def extract_multi(to_fetch, seen_urls):
results = []
for url in to_fetch:
if url in seen_urls: continue
seen_urls.add(url)
try:
results.append(extract(url))
except Exception:
continue
return results
def extract_multi(to_fetch, seen_urls):
results = []
for url in to_fetch:
if url in seen_urls: continue
seen_urls.add(url)
try:
results.append(extract(url))
except Exception:
continue
return results
def fetcher(fetch_queue, output_queue):
logging.info('Starting fetcher thread')
while True:
state, depth, url = fetch_queue.get()
logging.info('%s: Fetching in thread: %s', id(state), url)
try:
try:
_, data, found_urls = extract(url)
except Exception:
data, found_urls = None, []
output_queue.put(FetchResult(state, depth, url, data, found_urls))
finally:
fetch_queue.task_done()
def fetcher(fetch_queue, max_depth, seen_urls, output_queue):
while True:
depth, url = fetch_queue.get()
try:
if depth > max_depth: continue # Ignore URLs that are too deep
if url in seen_urls: continue # Prevent infinite loops
seen_urls.add(url) # GIL :/
try:
_, data, found_urls = extract(url)
except Exception:
continue
output_queue.put((url, data))
for found in found_urls:
fetch_queue.put((depth + 1, found))
finally:
fetch_queue.task_done()
def consumer(fetch_queue, max_depth, seen_urls, result):
while True:
depth, url = fetch_queue.get()
try:
if depth > max_depth: continue
if url in seen_urls: continue # GIL :|
seen_urls.add(url) # GIL :/
try:
_, data, found_urls = extract(url)
except Exception:
continue
result.append((depth, url, data)) # GIL :(
for found in found_urls:
fetch_queue.put((depth + 1, found))
finally:
fetch_queue.task_done()
