b@146: import xml.etree.ElementTree as ET
b@146: import os
b@146: import csv
b@146: 
b@146: # get every XML file in folder
b@146: for file in os.listdir("."): # You have to put this script in folder where output XML files are.
b@146:     if file.endswith(".xml"):
b@146:         tree = ET.parse(file)
b@146:         root = tree.getroot()
b@146: 
b@147:         # get list of all page names
b@147:         for audioholder in root.findall("./audioholder"):   # iterate over pages
b@147:             page_name = audioholder.get('id')               # get page name
b@206:             
b@206:             if page_name is None: # ignore 'empty' audio_holders
b@206:                 break
b@146: 
b@147:             # create folder [page_name] if not yet created
b@147:             if not os.path.exists(page_name):
b@147:                 os.makedirs(page_name)
b@146: 
b@147:             # for page [page_name], print comments related to fragment [id]
b@147:             for audioelement in root.findall("*/[@id='"+page_name+"']/audioelement"):
b@206:                 if audioelement is not None: # Check it exists
b@206:                     audio_id = str(audioelement.get('id'))
b@146: 
b@206:                     # append to file [page_name]/[page_name]-comments-[id].csv
b@206:                     with open(page_name+'/'+page_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
b@206:                         writer = csv.writer(csvfile, delimiter=',')
b@206:                         commentstr = root.find("*/[@id='"
b@206:                                                + page_name
b@206:                                                + "']/audioelement/[@id='"
b@206:                                                + audio_id
b@206:                                                + "']/comment/response").text
b@206:                         if commentstr is None:
b@206:                             writer.writerow([''])
b@206:                         else:
b@206:                             writer.writerow([commentstr.encode("utf-8")])
b@206:                         #TODO Comma doesn't act as delimiter now!
b@206:                         # (when adding more than just a comment per line):
b@206:                         # writer.writerow([file + ',' + commentstr.encode("utf-8")])
b@146: 
b@206:                         #TODO Replace 'new line' with something else?
nicholas@150: 
b@206:                         #TODO 'Append' means duplicate entries if run several times...
b@206: