annotate scripts/comment_parser.py @ 883:cd20f076f6a3

Readme: reference to issue tracker; Scripts: comment_parser and score_parser read and write in '/saves' from 'scripts/', condensing XML queries (references from children instead of from root)
author Brecht De Man <BrechtDeMan@users.noreply.github.com>
date Mon, 29 Jun 2015 13:15:00 +0100
parents 24d0d3111c00
children 7a8fcf04aad3
rev   line source
BrechtDeMan@878 1 #!/usr/bin/python
BrechtDeMan@880 2 # -*- coding: utf-8 -*-
BrechtDeMan@878 3
nicholas@872 4 import xml.etree.ElementTree as ET
nicholas@872 5 import os
nicholas@872 6 import csv
nicholas@872 7
BrechtDeMan@883 8 # XML results files location (modify as needed):
BrechtDeMan@883 9 folder_name = "../saves" # Looks in 'saves/' folder from 'scripts/' folder
BrechtDeMan@883 10
nicholas@872 11 # get every XML file in folder
BrechtDeMan@883 12 for file in os.listdir(folder_name):
nicholas@872 13 if file.endswith(".xml"):
BrechtDeMan@883 14 tree = ET.parse(folder_name + '/' + file)
nicholas@872 15 root = tree.getroot()
nicholas@872 16
nicholas@872 17 # get list of all page names
nicholas@872 18 for audioholder in root.findall("./audioholder"): # iterate over pages
nicholas@872 19 page_name = audioholder.get('id') # get page name
nicholas@872 20
nicholas@872 21 if page_name is None: # ignore 'empty' audio_holders
nicholas@872 22 break
nicholas@872 23
nicholas@872 24 # create folder [page_name] if not yet created
BrechtDeMan@883 25 if not os.path.exists(folder_name + "/" + page_name):
BrechtDeMan@883 26 os.makedirs(folder_name + "/" + page_name)
nicholas@872 27
nicholas@872 28 # for page [page_name], print comments related to fragment [id]
BrechtDeMan@883 29 for audioelement in audioholder.findall("./audioelement"):
nicholas@872 30 if audioelement is not None: # Check it exists
nicholas@872 31 audio_id = str(audioelement.get('id'))
BrechtDeMan@878 32
BrechtDeMan@883 33 csv_name = folder_name +'/' + page_name+'/'+page_name+'-comments-'+audio_id+'.csv'
nicholas@872 34
BrechtDeMan@878 35 # append (!) to file [page_name]/[page_name]-comments-[id].csv
BrechtDeMan@878 36 with open(csv_name, 'a') as csvfile:
BrechtDeMan@878 37 writer = csv.writer(csvfile,
BrechtDeMan@878 38 delimiter=',',
BrechtDeMan@878 39 dialect="excel",
BrechtDeMan@878 40 quoting=csv.QUOTE_ALL)
BrechtDeMan@883 41 commentstr = audioelement.find("./comment/response").text
BrechtDeMan@883 42
nicholas@872 43 if commentstr is None:
BrechtDeMan@883 44 commentstr = '';
BrechtDeMan@883 45
BrechtDeMan@883 46 # anonymous comments:
BrechtDeMan@883 47 #writer.writerow([commentstr.encode("utf-8")])
BrechtDeMan@883 48 # comments with (file) name:
BrechtDeMan@883 49 writer.writerow([file[:-4]] + [commentstr.encode("utf-8")])
nicholas@872 50
BrechtDeMan@878 51 #TODO Replace 'new line' in comment with something else?
BrechtDeMan@878 52
BrechtDeMan@878 53 # PRO TIP: Change from csv to txt by running this in bash:
BrechtDeMan@878 54 # $ cd folder_where_csvs_are/
BrechtDeMan@878 55 # $ for i in *.csv; do mv "$i" "${i/.csv}".txt; done