def test_parseWeiboLst(self, uid):
fd = '../BigVs/' + str(uid)
if os.path.exists(fd):
with open('../BigVs/' + str(uid), 'r') as f:
rawdoc = f.read()
d = PyQuery(rawdoc.decode('utf-8'))
fg = FigureItem()
fg.follow = d('strong').filter(lambda i, this: PyQuery(this).attr(
'node-type') == 'follow').text()
fg.fans = d('strong').filter(lambda i, this: PyQuery(this).attr(
'node-type') == 'fans').text()
fg.weibo = d('strong').filter(lambda i, this: PyQuery(this).attr(
'node-type') == 'weibo').text()
fg.name = d('span').filter('.name').text()
fg.verify = d('.pf_verified_info').contents()[0]
fg.intro = d('.pf_intro').text()
for i in d('.layer_menulist_tags').items('a'):
fg.tags.append(i.text())
return fg
else:
print 'file not exists'
def _parseHeadinfo(self, doc):
fg = FigureItem()
strimdata = ''
jdiclst = []
scripts = re.findall('<script>FM\.view\((.*)\);?</script>', doc)
if scripts:
for i in scripts:
jdiclst.append( json.loads(i) )
else:
print '_fetch_manload: raw doc parse error'
for jdic in jdiclst:
if 'ns' in jdic:
if jdic['ns'] == 'pl.header.head.index':
strimdata = jdic['html']
d = PyQuery( strimdata )
break
else:
raise Exception('_parseHeadinfo error')
info = self.remoteReader.getDoc( self.remoteReader.makeUrl_hostinfo(self.uid) )
m = re.search(r'注册时间[.\s\S]+(\d{4})-(\d{2})-(\d{2})', info)
if m:
t = time.mktime(time.strptime('%s-%s-%s' % (m.group(1), m.group(2), m.group(3)), '%Y-%m-%d'))
else:
t = 0 #2012-07-06
fg.uid = self.uid
fg.domainid = self.remoteReader.domain
fg.establish = t
fg.follow = re.search(self.followmask, strimdata).group(1)
fg.fans = re.search(self.fansmask, strimdata).group(1)
fg.weibo = re.search(self.weibomask, strimdata).group(1)
text1 = d('span').filter('.name').text()
text2 = d('strong').filter('.W_f20.W_Yahei').text()
if text1:
fg.name = text1
else:
fg.name = text2
try:
fg.verify = d('.pf_verified_info').contents()[0]
except:
fg.verify = ''
fg.intro = d('.pf_intro').text()
for i in d('.layer_menulist_tags').items('a'):
fg.tags.append( i.text() )
if not fg.isValid():
print ' - Thread {0} weibo figure info not enough'.format(self.no)
else:
return fg
def run(self):
fg = FigureItem()
while not self.q.empty():
uid = int(self.q.get()[0])
doc = self.remoteReader.getDoc( self.remoteReader.makeUrl_hostweibo(uid) )
self._parseHeadinfo(doc)
self.remoteReader.finishFetching()
self.localReader.record( fg )
def itemCast(self, row):
fg = FigureItem()
fg.uid = row[0]
fg.domainid = row[1]
fg.name = row[2]
fg.follow = row[3]
fg.fans = row[4]
fg.weibo = row[5]
fg.establish = row[6]
return fg
def __init__(self, queue=None, no=0, skip=False):
super(WeiboFetcher, self).__init__()
self.initRemask()
self.remoteReader = Page()
self.figure = FigureItem()
self.q = queue
self.no = no
self.repeat = 0
self.skip = skip
def __init__(self, queue):
super(FigureFetcher, self).__init__()
self.localReader = FigureDatabase()
self.figure = FigureItem()
self.q = queue