使用 XML::LibXML 获取 XML 元素

Question

I need to process xml-file in a different way depending on type element. I use findnodes($xpath,$contextNode) method to find elements that are part of one structure. Elements can have two types <type>A</type> and <type>B</type> . Is there any way, when I process node with type B to "jump" to element with type A in order to get ID-value?

Basically I need to be able to find out that element with id 2 and type B is in the same structure as element with id 1 and type A. Any ideas how can I jump in structure upwards?

<structure>
    <element>
        <def>
            <id>1</id>
            <type>A</type>
        </def>
    </element>
    <element>
        <def>
            <id>2</id>
            <type>B</type>
        </def>
    </element>
</structure>

Answer 1

You can use the parent axis.

my ( $elementB ) = $xpc->findnodes(
   "parent::*/element[ def/type='B' ]",
   $elementA,
);

Demo on xpather.com ^[1]

---

XPath provides the convenient .. shortcut.

my ( $elementB ) = $xpc->findnodes(
   "../element[ def/type='B' ]",
   $elementA,
);

Demo on xpather.com ^[1]

---

Alternatively, you can use a combination of the preceding-sibling and following-sibling axes.

my ( $elementB ) = $xpc->findnodes(
   " preceding-sibling::*[ def/type='B' ]
   | following-sibling::*[ def/type='B' ]
   ",
   $elementA,
);

Demo on xpather.com ^[1]

---

Outside of XPath, $node->parentNode will get the parent node. This answer focused on XPath since the OP mentioned findnodes , and because it's terser and simpler.

---

1. I think this supports XPath 2, while libxml2 (used by XML::LibXML) only supports XPath 1, but these paths should be valid for both.