Toby Inkster

NAME

HTML::HTML5::Builder - erect some scaffolding for your documents

SYNOPSIS

  use HTML::HTML5::Builder qw[:standard JQUERY];

  open my $fh, '<', 'inline-script.js';
  
  print html(
    -lang => 'en',
    head(
      title('Test'),
      Meta(-charset => 'utf-8'),
    ),
    body(
      h1('Test'),
      p('This is a test.'),
      JQUERY(-version => '1.6.4'),
      script(-type => 'text/javascript', $fh),
    ),
  );

DESCRIPTION

This module can export function names corresponding to any HTML5 element.

Each function returns an XML::LibXML::Element with the same name as the function. The arguments to each function are processed as a list, and used to set the attributes and contents of that element.

For each item on the list:

  • if it's an XML::LibXML::Element, XML::LibXML::TextNode, XML::LibXML::Comment, or XML::LibXML::PI, it's appended as a child of the returned element.

  • if it's an XML::LibXML::NodeList, each item on the list is appended as a child of the returned element.

  • if it's an XML::LibXML::Attr, it's set as an attribute on the returned element

  • if it's an IO::Handle, then it will be slurped and appended to the returned element as a text node.

  • if it's a scalar reference, then the returned element is also assigned to it. (This feature is at risk.)

  • if it's a scalar (string) some guesswork is conducted to figure out whether you're setting an attribute and value, or whether the string should be used as a text node. The presence of a hyphen at the start of the string is the main deciding factor.

      p('-class', 'warning', '$LordLucan not found.');

    In this example, a paragraph element is returned, with the class attribute set to 'warning' and the textual contents '$LordLucan not found.'.

    Sometimes it's necessary to protect values against this guesswork. By passing a hashref, all the keys and values are interpreted as setting attributes; by passing an arrayref, all values are interpreted as setting the contents of the element.

      p(['-class'], { warning => '$LordLucan not found.' });

    In this example, a paragraph element is returned, with the warning attribute set to '$LordLucan not found.' and the textual contents '-class'.

  • Anything else is stringified and added as a text node. This is useful for things with sensible stringification defined, such as DateTime and URI objects, but less so for some other objects, so you will sometimes get a warning if warnings are enabled. Warnings can be disabled using:

      no warnings 'HTML::HTML::Builder';

Exceptional Cases

The html function does not return an XML::LibXML::Element, but rather a HTML::HTML5::Builder::Document object.

There is special handling for time (or Time). If the first parameter passed to it is a DateTime object, then that object is used to set its datetime attribute. If there are no subsequent parameters, then the stringified form of the object is also used to form the content of the <time> element.

Note that the functions that generate <meta>, <link>, <q>, <time>, <sub>, <s> and <map> HTML elements are named Meta(), Link(), Q(), Time(), Sub(), S() and Map() respectively, with an upper-case first letter. This is because each of these names corresponds to a built-in perl keyword (except meta, which is used by Moose). The lower-case versions of these do exist, and can be exported if you ask for them explicitly. The lower-case versions are also available as methods using the object-oriented syntax. (In fact, lower case and ucfirst versions exist for all HTML elements - they're just not always exportable.)

General Purpose Functions

ELEMENT($tagname, @arguments)

If you need to insert an element which doesn't have its own function.

TEXT($string)

Produces a text node.

COMMENT($string)

Produces an HTML comment.

CHUNK($string)

Parses the string as HTML, and produces a list of elements, text nodes and comments.

This should be a so-called "balanced chunk". Due to limitations in HTML::HTML5::Parser, this only works for body content. Croaks if HTML::HTML5::Parser is not installed.

XML_CHUNK($string)

More useful version of CHUNK, without the restriction on content, but input needs to be a balanced and well-formed XML chunk.

RAW_CHUNK($string)

This allows you to include stuff that isn't anything close to valid HTML into the output document, such as a PHP block. e.g.

  html(
    head(
      title('Funny test'),
      ),
    body(
      h1('Funny test'),
      RAW_CHUNK("<p>Here's a fish: <=><"),
      ),
    );

A processing instruction is used to represent this data in the DOM. HTML::HTML5::Writer can detect that processing instruction and use it to output the raw data. If you're not using HTML::HTML5::Writer to serialise the document, then you may need to post-process the serialised document.

With great power comes great responsibility.

Boiler-Plate Functions

There are also a number of functions that create lists of multiple HTML elements, for boiler-plate code.

JQUERY(-version => $ver, %options)

Link to jQuery at a CDN.

Other options include -source, to indicate where to link to jQuery (currently allowed values are "Google", "Microsoft" and "official"); and -min, a boolean which indicates whether the minified version should be linked to (true by default).

Setting option -ui to true, also includes jQuery UI. A version number can be indicated using -ui_version. A theme can be included setting -theme. Setting either -ui_version or -theme will imply -ui.

  JQUERY(
    -source     => 'official',
    -version    => '1.6.4',
    -ui_version => '1.8.16',
    -theme      => 'eggplant',
    );

If versions aren't provided, defaults to the latest versions of the libraries that the author of HTML::HTML5::Builder was aware of at the time of publication. If you choose a version which is known to be unavailable at the selected CDN, the function should automatically choose a slightly later version.

CREATIVE_COMMONS($licence)
CREATIVE_COMMONS(-licence => $licence, %options)

$licence can be one of 'by', 'by-nd', 'by-nc', 'by-sa', 'by-sa-nc', or 'by-sa-nd'.

Other options supported are:

  • -url - URL of the thing being licensed (if not the page itself)

  • -size - 'large' or 'small' for the image

  • -title - title of the work

  • -attributionName - name people should use for attribution

  • -attributionURL - link people should use for attribution

OPENGRAPH(%data)

Returns a list of <meta> elements providing Open Graph Protocol data for your page.

  OPENGRAPH(
    -title       => "Hello World",
    -type        => "example",
    -description => "A global greeting.",
    );

Exporting Functions

None by default. Pretty much anything can be exported on request.

Export tags:

  • :all - everything

  • :standard - elements that are not obsolete in HTML5, plus ELEMENT, TEXT, COMMENT, CHUNK, XML_CHUNK and RAW_CHUNK

  • :metadata - head title base Link Meta style

  • :sections - body div section nav article aside h1 h2 h3 h4 h5 h6 header footer address

  • :grouping - p hr br pre dialog blockquote ol ul li dl dt dd

  • :text - a Q cite em strong small mark dfn abbr progress meter code var samp kbd Sub sup span i b bdo ruby rt rp Time

  • :embedded - figure img iframe embed object param video audio source canvas area

  • :tabular - table thead tbody tfoot th td colgroup col caption

  • :form - form fieldset label input button select datalist optgroup option textarea output

Object Oriented Interface

You can also use these functions as methods of an object blessed into the HTML::HTML5::Builder package.

  my $b = HTML::HTML5::Builder->new;
  my $document = $b->html(
    -lang => 'en',
    $b->head(
      $b->title('Test', \(my $foo)),
      $b->meta(-charset => 'utf-8'),
    ),
    $b->body(
      $b->h1('Test'),
      $b->p('This is a test.')
    ),
  );

Using with RDF::RDFa::Generator

RDF::RDFa::Generator has a nodes method which returns a handy list of XML::LibXML::Node objects.

  use DateTime;
  use HTML::HTML5::Builder qw[:standard];
  use RDF::RDFa::Generator;
  use RDF::Trine;
  
  my $url   = 'http://dbpedia.org/data/Charles_Darwin';
  my $model = RDF::Trine::Model->new;
  RDF::Trine::Parser->parse_url_into_model($url, $model);
  
  my $gen = RDF::RDFa::Generator->new(style=>'HTML::Pretty');
  
  print html(
    head(
      title("Some Data About Charles Darwin"),
      ),
    body(
      h1("Some Data About Charles Darwin"),
      $gen->nodes($model),
      hr(),
      address(
        "Source: $url", br(),
        "Generated: ", Time(DateTime->now),
        ),
      ),
    );

Nice?

Using with XML::LibXML::PrettyPrint

HTML::HTML5::Builder doesn't nicely indent your markup, but XML::LibXML::PrettyPrint can.

  use HTML::HTML5::Builder qw(:standard);
  use XML::LibXML::PrettyPrint qw(print_xml);
  print_xml html(
    head(title("Test")),
    body(h1("Test"), p("This is a test.")),
    );

BUGS

Please report any bugs to http://rt.cpan.org/Dist/Display.html?Queue=HTML-HTML5-Builder.

SEE ALSO

XML::LibXML, HTML::HTML5::Writer, HTML::HTML5::Builder::Document.

AUTHOR

Toby Inkster <tobyink@cpan.org>.

COPYRIGHT AND LICENCE

This software is copyright (c) 2011 by Toby Inkster.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.

DISCLAIMER OF WARRANTIES

THIS PACKAGE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.




Hosting generously
sponsored by Bytemark