Package checkxml provides functions for validating XML data against a struct definition.

The MissingXMLTags functions decode XML data and return a slice of struct fields that will not be set using the encoding/xml Unmarshal function. The UnknownXMLTags functions decode XML data and return a slice of XML elements that will not be decoded using the encoding/xml Unmarshal function for the specified struct definition.

Example:

data := `<doc>
           <elem1>a simple element</elem1>
           <elem2>
             <subelem>something more complex</subelem>
             <notes>take a look at this</notes>
           </elem2>
           <elem4>extraneous</elem4>
         </doc>`

type sub struct {
	Subelem string `xml:"subelem,omitempty"`
	Another string `xml:"another"`
}
type elem struct {
	Elem1 string `xml:"elem1"`
	Elem2 sub    `xml:"elem2"`
	Elem3 bool   `xml:"elem3"`
}

e := new(elem)
result, root, _ := MissingXMLTags([]byte(data), e)
// result: [elem2.another elem3]
// root: doc

result, root, _ = UnknownXMLTags([]byte(data), e)
// result: [elem2.notes elem4]
// root: doc

NOTE: this package is dependent upon clbanning/mxj.

USAGE

https://godoc.org/github.com/clbanning/checkxml

LIMITATION

The MissingXMLTags function does not support recursive structs.

MOTIVATION

I was having a conversation and the topic of validating streams of XML data from various sources came up. What the developer was looking for was a "signature" that could be used to route the data for refactoring to meet the production application requirements. Having already done something similar for JSON with the checkjson package and having the mxj package available, it was a simple exercise to do a mashup of the two packages.

RELATED

There is a complementary package for checking JSON objects against structs at https://github.com/clbanning/checkjson.