Python Splinter Return Text and Blank Values with Delimiter

I'm currently using find_by_xpath in splinter to retrieve all values of a table. It works great for getting all non-blank values and taking little time to do so. However, some cells of the table are blank and the following code is ignoring those cells. Also, I need a delimiter (perhaps a pipe - '|'?) between each value.

browser.find_by_xpath("//*[contains(text(),'Table of Data')]/..").value

Here's a sample result from the first row:

'col1 data col2 data col3 data'

What I need is this, because the 4th column (but sometimes other columns) has an empty cell:

'col1 data|col2 data|col3 data|""'

Thanks in advance!

HTML:

<td class="padtd" height="150" valign="top" width="75%" colspan="2">
   <div class="headingSum">Table of Data </div>
   <table style="width:100%;height=10;valign:top">
<tbody>
   <tr>
      <td height="15" width="50%" class="selTabSum">
         <div>
         <table style="width:100%;" valign="top">
            <tbody>
               <tr>
                  <td width="10%" class="tableheading">Column 1</td>
                  <td width="15%" class="tableheading">Column 2 </td>
                  <td width="25%" class="tableheading">Column 3 </td>
                  <td width="50%" class="tableheading">Column 4 </td>
               </tr>
               <tr>
                  <td width="10%" valign="top" class="tableCell"><a href=""><span class=“data” id="160042">col1 data</span></a></td>
                  <td width="15%" valign="top" class="tableCell">col2 data</td>
                  <td width="25%" valign="top" class="tableCell">col3 data</td>
                  <td width="50%" class="tableCell"></td>
               </tr>
               <tr>
                  <td width="10%" valign="top" class="tableCell"><a href=""><span class=“data” id="160042">col1 data</span></a></td>
                  <td width="15%" valign="top" class="tableCell">col2 data</td>
                  <td width="25%" valign="top" class="tableCell">col3 data</td>
                  <td width="50%" class="tableCell"></td>
               </tr>
               <tr>
                  <td width="10%" valign="top" class="tableCell"><a href=""><span class=“data” id="97851">col1 data</span></a></td>
                  <td width="15%" valign="top" class="tableCell">col2 data</td>
                  <td width="25%" 
                     valign="top" class="tableCell">col3 data</td>
                  <td width="50%" class="tableCell">
                     col4 data
                     <table width="100%">
                        <tbody>
                           <tr></tr>
                        </tbody>
                     </table>
                  </td>
               </tr>
            </tbody>
         </table>
      </td>

I ended up getting the HTML from the table (via xpath) and passing it into pandas via pd.read_html.

import pandas as pd
from splinter import Browser
...
xp = "//*[contains(text(),'Table of Data')]/.."
df = pd.read_html(browser.find_by_xpath(xp).html)[1]