Python采集百度地图数据

您的评价:

收藏该经验

百度利用其强大的中文搜索引擎数据，结合地图应用，包含了海量的公司联系方式，比Google要强，更别说什么黄页网站了。
因为一些业务需要，写了这个行业公司地址采集程序，使用方便，直接运行，支持命令行设定查询参数。

使用方法：
把代码保存成bmap.py
python bmap.py
或
python bmap.py 服饰厂

运行后会自动采集百度地图中所有的结果，保存为以tab分割的txt文件，方便导入各种数据库。 

001	#!/usr/bin/env python

002	# -*- coding: utf-8 -*-

003

#

004	# Copyright 2012 Channing Wong

005

#

006	# @mail: [e-mail]

007	# @home: http://blog.3363.me/

008	# @date: Mar 3, 2012

009

#

010

011	import json

012	import sys

013	import time

014	import types

015	import urllib

016

017	reload(sys)

018	sys.setdefaultencoding('utf-8')

019

020

021	class BaiduMap:

022

"""

023

"""

024	def __init__(self, keyword):

025	self.keyword = keyword

026	self.query = [

027	('b', '(-1599062.039999999,811604.75;24779177.96,8168020.75)'),

028	('c', '1'),

029	('from', 'webmap'),

030	('ie', 'utf-8'),

031	('l', '4'),

032	('newmap', '1'),

033	('qt', 's'),

034	('src', '0'),

035	('sug', '0'),

036	('t', time.time().__int__()),

037	('tn', 'B_NORMAL_MAP'),

038	('wd', keyword),

039	('wd2', '')

040

]

041	self.mapurl = 'http://map.baidu.com/'

042	self.file = open('%s.txt' % keyword, 'w')

043	self.count = 0

044	self.count_c = 0

045	self.total_num = 0

046

047	self._get_city()

048

049	def _fetch(self, query=None, json=True):

050	data = urllib.urlencode(query)

051	url = self.mapurl + '?' + data

052	opener = urllib.FancyURLopener()

053	data = opener.open(url).read()

054

055

if json:

056	return self._tojson(data)

057

else:

058	return data

059

060	def _tojson(self, data):

061

try:

062	js = json.loads(data, 'utf-8')

063

except:

064

js = None

065

066

return js

067

068	def _get_city(self):

069	data = self._fetch(self.query)

070

071	if type(data['content']) is not types.ListType:

072	print 'keyworld error.'

073	sys.exit()

074

075	self.city = data['content']

076

077	if data.has_key('more_city'):

078	for c in data['more_city']:

079	self.city.extend(c['city'])

080

081	for city in self.city:

082	self.total_num += city['num']

083

084	def _get_data(self, city, page=0):

085

query = [

086	('addr', '0'),

087	('b', '(%s)' % city['geo'].split('|')[1]),

088	('c', city['code']),

089	('db', '0'),

090	('gr', '3'),

091	('ie', 'utf-8'),

092	('l', '9'),

093	('newmap', '1'),

094	('on_gel', '1'),

095	('pn', page),

096	('qt', 'con'),

097	('src', '7'),

098	('sug', '0'),

099	('t', time.time().__int__()),

100	('tn', 'B_NORMAL_MAP'),

101	('wd', self.keyword),

102	('wd2', ''),

103

]

104	data = self._fetch(query)

105	return data

106

107	def _save(self, content, city):

108	for c in content:

109	self.count += 1

110	self.count_c += 1

111	if c.has_key('tel'):

112	tel = c['tel']

113

else:

114

tel = ''

115

116	_data = '%s\t%s\t%s\t%s\n' % (city['name'], c['name'], c['addr'], tel)

117	self.file.write(_data)

118	print '(%s/%s) %s[%s/%s]' % (self.count, self.total_num, city['name'], self.count_c, city['num'])

119

120	def get(self, city):

121	self.count_c = 0

122	pages = abs(-city['num'] / 10)

123	for page in range(0, pages):

124	data = self._get_data(city, page)

125	if data.has_key('content'):

126	self._save(data['content'], city)

127

128	def get_all(self):

129	for city in self.city:

130	self.get(city)

131

132	self.file.close()

133

134

135	if __name__ == '__main__':

136	if sys.argv.__len__() > 1:

137	keyword = sys.argv[1]

138

else:

139	keyword = '钻石'

140

141	baidumap = BaiduMap(keyword)

142	print '_' * 20

143	print 'CITY: %s' % baidumap.city.__len__()

144	print 'DATA: %s' % baidumap.total_num

145	baidumap.get_all()