CODEWITHSUNDEEP
javascriptpythonweb-scraping
TTT
TTT

Reputation: 4434

Python Web Crawler for JavaScript Generated URLs

I am trying to use some Python web crawler to download about 3000 PDFs from a website. However, the URLs of those PDFs are generated by JavaScript function. So, I am wondering if there is any tutorial on how to achieve this?

For example, the URL linked to Alberto European Hairspray (Aerosol) - All Variants will be generated after clicking onclick="javascript:__doPostBack('ctl00$placeBody$gridView$gridView','DocumentCenter.aspx?did={0}$0&#39. So the question is how to let the web crawler to get the computed URL. 

function __doPostBack(eventTarget, eventArgument) {
    if (!theForm.onsubmit || (theForm.onsubmit() != false)) {
        theForm.__EVENTTARGET.value = eventTarget;
        theForm.__EVENTARGUMENT.value = eventArgument;
        theForm.submit();
    }
}
<tbody>
    <tr>
        <td>
            <input type="image" src="App_Graphics/PDFDocument.gif" alt="MSDS" onclick="javascript:__doPostBack(&#39;ctl00$placeBody$gridView$gridView&#39;,&#39;DocumentCenter.aspx?did={0}$0&#39;);return false;" />
        </td>
        <td><a href="javascript:__doPostBack(&#39;ctl00$placeBody$gridView$gridView&#39;,&#39;MSDSDetail.aspx?did={0}$0&#39;)">Alberto European Hairspray (Aerosol) - All Variants</a>
        </td>
        <td>Unilever PLC</td>
        <td>8131-01</td>
    </tr>
    <tr class="row-alternate">
        <td>
            <input type="image" src="App_Graphics/PDFDocument.gif" alt="MSDS" onclick="javascript:__doPostBack(&#39;ctl00$placeBody$gridView$gridView&#39;,&#39;DocumentCenter.aspx?did={0}$1&#39;);return false;" />
        </td>
        <td><a href="javascript:__doPostBack(&#39;ctl00$placeBody$gridView$gridView&#39;,&#39;MSDSDetail.aspx?did={0}$1&#39;)">Alberto European Mousse (Aerosol) - All Variants</a>
        </td>
        <td>Unilever PLC</td>
        <td>8132-01</td>
    </tr>
</tbody>

Upvotes: 1

Views: 421

Answers (2)

Igor Savinkin
Igor Savinkin

Reputation: 6277

Another option is that you might use Selenium to execute js and get computed urls.

Upvotes: 1

cdonts
cdonts

Reputation: 9599

You can't. Use a JavaScript interpreter (SpiderMonkey, for example) to execute the code and then go ahead with HTML parsing. Using Qt's WebKit is a good approach also, but probably slower.

Upvotes: 1

Related Questions