nickjillings@1528: #!/usr/bin/python
nickjillings@1528: # -*- coding: utf-8 -*-
nickjillings@1528: 
nickjillings@1528: import xml.etree.ElementTree as ET
nickjillings@1528: import os
nickjillings@1528: import csv
nickjillings@1528: 
nickjillings@1528: # XML results files location (modify as needed):
nickjillings@1528: folder_name = "../saves"    # Looks in 'saves/' folder from 'scripts/' folder
nickjillings@1528: 
nickjillings@1528: # get every XML file in folder
nickjillings@1528: for file in os.listdir(folder_name): 
nickjillings@1528:     if file.endswith(".xml"):
nickjillings@1528:         tree = ET.parse(folder_name + '/' + file)
nickjillings@1528:         root = tree.getroot()
nickjillings@1528:         
nickjillings@1528:         # get list of all page names
nickjillings@1528:         for audioholder in root.findall("./audioholder"):   # iterate over pages
nickjillings@1528:             page_name = audioholder.get('id')               # get page name
nickjillings@1528:             
nickjillings@1528:             if page_name is None: # ignore 'empty' audio_holders
nickjillings@1528:                 print "WARNING: " + file + " contains empty audio holder. (comment_parser.py)"
nickjillings@1528:                 break
nickjillings@1528: 
nickjillings@1528:             # create folder [page_name] if not yet created
nickjillings@1528:             if not os.path.exists(folder_name + "/" + page_name):
nickjillings@1528:                 os.makedirs(folder_name + "/" + page_name)
nickjillings@1528: 
nickjillings@1528:             # for page [page_name], print comments related to fragment [id]
nickjillings@1528:             for audioelement in audioholder.findall("./audioelement"):
nickjillings@1528:                 if audioelement is not None: # Check it exists
nickjillings@1528:                     audio_id = str(audioelement.get('id'))
nickjillings@1528:                     
nickjillings@1528:                     csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
nickjillings@1528: 
nickjillings@1528:                     # append (!) to file [page_name]/[page_name]-comments-[id].csv
nickjillings@1528:                     with open(csv_name, 'a') as csvfile:
nickjillings@1528:                         writer = csv.writer(csvfile, 
nickjillings@1528:                                             delimiter=',', 
nickjillings@1528:                                             dialect="excel",
nickjillings@1528:                                             quoting=csv.QUOTE_ALL)
nickjillings@1528:                         commentstr = audioelement.find("./comment/response").text
nickjillings@1528:                         
nickjillings@1528:                         if commentstr is None:
nickjillings@1528:                            commentstr = '';
nickjillings@1528:                             
nickjillings@1528:                         # anonymous comments:
nickjillings@1528:                         #writer.writerow([commentstr.encode("utf-8")]) 
nickjillings@1528:                         # comments with (file) name:
nickjillings@1528:                         writer.writerow([file[:-4]] + [commentstr.encode("utf-8")]) 
nickjillings@1528: 
nickjillings@1528:                         #TODO Replace 'new line' in comment with something else?
nickjillings@1528:                         
nickjillings@1528: # PRO TIP: Change from csv to txt by running this in bash: 
nickjillings@1528: # $ cd folder_where_csvs_are/
nickjillings@1528: # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done