def enterprise_record_raw_1_function(conn, start, end):
'''
规则：看字段转换表内内容
把enterprise_record_raw记录转换转入自己表中的其他字段（轻度数据提取）
'''
tableName = 'enterprise_record_raw_1'
cursor = conn.cursor()
cursor2 = conn.cursor()
qSql = "select id, eid, enterprise_name, content from " + tableName + ' where id >='+str(start)+' and id <='+str(end)
cursor.execute(qSql)
print "start extract data in "+str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
count = 0
while 1:
record = cursor.fetchone()
if not record or record[0]>end:
continue
count += 1
id = record[0]
eid = record[1]
content = record[3].strip()
# print id
# print eid
clist = content.split('r_obj = t_obj.add_record(true);\n')
print '企业名称：'+ record[2]+', eid:'+str(eid)+', id:'+str(id)
tmp = clist[0].split('：')
publisher = tmp[0].split(',')[-1].strip()
print publisher
try:
category = tmp[1].split(',')[0].strip()
print category
except:
continue
if len(clist)<2:
recs = clist[0]
else:
recs = ''.join(clist[1].split(', false);'))
tmp = recs.split('\n')
for i in range(len(tmp)):
tt = tmp[i].split(',')
del tt[0]
tmp[i] = ''.join(tt)
records = ';'.join(tmp)
records = jTool.clearX(['(', ')', 'true'], records).strip()
print records
print '*'*30
tmpDic = {}
tmpDic['publisher'] = publisher
tmpDic['category'] = category
tmpDic['records'] = records
where = ' where id = '+str(id)
jTool.updateData(cursor2, where, tableName, tmpDic)
tmpDic = {}
conn.commit()
print "complete extract "+str(count)+" records in "+str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
cursor2.close()
cursor.close()
conn.close()
return True
def cutContent(content):
'''
剪切post获取的页面内容
'''
result = ' '
if content:
start = jTool.getStrIndex('function my_init()', content)
end = jTool.getStrIndex('</script>', content)
if start>0 and end>0:
result = jTool.getBetween(start, end, 'function my_init()', content)
result = jTool.clearX(["'"], result)
return result
def crawlUrl(conn, param, proxyList):
'''
访问url并获得企业基本信息和信用记录并入库
'''
cursor2 = conn.cursor()
cursor3 = conn.cursor()
pcount = len(proxyList)-1
proxy = proxyList[random.randint(0,pcount)]
content = jTool.getContentByProxy(proxy, param['url'])
if not content:
jTool.logError('Fail to get page content, url:'+ param['url'])
return False
dataSupplier = "法院记录/工商记录/国税记录/质监记录/经信记录/安监记录/统计记录/环保记录/民政记录/司法记录/劳动记录/建设记录/国土记录/交通记录/发改记录/信息产业/科技记录/农业记录/林业记录/海洋渔业/物价记录/食品药品/文化记录/出版记录/广电记录/公安记录/外贸记录/外汇记录/海关记录/检验检疫/人防记录/证监记录/银监记录/保监记录/金融记录/其他记录/行业协会/机构评级/社会中介/阿里巴巴/企业自报/投诉记录/异议记录"
post_data_dic = {'corpName': param['ename'], 'creditID': param['eid'], 'dataSupplier': dataSupplier, 'isAllInfo': 'False', 'organizeCode': '', 'returnFunction': 'parent.putDatasAndLoad'}
recEntBaseDic = getEntBase(content)
if not recEntBaseDic:
return False
recEntBaseDic['url'] = param['url']
recEntBaseDic['eid'] = param['eid']
recEntBaseDic['enterprise_name'] = param['ename'].strip('')
recEntBaseDic['ctime'] = str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
try:
jTool.insertData(cursor2, 'enterprise_raw', recEntBaseDic)
conn.commit()
print ' Insert enterprise baseinfo successfully in ' + str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
except:
print 'Fail to insert record '+ ', eid is '+str(param['eid'])
jTool.logError('Fail to insert record on '+ 'url is '+param['url']+', id is '+str(id))
return False
recEntDetailDic = {}
recEntDetailDic['url'] = param['url']
recEntDetailDic['eid'] = param['eid']
recEntDetailDic['enterprise_name'] = param['ename'].strip()
recEntDetailDic['records'] = ' '
recEntDetailDic['ctime'] = str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
# detailRecUrl = param['basePostUrl']
detailRecUrl = param['basePostUrlip']
contentRecord = jTool.getContentByProxy(proxy, detailRecUrl, post_data_dic)
if not contentRecord:
return True
recList = getEntDetail(contentRecord)
scount = 0
for rec in recList:
scount += 1
rec = jTool.clearX(["'"], rec)
recEntDetailDic['content'] = str(rec)
jTool.insertData(cursor3, 'enterprise_record_raw', recEntDetailDic)
conn.commit()
print ' Insert enterprise detail info No. '+str(scount)+' successfully in ' + str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
print 'Fetch and insert successfully in ' + str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
return True
def extractPostContent(conn, table, start, end):
'''
提取enterprise_raw表中的postContent字段到enterprse_record_raw中
每条包括多种信用记录，每种信用记录包括多条
'''
cursor = conn.cursor()
cursor2 = conn.cursor()
id = int(start)
while id>=int(start) and id<=int(end):
print '*'*30
sql = 'select eid, enterprise_name, url, postContent, id from '+str(table)+' where id ='+str(id)
id += 1
cursor.execute(sql)
record = cursor.fetchone()
if not record:
continue
recList = rp.getEntDetail(record[3])
scount = 0
recEntDetailDic = {}
recEntDetailDic['url'] = record[2]
recEntDetailDic['eid'] = record[0]
recEntDetailDic['enterprise_name'] = record[1].strip()
print 'id:'+str(record[4])+', '+recEntDetailDic['enterprise_name']
for rec in recList:
scount += 1
rDic = contentToRecords(str(rec).strip())
if not rDic:
continue
rec = jTool.clearX(["'"], rec)
recEntDetailDic['content'] = (str(rec)).strip()
recEntDetailDic['records'] = ' '
recEntDetailDic['ctime'] = str(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()))
fDic = dict(recEntDetailDic, **rDic)
jTool.insertData(cursor2, 'enterprise_record_raw', fDic)
conn.commit()
cursor.close()
cursor2.close()
conn.close()
except Exception, e:
print __name__, e
return tmpDic
del clist[0]
ccount = len(clist)
recs = ''
for c in range(ccount):
recs = recs+''.join(clist[c].split(', false);'))
tmp = recs.split('\n')
for i in range(len(tmp)):
tt = tmp[i].split(',')
del tt[0]
tmp[i] = ''.join(tt)
records = ';'.join(tmp)
records = jTool.clearX(['(', ')', 'true'], records).strip()
tmpDic['publisher'] = publisher
tmpDic['category'] = category
tmpDic['records'] = records
return tmpDic
def extractPostContent(conn, table, start, end):
'''
提取enterprise_raw表中的postContent字段到enterprse_record_raw中
每条包括多种信用记录，每种信用记录包括多条
'''
cursor = conn.cursor()
cursor2 = conn.cursor()
id = int(start)
while id>=int(start) and id<=int(end):