如何将具有不同名称空间的相同XML元素反序列化为结构中的不同元素

Question

I am working with the following kind of an XML structure

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" 
    xmlns:content="http://purl.org/rss/1.0/modules/content/" 
    xmlns:media="http://search.yahoo.com/mrss/">
    <channel>
        <title>TITLE</title>
        <link>http://something.com</link>
        <description>description</description>
        <lastBuildDate>Mon, 19 Dec 2016 16:48:54 +0000</lastBuildDate>
        <language>en</language>
        <item>
            <title>Title</title>
            <description>description</description>
            <author>
                <name>Name</name>
                <uri></uri>
            </author>
            <pubDate>Mon, 19 Dec 2016 16:42:32 +0000</pubDate>
            <link>http://google.com</link>
            <image>...</image>
            <media:description><![CDATA[Natalie Massenet]]></media:description>
            <media:credit>David Fisher/REX/Shutterstock</media:credit>
            <category>Category1</category>
            <category>Category2</category>
            <guid isPermaLink="false">http://something.com/?p=10730393</guid>
            <media:group></media:group>
            <content:encoded>content</content:encoded>
        </item>
    </channel>
</rss>

I am having trouble figuring out how to deserialize <description> and <media:description> into two different element in a struct.

I've tried a following kind of a struct to represent an <item>, but the value of media:description ends up in the struct.

type Item struct {
    // ...other fields
    Description           string   `xml:"description"`
    MediaDescription      string   `xml:"media:description"`
    // ...other fields
}

What would be the best way to do this?

Answer 1

So Go currently is not able to support this (as pointed out by @JimB), but you can still use some built in capabilities to pull out the right element. Start off by defining a new struct XMLElement

type XMLElement struct {
    XMLName xml.Name
    Data    string `xml:",chardata"`
}

This struct can capture the xml data along with the namespace and element name. Next map xml:"description" to an array of XMLElement. This will end up capturing all <*:description> elements in a list. You can then define functions to pull out the right one.

type Item struct {
    // ...other fields
    Descriptions           []XMLElement   `xml:"description"`
    // ...other fields
}

// GetDescription returns the description in the <description> tag
func (i *Item) GetDescription() string {
    for _, elem := range i.Descriptions {
        if elem.XMLName.Space == "" {
            return elem.Data
        }
    }
    return ""
}

// GetMediaDescription returns the description in the <media:description> tag
func (i *Item) GetMediaDescription() string {
    for _, elem := range i.Descriptions {
        if elem.XMLName.Space == "http://search.yahoo.com/mrss/" {
            return elem.Data
        }
    }
    return ""
}