def init_data():
br = BasicRequests()
br.select_user_agent('firefox')
s = open('af').read(64 * 1024)
odata = []
for m in re.finditer('<a (.*?)>(.*?)</a>', s, re.S):
name = m.group(2)
attrs = m.group(1)
url = None
prov = None
for m1 in re.finditer('([a-z][a-z0-9]+)="(.*?)"', attrs, re.S):
n, v = m1.group(1), m1.group(2)
if n == 'href':
url = v
if n == "prov":
prov = v
con = br.request_url(url)
siteurl = 'unknown'
if con is not None:
siteurl = con.request.url
print name, con.request.url
odata.append({
'name': name,
'url': siteurl,
'imgurl': '',
'prov': prov
})
print json.dumps(odata, ensure_ascii=0, indent=4)
def post_for_proxy():
req = BasicRequests()
con = req.request_url(
'http://dev.kuaidaili.com/api/getproxy/?orderid=925817981728018&num=50&b_pcchrome=1&b_pcie=1&b_pcff=1&protocol=1&method=2&an_an=1&an_ha=1&sp1=1&quality=1&sort=1&format=json&sep=1'
)
if con:
return eval(con.text)
def resolve(self, getimg, fc=None):
while True:
imgcon = getimg(dbgdata=fc)
f = StringIO.StringIO()
f.write(imgcon)
f.seek(0)
h = BasicRequests()
imgcode = None
if fc is not None and isinstance(fc, dict) and "type" in fc:
response = h.request_url(self.server, files={'file': imgcon}, data={"province": self._type})
result = response.text.strip()
if '"valid":true' in result:
try:
result = json.loads(result)
imgcode = result["answer"]
if imgcode == None or imgcode == "":
print "验证码图片识别错误 imgcode==None或''"
continue
except Exception as err:
print "验证码图片识别错误，重新校验...,result:", result, "错误原因:", err
time.sleep(1)
continue
else:
url = "%s?type=%s" % (self.server, self._type)
response = h.request_url(url, files={'file': f})
imgcode = response.text.strip()
if imgcode == '<fail>':
print "验证码图片识别错误 imgcode==<fail>"
continue
if isinstance(fc, dict):
fc['content'] = imgcon
fc['code'] = imgcode
return imgcode
def find_ipin_proxy():
ff = IpinFactory()
prs = ff.getProxyList()
s = BasicRequests()
#print json.dumps(prs, ensure_ascii=0).encode('utf-8')
res = {}
for p in prs:
print "trying", p
auth, proxies = ff.genRequestsParam(p)
con = s.request_url("http://ip.cn/",
auth=auth,
proxies=proxies,
timeout=6)
if con is None:
continue
m = re.search("<code>(.*?)</code>", con.text)
if m:
sys.stderr.write("%s %s\n" % (p['host'], m.group(1)))
res[m.group(1)] = "%s:%s:%s" % (p.get('host'), p.get('port'),
p.get('password'))
p2, proxies = ff.genTinyProxy(p)
print proxies
con = s.request_url("http://ip.cn/", proxies=proxies, timeout=5)
if con is None:
continue
m = re.search("<code>(.*?)</code>", con.text)
if m:
sys.stderr.write("%s %s\n" % (p['host'], m.group(1)))
res[m.group(1)] = p2
print "\n".join(res.values())
def __init__(self, sf='jobs'):
BasicRequests.__init__(self)
self._main_url = 'http://www.zjsfgkw.cn/Document/JudgmentBook'
self._court_search_url = 'http://www.zjsfgkw.cn/Judges/GetCountByCountId'
self._book_search_url = 'http://www.zjsfgkw.cn/document/JudgmentSearch'
self.start_date = '19700101'
self.end_date = time.strftime('%Y%m%d', time.localtime())
self.save_file = sf
def get_search_url(opts):
# SF_1_1_27=0为中文简历 SF_1_1_27=1为英文简历, 但是从js看并不能搜英文的.
b = BasicRequests()
# surl = 'http://rdsearch.zhaopin.com/Home/ResultForCustom?SF_1_1_7=7,9&orderBy=DATE_MODIFIED,1&pageSize=60&SF_1_1_27=0&exclude=1'
surl = 'http://rdsearch.zhaopin.com/Home/ResultForCustom?SF_1_1_7=8,9&orderBy=DATE_MODIFIED,1&pageSize=60&SF_1_1_27=0&exclude=1'
for name, value in opts.items():
surl = b.compose_url(surl, name, value)
return surl
def test_ps():
ps = PageStore51()
ps.testmode = True
br = BasicRequests()
br.select_user_agent('firefox')
url = "http://jobs.51job.com/beijing-hdq/70320056.html?s=0"
con = br.request_url(url)
ps.save(int(time.time()), "jd_51job://", url, con.text)
def test_login(self):
req = BasicRequests()
con = req.request_url(self.main_url + 'login',
date={
'username': '******',
'password': '******'
})
print con.headers
print con.text
def test_parse_time():
request = BasicRequests()
con = request.request_url(
'http://www.zjsfgkw.cn/attachment/documentbook/2016-04-05/0225-0229/html/671a34a7-b068-4025-af13-d9fe4c28ce6a.html'
)
m = re.search(
ur'[一二三四五六七八九〇零○十]{4}年[一二三四五六七八九〇十○]{1,2}月[一二三四五六七八九〇零○十]{1,3}日',
con.text)
if m:
print date_cs2num(m.group())
def thread_init(self, tid):
# self.proxyq is threading-safe
proxy = self.proxyq.get(True)
basicreq = BasicRequests()
basicreq.sp_proxies[proxy] = 0
basicreq._cur_proxy_index = 0
basicreq._auto_change_proxy = False
setattr(self._tls, "req", basicreq)
with self.locker:
Log.info("Thread%d's request prepared..Proxy:%s" % (tid, proxy))
def try_proxy(proxy, url='http://gaokao.chsi.com.cn', tag=u'阳光高考'):
req = BasicRequests()
req.set_proxy(proxy, 0, False)
# con = req.request_url('http://gk.chsi.com.cn/recruit/listSpecBySchool.do?yxdm=11055&start=0 ')
con = req.request_url(url, timeout=5)
if con:
m = re.search(r'<title>[^<]*<\/title>', con.text)
if m:
print m.group()
return re.search(tag, con.text)
def sub_pages(url, con):
count = CVZhilianUtil.get_count(url, con)
if count > 4000:
count = 4000
npages = (count + 60 - 1) / 60
if npages >= 2:
b = BasicRequests()
for p in range(2, npages + 1):
url1 = b.compose_url(url, 'pageIndex', p)
yield url1
def test_extract_inner_paper_url():
rq = BasicRequests()
con = rq.request_url(
'http://www.zjsfgkw.cn/document/JudgmentDetail/4177773')
content = re.search(r'<div class="books_detail_header">.*</IFRAME>',
con.text, re.S)
m = re.search(r'src="([^"]+)"', content.group())
if m:
print m.group(1)
else:
print content
def test_search():
a = BasicRequests()
while True:
data = {
'docids1': "1,2,3,4,5,6,6109234,6110168,11070364",
"keywords": "武汉"
}
con = a.request_url("http://localhost:4096/search?hehe=1", data=data)
if con is not None:
print con.code, con.text
time.sleep(10)
def filter_with_speed(proxies, url='http://www.baidu.com', timeout=10):
results = []
req = BasicRequests()
for proxy in proxies:
req.set_proxy(proxy, len(req.sp_proxies), False)
try:
con = req.request_url(url, timeout=timeout)
except Exception:
con = None
if con:
results.append(proxy)
return results
def get_child_court(self, court):
req = BasicRequests()
req.set_proxy(self.proxy)
time.sleep(1)
print 'fetching child court', court['key']
con = req.request_url('http://wenshu.court.gov.cn/Index/GetChildAllCourt',
data={'keyCodeArrayStr': court['key']})
if '<' in con.text:
return
court = self.parse_results(con)
for c in court:
self.child_courts.append(c)
def get_court(self):
req = BasicRequests()
req.set_proxy(self.proxy)
for p in self.provinces:
time.sleep(1)
print 'fetch province', p['name']
con = req.request_url('http://wenshu.court.gov.cn/Index/GetCourt', data={'province': p['name']})
if '<' in con.text:
print 'invalid response'
continue
court = self.parse_results(con)
for c in court:
self.courts.append(c)
def runjs(url):
rq = BasicRequests()
con = rq.request_url(url, data={})
if con:
print con.text
m = re.findall(r'<script[^>]*>(.+?)</script>', con.text, re.S)
if m:
for js in m:
if js == '':
continue
print js
sc = "document = {set cookie(a){console.log(a);}}, window = {innerWidth: 1024, innerHeight: 768, screenX: 200, screenY: 100, screen: {width: 1024, height: 768}}\n"
sc += js
rv = spider.util.runjs(sc)
print 'my results:'
print rv
def _do_requests(self, url, **kwargs):
rv = BasicRequests._do_requests(self, url, **kwargs)
# TODO: replace SimpleCookie with someone better.
if rv is not None:
curlckjar = getattr(self._curltls, 'cookies', None)
if curlckjar is None:
curlckjar = CurlCookieJar()
curlckjar.add_list(rv.cookies)
setattr(self._curltls, 'cookies', curlckjar)
return rv
def test_proxy(proxy, url, count=10):
c = count
rq = BasicRequests()
rq.set_proxy(proxy['p'])
total = 0
success = 0
while c > 0:
try:
s = time.time()
con = rq.request_url(url)
t = time.time() - s
except:
con = None
t = 0
pass
c -= 1
if con:
success += 1
total += t
if success > 0:
proxy['v'] = total / success
def test_proxy_speed(url, proxy, t=60):
req = BasicRequests()
req.set_proxy(proxy, 0, False)
s = time.time()
count = 0
while time.time() - s <= t:
try:
req.request_url(url, timeout=5)
count += 1
except Exception:
pass
return count
#!/usr/bin/env python
# -*- coding:utf8 -*-
from spider.httpreq import BasicRequests
if "__main__" == __name__:
rq = BasicRequests()
rq.set_proxy('106.75.134.190:18888:ipin:ipin1234')
con = rq.request_url('http://www.zjsfgkw.cn/document/JudgmentDetail/4062962')
if con:
print con.text
def __init__(self):
BasicRequests.__init__(self)
self.url = 'http://www.shenpan.cn/cpws/writopenlist.aspx?typeString='
self.params = {}
self.count = 0
def __init__(self):
BasicRequests.__init__(self)
self._url_format = 'http://gaokao.chsi.com.cn/zyk/pub/myd/schAppraisalTop.action?start=%d'
self._url_format2 = 'http://gaokao.chsi.com.cn/zsgs/zhangcheng/listVerifedZszc--method-index,lb-1,start-%d.dhtml'
self._schools = []
def __init__(self, url):
BasicRequests.__init__(self)
self.url = url
'format':
'http://gaokao.chsi.com.cn/zsgs/zhangcheng/listVerifedZszc--method-index,lb-1,start-%d.dhtml',
'url_reg':
r'<a href="\/zsgs\/zhangcheng\/listZszc\-\-schId\-\d+\.dhtml".*>([^<]*)<\/a>',
'page_reg': r'<a[^>]*start\-\d+[^>]*>(\d+)<',
'pagesize': 100
}]
url_format = [
'http://gaokao.chsi.com.cn/zyk/pub/myd/schAppraisalTop.action?start=%d',
'http://gaokao.chsi.com.cn/zsgs/zhangcheng/listVerifedZszc--method-index,lb-1,start-%d.dhtml'
]
url_pattern = [
r'<a href="#" onclick="doDialog\(\'\d+\',\'([^\']+)\'\);',
r'<a href="\/zsgs\/zhangcheng\/listZszc\-\-schId\-\d+\.dhtml".*>([^<]*)<\/a>'
]
page_pattern = [
r'<a[^>]*start=\d+[^>]*>(\d+)<', r'<a[^>]*start\-\d+[^>]*>(\d+)<'
]
request = BasicRequests()
schools = []
for site in sites:
schools += GetSchoolName.fetch_schools(request, site['format'],
site['pagesize'],
site['url_reg'],
site['page_reg'])
ss = []
for c in schools:
if c not in ss:
ss.append(c)
GetSchoolName.save(schools)
def __init__(self):
BasicRequests.__init__(self)
self.url = 'http://www.ftcourt.gov.cn/cpwspt/writopenlist.aspx?cls=0'
self.params = {}
self.count = 0
def __init__(self):
BasicRequests.__init__(self)
self.session = requests.Session()
def test_find(self):
req = BasicRequests()
con = req.request_url(self.main_url + '?key=' + 'mumas')
print con.text
con = req.request_url(self.main_url + '?key=' + 'skiloop')
print con.text
def load_url(url):
br = BasicRequests()
con = br.request_url(url)
print con.text
