def __init__(self, name, password, *args, **kwargs):
super(WeiboWapHomepage, self).__init__(*args, **kwargs)
self.start_urls = ["http://weibo.com"]
self.allowed_domains = ["weibo.com"]
self.weibo = WeiboRequests()
self.session, self.uid = self.weibo.login(name, password)
cookiejar = requests.utils.dict_from_cookiejar(self.session.cookies)
# Set sina weibo cookie
self.cookie = {
"ALF": cookiejar["ALF"],
"sso_info": cookiejar["sso_info"],
"SUB": cookiejar["SUB"],
"SUBP": cookiejar["SUBP"],
"SUE": cookiejar["SUE"],
"SUHB": cookiejar["SUHB"],
"SUP": cookiejar["SUP"],
"SUS": cookiejar["SUS"],
}
class WeiboWapHomepage(Spider):
name = "weibo_wap_homepage"
def __init__(self, name, password, *args, **kwargs):
super(WeiboWapHomepage, self).__init__(*args, **kwargs)
self.start_urls = ["http://weibo.com"]
self.allowed_domains = ["weibo.com"]
self.weibo = WeiboRequests()
self.session, self.uid = self.weibo.login(name, password)
cookiejar = requests.utils.dict_from_cookiejar(self.session.cookies)
# Set sina weibo cookie
self.cookie = {
"ALF": cookiejar["ALF"],
"sso_info": cookiejar["sso_info"],
"SUB": cookiejar["SUB"],
"SUBP": cookiejar["SUBP"],
"SUE": cookiejar["SUE"],
"SUHB": cookiejar["SUHB"],
"SUP": cookiejar["SUP"],
"SUS": cookiejar["SUS"],
}
def start_requests(self):
# Parse weibo homepage
home_url = "http://weibo.cn/u/%s" % self.uid
yield Request(url=home_url, cookies=self.cookie, callback=self._parse_homepage, errback=self._parse_fail)
def _parse_homepage(self, response):
"""
Parse the home page of user's weibo and get some parameters
:param response: Http response
:return: Some parameter
"""
print "Parse weibo homepage."
user_name = response.xpath('//div[@class="ut"]/text()').extract()
param = response.xpath('//div[@class="tip2"]/a/text()').extract()
# param = re.findall(u'\d+', "".join(param).decode("utf-8"))
param = re.findall(u"\d+", "".join(param))
item = ParamItem(name=user_name[0], weibo_num=param[0], fellowing_num=param[1], fellower_num=param[2])
print "Parse weibo content."
weibo_num = param[0]
print weibo_num
for page in range(0, (int(weibo_num)) / 10 + 1):
content_url = "http://weibo.cn/%s/profile?page=%s" % (self.uid, page + 1)
yield Request(
url=content_url.encode("utf-8"),
meta={"uid": self.uid},
cookies=self.cookie,
dont_filter=True,
callback=self._parse_weibo_content,
errback=self._parse_fail,
)
def _parse_weibo_content(self, response):
"""
Parse the home page html and get some important parameter.
:param response: http response
:return: a list of parameters
"""
print "parse weibo content."
uid = str(response.meta["uid"]).decode("utf-8")
contents = response.xpath('//div[@class="c"]/div/span[@class="ctt"]').extract()
feeds = response.xpath('//div[@class="c"]/div[last()]').extract() # last():the last elements
# parse the weibo content
contents_array = []
for content in contents:
p = re.compile(u"<[^>]+>") # Filter html tag
weibo_content = p.sub("", content)
contents_array.append(weibo_content)
# Get the number of good, retweet and reply. Get data time.
feeds_array = []
for feed in feeds:
# good = re.search(u"赞\[\d+\]", feed.decode("utf-8"))
# retweet = re.search(u"转发\[\d+", feed.decode("utf-8"))
# reply = re.search(u"评论\[\d+", feed.decode("utf-8"))
# data_time = re.search(u"\d+-\d+-\d+ \d+:\d+:\d+|\d+分钟前|\d+月\d+日 \d+:\d+|今天 \d+:\d+", feed.decode("utf-8"))
good = re.search(u"赞\[\d+\]", feed)
retweet = re.search(u"转发\[\d+", feed)
reply = re.search(u"评论\[\d+", feed)
data_time = re.search(u"\d+-\d+-\d+ \d+:\d+:\d+|\d+分钟前|\d+月\d+日 \d+:\d+|今天 \d+:\d+", feed)
# Format data time
data_type = "".join(re.findall(u"[\u4e00-\u9fa5]+", data_time.group(0)))
if data_type.decode("utf-8") == u"今天":
data_time = data_time.group(0).replace(u"今天 ", "")
cur_time = time.localtime()
month = str(cur_time.tm_mon / 10) + str(cur_time.tm_mon - cur_time.tm_mon / 10)
data_time = "%s-%s-%s %s:%s" % (cur_time.tm_year, month, cur_time.tm_mday, data_time, cur_time.tm_sec)
elif data_type.decode("utf-8") == u"分钟前":
data_time = data_time.group(0).replace(u"分钟前", "")
data_time = int(data_time)
data_time = time.localtime(time.time() - int(data_time) * 60)
data_time = time.strftime("%Y-%m-%d %H:%M:%S", data_time)
elif data_type.decode("utf-8") == u"月日":
data_time = data_time.group(0).replace(u"月", "-").replace(u"日", "")
cur_time = time.localtime()
data_time = "%s-%s:%s" % (cur_time.tm_year, data_time, cur_time.tm_sec)
else:
data_time = data_time.group(0)
print data_time
# Get the number of retweet, good, reply, feed
if retweet and good and reply:
p = re.compile(u"\d+")
# retweet_num = p.search(retweet.group(0).decode("utf-8")).group(0)
# good_num = p.search(good.group(0).decode("utf-8")).group(0)
# reply_num = p.search(reply.group(0).decode("utf-8")).group(0)
retweet_num = p.search(retweet.group(0)).group(0)
good_num = p.search(good.group(0)).group(0)
reply_num = p.search(reply.group(0)).group(0)
feeds_array.append([reply_num, retweet_num, good_num])
# Pipeline item
if len(contents_array) == len(feeds_array):
for i in range(0, len(contents_array)):
weibo_item = WeiboItem()
weibo_item["retweets"] = feeds_array[i][1]
weibo_item["replies"] = feeds_array[i][0]
weibo_item["goods"] = feeds_array[i][2]
weibo_item["content"] = contents_array[i]
weibo_item["uid"] = uid
yield weibo_item
# def parse_weibo(self, response):
# params = response
# sel = Selector(response)
# weibo_content = response.selector.xpath('//div[@class="c"]/div/span/text()').extract()
# for item in weibo_content:
# item = item.encode("utf-8")
# try:
# cursor = self.db.cursor()
# sql = u" INSERT INTO weibo(contant) VALUES(%s)"
# temp = item.decode("utf-8")
# cursor.execute(sql, item)
# chinese = u"[\u4e00-\u9fa5]+"
# pattern = re.compile(chinese)
# all = pattern.findall(temp)
# for i in all:
# print i.encode("utf-8")
# self.db.commit()
# except:
# print "Could not insert an item into mysql."
def _parse_photo(self, response):
"""
Parse and download some photos
:param response: Http response
:return: None
"""
photo_url = response.xpath('//a[contains(@href, "album")]/img/@src').extract()
i = 10
for photo in photo_url:
photo = photo.encode("utf-8").replace("square", "large")
conn = urllib.urlopen(photo)
f = open("lady_professor/pic_" + str(i) + ".jpg", "wb")
i += 1
f.write(conn.read())
f.close()
# print photo
def _get_weibo_url(self, pid, page):
"""
Get the weibo url to searcg
:param pid:
:param page:
:return:
"""
search_url = []
url = "http://weibo.cn/%s/profile?page=%s"
for i in range(0, page):
search_url.append(url % (pid, i + 1))
return search_url
def get_albums_url(self, pid, page):
albums_url = "http://photo.weibo.com/%s/talbum/index#!/mode/1/page/%s"
return albums_url % (pid, page)
def _parse_fail(self, response):
logging.log(logging.ERROR, "Fail to parse the http response file.")
# def process_request(self, request):
# request = request.replace(**{'cookies': self.cookie})
# return request
def _parse_friend_pid(self, response):
"""
Get friends' pid from the login page
:param response: http response
:return: list
"""
uid_url = response.xpath('//td[contains(@valign, "top")]/a[contains(@href, "del?uid=")]/@href').extract()
user_name = response.xpath(
'//td[contains(@valign, "top")]/a[contains(@href, "http://weibo.cn")][1]/text()'
).extract()
if len(uid_url) == len(user_name):
for i in range(0, len(user_name)):
uid = re.search("\d+", uid_url[i]).group(0)
name = user_name[i]
url = "http://weibo.cn/u/%s" % uid # Reform the home page url using uid
# Parse the home page of friends's content
print uid, name, url
else:
logging.log(logging.ERROR, "error in parse uid.")