b@2264: #!/usr/bin/python
b@2264: # -*- coding: utf-8 -*-
b@2264: 
b@2264: import xml.etree.ElementTree as ET
b@2264: import os
b@2264: import csv
b@2264: import sys
b@2264: 
b@2264: # COMMAND LINE ARGUMENTS
b@2264: 
b@2264: assert len(sys.argv)<3, "comment_parser takes at most 1 command line argument\n"+\
b@2533:                         "Use: python comment_parser.py [rating_folder_location]"
b@2264: 
b@2264: # XML results files location
b@2264: if len(sys.argv) == 1:
b@2264:     folder_name = "../saves"    # Looks in 'saves/' folder from 'scripts/' folder
b@2270:     print("Use: python comment_parser.py [XML_files_location]")
b@2270:     print("Using default path: " + folder_name)
b@2264: elif len(sys.argv) == 2:
b@2264:     folder_name = sys.argv[1]   # First command line argument is folder
b@2264: 
b@2264: # check if folder_name exists
b@2264: if not os.path.exists(folder_name):
b@2264:     #the file is not there
b@2270:     print("Folder '"+folder_name+"' does not exist.")
b@2264:     sys.exit() # terminate script execution
b@2264: elif not os.access(os.path.dirname(folder_name), os.W_OK):
b@2264:     #the file does exist but write privileges are not given
b@2270:     print("No write privileges in folder '"+folder_name+"'.")
b@2264: 
b@2264: 
b@2264: # CODE
b@2264: 
b@2264: # remember which files have been opened this time
b@2264: file_history = []
b@2264: 
b@2264: # get every XML file in folder
b@2264: for file in os.listdir(folder_name): 
b@2264:     if file.endswith(".xml"):
b@2264:         tree = ET.parse(folder_name + '/' + file)
b@2264:         root = tree.getroot()
b@2264:         
b@2264:         # get list of all page names
b@2264:         for audioholder in root.findall("./page"):   # iterate over pages
b@2264:             page_name = audioholder.get('ref')               # get page name
b@2264:             
b@2264:             if page_name is None: # ignore 'empty' audio_holders
b@2533:                 print("WARNING: " + file + " contains empty page.")
b@2264:                 break
b@2264:                 
b@2264:             if audioholder.get("state") != "complete":
b@2270:                 print("WARNING: " + file + "test page " + page_name + " is not complete, skipping.")
b@2264:             else:
b@2264:                 # create folder [page_name] if not yet created
b@2264:                 if not os.path.exists(folder_name + "/" + page_name):
b@2264:                     os.makedirs(folder_name + "/" + page_name)
b@2264: 
b@2264:                 # for page [page_name], print comments related to fragment [id]
b@2264:                 for audioelement in audioholder.findall("./audioelement"):
n@2740:                     if audioelement is not None and audioelement.get('type') != "outside-reference":
b@2264:                         audio_id = str(audioelement.get('ref'))
b@2264: 
b@2264:                         csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
b@2264: 
b@2264:                         # If file hasn't been opened yet this time, empty
b@2264:                         if csv_name not in file_history:
b@2264:                             csvfile = open(csv_name, 'w')
b@2264:                             file_history.append(csv_name) # remember this file has been written to this time around
b@2264:                         else: 
b@2264:                             # append (!) to file [page_name]/[page_name]-comments-[id].csv
b@2264:                             csvfile = open(csv_name, 'a')
b@2264:                         writer = csv.writer(csvfile, 
b@2264:                                             delimiter=',', 
b@2264:                                             dialect="excel",
b@2264:                                             quoting=csv.QUOTE_ALL)
n@2740:                         try:
n@2740:                             commentstr = audioelement.find("./comment/response").text
n@2740:                         except AttributeError:
n@2740:                             commentstr = ""
b@2441:                         valuestr = audioelement.find("./value").text
b@2264: 
b@2264:                         if commentstr is None:
b@2264:                            commentstr = ''
b@2441:                         if valuestr is None:
b@2441:                            valuestr = ''
b@2264: 
b@2264:                         # anonymous comments:
b@2957:                         # writer.writerow([commentstr.encode("utf-8")])
b@2264:                         # comments with (file) name:
b@2441:                         writer.writerow([file[:-4]] + [valuestr] + [commentstr.encode("utf-8")]) 
b@2264: 
b@2264:                     #TODO Replace 'new line' in comment with something else?
b@2264:                         
b@2264: # PRO TIP: Change from csv to txt by running this in bash: 
b@2264: # $ cd folder_where_csvs_are/
b@2264: # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done