def crawl_all():
articles = Articles()
visited = set()
coroutines = [parser.feed_urls() for parser in parsers]
for coroutine in asyncio.as_completed(coroutines):
urls = list(map(canonicalize_url, (yield from coroutine)))
if len(urls) < 1:
continue
parser = get_parser(urls[0])
log.info('Got {} URLs for {}'.format(len(urls), parser.domain))
to_get = [parser(x).parse() for x in urls if x not in visited]
visited = visited.union(urls)
for get_page in asyncio.as_completed(to_get):
try:
page = yield from get_page
articles.save_entry(page)
except Exception as e:
log.error(e)
urls = get_existing_urls(articles)
to_get = [get_parser(x)(x).parse() for x in urls if x not in visited]
log.info("updating {} existing unvisited URLs".format(len(to_get)))
for get_page in asyncio.as_completed(to_get):
try:
page = yield from get_page
articles.save_entry(page)
except Exception as e:
log.error(e)
def parser(self, value):
if isinstance(value, parsers.Parser):
self._parser = value
else:
try:
self.parser = parsers.get_parser(value)
except NameError as e:
raise ValueError("parser value '"+str(value)+"' must be an instance of parsers.Parser or a string allowed by parsers.get_parser.")
def load_article(url):
try:
parser = parsers.get_parser(url)
except KeyError:
logger.info('Unable to parse domain, skipping')
return
parsed_article = parser(url)
if not parsed_article.real_article:
return
return parsed_article
def parser(self, value):
if isinstance(value, parsers.Parser):
self._parser = value
else:
try:
self.parser = parsers.get_parser(value)
except NameError as e:
raise ValueError(
"parser value '" + str(value) +
"' must be an instance of parsers.Parser or a string allowed by parsers.get_parser."
)
def test_parsing(self):
"""Simple format, simple test"""
parser = parsers.get_parser('loadavg')
results = dict()
parser.parse(self.example, results)
stats = results['stats']
self.assertEqual(0.15, stats.one_minute_load)
self.assertEqual(0.22, stats.five_minute_load)
self.assertEqual(0.43, stats.fifteen_minute_load)
self.assertEqual(1, stats.running_threads)
self.assertEqual(650, stats.total_threads)
self.assertEqual(3914, stats.last_pid)
def load_article(url):
try:
parser = parsers.get_parser(url)
except KeyError:
logger.info('Unable to parse domain, skipping')
return
try:
parsed_article = parser(url)
except (AttributeError, urllib2.HTTPError, httplib.HTTPException), e:
if isinstance(e, urllib2.HTTPError) and e.msg == 'Gone':
return
logger.error('Exception when parsing %s', url)
logger.error(traceback.format_exc())
logger.error('Continuing')
return
def load_article(url):
try:
parser = parsers.get_parser(url)
except KeyError:
logger.info('Unable to parse domain, skipping')
return
try:
parsed_article = parser(url)
except (AttributeError, urllib2.HTTPError, httplib.HTTPException), e:
if isinstance(e, urllib2.HTTPError) and e.msg == 'Gone':
return
logger.error('Exception when parsing %s', url)
logger.error(traceback.format_exc())
logger.error('Continuing')
return
def test_parse_meminfo(self):
print(self.example)
parser = parsers.get_parser('meminfo')
results = parser.parse(self.example, dict())
meminfo = results['meminfo']
self.assertEqual(45, len(meminfo.meminfo))
self.assertEqual(20507388 * 1024, meminfo.get('MemTotal'))
self.assertEqual(8326068 * 1024, meminfo.get('MemFree'))
self.assertEqual(20559872 * 1024, meminfo.get('DirectMap2M'))
self.assertEqual(1, meminfo.get('HugePages_Total'))
self.assertEqual(2, meminfo.get('HugePages_Free'))
self.assertEqual(3, meminfo.get('HugePages_Rsvd'))
self.assertEqual(4, meminfo.get('HugePages_Surp'))
def load_article(url):
try:
parser = get_parser(url)
except KeyError:
log.info('Unable to parse domain, skipping')
return
try:
parsed_article = parser(url)
except (AttributeError, urllib.request.HTTPError, Exception) as e:
if isinstance(e, urllib.request.HTTPError) and e.msg == 'Gone':
return
log.error('Exception when parsing %s', url)
log.error(traceback.format_exc())
log.error('Continuing')
return
if not parsed_article.real_article:
return
return parsed_article
def _parse_section(section_name, current_process, current_thread, data, out):
try:
parser = parsers.get_parser(section_name)
parser.parse(data, out)
except:
pass
if current_thread and section_name == 'stat':
_save_stat(current_thread, out['stat'])
elif current_process and section_name != '':
# Hit a new file, consolidate what we have so far.
if 'smaps' == section_name:
_save_smaps_region(current_process.maps, out['meminfo'],
current_process.pid, data)
elif 'cmdline' == section_name:
# Some command lines have a number of empty arguments. Ignore
# that because it's not interesting here.
current_process.argv = filter(len, data.strip().split('\0'))
elif 'stat' == section_name:
_save_stat(current_process, out['stat'])
else:
LOGGER.error('Unrecognised section name: %s' % section_name)
def test_process_stat_parsing(self):
parser = parsers.get_parser('uptime')
stats = parser.parse(self.example, dict())['stats']
self.assertEqual(84983.36, stats.uptime)
self.assertEqual(434057.28, stats.uptime_idle)
from methods import Bisection, FixedPoint, NewtonRaphson
from parsers import get_parser
if __name__ == '__main__':
parser = get_parser()
args = parser.parse_args()
function = lambda x: 3143680 - 2 * 2**x - 51200 * x
derivative = lambda x: (-2**x) * (0.69314718) - 51200
error_esperado = 0.00001
if args.method == 'biseccion':
method = Bisection(function, args.a, args.b)
if args.method == 'punto-fijo':
method = FixedPoint(function, args.a, args.b)
if args.method == 'newton-raphson':
method = NewtonRaphson(function, derivative, args.inicial)
for i, valor in enumerate(method, 1):
if method.get_error() <= error_esperado:
break
print('Se obtuvo el valor ', valor, ' en ', i, ' iteraciones')
def test_stat_parsing(self):
parser = parsers.get_parser('stat')
res = parser.parse(self.example, dict())['stat']
self.assertEqual(res['comm'], 'tracker-miner-f', 'comm not as expected? {0}'.format(res['comm']))
self.assertEqual(res, self.expected, 'Did not parse /proc/stat correctly?')
$ python test_parser.py nyt.NYTParser <one of those URLs>
[text of article to store]
"""
import sys
import parsers
try:
parsername = sys.argv[1]
except IndexError:
print 'Usage: test_parser.py <modulename>.<classname> [<url_to_check>]'
sys.exit()
try:
url = sys.argv[2]
except IndexError:
url = None
module, classname = parsername.rsplit('.', 1)
parser = getattr(__import__(module, globals(), fromlist=[classname]),
classname)
if url:
assert type(parsers.get_parser(url)) == type(parser)
parsed_article = parser(url)
print unicode(parsed_article)
else:
links = parser.feed_urls()
links = [link for link in links if parser.filter(link)]
print '\n'.join(links)
def setUp(self):
self.parser = get_parser()
self.function = lambda x: x**2 - 2
self.derivative = lambda x: 2 * x
