概述
通常抓取级联数数据情况不多，但要是真需要时，确多了一些麻烦，比如抓取商品分类级别信息等。本内容将讲述如何采集无限级联联数据，并以GoldData来抓取2019年最新的省市县三级为示例。
创建数据集
在数据集管理里，添加数据集area。如下图所示：
数据集相当于数据库中的表，只是字段是灵活的，可以随着需要而添加和变更。
创建规则
在规则管理里，添加规则arearule，并将地址http://xzqh.mca.gov.cn/map 填为抓取入口地址。
我们通过分析可知，我们可从http://xzqh.mca.gov.cn/map 获取省级数据，然后通过级级数据请求http://xzqh.mca.gov.cn/selectJson 带有省级名称来获得市级数据，最后再通过市级数据分别请求http://xzqh.mca.gov.cn/selectJson 带有市级数据来获得县级数据。
并且发现请求 http://xzqh.mca.gov.cn/selectJson 是需要发POST请求的，因此我们需要将url加个前缀fake:,然后用规则里用JavaScript去请求URL。
我们在此编写的area数据集，有以下字段：
| 名称 | 说明 |
|---|---|
| sn | 取区域编码作为记录唯一字段 |
| name | 名称 |
| code | 取区域编码 |
| abbr | 省名简写 |
| parent_code | 取父区域编码 |
因此编写规则如下：
（注：具体规则内容请见文章最后）
然后编写完成后，我们就可以启动抓取器进行抓取。
查看数据
打开数据管理，选择area数据集查看如下图所示：
导出数据
回到数据管理，选择条件，选择需要导出的字段，进行导出。像这里数据比较多，GoldData将会以打包excel文件并压缩为zip文件下载。解压到本地，然后打开excel就可以看到抓取的数据，如下图所示：
结尾
通过这节内容，可以了解了如何通过GoldData 抓取级联数据。但是下一个问题是如何将数据导入自关联列表呢，且看下一往篇将会讲述如何将级联数据融合到数据库自关联表当中。
咐录：
（抓取规则）
[
{
__sample: http://xzqh.mca.gov.cn/map
match0: http/:////xzqh/.mca/.gov/.cn//map
fields0:
{
__model: true
__node: js
__js:
'''
var exp11=/json/s=/s(.+)/s+/$/(doc/
var ret=exp11.exec(html)
var ss=eval(ret[1])
for(var i=0;i<ss.length;i++){
var ele=ss[i]
var exp12=//((.+)/)/.exec(ele.shengji)
var exp13=/(.+)/(/.exec(ele.shengji)
out.add({sn:ele.quHuaDaiMa,name:exp13[1],code:ele.quHuaDaiMa,abbr:exp12[1],
parent_code:null,
})
}
'''
name:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
abbr:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
sn:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
__dataset: area
parent_code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
}
fields1:
{
__node: js
__js:
'''
var exp11=/json/s=/s(.+)/s+/$/(doc/
var ret=exp11.exec(html)
var ss=eval(ret[1])
for(var i=0;i<ss.length;i++){
var ele=ss[i]
var url='fake:http://xzqh.mca.gov.cn/selectJson?shengji='+ele.shengji+"&code="+ele.quHuaDaiMa
//var $ajax(url,[__method:'POST',data:'shengji='+ele.shengji]).content
out.add({href:url})
}
'''
href:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
}
}
{
__sample: fake:http://xzqh.mca.gov.cn/selectJson?shengji=北京市(京)&code=110000
match0: fake/:http/:////xzqh/.mca/.gov/.cn//selectJson/?shengji=.+/&code=/d+
fields0:
{
__model: true
__dataset: area
sn:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
__node: js
__js:
'''
var urlRet= /http/:////xzqh/.mca/.gov/.cn//selectJson/?shengji=(.+)/&code=(/d+)/.exec(baseUri);
var shengji=urlRet[1]
var sjCode=urlRet[2]
var url='http://xzqh.mca.gov.cn/selectJson'
var content=$ajax(url,{__method:'POST',data:'shengji='+shengji }).content
var arr=eval(content);
for(var i=0;i<arr.length;i++){
var dj=arr[i];
var area={
name:dj.diji,
code:dj.quHuaDaiMa,
parent_code:sjCode,
sn:dj.quHuaDaiMa,
}
out.add(area);
}
'''
name:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
abbr:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
sn:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
parent_code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
}
fields1:
{
__node: js
__js:
'''
var urlRet= /http/:////xzqh/.mca/.gov/.cn//selectJson/?shengji=(.+)/&code=(/d+)/.exec(baseUri);
var shengji=urlRet[1]
var sjCode=urlRet[2]
var url='http://xzqh.mca.gov.cn/selectJson'
var content=$ajax(url,{__method:'POST',data:'shengji='+shengji }).content
var arr=eval(content);
for(var i=0;i<arr.length;i++){
var dj=arr[i];
var url='fake:http://xzqh.mca.gov.cn/selectJson?shengji='+shengji+"&sjcode="+sjCode+"&diji="+dj.diji+"&djcode="+dj.quHuaDaiMa
out.add({href:url})
}
'''
href:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
}
}
{
__sample: fake:href :fake:http://xzqh.mca.gov.cn/selectJson?shengji=北京市(京)&sjcode=110000&diji=北京市&djcode=110000
match0: fake/:http/:////xzqh/.mca/.gov/.cn/selectJson/?shengji=([^&]+)/&sjcode=([^&]+)/&diji=([^&]+)/&djcode=([^&]+)
fields0:
{
__model: true
__dataset: area
__node: js
sn:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
__js:
'''
var urlRet=/fake/:http/:////xzqh/.mca/.gov/.cn//selectJson/?shengji=([^/&]+)/&sjcode=([^/&]+)/&diji=([^/&]+)/&djcode=([^/&]+)/.exec(baseUri)
var shengji=urlRet[1]
var sjCode=urlRet[2]
var dj=urlRet[3]
var djCode=urlRet[4]
var url='http://xzqh.mca.gov.cn/selectJson'
var content=$ajax(url,{__method:'POST',data:'shengji='+shengji+'&diji='+dj }).content
var arr=eval(content);
for(var i=0;i<arr.length;i++){
var dj=arr[i];
var area={
name:dj.xianji,
code:dj.quHuaDaiMa,
parent_code:djCode,
sn:dj.quHuaDaiMa,
}
out.add(area);
}
'''
name:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
abbr:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
sn:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
parent_code:
{
expr: ""
attr: ""
js: ""
__label: ""
__showOnList: false
__type: ""
down: "0"
accessPathJs: ""
uploadConf: ""
}
}
}
]原创文章，作者：奋斗，如若转载，请注明出处：https://blog.ytso.com/193000.html