|
d@0
|
1 #!/usr/bin/python
|
|
d@0
|
2 # -*- coding: utf-8 -*-
|
|
d@0
|
3
|
|
d@0
|
4 import xml.etree.ElementTree as ET
|
|
d@0
|
5 import os
|
|
d@0
|
6 import csv
|
|
d@0
|
7
|
|
d@0
|
8 # XML results files location (modify as needed):
|
|
d@0
|
9 folder_name = "../saves" # Looks in 'saves/' folder from 'scripts/' folder
|
|
d@0
|
10
|
|
d@0
|
11 # get every XML file in folder
|
|
d@0
|
12 for file in os.listdir(folder_name):
|
|
d@0
|
13 if file.endswith(".xml"):
|
|
d@0
|
14 tree = ET.parse(folder_name + '/' + file)
|
|
d@0
|
15 root = tree.getroot()
|
|
d@0
|
16
|
|
d@0
|
17 # get list of all page names
|
|
d@0
|
18 for audioholder in root.findall("./audioholder"): # iterate over pages
|
|
d@0
|
19 page_name = audioholder.get('id') # get page name
|
|
d@0
|
20
|
|
d@0
|
21 if page_name is None: # ignore 'empty' audio_holders
|
|
d@0
|
22 print "WARNING: " + file + " contains empty audio holder. (comment_parser.py)"
|
|
d@0
|
23 break
|
|
d@0
|
24
|
|
d@0
|
25 # create folder [page_name] if not yet created
|
|
d@0
|
26 if not os.path.exists(folder_name + "/" + page_name):
|
|
d@0
|
27 os.makedirs(folder_name + "/" + page_name)
|
|
d@0
|
28
|
|
d@0
|
29 # for page [page_name], print comments related to fragment [id]
|
|
d@0
|
30 for audioelement in audioholder.findall("./audioelement"):
|
|
d@0
|
31 if audioelement is not None: # Check it exists
|
|
d@0
|
32 audio_id = str(audioelement.get('id'))
|
|
d@0
|
33
|
|
d@0
|
34 csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
|
|
d@0
|
35
|
|
d@0
|
36 # append (!) to file [page_name]/[page_name]-comments-[id].csv
|
|
d@0
|
37 with open(csv_name, 'a') as csvfile:
|
|
d@0
|
38 writer = csv.writer(csvfile,
|
|
d@0
|
39 delimiter=',',
|
|
d@0
|
40 dialect="excel",
|
|
d@0
|
41 quoting=csv.QUOTE_ALL)
|
|
d@0
|
42 commentstr = audioelement.find("./comment/response").text
|
|
d@0
|
43
|
|
d@0
|
44 if commentstr is None:
|
|
d@0
|
45 commentstr = '';
|
|
d@0
|
46
|
|
d@0
|
47 # anonymous comments:
|
|
d@0
|
48 #writer.writerow([commentstr.encode("utf-8")])
|
|
d@0
|
49 # comments with (file) name:
|
|
d@0
|
50 writer.writerow([file[:-4]] + [commentstr.encode("utf-8")])
|
|
d@0
|
51
|
|
d@0
|
52 #TODO Replace 'new line' in comment with something else?
|
|
d@0
|
53
|
|
d@0
|
54 # PRO TIP: Change from csv to txt by running this in bash:
|
|
d@0
|
55 # $ cd folder_where_csvs_are/
|
|
d@0
|
56 # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done
|
