I am searching for hours a solution (in PHP) to this:

I have some XML files which structure may vary

<page id="this is what I want to extract">
  <boh>
     <bah>
        <other childs (maybe one, maybe ten)>
           <ref id="This is all I know!"> Some text Lorem Ipsum</ref>

I need two formulas which would be able:

to extract the page id from a search inside all the entire file for either a specific ref id, or some partial text inside the ref tag

In brief all I know about this file is: It has a ref tag, which sometimes has an id and always some text inside. I either have the ref id or some portions of the text. I need to find the id of the page node in which ref is contained.

So: Search for "This is all I know!" as ref id would output "this is what I want to extract"

as well as

Search for "Lorem" as text inside ref would output "this is what I want to extract"

How can I accomplish this? I've googled a lot, I think I should make something related with SimpleXML and XPATH, but I never used them in this way.

You can use this XPath expression in your code:

//page[contains(.//ref/text(), 'Lorem')]/@id

It will search all <ref> elements that are descendants of <page> and compare the text with the string 'Lorem' (which, in your code, you should pass as a variable). It will return a set containing all the ids of the <page> elements that contain the matching text.

more