使用xpath选择第一个结果的子项

I have a piece of XML where the same information can appear as a child of different nodes. Such as :

<root>
<category id=1>
        <product id="ABC123" >

              <sizes>
                    <size name="S"/>
                <size name="M"/>
                <size name="L"/>
                <size name="XL"/>
                <size name="2XL"/>
                <size name="3XL"/>
              </sizes>
            </product>

                 </products>
           </category>
<category id=2>
        <products>
        <product id="ABC123" >

              <sizes>
                    <size name="S"/>
                <size name="M"/>
                <size name="L"/>
                <size name="XL"/>
                <size name="2XL"/>
                <size name="3XL"/>
              </sizes>
            </product>

          <product id="PPP543" >

              <sizes>
                    <size name="S"/>
                <size name="M"/>
                <size name="L"/>
                <size name="XL"/>
              </sizes>
            </product>

                 </products>
           </category>

My goal is to select the sizes of product id ABC123 and store them as an array. The current code that I have is :

$arrTest=array();


    foreach($xml->xpath('//root/category/products/product[@id= "'.$productCall.'" ]/sizes/size') as $size){

              array_push($arrTest, $size["name"]);
      }

$productCall is the id I am looking for. In this case it's ABC123.

The output is S,M,L,XL,2XL,3XL,S,M,L,XL,2XL,3XL. Meaning that it is reading the two entries that were found. I expected this given the foreach loop, but I can't seem to find a way to just get the output of the first result. I have tried adding [0] and [1] :

$y=$xml->xpath('//root/category/products/product[@id= "'.$productCall.'" ][1]/sizes/size');

[0] returns nothing and [1] returns the same results I'm already getting.

I'm hoping this is a simple matter of me missing something something basic or just over-thinking, as I really haven't worked with xpath before.

Use:

(/*/*//product[@id='ABC123']/sizes)[1]/size

Or use:

((/*/*/product | /*/*/products/product)/sizes)[1]/size

XSLT - based verification:

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>

 <xsl:template match="/">
  <xsl:copy-of select=
    "(/*/*//product[@id='ABC123']/sizes)[1]/size"/>
========
  <xsl:copy-of select=
    "((/*/*/product | /*/*/products/product)/sizes)[1]/size"/>

 </xsl:template>
</xsl:stylesheet>

When this transformation is applied to the provided XML document (after correcting it into a well-formed one):

<root>
    <category id="1">
        <product id="ABC123" >
            <sizes>
                <size name="S"/>
                <size name="M"/>
                <size name="L"/>
                <size name="XL"/>
                <size name="2XL"/>
                <size name="3XL"/>
            </sizes>
        </product>
    </category>
    <category id="2">
        <products>
            <product id="ABC123" >
                <sizes>
                    <size name="S"/>
                    <size name="M"/>
                    <size name="L"/>
                    <size name="XL"/>
                    <size name="2XL"/>
                    <size name="3XL"/>
                </sizes>
            </product>
            <product id="PPP543" >
                <sizes>
                    <size name="S"/>
                    <size name="M"/>
                    <size name="L"/>
                    <size name="XL"/>
                </sizes>
            </product>
        </products>
    </category>
</root>

the two XPath expressions are evaluated and the selected nodes from each one, well visually delimited, are copied to the output -- both correct:

<size name="S"/>
<size name="M"/>
<size name="L"/>
<size name="XL"/>
<size name="2XL"/>
<size name="3XL"/>
========
  <size name="S"/>
<size name="M"/>
<size name="L"/>
<size name="XL"/>
<size name="2XL"/>
<size name="3XL"/>

Do note: I expect the second XPath expression to be more efficient (on sufficiently large documents), because it doesn't use the // pseudo-operator.