def __init__(self, info):
SessionRequests.__init__(self)
self.info = info
self.select_user_agent('firefox')
self._con = None
self.onl = OnlineOCR(info['prov'])
port = int(info['prov']) / 10000 + 8000
#self.onl.server = "http://win.haohaogame.com:%d/codeocr" % port
self.onl.server = "http://192.168.0.10:%d/codeocr" % port
self.code = None
def test_fetch_paper(url):
rq = SessionRequests()
con = rq.request_url(url)
print con.headers
print con.cookies
print con.text
con = rq.request_url(url)
print con.text
print con.headers
print con.cookies
def check_yantian_cookies():
sq = SessionRequests()
con = sq.request_url(
'http://www.shenpan.cn/cpws/writopenlist.aspx?typeString=')
print con.headers
print con.cookies
# print con.content
yt = YantianGenQueries()
yt.get_form_values()
yt.show()
def __init__(self, thcnt):
Spider.__init__(self, thcnt)
self.request = SessionRequests()
self.view_state = None
self.event_valid = None
self.rand = None
self.loc = "浙江"
self.data_file = FileSaver("浙江_data.txt")
self.have_get_url_file = FileSaver("浙江_get_url.txt")
self.init_already()
self.login("38037395", "773950")
def load_url_with_cookies(url, cookies):
rq = SessionRequests()
for c in cookies:
rq.add_cookie(c['domain'],
c['name'],
c['value'],
path=c['path'],
secure=c['secure'])
con = rq.request_url(url)
if con:
print con.text
print con.cookies
def init_req(self, proxies={}):
req = getattr(self._curltls, "req", None)
if req is None:
req = SessionRequests()
url = urls[random.randrange(0, len(urls))]
while True:
con_init = req.request_url(url, proxies=proxies)
if con_init is None or con_init.code != 200:
print "...初始化失败..."
else:
break
setattr(self._curltls, "req", req)
setattr(self._curltls, "init_url", url)
return req
def get_session_request(self):
sr = SessionRequests()
with self.locker:
if not isinstance(self.networker.sp_proxies, dict) or len(self.networker.sp_proxies.keys()) == 0:
return sr
if self.networker._auto_change_proxy:
prs = self.networker.sp_proxies.keys()
for i in range(0, len(prs)):
self.networker._cur_proxy_index = (self.networker._cur_proxy_index+1) % len(prs)
selproxy = prs[self.networker._cur_proxy_index]
if self.networker.sp_proxies.get(selproxy, 0) <= 10:
sr.set_proxy(selproxy, index=0, auto_change=False)
break
elif self.networker._cur_proxy_index < 0:
pass
# don't auto change proxy, and the index < 0, no proxy is used.
# but don't report an error.
else:
prs = self.networker.sp_proxies.keys()
selproxy = prs[self.networker._cur_proxy_index % len(prs)]
self.networker.set_proxy(selproxy, index=0, auto_change=False)
return sr
def test_post_list():
ll = []
with open('jobs', 'r') as f:
for s in f:
ll.append(s)
l = ll[3]
print l
v = eval(l.encode('utf-8'))
if not isinstance(v, dict):
return
p = copy.deepcopy(v['param'])
if not isinstance(p, dict):
return
print p.keys()
# p['__EVENTARGUMENT'] = ''
p['hdPageIndex'] = '10'
# p.pop('hdPageIndex')
# p.pop('__VIEWSTATE')
if p.has_key('__LASTFOCUS'):
p.pop('__LASTFOCUS')
# p.pop('__EVENTVALIDATION')
p['__EVENTTARGET'] = 'btnNext'
print len(p)
rq = SessionRequests()
# con = rq.request_url(v['url'])
# if con and con.text:
# m = re.search(r'<title.*', con.text)
# if m:
# print m.group()
# else:
# print len(con.text)
data = urlencode(p)
con = rq.request_url(v['url'], data=p)
print p
if con and con.text:
m = re.search(r'<title.*', con.text)
print con.text
def test_post(self):
req = SessionRequests()
res = req.request_url(self.main_url + 'update',
data={
'encrypt':
'LuJAxGaUMqnDOARGzY9zIe0Rd41opkL7',
'key': 'mumas',
'value': '192.168.1.251'
})
print res.headers
print res.text
res = req.request_url(self.main_url + 'update',
data={
'encrypt':
'LuJAxGaUMqnDOARGzY9zIe0Rd41opkL7',
'key': 'mumaas',
'value': '192.168.1.251'
})
print res.headers
print res.text
class check(object):
def __init__(self):
self.sessionReq = SessionRequests()
self.sessionReq.load_proxy('../_51job/proxy')
def start(self):
print 'begin crawler!'
url = r"http://gaokao.chsi.com.cn/zsjh/search.do?ccdm=&jhxzdm=&kldm=&method=majorList&sySsdm=11&year=2016&yxdm=10001"
header = {
"User-Agent":
r"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:39.0) Gecko/20100101 Firefox/39.0",
"Host": r"gaokao.chsi.com.cn",
"Referer": r"http://gaokao.chsi.com.cn/zsjh/"
}
while True:
i = random.randint(1, 60)
print i
con = self.sessionReq.request_url(url, headers=header)
if con is None or con.text.strip() is "":
print "Nothing back!readd job"
time.sleep(i)
continue
if r"302 Found" in con.text:
print "302 Found!"
time.sleep(i)
continue
if r"403 Forbidden" in con.text:
print "403 Forbidden!"
time.sleep(i)
continue
if "专业名称" not in con.text or "计划数" not in con.text:
print "阳光高考还没更新！"
time.sleep(i)
continue
if "专业名称" in con.text and "计划数" in con.text:
print "阳光高考已经更新！"
break
spider.util.sendmail(['*****@*****.**'], '阳光高考可以爬了!', "请迅速去抓该网站！")
def test_session():
cnt = 10
random.seed = int(time.time())
threads = []
req = SessionRequests()
for i in range(0, cnt):
t = threading.Thread(target=_session_test, args=(req, i))
threads.append(t)
for t in threads:
t.start()
time.sleep(5)
for t in threads:
t.join()
def get_code(image_content):
retry = 3
while retry >= 0:
try:
#联众帐号，密码
user = "******"
pwd = "15004622415"
#软件key 或者作者帐号
soft_key = "fpTof8NxP4FTFOTp6Tfi3ik6TxtfEOkE68nE3foK"
s = SessionRequests()
lz = Lianzhong(s, user, pwd, soft_key)
vcode, vid = lz.create(image_content)
return vcode
except:
traceback.print_exc()
retry -= 1
raise RuntimeError()
class X315Spider(Spider):
def init_req(self):
self.sesnreq.load_proxy("../gongshangju/proxy_all.txt")
self.sesnreq.select_user_agent("firefox")
self.sesnreq.request_url("http://www.x315.com/index")
con = self.sesnreq.request_url(
"http://s4.cnzz.com/stat.php?id=1256666136&show=pic1")
jsessionid = con.cookies[0].split("\t")[-1]
jsscript = "var document={};var window = {};" + con.text + "console.log(document.cookie);"
f = open("login.js", "w+b")
f.write(jsscript)
f.close()
os.system("nodejs login.js>cookie.txt")
f = open("cookie.txt", "r+b")
cookiestr = f.read()
self.cookiestr = urllib.unquote(
re.search("(CNZZDATA.*?;)", cookiestr).group(1) + "JSESSIONID=" +
jsessionid + ";")
print self.cookiestr
def __init__(self, threadcnt):
Spider.__init__(self, threadcnt)
self.sesnreq = SessionRequests()
self.sesnreq.load_proxy("../gongshangju/proxy1.txt", 0, False)
self.sesnreq.select_user_agent("firefox")
# self.init_req()
def dispatch(self):
currline = 0
skipto = 0
endline = 100000
with open(os.environ["HOME"] + "/r1.txt", "rb") as f:
while currline < skipto:
line = f.readline()
currline += 1
while currline < endline:
line = f.readline().strip()
key = line.split(" ")[2]
job = {
"type": "t1",
"key": key,
"line": line,
"lineno": currline
}
self.add_main_job(job)
currline += 1
self.add_main_job(None)
def jobrunner1(self, job):
con = self.sesnreq.request_url(
"http://www.x315.com/",
headers={"Accept-Language": "zh-CN,zh;q=0.8,en-US;q=0.5,en;q=0.3"})
# self.sesnreq.request_url("http://s4.cnzz.com/stat.php?id=1256666136&show=pic1")
url = r"http://www.x315.com/quicksearch?qk=%s&t=1&z=×tamp=%s" % (
"富士康", str(time.time()).split(".")[0] +
str(time.time()).split(".")[1][:3])
header = {
"Accept":
r"text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
"Connection": "Keep-alive",
"Content-Type": "application/x-www-form-urlencoded",
"Accept-Language": "zh-CN,zh;q=0.8,en-US;q=0.5,en;q=0.3",
"Referer": "http://www.x315.com/",
"X-Requested-With": "XMLHttpRequest"
}
con = self.sesnreq.request_url(url, headers=header)
print con.text
if ur"查询过于频繁" in con.text:
if not self.re_add_job(job):
Log.error("readd job failed.==>" + utf8str())
Log.info("查询过于频繁,sleep 10 s.")
time.sleep(10)
def __init__(self):
self.sessionReq = SessionRequests()
self.sessionReq.load_proxy('../_51job/proxy')
class GK66(Spider):
"""这是使用自己的多线程框架重构的gk66网站爬取代码，没有针对所有省份进行处理，只能针对特定省份的账号密码进行爬取，验证码也是手动的，未进行处理"""
def __init__(self, thcnt):
Spider.__init__(self, thcnt)
self.request = SessionRequests()
self.view_state = None
self.event_valid = None
self.rand = None
self.loc = "浙江"
self.data_file = FileSaver("浙江_data.txt")
self.have_get_url_file = FileSaver("浙江_get_url.txt")
self.init_already()
self.login("38037395", "773950")
def init_already(self):
cnt = 0
with open("浙江_get_url.txt") as f:
for line in f:
line = line.strip()
have_get_url.add(line)
cnt += 1
print "初始化已经爬过的链接 ", cnt
def login(self, username, password):
# 验证码
captcha_url = "http://14wj.gk66.cn/ashx/code.ashx"
rand = None
while True:
res = self.request.request_url(captcha_url)
if res is not None and res.code == 200:
spider.util.FS.dbg_save_file("captcha.jpeg", res.content)
rand = raw_input("login －－－＞ 请输入验证码：") #get_code(res.text) #
if rand == "retry":
continue
#break
login_url = "http://14wj.gk66.cn/ashx/login.ashx"
data = {
"username": username,
"password": password,
"rand": rand,
"rempass": "******"
}
while True:
con = self.request.request_url(login_url, data=data)
if con is None or con.code != 200:
continue
content = con.content
r_type = eval(content)[0]["type"]
if r_type == "2":
print "登陆失败：", spider.util.utf8str(content)
if u"验证码错误".encode("gb2312") in content:
break
continue
else:
self.view_state, self.event_valid = self.prepare_param(
)
self.rand = self.prepare_rand()
if self.view_state is None or self.event_valid is None or self.rand is None:
print "－－－未获取到view_state event_valid rand －－－"
continue
print "登陆成功..."
return
continue
else:
print "login 获取验证码图片失败,", "res is None" if res is None else "res.code = %d" % res.code
# login_url = "http://14wj.gk66.cn/ashx/login.ashx"
# data = {"username": username,
# "password": password,
# "rand": rand,
# "rempass": "******"}
# while True:
# con = self.request.request_url(login_url, data=data)
# if con is None or con.code != 200:
# continue
# content = con.content
# r_type = eval(content)[0]["type"]
# if r_type == "2":
# print "登陆失败：", spider.util.utf8str(content)
# continue
# else:
# self.view_state, self.event_valid = self.prepare_param()
# self.rand = self.prepare_rand()
# if self.view_state is None or self.event_valid is None or self.rand is None:
# print "－－－未获取到view_state event_valid rand －－－"
# continue
# print "登陆成功..."
# break
def prepare_param(self):
search_url = "http://14wj.gk66.cn/wj/fs.aspx"
res = self.request.request_url(search_url)
if res is not None and res.code == 200:
fs_page = res.content
soup = BeautifulSoup(fs_page, 'html5lib')
view_state = soup.find(attrs={'id': "__VIEWSTATE"}).get("value")
event_valid = soup.find(attrs={
'id': "__EVENTVALIDATION"
}).get("value")
print "view_state=", view_state, "event_valid=", event_valid
return view_state, event_valid
return None, None
def prepare_rand(self):
global FILE_NAME_1
captcha_url_2 = "http://14wj.gk66.cn/ashx/codewj.ashx"
while True:
con = self.request.request_url(captcha_url_2)
if con is None or con.code != 200:
print "prepare_rand 请求错误...", "结果为空" if con is None else "http code = %d" % con.code
continue
spider.util.FS.dbg_save_file("captcha2.jpeg", con.content)
rand = raw_input(
"prepare_rand －－－＞ 请输入验证码：") #get_code(con.content)
if rand == "retry":
continue
return rand
def logout(self):
logout_url = "http://www.gk66.cn/loginout.aspx"
self.request.request_url(logout_url)
def wait_q_breakable(self):
lt = 0
while True:
if not self.job_queue.empty() or not self.job_queue2.empty(
) or not self.job_queue3.empty():
time.sleep(5)
if time.time() < lt + 1 and self._running_count == 0:
return True
time.sleep(2)
lt = time.time()
if self._worker_count == 0:
return False
def dispatch(self):
nf_list = ["14"] #["06" , "07", "08", "09", "10", "11"] # 年份
wl_list = ["w", "l"] # 文科 理科
bz_list = ["b", "z"] # 本科 专科
for fs in range(732, 810):
print "分数：", fs
for nf in nf_list:
for wl in wl_list:
for bz in bz_list:
data = {
"fs": fs,
"nf": nf,
"wl": wl,
"bz": bz,
"pc": "",
"ImageButton1.x": 98,
"ImageButton1.y": 13
}
#print "合成请求数据：", data
job = {"data": data}
self.add_main_job(job)
time.sleep(0.1)
self.wait_q_breakable()
self.add_job(None, True)
def run_job(self, job):
data = job["data"]
data["__VIEWSTATE"] = self.view_state
data["__EVENTVALIDATION"] = self.event_valid
data["rand"] = self.rand
retry = 3
while retry >= 0:
try:
self.loop_exec(data)
break
except:
traceback.print_exc()
print "出错,sleep 1s"
time.sleep(1)
retry -= 1
try:
self.logout()
except:
pass
self.login("38037395", "773950")
def loop_exec(self, data):
try:
while True:
url = self.build_search_url(data)
if url != None:
break
else:
self.login("38037395", "773950")
except Exception as e:
print "build_search_url failure ...", e
return
page_break = False
last_v = {}
if url in have_get_url:
print "已经爬取，pass"
return
if "http://14wj.gk66.cn/login.aspx?" in url:
raise RuntimeError()
for page in range(1, 1000):
if page_break:
break
exec_url = url + "&s=0&page=" + str(page)
print "执行链接：", exec_url
datas = self.get_score_data(exec_url, page_break=page_break)
if len(datas) < 20:
page_break = True
for v in datas:
if v is None:
page_break = True
break
v["location"] = self.loc
v["year"] = data["nf"]
v["wl"] = data["wl"]
v["bz"] = data["bz"]
if (str(last_v) == str(v)):
page_break = True
break
last_v = v
k = {
"location": v["location"],
"school": v["school"],
"spec": v["spec"],
"batch": v["batch"], # 批次
"score": v["score"],
"year": v["year"],
"wl": v["wl"],
"bz": v["bz"]
}
print v
self.data_file.append(spider.util.utf8str(v))
#store_score(k, v)
self.recorde_spided(url)
def recorde_spided(self, url):
self.have_get_url_file.append(url)
have_get_url.add(url)
def get_score_data(self, data_url, page_break=False):
try:
page_content = None
while True:
res = self.request.request_url(data_url)
if res is not None and res.code == 200:
page_content = res.content
if u"对不起,请先登录".encode("gb2312") in page_content:
self.logout()
self.login("38037395", "773950")
continue
break
else:
print "获取页面出错＞．．", "res is None" if res is None else "res.code == %d " % res.code
continue
datas = []
if string.find(page_content, u"相近分数".encode("gb2312")) > 0:
print "该页面没有数据"
return datas
soup = BeautifulSoup(page_content, 'html5lib')
rows = soup.findAll("tr")
if rows is not None and len(rows) > 0:
if len(rows) != 20:
page_break = True
for row in rows:
cols = row.findAll("td")
if cols is not None and len(cols) == 13:
data = {}
data["school"] = cols[0].getText()
data["spec"] = cols[1].getText()
data["rank"] = cols[2].getText()
data["score"] = cols[3].getText()
data["batch"] = cols[4].getText()
data["score_number"] = cols[5].getText()
data["spec_number"] = cols[6].getText()
data["high_score"] = cols[7].getText()
data["high_score_rank"] = cols[8].getText()
data["low_score"] = cols[9].getText()
data["low_score_rank"] = cols[10].getText()
data["average_score"] = cols[11].getText()
data["average_score_rank"] = cols[12].getText()
datas.append(data)
return datas
else:
print "页面无内容：", page_content
except Exception as e:
print "get_score_data 发生异常", e
return None
def build_search_url(self, data):
search_url = "http://14wj.gk66.cn/wj/fs.aspx"
headers = {
"User-Agent":
"Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; Trident/4.0; .NET CLR 2.0.50727; .N ET4.0C; .NET4.0E)"
}
resp = self.request.request_url(search_url, data=data, headers=headers)
headers = resp.headers
m = re.search("Location:(.*)\r\n", headers)
if m:
location = m.group(1).strip()
return "http://14wj.gk66.cn" + location
else:
return None
#location = resp.headers["Location"]
def store_score(self, value):
print filename + ' being write-->', value
obj = Score_gk66.objects(
location=value["location"],
year=value["year"],
bz=value["bz"],
wl=value["wl"],
school=value['school'],
spec=value['spec'],
rank=value['rank'],
score=value['score'],
batch=value["batch"],
score_number=value['score_number'],
spec_number=value['spec_number'],
high_score=value['high_score'],
high_score_rank=value['high_score_rank'],
low_score=value['low_score'],
low_score_rank=value['low_score_rank'],
average_score=value['average_score'],
average_score_rank=value['average_score_rank']).no_cache().timeout(
False).first()
if not obj:
obj = Score_gk66(location=value["location"],
year=value["year"],
bz=value["bz"],
wl=value["wl"],
school=value['school'],
spec=value['spec'],
rank=value['rank'],
score=value['score'],
batch=value["batch"],
score_number=value['score_number'],
spec_number=value['spec_number'],
high_score=value['high_score'],
high_score_rank=value['high_score_rank'],
low_score=value['low_score'],
low_score_rank=value['low_score_rank'],
average_score=value['average_score'],
average_score_rank=value['average_score_rank'])
obj.save()
self.num_count += 1
print "保存成功：", value
else:
print u"数据已存在"
def event_handler(self, evt, msg, **kwargs):
if evt == 'DONE':
spider.util.sendmail('*****@*****.**', '%s DONE' % sys.argv[0],
msg)
class YggaokaoRetry(Spider):
def __init__(self, threadcnt):
super(Yggaokao, self).__init__(threadcnt)
self.sessionReq = SessionRequests()
def dispatch(self):
f = open("prov_list", "r+b")
currline = 0
skipto = 0
endline = 10000
for line in f:
currline += 1
if currline >= skipto:
sySsdm = line.split(" ")[0].strip()
job = {
"sySsdm": sySsdm,
"year": "2014",
"start": 0,
"type": "u1"
}
self.add_main_job(job)
if currline >= endline:
break
self.wait_q()
self.add_job(None, True)
def retry(self, con, job):
if re.search(u'<h1>An error occurred.</h1>', con.text) or re.search(
u'Tinyproxy was unable to', con.text):
#should reload this page.
if int(job["retrycnt"]) < 5:
job["retrycnt"] = int(job["retrycnt"]) + 1
self.add_job(job)
return True
return False
def save_sch_list(self, job, res):
with self.locker:
fr = open("prov/" + job["sySsdm"] + ".txt", "r+b")
f = open("prov/" + job["sySsdm"] + ".txt", "a+b")
schlist = re.findall(
r'<tr bgcolor="#FFFFFF" onMouseOver="this.style.background=\'#FFFFEE\'" onMouseOut=\"this.'
r'style.background=\'#ffffff\'">(.*?)</tr>', res, re.S)
for schinfo in schlist:
schstr = ""
tds = re.findall(r"<td.*?>(.*?)</td>", schinfo, re.S)
if len(tds) is 0:
spider.runtime.Log.error(
job["sySsdm"] + ", start at " + str(job["start"]) +
" match error! No td tag! Readd..\n")
self.re_add_job(job)
ferr = open(
"errors/" + job["sySsdm"] + "_" + str(job["start"]) +
".html", "w+b")
ferr.write(res)
ferr.close()
f.close()
return
schnamelist = re.findall(r'dhtml">(.*?)</a>', tds[0], re.S)
if len(schnamelist) is 0:
schnamelist = []
schnamelist.append(tds[0].strip())
if schnamelist[0] is "":
spider.runtime.Log.error(
job["sySsdm"] + ", start at " + str(job["start"]) +
" match error! No school name! Readd..\n")
self.re_add_job(job)
ferr = open(
"errors/" + job["sySsdm"] + "_" +
str(job["start"]) + ".html", "w+b")
ferr.write(res)
ferr.close()
f.close()
return
schname = schnamelist[0]
if schname in fr.read():
print 'skip...', schname, ", in", job["sySsdm"], ".txt"
fr.close()
f.close()
return
schstr += schname
if r'span985">985</span>' in tds[0]:
schstr += " 985"
if r'span211">211</span>' in tds[0]:
schstr += " 211"
if r'spanyan">研</span>' in tds[0]:
schstr += " 研"
for i in range(len(tds))[1:(len(tds) - 1)]:
schstr += " " + tds[i]
stucnt = re.findall(r"doDialog.*?\">(.*?)</a>",
tds[len(tds) - 1], re.S)[0].strip()
schstr += " " + stucnt
f.write(schstr + "\n")
f.flush()
f.close()
def save_sch_detail(self, job, res):
if not os.path.exists("detail/" + job["sySsdm"]):
os.makedirs("detail/" + job["sySsdm"])
f = open("detail/" + job["sySsdm"] + "/" + job["yxdm"] + ".html",
"w+b")
f.write(res)
f.flush()
f.close()
def check_should_fetch(self, job):
if (job["type"] is "u1"):
return True
else:
if os.path.exists("detail/" + job["sySsdm"] + r"/" + job["yxdm"] +
".html"):
return False
else:
return True
def run_job(self, job):
if job["type"] is "u1":
print "searching %s, start at %d" % (job["sySsdm"], job["start"])
url = "http://gaokao.chsi.com.cn/zsjh/searchZsjh.do?ccdm=&jhxzdm=&kldm=&searchType=1&ssdm=&" \
"sySsdm=%s&year=%s&yxmc=&start=%d" % (job["sySsdm"], job["year"], job["start"])
header = {
"User-Agent":
r"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.73 Safari/537.36",
"Referer": r"http://gaokao.chsi.com.cn/zsjh/zsjh2014.jsp",
"Origin": r"http://gaokao.chsi.com.cn",
"X-Requested-With": r"XMLHttpRequest",
"Pragma": r"no-cache"
}
con = self.sessionReq.request_url(url, headers=header)
if con is None or con.text.strip() == "":
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" nothing return ! Readd..\n")
time.sleep(10)
self.re_add_job(job)
return
res = con.text
if r"302 Found" in res:
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" 302 Found ! Readd..\n")
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
elif re.search(u'无查询结果', res):
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" no data! Readd..\n")
self.re_add_job(job)
return
elif re.search(ur'<H1>错误</H1>', res):
time.sleep(3)
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" error occur ! Readd..\n")
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
else:
if int(job["start"]) is 0:
m = re.search(r"if \(Num > (\d+)", res)
if m:
pgcnt = int(m.group(1))
while pgcnt > 1:
jobb = {
"sySsdm": job["sySsdm"],
"year": job["year"],
"start": (pgcnt - 1) * 20,
"type": "u1"
}
self.add_job(jobb)
pgcnt -= 1
else:
spider.runtime.Log.error(job["sySsdm"] +
", start at " +
str(job["start"]) +
" no more page! Readd!\n")
self.re_add_job(job)
return
yxdms = re.findall(r"doDialog\('(\d+)'", res, re.S)
if len(yxdms) == 0:
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" no url! Readd.\n")
self.re_add_job(job)
return
for yxdm in yxdms:
job2 = {
"yxdm": yxdm,
"sySsdm": job["sySsdm"],
"year": "2014",
"type": "u2"
}
if not self.check_should_fetch(job2):
print "skip...", job['sySsdm'], "/", job["yxdm"]
else:
self.add_job(job2)
self.save_sch_list(job, res)
elif job["type"] is "u2":
url = r"http://gaokao.chsi.com.cn/zsjh/search.do?" \
r"ccdm=&jhxzdm=&kldm=&method=majorList&sySsdm=%s&year=%s&yxdm=%s" % (job["sySsdm"], job["year"], job["yxdm"])
header = {
"User-Agent":
r"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.73 Safari/537.36"
}
con = self.sessionReq.request_url(url, headers=header)
if con is None or con.text.strip() is "":
print "Nothing back!readd job" + job.__str__()
time.sleep(10)
self.re_add_job(job)
return
if r"302 Found" in con.text:
spider.runtime.Log.error(job.__str__() +
" 302 Found ! Readd..\n")
self.re_add_job(job)
return
res = con.text
self.save_sch_detail(job, res)
def __init__(self, threadcnt):
Spider.__init__(self, threadcnt)
self.sesnreq = SessionRequests()
self.sesnreq.load_proxy("../gongshangju/proxy1.txt", 0, False)
self.sesnreq.select_user_agent("firefox")
def __init__(self, threadcnt):
super(Yggaokao, self).__init__(threadcnt)
self.sessionReq = SessionRequests()
self.sessionReq.load_proxy('proxy')
class Yggaokao(Spider):
def __init__(self, threadcnt):
super(Yggaokao, self).__init__(threadcnt)
self.sessionReq = SessionRequests()
self.sessionReq.load_proxy('proxy')
def dispatch(self):
f = open("prov_list", "r+b")
currline = 0
skipto = 0
endline = 100
for line in f:
currline += 1
if currline >= skipto:
sySsdm = line.split(" ")[0].strip()
job = {
"sySsdm": sySsdm,
"year": "2015",
"start": 0,
"type": "u1"
}
self.add_main_job(job)
if currline >= endline:
break
self.wait_q()
self.add_job(None, True)
def check_sch_list(self, url, flag=True):
if not os.path.exists("prov"):
os.makedirs("prov")
if not os.path.exists("prov/check.txt"):
f = open('prov/check.txt', 'w')
f.close()
if (flag):
with open("prov/check.txt") as file_:
for line in file_:
if url in line:
return False
return True
f = open("prov/check.txt", "a+")
f.write(url + "\n")
f.close()
def check_sch_detail(self, job):
if not os.path.exists("detail/" + job["sySsdm"]):
os.makedirs("detail/" + job["sySsdm"])
return True
files = os.listdir("detail/" + job["sySsdm"])
if files.count(job["yxdm"] + ".html") == 0:
return True
return False
def retry(self, con, job):
if re.search(u'<h1>An error occurred.</h1>', con.text) or re.search(
u'Tinyproxy was unable to', con.text):
#should reload this page.
if int(job["retrycnt"]) < 5:
job["retrycnt"] = int(job["retrycnt"]) + 1
self.add_job(job)
return True
return False
def save_sch_list(self, job, res):
if not os.path.exists("prov"):
os.makedirs("prov")
with self.locker:
f = open("prov/" + job["sySsdm"] + ".txt", "a+b")
schlist = re.findall(
r'<tr bgcolor="#FFFFFF" onMouseOver="this.style.background=\'#FFFFEE\'" onMouseOut=\"this.'
r'style.background=\'#ffffff\'">(.*?)</tr>', res, re.S)
for schinfo in schlist:
schstr = ""
tds = re.findall(r"<td.*?>(.*?)</td>", schinfo, re.S)
if len(tds) is 0:
spider.runtime.Log.error(
job["sySsdm"] + ", start at " + str(job["start"]) +
" match error! No td tag! Readd..\n")
self.re_add_job(job)
ferr = open(
"errors/" + job["sySsdm"] + "_" + str(job["start"]) +
".html", "w+b")
ferr.write(res)
ferr.close()
f.close()
return
schnamelist = re.findall(r'dhtml">(.*?)</a>', tds[0], re.S)
if len(schnamelist) is 0:
schnamelist = []
schnamelist.append(tds[0].strip())
if schnamelist[0] is "":
spider.runtime.Log.error(
job["sySsdm"] + ", start at " + str(job["start"]) +
" match error! No school name! Readd..\n")
self.re_add_job(job)
ferr = open(
"errors/" + job["sySsdm"] + "_" +
str(job["start"]) + ".html", "w+b")
ferr.write(res)
ferr.close()
f.close()
return
schname = schnamelist[0]
schstr += schname
if r'span985">985</span>' in tds[0]:
schstr += " 985"
if r'span211">211</span>' in tds[0]:
schstr += " 211"
if r'spanyan">研</span>' in tds[0]:
schstr += " 研"
for i in range(len(tds))[1:(len(tds) - 1)]:
schstr += " " + tds[i]
stucnt = re.findall(r"doDialog.*?\">(.*?)</a>",
tds[len(tds) - 1], re.S)[0].strip()
schstr += " " + stucnt
f.write(schstr + "\n")
f.flush()
f.close()
def save_sch_detail(self, job, res):
if not os.path.exists("detail/" + job["sySsdm"]):
os.makedirs("detail/" + job["sySsdm"])
f = open("detail/" + job["sySsdm"] + "/" + job["yxdm"] + ".html",
"w+b")
f.write(res)
f.flush()
f.close()
def run_job(self, job):
i = random.randint(15, 30)
if job["type"] is "u1":
print "searching %s, start at %d" % (job["sySsdm"], job["start"])
url = "http://gaokao.chsi.com.cn/zsjh/searchZsjh--year-%s,searchType-1,sySsdm-%s,start-%d.dhtml" % (
job["year"], job["sySsdm"], job["start"])
#url = "http://gaokao.chsi.com.cn/zsjh/searchZsjh.do?ccdm=&jhxzdm=&kldm=&searchType=1&ssdm=&" \
# "sySsdm=%s&year=%s&yxmc=&start=%d" % (job["sySsdm"], job["year"], job["start"])
header = {
"User-Agent":
r"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:39.0) Gecko/20100101 Firefox/39.0",
#"Referer":r"http://gaokao.chsi.com.cn/zsjh/",
#"Origin":r"http://gaokao.chsi.com.cn",
"Host": r"gaokao.chsi.com.cn",
"Accept":
"text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
"Accept - Language": "zh-CN,zh;q=0.5",
"Accept - Encoding": "gzip, deflate",
"Connection": "keep-alive",
"Cache - Control": "max-age=0"
#"X-Requested-With":r"XMLHttpRequest",
#"Pragma":r"no-cache"
}
con = self.sessionReq.request_url(url, headers=header)
if con is None or con.text.strip() == "":
print "result is None!"
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" nothing return ! Readd..\n")
time.sleep(i)
self.re_add_job(job)
return
res = con.text
if r"302 Found" in res:
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" 302 Found ! Readd..\n")
time.sleep(i)
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
elif r"403 " in con.text:
print "403 Forbidden (操作被拒绝) 列表页"
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
"403 Forbidden ! Readd..\n")
time.sleep(i)
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
elif re.search(u'无查询结果', res):
print "无查询结果!"
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" no data! Readd..\n")
time.sleep(i)
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
elif re.search(ur'<H1>错误</H1>', res):
print "<H1>错误</H1>!"
time.sleep(i)
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" error occur ! Readd..\n")
self.re_add_job(job)
firstCraw = True
setattr(self._curltls, 'firstCraw', firstCraw)
return
else:
if int(job["start"]) is 0:
m = re.search(r"if \(Num > (\d+)", res)
if m:
pgcnt = int(m.group(1))
while pgcnt > 1:
jobb = {
"sySsdm": job["sySsdm"],
"year": job["year"],
"start": (pgcnt - 1) * 20,
"type": "u1"
}
self.add_job(jobb)
pgcnt -= 1
#else:
# spider.runtime.Log.error(job["sySsdm"]+", start at "+ str(job["start"]) + " no more page! Readd!\n")
# self.re_add_job(job)
yxdms = re.findall(r"doDialog\('(\d+)'", res, re.S)
if len(yxdms) == 0:
spider.runtime.Log.error(job["sySsdm"] + ", start at " +
str(job["start"]) +
" no url! Readd.\n")
time.sleep(i)
self.re_add_job(job)
return
for yxdm in yxdms:
job2 = {
"yxdm": yxdm,
"sySsdm": job["sySsdm"],
"year": "2015",
"type": "u2"
}
self.add_job(job2)
if self.check_sch_list(url):
self.save_sch_list(job, res)
self.check_sch_list(url, False)
else:
print "该列表页已抓取过！"
elif job["type"] is "u2":
url = r"http://gaokao.chsi.com.cn/zsjh/search.do?" \
r"ccdm=&jhxzdm=&kldm=&method=majorList&sySsdm=%s&year=%s&yxdm=%s" % (job["sySsdm"], job["year"], job["yxdm"])
header = {
"User-Agent":
r"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:39.0) Gecko/20100101 Firefox/39.0",
"Host": "gaokao.chsi.com.cn",
"Referer": "http://gaokao.chsi.com.cn/zsjh/",
"Accept":
"text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
"Accept - Language": "zh-CN,zh;q=0.5",
"Accept - Encoding": "gzip, deflate",
"Connection": "keep-alive",
"Cache - Control": "max-age=0"
}
if self.check_sch_detail(job):
con = self.sessionReq.request_url(url, headers=header)
if con is None or con.text.strip() is "":
print "Nothing back!readd job" + job.__str__()
time.sleep(i)
self.re_add_job(job)
return
if r"302 Found" in con.text:
print "302 Found!"
spider.runtime.Log.error(job.__str__() +
" 302 Found ! Readd..\n")
time.sleep(i)
self.re_add_job(job)
return
if r"403 " in con.text:
print "403 Forbidden (操作被拒绝) 详情页"
spider.runtime.Log.error(job.__str__() +
" 302 Found ! Readd..\n")
time.sleep(i)
self.re_add_job(job)
return
res = con.text
self.save_sch_detail(job, res)
else:
print job["sySsdm"] + ":" + str(
job["yxdm"]) + "-----该学校已抓取过！！！"