def __init__(self, resources_collection, opener, config, worker_pool_size=5):
super(WebPageLoader, self).__init__()
self._worker_pool = WorkerPool(worker_pool_size)
self._resources_collection = resources_collection
self._opener = opener
logging.config.dictConfig(config['logging'])
self._logger = logging.getLogger()
class WebPageLoader(Thread):
def __init__(self, resources_collection, opener, config, worker_pool_size=5):
super(WebPageLoader, self).__init__()
self._worker_pool = WorkerPool(worker_pool_size)
self._resources_collection = resources_collection
self._opener = opener
logging.config.dictConfig(config['logging'])
self._logger = logging.getLogger()
def run(self):
while True:
# add the web page processing task
for resource in self._resources_collection.find_models():
self._worker_pool.add_task(self._process_web_page, resource)
def _process_web_page(self, resource):
# if the 'http://' doesn't exist
if (resource.uri[:7] != 'http://'):
resource.uri = 'http://' + resource.uri
entire_content = ''
try:
handle = self._opener.open(resource.uri)
resource.uri = handle.url
encoding = detect_header_encoding(handle.headers.dict)
entire_content = decode_html(handle.read(), encoding)
resource.content = entire_content
handle.close()
self._logger.info('Reading %s. Success.' % resource.uri)
self._enqueue(resource)
except (IOError, HTTPException), e:
# mark for retry
self._logger.error('Reading %s. IO error %s.' % (resource.uri, e))
except UnicodeDecodeError, e:
# mark for no more retries
self._logger.error('Reading %s. Unicode error %s.' % (resource.uri, e))
