas@872: import xml.etree.ElementTree as ET
nicholas@872: import os
nicholas@872: import csv
nicholas@872: 
nicholas@872: # get every XML file in folder
nicholas@872: for file in os.listdir("."): # You have to put this script in folder where output XML files are.
nicholas@872:     if file.endswith(".xml"):
nicholas@872:         tree = ET.parse(file)
nicholas@872:         root = tree.getroot()
nicholas@872: 
nicholas@872:         # get list of all page names
nicholas@872:         for audioholder in root.findall("./audioholder"):   # iterate over pages
nicholas@872:             page_name = audioholder.get('id')               # get page name
nicholas@872:             
nicholas@872:             if page_name is None: # ignore 'empty' audio_holders
nicholas@872:                 break
nicholas@872: 
nicholas@872:             # create folder [page_name] if not yet created
nicholas@872:             if not os.path.exists(page_name):
nicholas@872:                 os.makedirs(page_name)
nicholas@872: 
nicholas@872:             # for page [page_name], print comments related to fragment [id]
nicholas@872:             for audioelement in root.findall("*/[@id='"+page_name+"']/audioelement"):
nicholas@872:                 if audioelement is not None: # Check it exists
nicholas@872:                     audio_id = str(audioelement.get('id'))
nicholas@872: 
nicholas@872:                     # append to file [page_name]/[page_name]-comments-[id].csv
nicholas@872:                     with open(page_name+'/'+page_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
nicholas@872:                         writer = csv.writer(csvfile, delimiter=',')
nicholas@872:                         commentstr = root.find("*/[@id='"
nicholas@872:                                                + page_name
nicholas@872:                                                + "']/audioelement/[@id='"
nicholas@872:                                                + audio_id
nicholas@872:                                                + "']/comment/response").text
nicholas@872:                         if commentstr is None:
nicholas@872:                             writer.writerow([''])
nicholas@872:                         else:
nicholas@872:                             writer.writerow([commentstr.encode("utf-8")])
nicholas@872:                         #TODO Comma doesn't act as delimiter now!
nicholas@872:                         # (when adding more than just a comment per line):
nicholas@872:                         # writer.writerow([file + ',' + commentstr.encode("utf-8")])
nicholas@872: 
nicholas@872:                         #TODO Replace 'new line' with something else?
nicholas@872: 
nicholas@872:                         #TODO 'Append' means duplicate entries if run several times...
nicholas@872: