HTML5-PHP
HTML5 is a standards-compliant HTML5 parser and writer written entirely in PHP.
It is stable and used in many production websites, and has
well over five million downloads.
HTML5 provides the following features.
- An HTML5 serializer
- Support for PHP namespaces
- Composer support
- Event-based (SAX-like) parser
- A DOM tree builder
- Interoperability with QueryPath
- Runs on PHP 5.3.0 or newer
Installation
Install HTML5-PHP using composer.
By adding the masterminds/html5 dependency to your composer.json file:
{
"require" : {
"masterminds/html5": "^2.0"
},
}
By invoking require command via composer executable:
composer require masterminds/html5
Basic Usage
HTML5-PHP has a high-level API and a low-level API.
Here is how you use the high-level HTML5 library API:
<?php
// Assuming you installed from Composer:
require "vendor/autoload.php";
use Masterminds\HTML5;
// An example HTML document:
$html = <<< 'HERE'
<html>
<head>
<title>TEST</title>
</head>
<body id='foo'>
<h1>Hello World</h1>
<p>This is a test of the HTML5 parser.</p>
</body>
</html>
HERE;
// Parse the document. $dom is a DOMDocument.
$html5 = new HTML5();
$dom = $html5->loadHTML($html);
// Render it as HTML5:
print $html5->saveHTML($dom);
// Or save it to a file:
$html5->save($dom, 'out.html');
The $dom created by the parser is a full DOMDocument object. And the
save() and saveHTML() methods will take any DOMDocument.
Options
It is possible to pass in an array of configuration options when loading
an HTML5 document.
// An associative array of options
$options = array(
'option_name' => 'option_value',
);
// Provide the options to the constructor
$html5 = new HTML5($options);
$dom = $html5->loadHTML($html);
The following options are supported:
encode_entities(boolean): Indicates that the serializer should aggressively
encode characters as entities. Without this, it only encodes the bare
minimum.disable_html_ns(boolean): Prevents the parser from automatically
assigning the HTML5 namespace to the DOM document. This is for
non-namespace aware DOM tools.target_document(\DOMDocument): A DOM document that will be used as the
destination for the parsed nodes.implicit_namespaces(array): An assoc array of namespaces that should be
used by the parser. Name is tag prefix, value is NS URI.
The Low-Level API
This library provides the following low-level APIs that you can use to
create more customized HTML5 tools:
- A SAX-like event-based parser that you can hook into for special kinds
of parsing. - A flexible error-reporting mechanism that can be tuned to document
syntax checking. - A DOM implementation that uses PHP's built-in DOM library.
The unit tests exercise each piece of the API, and every public function
is well-documented.
Parser Design
The parser is designed as follows:
- The
Scannerhandles scanning on behalf of the parser. - The
Tokenizerrequests data off of the scanner, parses it, clasifies
it, and sends it to anEventHandler. It is a recursive descent parser. - The
EventHandlerreceives notifications and data for each specific
semantic event that occurs during tokenization. - The
DOMBuilderis anEventHandlerthat listens for tokenizing
events and builds a document tree (DOMDocument) based on the events.
Serializer Design
The serializer takes a data structure (the DOMDocument) and transforms
it into a character representation -- an HTML5 document.
The serializer is broken into three parts:
- The
OutputRulescontain the rules to turn DOM elements into strings. The
rules are an implementation of the interfaceRulesInterfaceallowing for
different rule sets to be used. - The
Traverser, which is a special-purpose tree walker. It visits
each node node in the tree and uses theOutputRulesto transform the node
into a string. HTML5manages theTraverserand stores the resultant data
in the correct place.
The serializer (save(), saveHTML()) follows the
section 8.9 of the HTML 5.0 spec.
So tags are serialized according to these rules:
- A tag with children: <foo>CHILDREN</foo>
- A tag that cannot have content: <foo> (no closing tag)
- A tag that could have content, but doesn't: <foo></foo>
Known Issues (Or, Things We Designed Against the Spec)
Please check the issue queue for a full list, but the following are
issues known issues that are not presently on the roadmap:
- Namespaces: HTML5 only supports a selected list of namespaces
and they do not operate in the same way as XML namespaces. A:has no special
meaning.
By default the parser does not support XML style namespaces via:;
to enable the XML namespaces see the XML Namespaces section - Scripts: This parser does not contain a JavaScript or a CSS
interpreter. While one may be supplied, not all features will be
supported. - Rentrance: The current parser is not re-entrant. (Thus you can't pause
the parser to modify the HTML string mid-parse.) - Validation: The current tree builder is not a validating parser.
While it will correct some HTML, it does not check that the HTML
conforms to the standard. (Should you wish, you can build a validating
parser by extending DOMTree or building your own EventHandler
implementation.)- There is limited support for insertion modes.
- Some autocorrection is done automatically.
- Per the spec, many legacy tags are admitted and correctly handled,
even though they are technically not part of HTML5.
- Attribute names and values: Due to the implementation details of the
PHP implementation of DOM, attribute names that do not follow the
XML 1.0 standard are not inserted into the DOM. (Effectively, they
are ignored.) If you've got a clever fix for this, jump in! - Processor Instructions: The HTML5 spec does not allow processor
instructions. We do. Since this is a server-side library, we think
this is useful. And that means, dear reader, that in some cases you
can parse the HTML from a mixed PHP/HTML document. This, however,
is an incidental feature, not a core feature. - HTML manifests: Unsupported.
- PLAINTEXT: Unsupported.
- Adoption Agency Algorithm: Not yet implemented. (8.2.5.4.7)
XML Namespaces
To use XML style namespaces you have to configure well the main HTML5 instance.
use Masterminds\HTML5;
$html = new HTML5(array(
"xmlNamespaces" => true
));
$dom = $html->loadHTML('<t:tag xmlns:t="http://www.example.com"/>');
$dom->documentElement->namespaceURI; // http://www.example.com
You can also add some default prefixes that will not require the namespace declaration,
but its elements will be namespaced.
use Masterminds\HTML5;
$html = new HTML5(array(
"implicitNamespaces"=>array(
"t"=>"http://www.example.com"
)
));
$dom = $html->loadHTML('<t:tag/>');
$dom->documentElement->namespaceURI; // http://www.example.com
Thanks to...
The dedicated (and patient) contributors of patches small and large,
who have already made this library better.See the CREDITS file for
a list of contributors.
We owe a huge debt of gratitude to the original authors of html5lib.
While not much of the original parser remains, we learned a lot from
reading the html5lib library. And some pieces remain here. In
particular, much of the UTF-8 and Unicode handling is derived from the
html5lib project.
License
This software is released under the MIT license. The original html5lib
library was also released under the MIT license.
See LICENSE.txt
Certain files contain copyright assertions by specific individuals
involved with html5lib. Those have been retained where appropriate.