b@1597: import xml.etree.ElementTree as ET
b@1597: import os
b@1597: import csv
b@1597: 
b@1597: # get every XML file in folder
b@1597: for file in os.listdir("."): # You have to put this script in folder where output XML files are.
b@1597:     if file.endswith(".xml"):
b@1597:         tree = ET.parse(file)
b@1597:         root = tree.getroot()
b@1597: 
b@1598:         # get list of all page names
b@1598:         for audioholder in root.findall("./audioholder"):   # iterate over pages
b@1598:             page_name = audioholder.get('id')               # get page name
b@1597: 
b@1598:             # create folder [page_name] if not yet created
b@1598:             if not os.path.exists(page_name):
b@1598:                 os.makedirs(page_name)
b@1597: 
b@1598:             # for page [page_name], print comments related to fragment [id]
b@1598:             for audioelement in root.findall("*/[@id='"+page_name+"']/audioelement"):
b@1597:                 audio_id = str(audioelement.get('id'))
b@1598:                 # append to file [page_name]/[page_name]-comments-[id].csv
b@1598:                 with open(page_name+'/'+page_name+'-comments-'+audio_id+'.csv', 'a') as csvfile:
b@1597:                     commentstr = root.find("*/[@id='"
b@1598:                                            + page_name
b@1597:                                            + "']/audioelement/[@id='"
b@1597:                                            + audio_id
b@1597:                                            + "']/comment/response").text
b@1597:                     writer = csv.writer(csvfile, delimiter=',')
b@1597:                     writer.writerow([commentstr.encode("utf-8")])
b@1597:                     #TODO Comma doesn't act as delimiter now!
b@1597:                     # (when adding more than just a comment per line):
b@1597:                     # writer.writerow([file + ',' + commentstr.encode("utf-8")])
b@1597: 
b@1597:                     #TODO Replace 'new line' with something else?
b@1597: 
nickjillings@1601:                     #TODO 'Append' means duplicate entries if run several times...
nickjillings@1601: