因为工作需要，需要对京东地址和国标地址转换，特意采集了一下最新的国标地址库，以方便进行数据映射。
因为前端技术不太好，只能一点一点的页面分析（大神勿笑），故整理出来，一是为了存储方便以后再用，二是为了一样的方便同样需要这部分数据的小伙伴。
/// <summary>
/// 国标地址库采集
///
/// 作者：南丘伟
/// 时间：2017-06-21
/// 版本：V1.0.0
/// 版本说明：
/// 下次采集时很有可能因为页面格式变化导致采集不到
/// 此方法可作为下次采集的Demo,采集时另做调整
///
/// </summary>
public class GbAddressHelper
{
public GbAddressHelper() { }
/// <summary>
/// 国家统计局地址信息采集
/// 地址：http://www.stats.gov.cn/tjsj/tjbz/xzqhdm/201703/t20170310_1471429.html
/// 调试结果：
/// [{省，市，区县}]
///
/// </summary>
/// <param name="gbUrl"></param>
/// <returns></returns>
public string DownLoadGbAddress(string gbUrl)
{
string newjson = string.Empty;
string html = HttpUtilitys.SendHttpRequest(gbUrl, "");
if (html.Contains("xilan_con"))
{
//第一步：从页面上获取地址数据相关块
string regEx = "<div class=\"xilan_con\"([\\w\\W]+)</span></b></p></div>";//获取当前页面
MatchCollection mcPageIndex = Regex.Matches(html, regEx, RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (mcPageIndex.Count == 1)
{
html = mcPageIndex[0].Groups[0].Value.Trim();
var sb = new StringBuilder();
newjson += "[";
//第二步：页面地址数据字符串，根据省份切割
var prohtmlArry = Regex.Split(html, "<p class=\"MsoNormal\"><span lang=\"EN-US\">", RegexOptions.IgnoreCase);
foreach (var item in prohtmlArry)
{
var prolist = Regex.Split(item, "<p class=\"MsoNormal\"><b><span lang=\"EN-US\">", RegexOptions.IgnoreCase);
for (int i = 0; i < prolist.Length; i++)
{
string newhtml = "<p class=\"MsoNormal\"><b><span lang=\"EN-US\">" + prolist[i];
regEx = "<p class=\"MsoNormal\"><b><span lang=\"EN-US\">([\\d]+)<span>";//获取当前页面
MatchCollection mcproId = Regex.Matches(newhtml, regEx, RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (mcproId.Count > 0)
{
int proId = Utils.ObjToInt(mcproId[0].Groups[1].Value.Trim(), 0);
if (proId > 0)
{
//
regEx = "<b><span style=\"font-family: 宋体\">([\u4e00-\u9fa5]+)</span></b></p>";//获取当前页面
// regEx = "</span></span></b><b><span style=\"font-family: 宋体\">([\\w\\W]+)</span></b></p>";//获取当前页面
MatchCollection mcproName = Regex.Matches(newhtml, regEx, RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (mcproName.Count > 0)
{
string proName = mcproName[0].Groups[1].Value.Trim();
newjson += "{";
newjson += "\"PId\":\"0\",";
newjson += "\"AreaId\":\"" + proId + "\",";
newjson += "\"AreaName\":\"" + proName + "\"},";
//第三步：根据省数据，分割城市
//匹配城市
//"<span lang=\"EN-US\">([\\d]+)<span> (\\s*)</span></span><span style=\"font-family: 宋体\">(\\s*)([\u4e00-\u9fa5]+)</span></p>"
var citylist = Regex.Split(newhtml, "<p class=\"MsoNormal\"><span style=\"font-family: 宋体\"> </span>", RegexOptions.IgnoreCase);
for (int k = 0; k < citylist.Length; k++)
{
int cityId = 0;
//拆分出来的每个城市详情
newhtml = citylist[k];
regEx = "<span lang=\"EN-US\">([\\d]+)<span> (\\s*)</span></span><span style=\"font-family: 宋体\">(\\s*)([\u4e00-\u9fa5]+)</span></p>";
//拆分城市ID和名称
MatchCollection cityslist = Regex.Matches(newhtml, regEx, RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (cityslist.Count > 0)
{
cityId = Utils.ObjToInt(cityslist[0].Groups[1].Value.Trim(), 0);
newjson += "{";
newjson += "\"PId\":\"" + proId + "\",";
newjson += "\"AreaId\":\"" + cityId + "\",";
newjson += "\"AreaName\":\"" + cityslist[0].Groups[4].Value.Trim() + "\"},";
//拆分区县
regEx = "<span lang=\"EN-US\">([\\d]+)<span> </span></span><span style=\"font-family: 宋体\">(\\s*)([\u4e00-\u9fa5]+)</span></p>";
MatchCollection townlist = Regex.Matches(newhtml, regEx, RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (townlist.Count > 0)
{
for (int j = 0; j < townlist.Count; j++)
{
var town = townlist[j].Groups[0].Value.Trim();
newjson += "{";
newjson += "\"PId\":\"" + cityId + "\",";
newjson += "\"AreaId\":\"" + Utils.ObjToInt(townlist[j].Groups[1].Value.Trim(), 0) + "\",";
newjson += "\"AreaName\":\"" + townlist[j].Groups[3].Value.Trim() + "\"},";
}
}
}
}
}
}
else
{
newjson = "{\"status\":\"n\",\"info\":\"省份代码为空！，请核查采集源码！\"}";
}
}
}
}
newjson = newjson.Substring(0, newjson.Length - 1);
newjson += "]";
}
else
{
newjson = "{\"status\":\"n\",\"info\":\"页面数据标签位置变动，请核查采集源码！\"}";
}
}
else
{
newjson = "{\"status\":\"n\",\"info\":\"采集页面数据格式变动，请核查采集源码！\"}";
}
return newjson;
}
}
时间： 2024-11-10 16:33:20