如何在神箭手云爬虫上写爬虫

发表评论
992 次浏览

A+

摘要

上一篇博客跟大家详细介绍了如何写出《黄焖鸡米饭是怎么火起来的》这样的数据分析类的文章，相信很多人都对数据来源也就是如何爬取到黄焖鸡米饭商家信息很感兴趣。那么今天我就跟大家具体讲一讲怎么使用神箭手云爬虫写爬虫，以上篇博客的黄焖鸡米饭的代码为例。

首先我们先看一下这段从大众点评网上爬取黄焖鸡米饭商户信息的脚本代码：

// 大众点评上爬取所有"黄焖鸡米饭"的商户信息
var keywords = "黄焖鸡米饭";
var scanUrls = ["http://www.dianping.com/search/keyword/1/0_"+keywords];
//国内的城市id到2323，意味着种子url有2323个
//作为sample，这里改成1，只爬取上海的黄焖鸡米饭门店
//for (var i = 1; i <= 2323; i++) {
for (var i = 1; i <= 1; i++) {
scanUrls.push("http://www.dianping.com/search/keyword/"+i+"/0_"+keywords);
}

var configs = {
    domains: ["dianping.com"],
    scanUrls: scanUrls,
    helperUrlRegexes: ["http://www.dianping.com/search/keyword/\\d+/0_.*"],
    contentUrlRegexes: ["http://www.dianping.com/shop/\\d+/editmember"],
    enableProxy: true,
    interval: 5000,
    fields: [
        {
            name: "shop_name",
            selector: "//div[contains(@class,'shop-review-wrap')]/div/h3/a/text()"
        },
        {
            name: "id",
            selector: "//div[contains(@class,'shop-review-wrap')]/div/h3/a/@href"
        },
        {
            name: "create_time",
            selector: "//div[contains(@class,'block raw-block')]/ul/li[1]/span"
        },
        {
            name: "region_name",
            selector: "//div[@class='breadcrumb']/b[1]/a/span/text()",
            required: true
        },
        {
            name: "province_name",
            selector: "//div[@class='breadcrumb']/b[1]/a/span/text()"
        }
    ]
};

configs.onProcessHelperUrl = function(url, content) {
    var urls = extractList(content, "//div[@class='tit']/a[not(contains(@class,'shop-branch'))]/@href");
    for (var i = 0; i < urls.length; i++) {
        addUrl(urls[i]+"/editmember");
    }
    var nextPage = extract(content,"//div[@class='page']/a[@class='next']/@href");
    if (nextPage) {
        addUrl(nextPage);
        var result = /\d+$/.exec(nextPage);
        if (result) {
            var data = result[0];
            var count = nextPage.length-data.length;
            var lll = nextPage.substr(0, count)+(parseInt(data)+1);
            addUrl(nextPage.substr(0, count)+(parseInt(data)+1));
            addUrl(nextPage.substr(0, count)+(parseInt(data)+2));
        }
    }
    return false;
}

configs.afterExtractField = function(fieldName, data) {
    if (fieldName == "id") {
        var result = /\d+$/.exec(data);
        if (result) {
            data = result[0];
        }
    }
    else if (fieldName == "shop_name") {
        if (data.indexOf("黄焖鸡米饭") == -1) {
            skip();
        }
    }
    else if (fieldName == "create_time") {
        var result = /\d{2}-\d{2}-\d{2}$/.exec(data);
        data = "20"+result[0];
    }
    else if (fieldName == "province_name" || fieldName == "region_name") {
        var position = data.indexOf("县");
        if (position != -1 && position < data.length -1) {
            data = data.substr(0,position+1);
        }
        position = data.indexOf("市");
        if (position != -1 && position < data.length -1) {
            data = data.substr(0,position+1);
        }
        data = data.replace("餐厅","");
        if (fieldName == "province_name") {
            data = getProvinceNameByRegion(data);
        }
    }
    return data;
}