Extract specific xml node from duplicate ones in oracle

Given is the value of "my_xml" column in "XYZ" table

<?xml version="1.0" encoding="UTF-8"?>
<India>
  <city>
    <string>ADI</string>
    <string>Ahmedabad</string>
  </city>
  <city>
    <string>GNR</string>
    <string>Gandhinagar</string>
  </city>
  <city>
    <string>PUN</string>
    <string>Pune</string>
  </city>
  <city>
    <string>RJT</string>
    <string>Rajkot</string>
  </city>
</India>

I am trying to extract value of second string node where first string node value is ADI

Output should be "Ahmedabad" only

Failed attempts:

select t.my_xml.extract('/India/city/string[2]/text()').getStringVal() from XYZ t where t.my_xml.existsNode('/India/city[string[1] = "ADI"]') = 1;

Output for above query is AhmedabadGandhinagarPuneRajkot

Expected output: Ahmedabad

How to extract specific node value for string node here?

You want to select the node that has the ADI text as first string.

Try this:

select 
    t.my_xml.extract('//city[./string[1]/text() = "ADI"]/string[2]/text()').getStringVal()
from XYZ t
    where t.my_xml.existsNode('/India/city[string[1] = "ADI"]') = 1;