BeautifulSoup gives unexpected results

时间:2018-12-19 11:20:25

标签: python-3.x beautifulsoup python-requests

I am new to the field of web-scraping and while running the following code, I got unexpected results:

from bs4 import BeautifulSoup as soup
import requests

header = {'User-Agent':'Chrome/71.0.3578.98 Mozilla/64.0'}
url = 'https://www.amazon.in/Philips-Trimmer-Cordless-QT4001-15/dp/B00L8PEEAI'
res = requests.get(url, headers = header)
bs = soup(res.text, 'html.parser')
print(bs.div)

which yields the following output:

<!-- From remote config v3-->
<script type="text/javascript">
    (function(d) {
        document.createElement("header");

        function b(e) {
            return [].slice.call(e)
        }

        function c(f, e) {
            return {
                m: f,
                a: b(e)
            }
        }
        var a = function(f) {
            var g = {};
            g._sourceName = f;
            g._replay = [];
            g.getNow = function(i, h) {
                return h
            };

            function e(i, h, j) {
                i[j] = function() {
                    g._replay.push(h.concat(c(j, arguments)))
                }
            }
            g.when = function() {
                var i = [c("when", arguments)];
                var h = {};
                e(h, i, "run");
                e(h, i, "declare");
                e(h, i, "publish");
                e(h, i, "build");
                return h
            };
            e(g, [], "declare");
            e(g, [], "build");
            e(g, [], "publish");
            e(g, [], "importEvent");
            a._shims.push(g);
            return g
        };
        a._shims = [];
        if (!d.$Nav) {
            d.$Nav = a("rcx-nav")
        }
        if (!d.$Nav.make) {
            d.$Nav.make = a
        }
    }(window));

which clearly isn't what I was looking for. And I cannot work with this, for example if I need to fetch the product name, I would write something like:

bs.findAll("div",{"id":"titleSection"})[0].span.text.strip()

which, given the output above throws an index error. Oddly, when I run this code through a loop for several similar products on Amazon across a terminal, it works fine for the first 10-15 products and then throws the same error and sometimes goes up to 130 items before breaking down and on other occasions gives the error on the very first iteration. I would be very thankful for an explanation(and possibly a solution) for this random behaviour.

1 个答案:

答案 0 :(得分:1)

请求无法提取带有javascript呈现的页面,因此要解决问题,您可以选择:

  1. 使用硒,启动或html请求呈现javascript;
  2. 使用浏览器的请求和网络工具获取json文件。