Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Extracting table contents from html with python and BeautifulSoup

I want to extract certain information out of an html document. E.g. it contains a table (among other tables with other contents) like this:

    <table class="details">
            <tr>
                    <th>Advisory:</th>
                    <td>RHBA-2013:0947-1</td>
            </tr>
            <tr>    
                    <th>Type:</th>
                    <td>Bug Fix Advisory</td>
            </tr>
            <tr>
                    <th>Severity:</th>
                    <td>N/A</td>
            </tr>
            <tr>    
                    <th>Issued on:</th>
                    <td>2013-06-13</td>
            </tr>
            <tr>    
                    <th>Last updated on:</th>
                    <td>2013-06-13</td>
            </tr>

            <tr>
                    <th valign="top">Affected Products:</th>
                    <td><a href="#Red Hat Enterprise Linux ELS (v. 4)">Red Hat Enterprise Linux ELS (v. 4)</a></td>
            </tr>


    </table>

I want to extract Information like the date of "Issued on:". It looks like BeautifulSoup4 could do this easyly, but somehow I don't manage to get it right. My code so far:

    from bs4 import BeautifulSoup
    soup=BeautifulSoup(unicodestring_containing_the_entire_htlm_doc)
    table_tag=soup.table
    if table_tag['class'] == ['details']:
            print table_tag.tr.th.get_text() + " " + table_tag.tr.td.get_text()
            a=table_tag.next_sibling
            print  unicode(a)
            print table_tag.contents

This gets me the contents of the first table row, and also a listing of the contents. But the next sibling thing is not working right, I guess I am just using it wrong. Of course I could just parse the contents thingy, but it seems to me that beautiful soup was designed to prevent us from doing exactly this (if I start parsing myself, I might as well parse the whole doc ...). If someone could enlighten me on how to acomplish this, I would be gratefull. If there is a better way then BeautifulSoup, I would be interested to hear about it.

like image 426
Isaac Avatar asked Jun 19 '13 16:06

Isaac


1 Answers

>>> from bs4 import BeautifulSoup
>>> soup = BeautifulSoup(unicodestring_containing_the_entire_htlm_doc)
>>> table = soup.find('table', {'class': 'details'})
>>> th = table.find('th', text='Issued on:')
>>> th
<th>Issued on:</th>
>>> td = th.findNext('td')
>>> td
<td>2013-06-13</td>
>>> td.text
u'2013-06-13'
like image 93
falsetru Avatar answered Oct 01 '22 17:10

falsetru