How can I check if the contents of a result set are the same in the XML?-CodePudding

I have many XMLs, which can contain a specific element, functioning as an identifier for the record. It's possible that it's missing from the file, or there is only one occurence, this is the easiest case, that I could solve already. But it's also possible that there are many of these elements, in that case I need to check if their values are the same. I attached some example XMLs with the expected output:

Input 1:

<root>
  <daily>
    <group>ID@12</group>
    <surrounded>rich</surrounded>
    <clothing>rod</clothing>
    <outside>-1084855717</outside>
    <section>-1103031959</section>
  </daily>
  <group>ID@13</group>
  <account>remain</account>
  <point>-1624875729</point>
  <cotton>941344054.3731294</cotton>
  <group>ID@12</group>
  <scale>almost</scale>
</root>

Output: false, because there is three occurences, but one of them is different from the other 2.

Input 2:

<root>
  <daily>
    <group>ID@12</group>
    <mill>spread</mill>
    <surrounded>rich</surrounded>
    <clothing>rod</clothing>
    <outside>-1084855717</outside>
    <section>-1103031959</section>
  </daily>
  <group>ID@12</group>
  <account>remain</account>
  <point>-1624875729</point>
  <cotton>941344054.3731294</cotton>
  <scale>almost</scale>
</root>

Output: true, because there are two occurences, and both are the same value.

I would have to use XPath expressions, but it could be also okay with XSLT file.

I could manage to find the relevant elements with the following expression, but I'm stuck at that point:

//group[starts-with(text(), 'ID@')]

CodePudding user response：

If //group[starts-with(text(), 'ID@')] selects the elements I would shorten that to //group[starts-with(., 'ID@')] first and then use count(distinct-values(//group[starts-with(., 'ID@')])) = 1, for instance. That assumes XPath 2 and distinct-values is available, as your XQuery tag suggests.

In both XSLT or XQuery it could of course be also considered a grouping of those elements by their own value and checking that there is exactly one group.

CodePudding user response：

This may look strange, but I will explain it:

not(//group[starts-with(text(), 'ID@')] != //group[starts-with(text(), 'ID@')])

The expression selects the set of ID nodes and compares it with the same set of ID nodes using the != operator. You might think that $x != $x would always return a false value, but that's not how it works, actually. See the section on boolean expressions in the XPath 1.0 specification:

If both objects to be compared are node-sets, then the comparison will be true if and only if there is a node in the first node-set and a node in the second node-set such that the result of performing the comparison on the string-values of the two nodes is true.

In other words, if any pair of elements from the first node-set and the second node-set satisfy the comparison operator (!=, in our case), then the expression returns true.

In our case, if you have a node-set $x and you evaluate the expression $x != $x it will return true if there is a member of $x which is not equal to some (other) member of $x.

Using the not() function you can then negate the result, e.g. not($x != $x) will return false if the members of $x are not all equal to each other.