README
¶
libxml2
Interface to libxml2, with DOM interface.
Why?
I needed to write go-xmlsec. This means we need to build trees using libxml2, and then muck with it in xmlsec: Two separate packages in Go means we cannot (safely) pass around C.xmlFooPtr objects (also, you pay a penalty for pointer types). This package carefully avoid references to C.xmlFooPtr types and uses uintptr to pass data around, so other libraries that needs to interact with libxml2 can safely interact with it.
Status
- This library should be considered alpha grade. API may still change.
- Much of commonly used functionalities from libxml2 that I use are there already, and are known to be functional
Package Layout:
| Name | Description |
|---|---|
| libxml2 | Globally available utility functions, such as ParseString |
| types | Common data types, such as types.Node |
| parser | Parser routines |
| dom | DOM-like manipulation of XML document/nodes |
| xpath | XPath related tools |
| xsd | XML Schema related tools |
| clib | Wrapper around C libxml2 library - DO NOT TOUCH IF UNSURE |
Features
Create XML documents using DOM-like interface:
d := dom.CreateDocument()
e, err := d.CreateElement("foo")
if err != nil {
println(err)
return
}
d.SetDocumentElement(e)
...
Parse documents:
d, err := libxml2.ParseString(xmlstring)
if err != nil {
println(err)
return
}
Use XPath to extract node values:
text := xpath.String(node.Find("//xpath/expression"))
Examples
Basic XML Example
import (
"log"
"net/http"
"github.com/KemalovMaulen/libxml2"
"github.com/KemalovMaulen/libxml2/parser"
"github.com/KemalovMaulen/libxml2/types"
"github.com/KemalovMaulen/libxml2/xpath"
)
func ExampleXML() {
res, err := http.Get("http://blog.golang.org/feed.atom")
if err != nil {
panic("failed to get blog.golang.org: " + err.Error())
}
p := parser.New()
doc, err := p.ParseReader(res.Body)
defer res.Body.Close()
if err != nil {
panic("failed to parse XML: " + err.Error())
}
defer doc.Free()
doc.Walk(func(n types.Node) error {
log.Printf(n.NodeName())
return nil
})
root, err := doc.DocumentElement()
if err != nil {
log.Printf("Failed to fetch document element: %s", err)
return
}
ctx, err := xpath.NewContext(root)
if err != nil {
log.Printf("Failed to create xpath context: %s", err)
return
}
defer ctx.Free()
ctx.RegisterNS("atom", "http://www.w3.org/2005/Atom")
title := xpath.String(ctx.Find("/atom:feed/atom:title/text()"))
log.Printf("feed title = %s", title)
}
Basic HTML Example
func ExampleHTML() {
res, err := http.Get("http://golang.org")
if err != nil {
panic("failed to get golang.org: " + err.Error())
}
doc, err := libxml2.ParseHTMLReader(res.Body)
if err != nil {
panic("failed to parse HTML: " + err.Error())
}
defer doc.Free()
doc.Walk(func(n types.Node) error {
log.Printf(n.NodeName())
return nil
})
nodes := xpath.NodeList(doc.Find(`//div[@id="menu"]/a`))
for i := 0; i < len(nodes); i++ {
log.Printf("Found node: %s", nodes[i].NodeName())
}
}
XSD Validation
import (
"io/ioutil"
"log"
"os"
"path/filepath"
"github.com/KemalovMaulen/libxml2"
"github.com/KemalovMaulen/libxml2/xsd"
)
func ExampleXSD() {
xsdfile := filepath.Join("test", "xmldsig-core-schema.xsd")
f, err := os.Open(xsdfile)
if err != nil {
log.Printf("failed to open file: %s", err)
return
}
defer f.Close()
buf, err := ioutil.ReadAll(f)
if err != nil {
log.Printf("failed to read file: %s", err)
return
}
s, err := xsd.Parse(buf)
if err != nil {
log.Printf("failed to parse XSD: %s", err)
return
}
defer s.Free()
d, err := libxml2.ParseString(`<foo></foo>`)
if err != nil {
log.Printf("failed to parse XML: %s", err)
return
}
if err := s.Validate(d); err != nil {
for _, e := range err.(xsd.SchemaValidationError).Errors() {
log.Printf("error: %s", e.Error())
}
return
}
log.Printf("validation successful!")
}
Caveats
Other libraries
There exists many similar libraries. I want speed, I want DOM, and I want XPath.When all of these are met, I'd be happy to switch to another library.
For now my closest contender was xmlpath, but as of this writing it suffers in the speed (for xpath) area a bit:
shoebill% go test -v -run=none -benchmem -benchtime=5s -bench .
PASS
BenchmarkXmlpathXmlpath-4 500000 11737 ns/op 721 B/op 6 allocs/op
BenchmarkLibxml2Xmlpath-4 1000000 7627 ns/op 368 B/op 15 allocs/op
BenchmarkEncodingXMLDOM-4 2000000 4079 ns/op 4560 B/op 9 allocs/op
BenchmarkLibxml2DOM-4 1000000 11454 ns/op 264 B/op 7 allocs/op
ok github.com/KemalovMaulen/libxml2 37.597s
See Also
Credits
- Work on this library was generously sponsored by HDE Inc (https://www.hde.co.jp)
Documentation
¶
Overview ¶
Package libxml2 is an interface to libxml2 library, providing XML and HTML parsers with DOM interface. The inspiration is Perl5's XML::LibXML module.
This library is still in very early stages of development. API may still change without notice.
For the time being, the API is being written so that thye are as close as we can get to DOM Layer 3, but some methods will, for the time being, be punted and aliases for simpler methods that don't necessarily check for the DOM's correctness will be used.
Also, the return values are still shaky -- I'm still debating how to handle error cases gracefully.
Index ¶
- func Parse(buf []byte, o ...parser.Option) (types.Document, error)
- func ParseHTML(content []byte, options ...parser.HTMLOption) (types.Document, error)
- func ParseHTMLReader(in io.Reader, options ...parser.HTMLOption) (types.Document, error)
- func ParseHTMLString(content string, options ...parser.HTMLOption) (types.Document, error)
- func ParseReader(rdr io.Reader, o ...parser.Option) (types.Document, error)
- func ParseString(s string, o ...parser.Option) (types.Document, error)
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func ParseHTML ¶
ParseHTML parses an HTML document. You can omit the options argument, or you can provide one bitwise-or'ed option
func ParseHTMLReader ¶
ParseHTMLReader parses an HTML document. You can omit the options argument, or you can provide one bitwise-or'ed option
func ParseHTMLString ¶
ParseHTMLString parses an HTML document. You can omit the options argument, or you can provide one bitwise-or'ed option
func ParseReader ¶
ParseReader parses XML from the given io.Reader and returns a Document.
Types ¶
This section is empty.
Source Files
Directories
¶
| Path | Synopsis |
|---|---|
|
Package clib holds all of the dirty C interaction for go-libxml2.
|
Package clib holds all of the dirty C interaction for go-libxml2. |
|
internal
|
|
|
Package types exist to provide with common types that are used through out in go-libxml2.
|
Package types exist to provide with common types that are used through out in go-libxml2. |
|
Package xpath contains tools to handle XPath evaluation.
|
Package xpath contains tools to handle XPath evaluation. |
|
Package xsd contains some of the tools available from libxml2 that allows you to validate your XML against an XSD This is basically all you need to do: schema, err := xsd.Parse(xsdsrc) if err != nil { panic(err) } defer schema.Free() if err := schema.Validate(doc); err != nil{ for _, e := range err.(SchemaValidationErr).Error() { println(e.Error()) } }
|
Package xsd contains some of the tools available from libxml2 that allows you to validate your XML against an XSD This is basically all you need to do: schema, err := xsd.Parse(xsdsrc) if err != nil { panic(err) } defer schema.Free() if err := schema.Validate(doc); err != nil{ for _, e := range err.(SchemaValidationErr).Error() { println(e.Error()) } } |