0

I want to scrapy and get the html by each herf in the page,then output to a csv.now I first time to use nightmare.so I have a problem about parament not defined.

var Nightmare = require('nightmare');
    var nightmare = Nightmare({ show: true });
    var fs  = require('fs');
    var result ;
    nightmare
      .goto('http://football-system.jp/fss/pub_kaijyolist.php?lid=h0xYuxqKQ+M=')
      .wait(1000)
      .evaluate(function () {
        var divs = document.querySelectorAll('a[target="_blank"]'),i;
        for (i = 0,result = ""; i < divs.length; ++i) {
            result += divs[i].href.toString()+"\n";
            }
        return divs;
      })
      .end()
      .then(function (divs) {
        console.log(divs)
        ///////////////////////////////////////////////////
        fs.writeFile('8.csv', result, function (err) {
      if (err) throw err;
      console.log('It\'s saved!');
    });
    ////////////////////////////////////////////////////
      })
      .then(function() {
        //since Nightmare has an internal `.then()`, return the instance returned by the final call in the chain
        return nightmare
          //click the next button to get the next page of search results
          .goto(divs[0].href)
          //get the first HREF from the second page of results
          .evaluate(function() {
            return document.querySelector('div[class="outputDate"]');
          })
      })
      .catch(function (error) {
        console.error('Search failed:', error);
      });

The second goto() have a Error: divs is not defined. Thanks a lot.

云鹏赵
  • 33
  • 1
  • 9

1 Answers1

0

The error message says it all: divs is not defined. That means that there isn't a divs variable in the scope of the second then callback.

You could "merge" the two then callbacks you have there into one, and have something like this:

var Nightmare = require('nightmare');
    var nightmare = Nightmare({ show: true });
    var fs  = require('fs');
    var result ;
    nightmare
      .goto('http://football-system.jp/fss/pub_kaijyolist.php?lid=h0xYuxqKQ+M=')
      .wait(1000)
      .evaluate(function () {
        var divs = document.querySelectorAll('a[target="_blank"]'),i;
        for (i = 0,result = ""; i < divs.length; ++i) {
            result += divs[i].href.toString()+"\n";
            }
        return divs;
      })
      .then(function (divs) {
        console.log(divs)
        ///////////////////////////////////////////////////
        fs.writeFile('8.csv', result, function (err) {
          if (err) throw err;
          console.log('It\'s saved!');
        });
        ////////////////////////////////////////////////////
        nightmare
          //click the next button to get the next page of search results
          console.log('this is the link I want to navigate to', divs[0].href)
          .goto(divs[0].href)
          //get the first HREF from the second page of results
          .evaluate(function() {
            return document.querySelector('div[class="outputDate"]');
          })
          .then(function(div) {
            console.log('I got the "div", now Ill do something with it', div)
          })
      })
      .catch(function (error) {
        console.error('Search failed:', error);
      });

      nightmare.end()

But this code will also fail, because your divs list is full of empty objects. I have changed a bit the code to get it to half work for you:

var Nightmare = require('nightmare');
    var nightmare = Nightmare({ show: true });
    var fs  = require('fs');
    var result ;
    nightmare
      .goto('http://football-system.jp/fss/pub_kaijyolist.php?lid=h0xYuxqKQ+M=')
      .wait(1000)
      .evaluate(function () {
        var divs = document.querySelectorAll('a[target="_blank"]'),i;
        var links = []
        for (i = 0,result = ""; i < divs.length; ++i) {
            links.push(divs[i].href)
            result += divs[i].href.toString()+"\n";
            }
        return links;
      })
      .then(function (links) {
        console.log(links)
        ///////////////////////////////////////////////////
        fs.writeFile('8.csv', links.join('\n'), function (err) {
          if (err) throw err;
          console.log('It\'s saved!');
        });
        ////////////////////////////////////////////////////
        console.log('this is the link I want to navigate to', links[0])
        nightmare
          //click the next button to get the next page of search results
          .goto(links[0])
          .wait(5000)
          //get the first HREF from the second page of results
          .evaluate(function() {
            return document.querySelector('div[class="outputDate"]');
          })
          .end()
          .then(function(div) {
            console.log('I got the "div", now Ill do something with it', div)
          })
      })
      .catch(function (error) {
        console.error('Search failed:', error);
      });

Now, you will have to iterate trough the links list and navigate to each and everyone of the links. This is a bit tricky to do with Nightmare, take a look at this resource and this answer. I believe you will find them quite helpful.

Community
  • 1
  • 1
felipe_gdr
  • 1,088
  • 2
  • 11
  • 27