我目前正在NodeJS中构建web scraper
,并且遇到了某些问题。运行代码后,我收到此错误:
未定义不是有效的uri或options对象。
我不确定如何绕过此错误,我看过以下示例:Example One,Example Two
这是我所有的代码:
var request = require('request');
var cheerio = require('cheerio');
var URL = require('url-parse');
var START_URL = "http://example.com";
var pagesVisited = {};
var numPagesVisited = 0;
var pagesToVisit = [];
var url = new URL(START_URL);
var baseUrl = url.protocol + "//" + url.hostname;
pagesToVisit.push(START_URL);
setInterval(crawl,5000);
function crawl() {
var nextPage = pagesToVisit.pop();
if (nextPage in pagesVisited) {
// We've already visited this page, so repeat the crawl
setInterval(crawl,5000);
} else {
// New page we haven't visited
visitPage(nextPage, crawl);
}
}
function visitPage(url, callback) {
// Add page to our set
pagesVisited[url] = true;
numPagesVisited++;
// Make the request
console.log("Visiting page " + url);
request(url, function(error, response, body) {
// Check status code (200 is HTTP OK)
console.log("Status code: " + response.statusCode);
if(response.statusCode !== 200) {
console.log(response.statusCode);
callback();
return;
}else{
console.log(error);
}
// Parse the document body
var $ = cheerio.load(body);
collectInternalLinks($);
// In this short program, our callback is just calling crawl()
callback();
});
}
function collectInternalLinks($) {
var relativeLinks = $("a[href^='/']");
console.log("Found " + relativeLinks.length + " relative links on page");
relativeLinks.each(function() {
pagesToVisit.push(baseUrl + $(this).attr('href'));
});
}
答案 0 :(得分:0)
您的pagesToVisit清空后,该网址将是未定义的,因为在空数组上调用pop会返回该值。
我会在visitPage中添加一个网址不是未定义的检查,例如
function visitPage(url, callback) {
if (!url) {
// We're done
return;
}
或者在抓取中,检查pagesToVisit是否包含元素,例如
function crawl() {
var nextPage = pagesToVisit.pop();
if (!nextPage) {
// We're done!
console.log('Crawl complete!');
} else if (nextPage in pagesVisited) {
// We've already visited this page, so repeat the crawl
setInterval(crawl,5000);
} else {
// New page we haven't visited
visitPage(nextPage, crawl);
}
}
答案 1 :(得分:0)
根据Terry Lennox's answer的提示,我对crawl()
函数进行了一些修改:
function crawl() {
var nextPage = pagesToVisit.pop();
if (nextPage in pagesVisited) {
// We've already visited this page, so repeat the crawl
setInterval(crawl, 5000);
} else if(nextPage) {
// New page we haven't visited
visitPage(nextPage, crawl);
}
}
我要做的就是在调用visitPage()
之前检查弹出的元素是否存在。
我得到以下输出:
Visiting page http://example.com
Status code: 200
response.statusCode: 200
null
Found 0 relative links on page
^C