how is it possible to select a node attribute based on an attribute of another same level node? What I would like to achieve is extract the links (href value) from all nodes which are followed by node span with attribute class="body-table-news-loss", i.e. the selector should return https://finance.yahoo.com/news/inspiremd-announces-planned-recapitalization-120000752.html in the example below. I have tried
"//tr[td/span[@class='body-table-news-loss']]/a/@href"
and similar variant thereof but I don't get the desired result as I don't know exactly how to explain the hierarchy level to the XPath selector. Many thanks for your help.
<tr>
<td>Nov-29-17 07:00AM
</td>
<td align="left">
<a href="https://finance.yahoo.com/news/inspiremd-announces-planned-
recapitalization-120000752.html">InspireMD Announces Planned
Recapitalization
</a>
<span class="body-table-news-loss">-51.29%</span>
</td>
</tr>
<tr>
<td >Nov-07-17 04:05PM
</td>
<td align="left">
<a href="https://finance.yahoo.com/news/inspiremd-announces-third-
quarter-2017-210500523.html">InspireMD Announces Third Quarter 2017
Results; CGuard(TM) Revenue Increases 90% Versus Same Period Last Year
</a>
<span class="body-table-news-gain">+5.94%</span>
</td>
</tr>
Your looking for the preceding-sibling:: axes. It selects all the sibling nodes of the context that precedes it.
Something like this should work.
//tr/td/span[@class='body-table-news-loss']/preceding-sibling::a/data(@href)