为了获得培训数据,我写了一个爬虫来跟踪Alexa上的前500个网站,深度为2,并将找到的所有链接写入文件。现在,它查找html中的所有链接并将它们写入文件。问题是,抓取工具错过了所有广告链接,其中一些链接位于iframe中或位于CSS文件中。如何更改我的网络抓取工具,以便抓取所有链接,包括广告?相关代码可以在下面找到。
class Crawler(object):
xRunnable::new
类Fetcher(对象):
def __init__(self, root, depth, locked=True):
self.root = root
self.depth = depth
self.locked = locked
self.host = urlparse.urlparse(root)[1]
self.urls = []
self.links = 0
self.followed = 0
def crawl(self):
#print " in crawl"
page = Fetcher(self.root)
q = Queue()
#print "made fetcher"
try:
page.fetch()
if page.urls == []:
print "Error: could not fetch urls for %s" % (self.root)
return
#raise KeyboardInterrupt
else:
target = open("output.txt", 'w')
for url in page.urls:
q.put(url)
target.write((url+'\n').encode('utf-8'))
followed = [self.root]
target.close()
except Exception as e:
print('Error: could not fetch urls')
raise KeyboardInterrupt
'''
q = Queue()
target = open("output.txt", 'w')
for url in page.urls:
q.put(url) f
target.write((url+'\n').encode('utf-8'))
followed = [self.root]
target.close()
#print followed
'''
n = 0
while True:
try:
url = q.get()
except QueueEmpty:
break
n += 1
if url not in followed:
try:
host = urlparse.urlparse(url)[1]
if self.locked and re.match(".*%s" % self.host, host):
followed.append(url)
#print url
self.followed += 1
page = Fetcher(url)
page.fetch()
for i, url in enumerate(page):
if url not in self.urls:
self.links += 1
q.put(url)
self.urls.append(url)
with open("data.out", 'w') as f:
f.write(url)
if n > self.depth and self.depth > 0:
break
except Exception, e:
print "ERROR: Can't process url '%s' (%s)" % (url, e)
print format_exc()
静态方法:
def __init__(self, url):
self.url = url
self.urls = []
def __getitem__(self, x):
return self.urls[x]
def _addHeaders(self, request):
request.add_header("User-Agent", AGENT)
def open(self):
url = self.url
try:
request = urllib2.Request(url)
handle = urllib2.build_opener()
except IOError:
return None
return (request, handle)
def fetch(self):
request, handle = self.open()
self._addHeaders(request)
if handle:
try:
content = unicode(handle.open(request).read(), "utf-8",
errors="replace")
soup = BeautifulSoup(content)
tags = soup('a')
except urllib2.HTTPError, error:
if error.code == 404:
print >> sys.stderr, "ERROR: %s -> %s" % (error, error.url)
else:
print >> sys.stderr, "ERROR: %s" % error
tags = []
except urllib2.URLError, error:
print >> sys.stderr, "ERROR: %s" % error
tags = []
for tag in tags:
href = tag.get("href")
if href is not None:
url = urlparse.urljoin(self.url, escape(href))
if url not in self:
self.urls.append(url)
def getLinks(url):
page = Fetcher(url)
page.fetch()
for i, url in enumerate(page):
print "%d. %s" % (i, url)
答案 0 :(得分:0)
通过在页面上执行的异步javascript提供大量广告。如果您只是抓取服务器初始输出,您将无法获得其他链接。一种方法是使用像PhantomJS这样的无头浏览器将html渲染到文件,然后使用你的脚本。还有其他可能性。