BrechtDeMan@927: import xml.etree.ElementTree as ET
BrechtDeMan@927: import os
BrechtDeMan@927: import csv
BrechtDeMan@927: 
BrechtDeMan@927: # get every XML file in folder
BrechtDeMan@927: for file in os.listdir("."): # You have to put this script in folder where output XML files are.
BrechtDeMan@927:     if file.endswith(".xml"):
BrechtDeMan@927:         tree = ET.parse(file)
BrechtDeMan@927:         root = tree.getroot()
BrechtDeMan@927: 
BrechtDeMan@928:         # get list of all page names
BrechtDeMan@928:         for audioholder in root.findall("./audioholder"):   # iterate over pages
BrechtDeMan@928:             page_name = audioholder.get('id')               # get page name
BrechtDeMan@1030:             
BrechtDeMan@1030:             if page_name is None: # ignore 'empty' audio_holders
BrechtDeMan@1030:                 break
BrechtDeMan@927: 
BrechtDeMan@928:             # create folder [page_name] if not yet created
BrechtDeMan@928:             if not os.path.exists(page_name):
BrechtDeMan@928:                 os.makedirs(page_name)
BrechtDeMan@927: 
BrechtDeMan@928:             # for page [page_name], print comments related to fragment [id]
BrechtDeMan@928:             for audioelement in root.findall("*/[@id='"+page_name+"']/audioelement"):
BrechtDeMan@1030:                 if audioelement is not None: # Check it exists
BrechtDeMan@1030:                     audio_id = str(audioelement.get('id'))
BrechtDeMan@927: 
BrechtDeMan@1030:                     # append to file [page_name]/[page_name]-comments-[id].csv
BrechtDeMan@1030:                     with open(page_name+'/'+page_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
BrechtDeMan@1030:                         writer = csv.writer(csvfile, delimiter=',')
BrechtDeMan@1030:                         commentstr = root.find("*/[@id='"
BrechtDeMan@1030:                                                + page_name
BrechtDeMan@1030:                                                + "']/audioelement/[@id='"
BrechtDeMan@1030:                                                + audio_id
BrechtDeMan@1030:                                                + "']/comment/response").text
BrechtDeMan@1030:                         if commentstr is None:
BrechtDeMan@1030:                             writer.writerow([''])
BrechtDeMan@1030:                         else:
BrechtDeMan@1030:                             writer.writerow([commentstr.encode("utf-8")])
BrechtDeMan@1030:                         #TODO Comma doesn't act as delimiter now!
BrechtDeMan@1030:                         # (when adding more than just a comment per line):
BrechtDeMan@1030:                         # writer.writerow([file + ',' + commentstr.encode("utf-8")])
BrechtDeMan@927: 
BrechtDeMan@1030:                         #TODO Replace 'new line' with something else?
nicholas@931: 
BrechtDeMan@1030:                         #TODO 'Append' means duplicate entries if run several times...
BrechtDeMan@1030: