<p>Say I have XML like this:</p>
<pre><code><ITEMS>
<ITEM INDEX="1">
<NAME>foo</NAME>
<TYPE ID="123">bar</TYPE>
</ITEM>
</ITEMS>
</code></pre>
<p>I can parse (unmarshal) this into a struct like this:</p>
<pre><code>type Items struct {
Items []Item `xml:"ITEM"`
}
type Item struct {
Index int `xml:"INDEX,attr"`
Name string `xml:"NAME"`
Type Type `xml:"TYPE"`
}
type Type struct {
ID int `xml:"ID,attr"`
Value string `xml:",chardata"`
}
</code></pre>
<p>Is there any way to get the ID from the TYPE tag without making a separate struct for it (ie: set Item.Type to "bar" and Item.TypeID to "123")?</p>
<p>It might not look too bad in that example, but when you're working with large XML responses with dozens of tags, some that have attributes and some that don't, it gets very tedious and confusing to program and look at, not to mention the inconsistency between accessing those values (Item.Name vs Item.Type.Value).</p>
<p>Obviously some kind of wrapper could be written over top of this to make accessing values and attributes more consistent, but that doesn't fix the underlying issue.</p>
<p>I've mostly read through the Go XML documentation, I don't think it mentions anything like this, though the long walls of text about how it goes about parsing the XML were somewhat confusing to keep track of, so I might have missed it somewhere in that.</p>
<hr/>**评论：**<br/><br/>Spirit_of_Stallman: <pre><p>It was thought that this should work like this: <code>xml:"TYPE>ID,attr"</code>. But for many years it does't work, and all don't care :) <a href="https://github.com/golang/go/issues/3688" rel="nofollow">Ticket</a>.</p></pre>Hakkin: <pre><p>That's unfortunate. Guess I'll have to put a wrapper around it instead of directly accessing the structs.</p>
<p>Thank you for the link to the issue.</p></pre>
