BrechtDeMan@1051: #!/usr/bin/python
BrechtDeMan@1051: # -*- coding: utf-8 -*-
BrechtDeMan@1051: 
BrechtDeMan@1051: import xml.etree.ElementTree as ET
BrechtDeMan@1051: import os
BrechtDeMan@1051: 
BrechtDeMan@1051: # XML results files location (modify as needed):
BrechtDeMan@1051: folder_name = "../saves"    # Looks in 'saves/' folder from 'scripts/' folder
BrechtDeMan@1051: 
BrechtDeMan@1051: # Turn number of seconds (int) to '[minutes] min [seconds] s' (string)
BrechtDeMan@1051: def seconds2timestr(time_in_seconds):
BrechtDeMan@1051:     time_in_minutes = int(time_in_seconds/60)
BrechtDeMan@1051:     remaining_seconds = int(time_in_seconds%60)
BrechtDeMan@1051:     return str(time_in_minutes) + " min " + str(remaining_seconds) + " s"
BrechtDeMan@1051: 
BrechtDeMan@1051: # stats initialisation
BrechtDeMan@1051: number_of_XML_files = 0
BrechtDeMan@1051: number_of_pages = 0
BrechtDeMan@1051: time_per_page_accum = 0
BrechtDeMan@1051: 
BrechtDeMan@1051: # arrays initialisation
BrechtDeMan@1051: page_names = []
BrechtDeMan@1051: page_count = []
BrechtDeMan@1051: duration_page = []      # duration of experiment in function of page
BrechtDeMan@1051: duration_subject = []
BrechtDeMan@1051: duration_order = []
BrechtDeMan@1051: 
BrechtDeMan@1051: # get every XML file in folder
BrechtDeMan@1051: files_list = os.listdir(folder_name)
BrechtDeMan@1051: for file in files_list: # iterate over all files in files_list
BrechtDeMan@1051:     if file.endswith(".xml"): # check if XML file
BrechtDeMan@1051:         number_of_XML_files += 1
BrechtDeMan@1051:         tree = ET.parse(folder_name + '/' + file)
BrechtDeMan@1051:         root = tree.getroot()
BrechtDeMan@1051:         
BrechtDeMan@1051:         print file # print file name (subject name)
BrechtDeMan@1051:         
BrechtDeMan@1051:         # reset for new subject
BrechtDeMan@1051:         total_duration = 0
BrechtDeMan@1051:         page_number = 0
BrechtDeMan@1051:         
BrechtDeMan@1051:         # get list of all page names
BrechtDeMan@1051:         for audioholder in root.findall("./audioholder"):   # iterate over pages
BrechtDeMan@1051:             page_name = audioholder.get('id')               # get page name
BrechtDeMan@1051:             
BrechtDeMan@1051:             if page_name is None: # ignore 'empty' audio_holders
BrechtDeMan@1051:                 print "WARNING: " + file + " contains empty audio holder. (evaluation_stats.py)"
BrechtDeMan@1051:                 break # move on to next
BrechtDeMan@1051:                 
BrechtDeMan@1051:             # keep list of audioholder ids and count how many times each audioholder id was tested
BrechtDeMan@1051:             if page_name in page_names:
BrechtDeMan@1051:                 page_index = page_names.index(page_name) # get index
BrechtDeMan@1051:                 page_count[page_index] += 1
BrechtDeMan@1051:             else: 
BrechtDeMan@1051:                 page_names.append(page_name)
BrechtDeMan@1051:                 page_count.append(1)
BrechtDeMan@1051:             
BrechtDeMan@1051:             # 'testTime' keeps total duration: subtract time so far for duration of this audioholder
BrechtDeMan@1051:             duration = float(audioholder.find("./metric/metricresult[@id='testTime']").text) - total_duration
BrechtDeMan@1051:             
BrechtDeMan@1051:             # total duration of test
BrechtDeMan@1051:             total_duration += duration
BrechtDeMan@1051:             
BrechtDeMan@1051:             # print audioholder id and duration
BrechtDeMan@1051:             print "    " + page_name + ": " + seconds2timestr(duration)
BrechtDeMan@1051:             
BrechtDeMan@1051:             # keep track of duration in function of page index
BrechtDeMan@1051:             if len(duration_order)>page_number:
BrechtDeMan@1051:                 duration_order[page_number].append(duration)
BrechtDeMan@1051:             else:
BrechtDeMan@1051:                 duration_order.append([duration])
BrechtDeMan@1051:             
BrechtDeMan@1051:             page_number += 1 # increase page count for this specific test
BrechtDeMan@1051:             number_of_pages += 1 # increase total number of pages
BrechtDeMan@1051:             time_per_page_accum += duration # total duration (for average time spent per page)
BrechtDeMan@1051:         
BrechtDeMan@1051:         # print total duration of this test
BrechtDeMan@1051:         print "    TOTAL: " + seconds2timestr(total_duration)
BrechtDeMan@1051:                     
BrechtDeMan@1051: # PRINT EVERYTHING
BrechtDeMan@1051: 
BrechtDeMan@1051: print "Number of XML files: " + str(number_of_XML_files)
BrechtDeMan@1051: print "Number of pages: " + str(number_of_pages)
BrechtDeMan@1051: print "Average time per page: " + seconds2timestr(time_per_page_accum/number_of_pages)
BrechtDeMan@1051: page_count_strings = list(str(x) for x in page_count)
BrechtDeMan@1051: count_list = page_names + page_count_strings
BrechtDeMan@1051: count_list[::2] = page_names
BrechtDeMan@1051: count_list[1::2] = page_count_strings
BrechtDeMan@1051: print "Pages tested: " + str(count_list)
BrechtDeMan@1051: 
BrechtDeMan@1051: # Average duration for first, second, ... page
BrechtDeMan@1051: for page_number in range(len(duration_order)): #TODO make maximum page number automatic
BrechtDeMan@1051:     print "Average duration page " + str(page_number+1) + ": " +\
BrechtDeMan@1051:           seconds2timestr(sum(duration_order[page_number])/len(duration_order[page_number])) +\
BrechtDeMan@1051:           " ("+str(len(duration_order[page_number]))+" subjects)"
BrechtDeMan@1051:           
BrechtDeMan@1051: 
BrechtDeMan@1051: #TODO
BrechtDeMan@1051: # time per page in function of number of fragments (plot)
BrechtDeMan@1051: # time per participant in function of number of pages
BrechtDeMan@1051: # plot total time for each participant
BrechtDeMan@1051: # plot total time 
BrechtDeMan@1051: # show 'count' per page (in order)
BrechtDeMan@1051: 
BrechtDeMan@1051: # clear up page_index <> page_count <> page_number confusion