convert html table to csv using python

Question

I have a string which contains the source code of a html file extracted through mechanize library. The html file will always contain a table like this. I want to convert the table to CSV Format.

Several SO questions which address the same problem have the table with a class name. But my table doesnt have a class attribute. So what should i do...?

<table border=1 cellPadding="2" cellSpacing="0" width="75%" bordercolor="#000000" >
 <tr bgcolor="mediumblue">
 <td width="20%"><p align="center"><font face="Arial" color="white" size="2"><strong>SUB CODE</strong></font></p></td>
 <td width="26%"><p align="left"><font face="Arial" color="white" size="2"><strong>SUB NAME</strong></font></p></td>
 <td width="13%"><p align="left"><font face="Arial" color="white" size="2"><strong>THEORY</strong></font></p> </td>
 <td width="10%"><p align="left"><font face="Arial" color="white" size="2"><strong>PRACTICAL</strong></font></p> </td>
 <td width="17%"><p align="left"><font face="Arial" color="white" size="2"><strong>MARKS</strong></font></p></td>
 <td width="14%"><p align="center"><font face="Arial" color="white" size="2"><strong>GRADE</strong></font></p></td>
 </tr>
 <tr bgColor="#ffffff">
 <td align="middle"><font face="Arial" size=2> 301</font></td>
 <td align="left" ><font face="Arial" size=2>ENGLISH CORE</font></td>
 <td align="left" ><font face="Arial" size=2>067</font></td>
 <td align="left" ><font face="Arial" size=2></font></td>
 <td align="left" ><font face="Arial" size=2>067&nbsp;&nbsp;&nbsp;&nbsp;</font></td>
 <td align="middle"><font face="Arial" size=2>C2</font></td>
 </tr>
 </table>

• Colum: Is there a reason why you cannot give it a class?
• Rohith R: because i am getting the code directly from a library called mechanize
• Colum: If there is only going to be one table in the string that you need to convert, can you identify it by the table element tag, rather than a class name?
• Rohith R: no...there are more than 1 tables

Answer 1

pandas has a neat way to read html tables.

import pandas as pd
html_data = '''
<table border=1 cellPadding="2" cellSpacing="0" width="75%" bordercolor="#000000" >
 <tr bgcolor="mediumblue">
 <td width="20%"><p align="center"><font face="Arial" color="white" size="2"><strong>SUB CODE</strong></font></p></td>
 <td width="26%"><p align="left"><font face="Arial" color="white" size="2"><strong>SUB NAME</strong></font></p></td>
 <td width="13%"><p align="left"><font face="Arial" color="white" size="2"><strong>THEORY</strong></font></p> </td>
 <td width="10%"><p align="left"><font face="Arial" color="white" size="2"><strong>PRACTICAL</strong></font></p> </td>
 <td width="17%"><p align="left"><font face="Arial" color="white" size="2"><strong>MARKS</strong></font></p></td>
 <td width="14%"><p align="center"><font face="Arial" color="white" size="2"><strong>GRADE</strong></font></p></td>
 </tr>
 <tr bgColor="#ffffff">
 <td align="middle"><font face="Arial" size=2> 301</font></td>
 <td align="left" ><font face="Arial" size=2>ENGLISH CORE</font></td>
 <td align="left" ><font face="Arial" size=2>067</font></td>
 <td align="left" ><font face="Arial" size=2></font></td>
 <td align="left" ><font face="Arial" size=2>067&nbsp;&nbsp;&nbsp;&nbsp;</font></td>
 <td align="middle"><font face="Arial" size=2>C2</font></td>
 </tr>
 </table>
'''
print pd.read_html(html_data)[0].to_csv(index=False, header=False)

When where's multiple tables in html, you can check column names of the table, to remove unneeded ones.

• Rohith R: print pd.read_html(html_data)[0].to_csv(index=False,header=False) TypeError: to_csv() takes at least 2 arguments (3 given)
• tdihp: @PRP using Python2, pandas 0.16.0