def test_crawl(self):
"""
:return:
"""
url1 = "http://localhost:8081"
q = Queue()
q.put(url1)
thread1 = CrawlerThread(q)
thread1.crawl((url1, 0))
self.assertEqual(CrawlerThreadPool.total_links, 5)
#depth exceed test
url2 = 'localhost:8081/mirror/page1.html'
thread2 = CrawlerThread(q)
thread2.crawl((url2, 101))
self.assertEqual(CrawlerThreadPool.total_links, 5)
url3 = 'http://www.baidu.com?query=10000'
thread3 = CrawlerThread(q)
CrawlerThreadPool.interval_links_cnt = \
ConfReader.instance().get_max_links_count() + 1
t1 = time.time()
thread3.crawl((url3, 0))
t2 = time.time()
self.assertAlmostEqual(t2 - t1,
ConfReader.instance().get_crawl_interval(), 0)
def main(self, argv):
Shared.options = self.defaults
Shared.th_condition = threading.Condition()
Shared.main_condition = threading.Condition()
deps_errors = check_dependences(self.base_dir)
if len(deps_errors) > 0:
print "Dependences errors: "
for err in deps_errors:
print " %s" % err
sys.exit(1)
start_cookies = []
start_referer = None
probe_options = ["-R", self.randstr(20)]
threads = []
num_threads = self.defaults['num_threads']
out_file = ""
out_file_overwrite = self.defaults['out_file_overwrite']
cookie_string = None
initial_checks = True
http_auth = None
get_robots_txt = True
save_html = False
try:
opts, args = getopt.getopt(
argv, 'hc:t:jn:x:A:p:d:BGR:U:wD:s:m:C:qr:SIHFP:OvelE:')
except getopt.GetoptError as err:
print str(err)
sys.exit(1)
if len(args) < 2:
self.usage()
sys.exit(1)
for o, v in opts:
if o == '-h':
self.usage()
sys.exit(0)
elif o == '-c':
cookie_string = v
elif o == '-C':
try:
with open(v) as cf:
cookie_string = cf.read()
except Exception as e:
print "error reading cookie file"
sys.exit(1)
elif o == '-r':
start_referer = v
elif o == '-n':
num_threads = int(v)
elif o == '-t':
Shared.options['process_timeout'] = int(v)
elif o == '-q':
self.display_progress = False
elif o == '-A':
http_auth = v
elif o == '-p':
try:
Shared.options['proxy'] = parse_proxy_string(v)
except Exception as e:
print e
sys.exit(1)
elif o == '-d':
for ad in v.split(","):
# convert *.domain.com to *.\.domain\.com
pattern = re.escape(ad).replace("\\*\\.", "((.*\\.)|)")
Shared.allowed_domains.add(pattern)
elif o == '-x':
for eu in v.split(","):
Shared.excluded_urls.add(eu)
elif o == "-G":
Shared.options['group_qs'] = True
elif o == "-w":
out_file_overwrite = True
elif o == "-R":
Shared.options['max_redirects'] = int(v)
elif o == "-U":
Shared.options['useragent'] = v
elif o == "-s":
if not v in (CRAWLSCOPE_DOMAIN, CRAWLSCOPE_DIRECTORY,
CRAWLSCOPE_URL):
self.usage()
print "* ERROR: wrong scope set '%s'" % v
sys.exit(1)
Shared.options['scope'] = v
elif o == "-m":
if not v in (CRAWLMODE_PASSIVE, CRAWLMODE_ACTIVE,
CRAWLMODE_AGGRESSIVE):
self.usage()
print "* ERROR: wrong mode set '%s'" % v
sys.exit(1)
Shared.options['mode'] = v
elif o == "-S":
initial_checks = False
elif o == "-I":
get_robots_txt = False
elif o == "-H":
save_html = True
elif o == "-D":
Shared.options['max_depth'] = int(v)
elif o == "-P":
Shared.options['max_post_depth'] = int(v)
elif o == "-O":
Shared.options['override_timeout_functions'] = False
elif o == "-F":
Shared.options['crawl_forms'] = False
elif o == "-v":
self.verbose = True
elif o == "-e":
Shared.options['deduplicate_pages'] = False
elif o == "-l":
Shared.options['headless_chrome'] = False
elif o == "-E":
if not Shared.options['extra_headers']:
Shared.options['extra_headers'] = {}
(hn, hv) = v.split("=", 1)
Shared.options['extra_headers'][hn] = hv
probe_cmd = get_node_cmd()
if not probe_cmd: # maybe useless
print "Error: unable to find node executable"
sys.exit(1)
if Shared.options['scope'] != CRAWLSCOPE_DOMAIN and len(
Shared.allowed_domains) > 0:
print "* Warinig: option -d is valid only if scope is %s" % CRAWLSCOPE_DOMAIN
if cookie_string:
try:
start_cookies = parse_cookie_string(cookie_string)
except Exception as e:
print "error decoding cookie string"
sys.exit(1)
if Shared.options['mode'] != CRAWLMODE_AGGRESSIVE:
probe_options.append("-f") # dont fill values
if Shared.options['mode'] == CRAWLMODE_PASSIVE:
probe_options.append("-t") # dont trigger events
if Shared.options['proxy']:
probe_options.extend([
"-y",
"%s:%s:%s" % (Shared.options['proxy']['proto'],
Shared.options['proxy']['host'],
Shared.options['proxy']['port'])
])
if not Shared.options['headless_chrome']:
probe_options.append("-l")
probe_cmd.append(os.path.join(self.base_dir, 'probe', 'analyze.js'))
if len(Shared.excluded_urls) > 0:
probe_options.extend(("-X", ",".join(Shared.excluded_urls)))
if save_html:
probe_options.append("-H")
probe_options.extend(("-x", str(Shared.options['process_timeout'])))
probe_options.extend(("-A", Shared.options['useragent']))
if not Shared.options['override_timeout_functions']:
probe_options.append("-O")
if Shared.options['extra_headers']:
probe_options.extend(
["-E", json.dumps(Shared.options['extra_headers'])])
Shared.probe_cmd = probe_cmd + probe_options
Shared.starturl = normalize_url(args[0])
out_file = args[1]
purl = urlsplit(Shared.starturl)
Shared.allowed_domains.add(purl.hostname)
for sc in start_cookies:
Shared.start_cookies.append(Cookie(sc, Shared.starturl))
start_req = Request(REQTYPE_LINK,
"GET",
Shared.starturl,
set_cookie=Shared.start_cookies,
http_auth=http_auth,
referer=start_referer)
if not hasattr(ssl, "SSLContext"):
print "* WARNING: SSLContext is not supported with this version of python, consider to upgrade to >= 2.7.9 in case of SSL errors"
stdoutw("Initializing . ")
start_requests = self.init_crawl(start_req, initial_checks,
get_robots_txt)
database = None
self.db_file = self.generate_filename(out_file, out_file_overwrite)
try:
database = self.init_db(self.db_file, out_file)
except Exception as e:
print str(e)
sys.exit(1)
database.save_crawl_info(htcap_version=get_program_infos()['version'],
target=Shared.starturl,
start_date=self.crawl_start_time,
commandline=cmd_to_str(argv),
user_agent=Shared.options['useragent'],
proxy=json.dumps(Shared.options['proxy']),
extra_headers=json.dumps(
Shared.options['extra_headers']),
cookies=json.dumps(start_cookies))
database.connect()
database.begin()
for req in start_requests:
database.save_request(req)
database.commit()
database.close()
print "done"
print "Database %s initialized, crawl started with %d threads (^C to pause or change verbosity)" % (
self.db_file, num_threads)
for n in range(0, num_threads):
thread = CrawlerThread()
threads.append(thread)
thread.start()
self.main_loop(threads, start_requests, database)
self.kill_threads(threads)
self.crawl_end_time = int(time.time())
print "Crawl finished, %d pages analyzed in %d minutes" % (
Shared.requests_index,
(self.crawl_end_time - self.crawl_start_time) / 60)
database.save_crawl_info(end_date=self.crawl_end_time)
def main(self, argv):
Shared.options = self.defaults
Shared.th_condition = threading.Condition()
Shared.main_condition = threading.Condition()
probe_cmd = get_phantomjs_cmd()
if not probe_cmd:
print "Error: unable to find phantomjs executable"
sys.exit(1)
start_cookies = []
start_referer = None
probe_options = ["-R", self.randstr(20)]
threads = []
num_threads = self.defaults['num_threads']
out_file = ""
out_file_overwrite = self.defaults['out_file_overwrite']
cookie_string = None
display_progress = True
verbose = False
initial_checks = True
http_auth = None
get_robots_txt = True
save_html = False
user_script = None
try:
opts, args = getopt.getopt(
argv, 'hc:t:jn:x:A:p:d:BGR:U:wD:s:m:C:qr:SIHFP:Ovu:')
except getopt.GetoptError as err:
print str(err)
sys.exit(1)
if len(args) < 2:
self.usage()
sys.exit(1)
for o, v in opts:
if o == '-h':
self.usage()
sys.exit(0)
elif o == '-c':
cookie_string = v
elif o == '-C':
try:
with open(v) as cf:
cookie_string = cf.read()
except Exception as e:
print "error reading cookie file"
sys.exit(1)
elif o == '-r':
start_referer = v
elif o == '-n':
num_threads = int(v)
elif o == '-t':
Shared.options['process_timeout'] = int(v)
elif o == '-q':
display_progress = False
elif o == '-A':
http_auth = v
elif o == '-p':
if v == "tor": v = "socks5:127.0.0.1:9150"
proxy = v.split(":")
if proxy[0] not in ("http", "socks5"):
print "only http and socks5 proxies are supported"
sys.exit(1)
Shared.options['proxy'] = {
"proto": proxy[0],
"host": proxy[1],
"port": proxy[2]
}
elif o == '-d':
for ad in v.split(","):
# convert *.domain.com to *.\.domain\.com
pattern = re.escape(ad).replace("\\*\\.", "((.*\\.)|)")
Shared.allowed_domains.add(pattern)
elif o == '-x':
for eu in v.split(","):
Shared.excluded_urls.add(eu)
elif o == "-G":
Shared.options['group_qs'] = True
elif o == "-w":
out_file_overwrite = True
elif o == "-R":
Shared.options['max_redirects'] = int(v)
elif o == "-U":
Shared.options['useragent'] = v
elif o == "-s":
if not v in (CRAWLSCOPE_DOMAIN, CRAWLSCOPE_DIRECTORY,
CRAWLSCOPE_URL):
self.usage()
print "* ERROR: wrong scope set '%s'" % v
sys.exit(1)
Shared.options['scope'] = v
elif o == "-m":
if not v in (CRAWLMODE_PASSIVE, CRAWLMODE_ACTIVE,
CRAWLMODE_AGGRESSIVE):
self.usage()
print "* ERROR: wrong mode set '%s'" % v
sys.exit(1)
Shared.options['mode'] = v
elif o == "-S":
initial_checks = False
elif o == "-I":
get_robots_txt = False
elif o == "-H":
save_html = True
elif o == "-D":
Shared.options['max_depth'] = int(v)
elif o == "-P":
Shared.options['max_post_depth'] = int(v)
elif o == "-O":
Shared.options['override_timeout_functions'] = False
elif o == "-F":
Shared.options['crawl_forms'] = False
elif o == "-v":
verbose = True
elif o == "-u":
if os.path.isfile(v):
user_script = os.path.abspath(v)
else:
print "error: unable to open USER_SCRIPT"
sys.exit(1)
if Shared.options['scope'] != CRAWLSCOPE_DOMAIN and len(
Shared.allowed_domains) > 0:
print "* Warinig: option -d is valid only if scope is %s" % CRAWLSCOPE_DOMAIN
if cookie_string:
try:
start_cookies = self.parse_cookie_string(cookie_string)
except Exception as e:
print "error decoding cookie string"
sys.exit(1)
if Shared.options['mode'] != CRAWLMODE_AGGRESSIVE:
probe_options.append("-f") # dont fill values
if Shared.options['mode'] == CRAWLMODE_PASSIVE:
probe_options.append("-t") # dont trigger events
if Shared.options['proxy']:
probe_cmd.append("--proxy-type=%s" %
Shared.options['proxy']['proto'])
probe_cmd.append("--proxy=%s:%s" %
(Shared.options['proxy']['host'],
Shared.options['proxy']['port']))
probe_cmd.append(self.base_dir + 'probe/analyze.js')
if len(Shared.excluded_urls) > 0:
probe_options.extend(("-X", ",".join(Shared.excluded_urls)))
if save_html:
probe_options.append("-H")
if user_script:
probe_options.extend(("-u", user_script))
probe_options.extend(("-x", str(Shared.options['process_timeout'])))
probe_options.extend(("-A", Shared.options['useragent']))
if not Shared.options['override_timeout_functions']:
probe_options.append("-O")
Shared.probe_cmd = probe_cmd + probe_options
Shared.starturl = normalize_url(args[0])
out_file = args[1]
purl = urlsplit(Shared.starturl)
Shared.allowed_domains.add(purl.hostname)
for sc in start_cookies:
Shared.start_cookies.append(Cookie(sc, Shared.starturl))
start_req = Request(REQTYPE_LINK,
"GET",
Shared.starturl,
set_cookie=Shared.start_cookies,
http_auth=http_auth,
referer=start_referer)
if not hasattr(ssl, "SSLContext"):
print "* WARNING: SSLContext is not supported with this version of python, consider to upgrade to >= 2.7.9 in case of SSL errors"
stdoutw("Initializing . ")
if user_script and initial_checks:
self.check_user_script_syntax(probe_cmd, user_script)
start_requests = self.init_crawl(start_req, initial_checks,
get_robots_txt)
database = None
fname = self.generate_filename(out_file, out_file_overwrite)
try:
database = self.init_db(fname, out_file)
except Exception as e:
print str(e)
sys.exit(1)
database.save_crawl_info(htcap_version=get_program_infos()['version'],
target=Shared.starturl,
start_date=self.crawl_start_time,
commandline=cmd_to_str(argv),
user_agent=Shared.options['useragent'])
database.connect()
database.begin()
for req in start_requests:
database.save_request(req)
database.commit()
database.close()
print "done"
print "Database %s initialized, crawl started with %d threads" % (
fname, num_threads)
for n in range(0, num_threads):
thread = CrawlerThread()
threads.append(thread)
thread.start()
self.main_loop(threads, start_requests, database, display_progress,
verbose)
self.kill_threads(threads)
self.crawl_end_time = int(time.time())
print "Crawl finished, %d pages analyzed in %d minutes" % (
Shared.requests_index,
(self.crawl_end_time - self.crawl_start_time) / 60)
database.save_crawl_info(end_date=self.crawl_end_time)
