Batch Results Retrieval in Python

 1 #! /usr/bin/python
 2 
 3 ########################################################################
 4 #
 5 #  Example client to submit a batch request to TextServer 
 6 # SERVICENAME service, and wait for the job to be finished.
 7 #
 8 #  Input must be a ZIP file containing one or more text files to analyze
 9 #  Output will be a ZIP file containing results for each input file in
10 # the requested format (XML, json, conll)
11 #
12 ########################################################################
13 
14 # import required libraries
15 import urllib2
16 # You may need to install "poster" python module for these two:
17 from poster.encode import multipart_encode
18 from poster.streaminghttp import register_openers
19 import xml.etree.ElementTree
20 import time
21 
22 # Register the streaming http handlers with urllib2
23 register_openers()
24 
25 # set query elements
26 user = raw_input('TextServer Username: ')
27 tkid = raw_input('TextServer Job Token ID: ')
28 outfname = raw_input('Output ZIP file: ')
29 
30 # service URL
31 TextServer_URL = "http://frodo.lsi.upc.edu:8080/TextWS/textservlet/ws"
32 
33 # prepare request to poll for completion and retrieve results
34 datagen, headers = multipart_encode({'username' : user,
35                                      'tokenID' : tkid
36                                      })
37 request = urllib2.Request(TextServer_URL + "/resultRetrieve", 
38                           datagen, 
39                           headers)
40 
41 try:
42   # send request to check for results
43   print "Polling server for job completion..."
44   resp =  urllib2.urlopen(request).read()
45   
46 except urllib2.HTTPError, e:
47   if (e.code == 503 and e.read()[0:8] == "[TS-125]" ) :
48     # http status 503, textserver code TS-125 means the job is not finished yet. 
49     print "Job not finished yet" 
50   else :      
51     # some actual error happened. abort
52     print e, "-", e.read()
53   exit()
54     
55 
56 # no error, save results to output file
57 outf=open(outfname, 'w+')
58 print >>outf, resp
59 outf.close()
60 print "Job finished. Results saved to",outfname