XPath syntax to make a query that exclude some specific element - xpath

Question

You can find my test html page at https://sabbiobet.netsons.org/test.html
This is the html markup of the page:
<table border="1" class="class_table">
<tbody>
<tr class="class_tr">
<td class="class_td"> </td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ok"></span>square</td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ko"></span>circle</td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ok"></span>triangle</td>
</tr>
</tbody>
</table>
I need to obtain all the text in <td> with class="class_td" minus the ones that have text empty or or that have a child <span> with class="class_ko".
In other words I want to obtain only these values:
Square
Triangle
Using the importXML function of google sheets and following the suggestion of another user I've tried:
//td[#class='class_td' and span and not(span[#class='class_ko'])]
but it works only if i put some text between "span" and "/span"
Without any text I'll obtain only an empty result.
Can somebody help me?

Answer 1

In case the provided non-well-formed document is corrected to a well-formed one by replacing the undefined entity with the equivalent character entity reference  :
<table border="1" class="class_table">
<tbody>
<tr class="class_tr">
<td class="class_td"> </td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ok"></span>square</td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ko"></span>circle</td>
</tr>
<tr class="class_tr">
<td class="class_td"><span class="class_span_ok"></span>triangle</td>
</tr>
</tbody>
</table>
then this XPath expression:
/*/*/*/td
[#class='class_td'
and not(span[#class='class_span_ko'])
and normalize-space(translate(., ' ', ''))
]/>
when evaluated, selects exactly the wanted td elements:
<td class="class_td">
<span class="class_span_ok"/>square</td>
<td class="class_td">
<span class="class_span_ok"/>triangle</td>
XSLT - based verification
This transformation evaluates the above XPath expression and copies the selected elements to the output:
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output omit-xml-declaration="yes" indent="yes"/>
<xsl:template match="/">
<xsl:copy-of select=
"/*/*/*/td
[#class='class_td'
and not(span[#class='class_span_ko'])
and normalize-space(translate(., ' ', ''))
]"/>
</xsl:template>
</xsl:stylesheet>
The wanted, correct result is produced:
<td class="class_td">
<span class="class_span_ok"/>square</td>
<td class="class_td">
<span class="class_span_ok"/>triangle</td>
Note:
If only the string values of the wanted elements are needed, then the XPath expression can be:
/*/*/*/td
[#class='class_td'
and not(span[#class='class_span_ko'])
and normalize-space(translate(., ' ', ''))
]/text()