giuliomoro@1088: #!/usr/bin/python
giuliomoro@1088: # -*- coding: utf-8 -*-
giuliomoro@1088: 
giuliomoro@1088: import xml.etree.ElementTree as ET
giuliomoro@1088: import os
giuliomoro@1088: import csv
giuliomoro@1088: import sys
giuliomoro@1088: 
giuliomoro@1088: # COMMAND LINE ARGUMENTS
giuliomoro@1088: 
giuliomoro@1088: assert len(sys.argv)<3, "comment_parser takes at most 1 command line argument\n"+\
giuliomoro@1088:                         "Use: python score_parser.py [rating_folder_location]"
giuliomoro@1088: 
giuliomoro@1088: # XML results files location
giuliomoro@1088: if len(sys.argv) == 1:
giuliomoro@1088:     folder_name = "../saves"    # Looks in 'saves/' folder from 'scripts/' folder
giuliomoro@1088:     print "Use: python  comment_parser.py [XML_files_location]"
giuliomoro@1088:     print "Using default path: " + folder_name
giuliomoro@1088: elif len(sys.argv) == 2:
giuliomoro@1088:     folder_name = sys.argv[1]   # First command line argument is folder
giuliomoro@1088: 
giuliomoro@1088: # check if folder_name exists
giuliomoro@1088: if not os.path.exists(folder_name):
giuliomoro@1088:     #the file is not there
giuliomoro@1088:     print "Folder '"+folder_name+"' does not exist."
giuliomoro@1088:     sys.exit() # terminate script execution
giuliomoro@1088: elif not os.access(os.path.dirname(folder_name), os.W_OK):
giuliomoro@1088:     #the file does exist but write privileges are not given
giuliomoro@1088:     print "No write privileges in folder '"+folder_name+"'."
giuliomoro@1088: 
giuliomoro@1088: 
giuliomoro@1088: # CODE
giuliomoro@1088: 
giuliomoro@1088: # remember which files have been opened this time
giuliomoro@1088: file_history = []
giuliomoro@1088: 
giuliomoro@1088: # get every XML file in folder
giuliomoro@1088: for file in os.listdir(folder_name): 
giuliomoro@1088:     if file.endswith(".xml"):
giuliomoro@1088:         tree = ET.parse(folder_name + '/' + file)
giuliomoro@1088:         root = tree.getroot()
giuliomoro@1088:         
giuliomoro@1088:         # get list of all page names
giuliomoro@1088:         for audioholder in root.findall("./page"):   # iterate over pages
giuliomoro@1088:             page_name = audioholder.get('id')               # get page name
giuliomoro@1088:             
giuliomoro@1088:             if page_name is None: # ignore 'empty' audio_holders
giuliomoro@1088:                 print "WARNING: " + file + " contains empty page. (comment_parser.py)"
giuliomoro@1088:                 break
giuliomoro@1088: 
giuliomoro@1088:             # create folder [page_name] if not yet created
giuliomoro@1088:             if not os.path.exists(folder_name + "/" + page_name):
giuliomoro@1088:                 os.makedirs(folder_name + "/" + page_name)
giuliomoro@1088: 
giuliomoro@1088:             # for page [page_name], print comments related to fragment [id]
giuliomoro@1088:             for audioelement in audioholder.findall("./audioelement"):
giuliomoro@1088:                 if audioelement is not None: # Check it exists
giuliomoro@1088:                     audio_id = str(audioelement.get('id'))
giuliomoro@1088:                     
giuliomoro@1088:                     csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
giuliomoro@1088: 
giuliomoro@1088:                     # If file hasn't been opened yet this time, empty
giuliomoro@1088:                     if csv_name not in file_history:
giuliomoro@1088:                         csvfile = open(csv_name, 'w')
giuliomoro@1088:                         file_history.append(csv_name) # remember this file has been written to this time around
giuliomoro@1088:                     else: 
giuliomoro@1088:                         # append (!) to file [page_name]/[page_name]-comments-[id].csv
giuliomoro@1088:                         csvfile = open(csv_name, 'a')
giuliomoro@1088:                     writer = csv.writer(csvfile, 
giuliomoro@1088:                                         delimiter=',', 
giuliomoro@1088:                                         dialect="excel",
giuliomoro@1088:                                         quoting=csv.QUOTE_ALL)
giuliomoro@1088:                     commentstr = audioelement.find("./comment/response").text
giuliomoro@1088:                         
giuliomoro@1088:                     if commentstr is None:
giuliomoro@1088:                        commentstr = ''
giuliomoro@1088:                         
giuliomoro@1088:                     # anonymous comments:
giuliomoro@1088:                     #writer.writerow([commentstr.encode("utf-8")]) 
giuliomoro@1088:                     # comments with (file) name:
giuliomoro@1088:                     writer.writerow([file[:-4]] + [commentstr.encode("utf-8")]) 
giuliomoro@1088: 
giuliomoro@1088:                     #TODO Replace 'new line' in comment with something else?
giuliomoro@1088:                         
giuliomoro@1088: # PRO TIP: Change from csv to txt by running this in bash: 
giuliomoro@1088: # $ cd folder_where_csvs_are/
giuliomoro@1088: # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done