[rustc.git] / vendor / markup5ever_rcdom / examples / html2html.rs

// Copyright 2014-2017 The html5ever Project Developers. See the
// COPYRIGHT file at the top-level directory of this distribution.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.

//! Parse and re-serialize a HTML5 document.
//!
//! This is meant to produce the exact same output (ignoring stderr) as
//!
//!   java -classpath htmlparser-1.4.jar nu.validator.htmlparser.tools.HTML2HTML
//!
//! where htmlparser-1.4.jar comes from http://about.validator.nu/htmlparser/

extern crate html5ever;
extern crate markup5ever_rcdom as rcdom;

use std::default::Default;
use std::io::{self, Write};

use html5ever::driver::ParseOpts;
use html5ever::tendril::TendrilSink;
use html5ever::tree_builder::TreeBuilderOpts;
use html5ever::{parse_document, serialize};
use rcdom::{RcDom, SerializableHandle};

fn main() {
    let opts = ParseOpts {
        tree_builder: TreeBuilderOpts {
            drop_doctype: true,
            ..Default::default()
        },
        ..Default::default()
    };
    let stdin = io::stdin();
    let dom = parse_document(RcDom::default(), opts)
        .from_utf8()
        .read_from(&mut stdin.lock())
        .unwrap();

    // The validator.nu HTML2HTML always prints a doctype at the very beginning.
    io::stdout()
        .write_all(b"<!DOCTYPE html>\n")
        .ok()
        .expect("writing DOCTYPE failed");
    let document: SerializableHandle = dom.document.clone().into();
    serialize(&mut io::stdout(), &document, Default::default())
        .ok()
        .expect("serialization failed");
}
Commit	Line	Data
83c7162d XL	1	// Copyright 2014-2017 The html5ever Project Developers. See the
	2	// COPYRIGHT file at the top-level directory of this distribution.
	3	//
	4	// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
	5	// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
	6	// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
	7	// option. This file may not be copied, modified, or distributed
	8	// except according to those terms.
	9
	10	//! Parse and re-serialize a HTML5 document.
	11	//!
	12	//! This is meant to produce the exact same output (ignoring stderr) as
	13	//!
	14	//! java -classpath htmlparser-1.4.jar nu.validator.htmlparser.tools.HTML2HTML
	15	//!
	16	//! where htmlparser-1.4.jar comes from http://about.validator.nu/htmlparser/
	17
	18	extern crate html5ever;
3dfed10e	19	extern crate markup5ever_rcdom as rcdom;
83c7162d	20
83c7162d	21	use std::default::Default;
dc9dc135	22	use std::io::{self, Write};
83c7162d	23
83c7162d	24	use html5ever::driver::ParseOpts;
83c7162d XL	25	use html5ever::tendril::TendrilSink;
83c7162d XL	26	use html5ever::tree_builder::TreeBuilderOpts;
dc9dc135	27	use html5ever::{parse_document, serialize};
3dfed10e	28	use rcdom::{RcDom, SerializableHandle};
83c7162d XL	29
	30	fn main() {
	31	let opts = ParseOpts {
	32	tree_builder: TreeBuilderOpts {
	33	drop_doctype: true,
	34	..Default::default()
	35	},
	36	..Default::default()
	37	};
	38	let stdin = io::stdin();
	39	let dom = parse_document(RcDom::default(), opts)
	40	.from_utf8()
	41	.read_from(&mut stdin.lock())
	42	.unwrap();
	43
	44	// The validator.nu HTML2HTML always prints a doctype at the very beginning.
dc9dc135 XL	45	io::stdout()
	46	.write_all(b"<!DOCTYPE html>\n")
	47	.ok()
	48	.expect("writing DOCTYPE failed");
3dfed10e XL	49	let document: SerializableHandle = dom.document.clone().into();
3dfed10e XL	50	serialize(&mut io::stdout(), &document, Default::default())
dc9dc135 XL	51	.ok()
dc9dc135 XL	52	.expect("serialization failed");
83c7162d	53	}