I am trying to click on a webelement that changes ID and xpath dynamically. I dont see any other information that can be used directly to identify the element uniquely everytime the page loads.
Here is the Xpath for the element:
.//*[@id='isc_SB']/table/tbody/tr/td
This will be updated along with ID at every page load.
Here is the HTML code for the same element:
<table width="74px" height="30px" cellspacing="0" cellpadding="0">
<tbody>
<tr>
<td class="OBToolbarTextButtonFocusedOver" valign="center" nowrap="true" align="center" onfocus="isc_OBToolbarActionButton_3.$47()" tabindex="-1">
Pu
<u>b</u>
lish
</td>
</tr>
</tbody>
</table>
I want to click on the button with Text called "publish" with 'u' having an underline
Here is what i have tried to no success:
Attempt 1:
publish = //td[. = 'Pu<u>b</u>lish']
Attempt 2:
publish = //td[contains(text(), 'lish')]
Attempt 3:
publish = .//*[@id='isc_S7']/table/tbody/tr/td # this is just an example to show id, xpath and everything is dynamic.
Please assist. Thank you.
In XPath the string value of an element node is defined as the concatenation of all its descendant text nodes, so if you have a
<td>Pu<u>b</u>lish</td>
then you can match it using simply
//td[. = 'Publish']
If, as in your question, there are newlines either side of the u element then you need to be a bit more creative, e.g.
//td[normalize-space() = 'Pu b lish']
If you want to cover both cases then
//td[translate(normalize-space(), ' ', '') = 'Publish']
As an aside, the reason your attempt 2 didn't work is that contains expects both its arguments to be strings, but in your example text() gives you a set of two text nodes ("Pu" and "lish"), and converting a node set to a string means taking the value of just the first node in the set in document order and ignoring the other(s).