me@1942: #!/usr/bin/python
me@1942: # -*- coding: utf-8 -*-
me@1942: 
me@1942: import xml.etree.ElementTree as ET
me@1942: import os
me@1942: import csv
me@1942: import sys
me@1942: 
me@1942: # COMMAND LINE ARGUMENTS
me@1942: 
me@1942: assert len(sys.argv)<3, "comment_parser takes at most 1 command line argument\n"+\
me@1942:                         "Use: python score_parser.py [rating_folder_location]"
me@1942: 
me@1942: # XML results files location
me@1942: if len(sys.argv) == 1:
me@1942:     folder_name = "../saves"    # Looks in 'saves/' folder from 'scripts/' folder
me@1942:     print "Use: python  comment_parser.py [XML_files_location]"
me@1942:     print "Using default path: " + folder_name
me@1942: elif len(sys.argv) == 2:
me@1942:     folder_name = sys.argv[1]   # First command line argument is folder
me@1942: 
me@1942: # check if folder_name exists
me@1942: if not os.path.exists(folder_name):
me@1942:     #the file is not there
me@1942:     print "Folder '"+folder_name+"' does not exist."
me@1942:     sys.exit() # terminate script execution
me@1942: elif not os.access(os.path.dirname(folder_name), os.W_OK):
me@1942:     #the file does exist but write privileges are not given
me@1942:     print "No write privileges in folder '"+folder_name+"'."
me@1942: 
me@1942: 
me@1942: # CODE
me@1942: 
me@1942: # remember which files have been opened this time
me@1942: file_history = []
me@1942: 
me@1942: # get every XML file in folder
me@1942: for file in os.listdir(folder_name): 
me@1942:     if file.endswith(".xml"):
me@1942:         tree = ET.parse(folder_name + '/' + file)
me@1942:         root = tree.getroot()
me@1942:         
me@1942:         # get list of all page names
me@1942:         for audioholder in root.findall("./audioholder"):   # iterate over pages
me@1942:             page_name = audioholder.get('id')               # get page name
me@1942:             
me@1942:             if page_name is None: # ignore 'empty' audio_holders
me@1942:                 print "WARNING: " + file + " contains empty audio holder. (comment_parser.py)"
me@1942:                 break
me@1942: 
me@1942:             # create folder [page_name] if not yet created
me@1942:             if not os.path.exists(folder_name + "/" + page_name):
me@1942:                 os.makedirs(folder_name + "/" + page_name)
me@1942: 
me@1942:             # for page [page_name], print comments related to fragment [id]
me@1942:             for audioelement in audioholder.findall("./audioelement"):
me@1942:                 if audioelement is not None: # Check it exists
me@1942:                     audio_id = str(audioelement.get('id'))
me@1942:                     
me@1942:                     csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
me@1942: 
me@1942:                     # If file hasn't been opened yet this time, empty
me@1942:                     if csv_name not in file_history:
me@1942:                         csvfile = open(csv_name, 'w')
me@1942:                         file_history.append(csv_name) # remember this file has been written to this time around
me@1942:                     else: 
me@1942:                         # append (!) to file [page_name]/[page_name]-comments-[id].csv
me@1942:                         csvfile = open(csv_name, 'a')
me@1942:                     writer = csv.writer(csvfile, 
me@1942:                                         delimiter=',', 
me@1942:                                         dialect="excel",
me@1942:                                         quoting=csv.QUOTE_ALL)
me@1942:                     commentstr = audioelement.find("./comment/response").text
me@1942:                         
me@1942:                     if commentstr is None:
me@1942:                        commentstr = ''
me@1942:                         
me@1942:                     # anonymous comments:
me@1942:                     #writer.writerow([commentstr.encode("utf-8")]) 
me@1942:                     # comments with (file) name:
me@1942:                     writer.writerow([file[:-4]] + [commentstr.encode("utf-8")]) 
me@1942: 
me@1942:                     #TODO Replace 'new line' in comment with something else?
me@1942:                         
me@1942: # PRO TIP: Change from csv to txt by running this in bash: 
me@1942: # $ cd folder_where_csvs_are/
me@1942: # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done