How can I get the XPATH of elements under all rows of the same rowspan? - selenium

Question

Test data:
<table>
<tbody>
<tr>
<td id="mainfield:1" rowspan="3">A1</td>
<td ><span class="searching_for_this"> AA1</span></td>
<td ><span class="not_searching_for_this">AA2</span></td>
</tr>
<tr>
<td ><span class="searching_for_this"> AA3 </span></td>
<td ><span class="not_searching_for_this">AA3 </span></td>
</tr>
<tr>
<td ><span class="searching_for_this"> AA1 </span></td>
<td ><span class="not_searching_for_this">AA4 </span></td>
</tr>
<tr>
<td id="main_field:2" rowspan="3">B1</td>
<td ><span class="searching_for_this"> BB1</span></td>
<td ><span class="not_searching_for_this">BB2</span></td>
</tr>
<tr>
<td ><span class="searching_for_this"> AA1 </span></td>
<td ><span class="not_searching_for_this">BB3 </span></td>
</tr>
<tr>
<td ><span class="searching_for_this"> BB2 </span></td>
<td ><span class="not_searching_for_this">BB3 </span></td>
</tr>
</tbody>
</table>
Premises
I know the content of the row and column where 3 rowspan is located, (in this example A1)
I now the content of one element of the class I want to look for, in this scenario AA1 and searching_for_this
I want to get the rows (tr) of AA1 under the rowspan of A1. So the result would be the first and third row
First try
So in a single row scenario this would be something like:
Main row: //tr[td[contains(text(), 'A1')]]
Search in the children from the row (relative search .//):
.//tr[td/span[class=searching_for_this and contains(text(), 'AA1')]]
Problem
With this rowspan scenario I don't know how can I get all elements taking into account "next rows" after colspan without including the rows outside the colspan (B1).
Update
After the last answer I tried to build from there, but I'm still not able to get the rows under the main row span row to build the query combine with the main row. This was my try
$x("//tr[ (preceding-sibling::tr[ .//td[ contains(#id, 'main_field')]])[1][.//td[contains(text(),'A1')]] ]")
I tried to get all tr that have a preceding sibling tr with the given known partial id, take the first one of that list with [0] (direct sibling with the given id) then filter with the content A1. But I do not get anything.

Answer 1

If you want to do this in a single expression in XPath 1.0, it gets a bit complex. You could approach it like this, building it piece by piece.
As a starting point, here's how you select your "main row":
//tr[td[contains(text(),'A1')]]
Building on this, you can select the following rows within the same rowspan:
//tr[td[contains(text(),'A1')]]
/following-sibling::tr[
position() < number(preceding-sibling::tr/td[contains(text(),'A1')]/#rowspan)
]
However, this does not include the "main row" itself. To get it also, you can take a union of both of the above with the union operator (|), so you get both the main row and the following rows that fall within the the same rowspan:
(//tr[td[contains(text(),'A1')]]
|//tr[td[contains(text(),'A1')]]
/following-sibling::tr[
position() < number(preceding-sibling::tr/td[contains(text(),'A1')]/#rowspan)
]
)
Now that you have the set of rows of interest at hand, you can further narrow down to the rows that you want within that set, e.g.:
(//tr[td[contains(text(),'A1')]]
|//tr[td[contains(text(),'A1')]]
/following-sibling::tr[
position() < number(preceding-sibling::tr/td[contains(text(),'A1')]/#rowspan)
]
)[td/span[#class='searching_for_this' and contains(text(), 'AA1')]]