def getDocument(self, url, file_path, file_name, deep, error_count):
if 0 <= recursion_deep < deep or error_count > document_error_max:
return
url = urlTools.dealUrl2Request(url, url)
if file_path == '' and file_name == '':
file_name = 'index.html'
try:
req = self.requestGet(url)
charset = self.getHTMLCharset(req.content)
req.encoding = charset
d = pq(req.text)
# print charset
linkList1 = d('link')
self.dealSourceLink(linkList1, Url(req.url), 'href')
linkList2 = d('script')
self.dealSourceLink(linkList2, Url(req.url), 'src')
linkList3 = d('img')
self.dealSourceLink(linkList3, Url(req.url), 'src')
linkList4 = d('a')
self.dealALink(linkList4, Url(req.url), 'href', deep)
self.source_task.start()
self.saveFile(file_path, file_name,
bytearray(source=d.outer_html(), encoding='utf-8'))
except requests.exceptions.ConnectionError, e:
print 'ConnectionError:', e
self.queue_document.push(
[url, file_path, file_name, deep, error_count + 1])
def __init__(self, url):
self.main_url = Url(self.requestGet(url).url)
self.host_option, self.host, self.host_url = self.main_url.getHost()
self.queue_resource = None
self.queue_document = None
self.set = None
self.document_task = None
self.source_task = None
def saveFile(self, file_path, file_name, bytes):
path = Url(output + file_path)
path = path.addUrlEnd()
if not os.path.exists(path):
os.makedirs(path)
try:
f = open(path + file_name, "wb")
f.write(bytes)
f.close()
except IOError, e:
print 'save Error: ', e, 'path: ', path, 'name: ', file_name
def saveFile(self, file_path, file_name, bytes):
path = Url(output + file_path)
path = path.addUrlEnd()
if not os.path.exists(path):
os.makedirs(path)
try:
f = open(path + file_name, "wb")
f.write(bytes)
f.close()
except IOError, e:
print 'save Error: ', e, 'path: ', path, 'name: ', file_name
def convLongPath(self, file_path, file_name):
if len(file_name) > 128:
file_name = Url(hashlib.sha1(file_name).hexdigest())
if len(file_path) > 128:
# 懒得管最前面有没有/了
file_path = file_path[0] + Url(hashlib.sha1(file_path).hexdigest())
# path_dirs = file_path.split('/')
# for i, it in enumerate(path_dirs):
# if len(it) > 250:
# path_dirs[i] = str(hashlib.sha1(it).hexdigest())
# file_path = '/'.join(path_dirs)
return file_path, file_name
def dealUrl2File(self, url, origin, host=None, is_req_url=False):
"""
:param
url: 待处理的url
origin: 请求发生时所在的url
host: 对于域名为host的url，资源存放目录为output根目录，而不是域名文件夹。默认不设置主host
is_req_url: url是否做过 dealUrl2Request 处理
:return
"""
if not is_req_url:
url = self.dealUrl2Request(url, origin)
# url = self.simplifyUrl(url)
# url除去最后的/
url = url.delUrlEnd()
if host is not None:
# 如果该url就是这个站点域名下的，那么无需新建域名目录存放
if url.cmpHost(host):
# 除去host 这里有可能超出output根目录
url = url.delHttp()
url = url.delUrlStart()
url = url.replace(host.getHost()[1], '')
# 除去头，变身成文件路径
url = url.delHttp()
url = url.delUrlStart()
for k, v in self.replaceChr.iteritems():
if k in url:
url = url.replace(k, v)
file_name = Url(os.path.basename(url))
file_path = Url(os.path.dirname(url))
# 如果文件名或文件路径过长
file_path, file_name = self.convLongPath(file_path, file_name)
# if file_path.startswith('/') or file_path.startswith('.'):
# file_path = file_path[1:]
# 为了解决同一目录下，文件和文件夹名不能重复的问题
if file_name != '':
file_name = 'f_'+file_name
url = file_path.addUrlEnd() + file_name
url = url.addUrlStart()
# 当file_path为""时,表示当前目录
return file_path, file_name, url
def dealUrl2Request(self, url, origin):
origin = Url(origin)
url = Url(url)
if not url.startswith('http://') and not url.startswith('https://'):
if url.startswith('//'):
url = 'http:' + url
elif url.startswith('/'):
url = origin.getHost()[2] + url
else:
origin = origin.getUrlDir()
url = origin + url
url = url.simplifyUrl()
return url
def dealUrl2File(self, url, origin, host=None, is_req_url=False):
"""
:param
url: 待处理的url
origin: 请求发生时所在的url
host: 对于域名为host的url，资源存放目录为output根目录，而不是域名文件夹。默认不设置主host
is_req_url: url是否做过 dealUrl2Request 处理
:return
"""
if not is_req_url:
url = self.dealUrl2Request(url, origin)
# url = self.simplifyUrl(url)
# url除去最后的/
url = url.delUrlEnd()
if host is not None:
# 如果该url就是这个站点域名下的，那么无需新建域名目录存放
if url.cmpHost(host):
# 除去host 这里有可能超出output根目录
url = url.delHttp()
url = url.delUrlStart()
url = url.replace(host.getHost()[1], '')
# 除去头，变身成文件路径
url = url.delHttp()
url = url.delUrlStart()
for k, v in self.replaceChr.iteritems():
if k in url:
url = url.replace(k, v)
file_name = Url(os.path.basename(url))
file_path = Url(os.path.dirname(url))
# 如果文件名或文件路径过长
file_path, file_name = self.convLongPath(file_path, file_name)
# if file_path.startswith('/') or file_path.startswith('.'):
# file_path = file_path[1:]
# 为了解决同一目录下，文件和文件夹名不能重复的问题
if file_name != '':
file_name = 'f_' + file_name
url = file_path.addUrlEnd() + file_name
url = url.addUrlStart()
# 当file_path为""时,表示当前目录
return file_path, file_name, url
def __init__(self, url):
self.main_url = Url(self.requestGet(url).url)
self.host_option, self.host, self.host_url = self.main_url.getHost()
self.queue_resource = None
self.queue_document = None
self.set = None
self.document_task = None
self.source_task = None
def dealSourceLink(self, linkList, origin_url, attr):
for li in linkList:
url = pq(li).attr(attr)
if url is not None:
url = Url(url)
request_url = urlTools.dealUrl2Request(url, origin_url)
if request_url in self.set:
file_path, file_name, html_url = self.set[request_url]
else:
file_path, file_name, html_url = urlTools.dealUrl2File(
request_url, origin_url, self.host, True)
error_count = 0
self.queue_resource.push(
[request_url, file_path, file_name, error_count])
self.set[request_url] = [file_path, file_name, html_url]
pq(li).attr(attr, html_url)
def dealALink(self, linkList, origin_url, attr, deep):
for li in linkList:
url = pq(li).attr(attr)
if url is not None:
url = Url(url)
request_url = urlTools.dealUrl2Request(url, origin_url)
# print 'A:', request_url
if outsite_page or request_url.getHost()[1] == self.host:
if request_url in self.set:
file_path, file_name, html_url = self.set[request_url]
else:
file_path, file_name, html_url = urlTools.dealUrl2File(
request_url, origin_url, self.host, True)
self.queue_document.push(
[request_url, file_path, file_name, deep + 1, 0])
self.set[request_url] = [
file_path, file_name, html_url
]
pq(li).attr(attr, html_url)
class Crawler:
def __init__(self, url):
self.main_url = Url(self.requestGet(url).url)
self.host_option, self.host, self.host_url = self.main_url.getHost()
self.queue_resource = None
self.queue_document = None
self.set = None
self.document_task = None
self.source_task = None
def __del__(self):
# del self.document_task
# del self.queue_document
# del self.source_task
# del self.queue_resource
print 'del c'
def start(self, is_continue=False):
if is_continue:
self.queue_resource = Queue.load(logs_path + 'queue_resource.json')
self.queue_document = Queue.load(logs_path + 'queue_document.json')
self.set = UrlSet.load(logs_path + 'url_set.json')
else:
self.queue_resource = Queue(logs_path + 'queue_resource.json')
self.queue_document = Queue(logs_path + 'queue_document.json')
self.set = UrlSet(logs_path + 'url_set.json')
self.document_task = Task(self.queue_document, doc_pool_max)
self.document_task.initTaskWork(self.getDocument)
self.source_task = Task(self.queue_resource, res_pool_max)
self.source_task.initTaskWork(self.requestSource)
self.main_url = urlTools.dealUrl2Request(self.main_url, self.main_url)
print self.main_url, self.host
file_path, file_name, html_url = urlTools.dealUrl2File(
self.main_url, self.main_url, self.host, True)
self.queue_document.push([self.main_url, file_path, file_name, 0, 0])
print "file_path:" + file_path, "file_name:" + file_name, "html_url:" + html_url
self.document_task.start()
def requestGet(self, url):
wait = random.random() * (wait_time[1] - wait_time[0])
sleep(wait)
timeout = Timeout(request_timeout)
timeout.start()
try:
req = requests.get(url=url,
verify=True,
headers=headers,
proxies=proxies)
except IncompleteRead:
pass
# todo:未知错误，暂还未查清
timeout.cancel()
return req
def saveFile(self, file_path, file_name, bytes):
path = Url(output + file_path)
path = path.addUrlEnd()
if not os.path.exists(path):
os.makedirs(path)
try:
f = open(path + file_name, "wb")
f.write(bytes)
f.close()
except IOError, e:
print 'save Error: ', e, 'path: ', path, 'name: ', file_name
self.set.save()
self.document_task.stop()
self.source_task.stop()
def work():
print 123
c.start()
time.sleep(10)
print 321
if __name__ == '__main__':
url0 = raw_input("input the Url:")
c = Crawler(Url(url0))
def work(is_continue):
c.start(is_continue)
print 'work stoped'
print "Enter 's' to start, 'c' to continue, 'e' to stop."
while True:
char = raw_input()
if char == 's':
p = threading.Thread(target=work, args=(False, ))
print 'Process will start.'
p.start()
elif char == 'c':
p = threading.Thread(target=work, args=(True, ))
print 'Process will start.'
class Crawler:
def __init__(self, url):
self.main_url = Url(self.requestGet(url).url)
self.host_option, self.host, self.host_url = self.main_url.getHost()
self.queue_resource = None
self.queue_document = None
self.set = None
self.document_task = None
self.source_task = None
def __del__(self):
# del self.document_task
# del self.queue_document
# del self.source_task
# del self.queue_resource
print 'del c'
def start(self, is_continue=False):
if is_continue:
self.queue_resource = Queue.load(logs_path + 'queue_resource.json')
self.queue_document = Queue.load(logs_path + 'queue_document.json')
self.set = UrlSet.load(logs_path + 'url_set.json')
else:
self.queue_resource = Queue(logs_path + 'queue_resource.json')
self.queue_document = Queue(logs_path + 'queue_document.json')
self.set = UrlSet(logs_path + 'url_set.json')
self.document_task = Task(self.queue_document, doc_pool_max)
self.document_task.initTaskWork(self.getDocument)
self.source_task = Task(self.queue_resource, res_pool_max)
self.source_task.initTaskWork(self.requestSource)
self.main_url = urlTools.dealUrl2Request(self.main_url, self.main_url)
print self.main_url, self.host
file_path, file_name, html_url = urlTools.dealUrl2File(self.main_url, self.main_url, self.host, True)
self.queue_document.push([self.main_url, file_path, file_name, 0, 0])
print "file_path:" + file_path, "file_name:" + file_name, "html_url:" + html_url
self.document_task.start()
def requestGet(self, url):
wait = random.random() * (wait_time[1] - wait_time[0])
sleep(wait)
timeout = Timeout(request_timeout)
timeout.start()
try:
req = requests.get(url=url, verify=True, headers=headers, proxies=proxies)
except IncompleteRead:
pass
# todo:未知错误，暂还未查清
timeout.cancel()
return req
def saveFile(self, file_path, file_name, bytes):
path = Url(output + file_path)
path = path.addUrlEnd()
if not os.path.exists(path):
os.makedirs(path)
try:
f = open(path + file_name, "wb")
f.write(bytes)
f.close()
except IOError, e:
print 'save Error: ', e, 'path: ', path, 'name: ', file_name
def dealUrl2Request(self, url, origin):
origin = Url(origin)
url = Url(url)
if not url.startswith('http://') and not url.startswith('https://'):
if url.startswith('//'):
url = 'http:' + url
elif url.startswith('/'):
url = origin.getHost()[2] + url
else:
origin = origin.getUrlDir()
url = origin + url
url = url.simplifyUrl()
return url
