def on_receive(self, message):
log.log_info("Orchestrator received message: {:}".format(message))
if message['method'] == 'load_file':
data = message['data']
if data['file']:
response = self.document_parser.ask(message)
if response['status'] == 0:
return msg.build_response(status=0)
else:
return msg.build_response(
status=-2,
error_msg="Orchestrator.load_file failed: {:}".format(
response['error_msg']))
else:
return msg.build_response(
status=-1,
error_msg="No file provided to Orchestrator.load_file")
elif message['method'] == 'search':
data = message['data']
if data['query']:
response = self.query_processor.ask(
msg.build_request(method='query', data=message['data']))
if response['status'] == 0:
return response
else:
return msg.build_response(
status=-1,
error_msg="No word provided to Orchestrator.search")
return msg.build_response(
status=-13,
error_msg="No method to process message: {:}".format(message))
def _get_doc_info(self, doc_id):
response = self.document_parser.ask(
msg.build_request(method='get_doc', data={'doc_id': doc_id}))
if response['status'] == 0:
data = response['data']
return data['title'], data['text'], data['author'], data['date']
else:
log.log_error("Could not get info for doc: {:}".format(
response['error_msg']))
def _stem_word(self, word):
response = self.document_parser.ask(
msg.build_request(method='stem_word', data={'word': word}))
if response['status'] == 0:
return response['data']['stem']
else:
log.log_error("Orchestrator could not stem word: {:}".format(
response['error_msg']))
return ""
def main(argv):
log.flush_log()
log.log_info("Hello")
# process arguments
document = None
if '-d' in argv:
document = argv[argv.index('-d') + 1]
# launch orchestrator
orchestrator = Orchestrator.start()
if document:
response = orchestrator.ask(
msg.build_request(method='load_file', data={'file': document}))
if response['status'] != 0:
log.log_error(response['error_msg'])
else:
log.log_info("Loaded file")
is_running = True
while is_running:
query = raw_input("Search: ")
log.log_info("Query: {:}".format(query))
if query == "q" or query == "quit":
is_running = False
else:
response = orchestrator.ask(
msg.build_request(method='search', data={'query': query}))
if response['status'] == 0:
# log.log_info("Found word: {:}".format(response['data']))
print response['data']
else:
log.log_error(response['error_msg'])
orchestrator.stop()
log.log_info("Goodbye")
def _load_file(self, _file):
try:
log.log_info("DocumentParser.load_file parsing xml...")
tree = ET.parse(_file)
root = tree.getroot()
log.log_info("Loading pages...")
for page in root.findall('page'):
page_id = int(page.find('id').text)
log.log_info("Loading page: {:}".format(page_id))
page_data = self._parse_xml_page(page)
self.index_handler.ask(
msg.build_request(method='store_page',
data={'page': page_data}))
log.log_info("Done loading pages")
self.parsed_docs.append(_file)
except:
log.log_error("DocumentParser.load_file error parsing xml")
log.log_debug(traceback.format_exc())
def _process_query(self, query):
# assume single word query
query = str(query).lower()
stem_word = self._stem_word(query)
if stem_word != "":
response = self.index_handler.ask(
msg.build_request(method='search', data={'word': stem_word}))
if response['status'] == 0:
# TODO(JC): Add document retrival
result = self._beautify_result(response['data'], query)
return msg.build_response(status=0, data=result)
else:
return msg.build_response(
status=-3,
error_msg="QueryProcessor.process_query failed: {:}".
format(response['error_msg']))
else:
return msg.build_response(status=-2,
error_msg="Stemmed word is empty string")