MAN page from Fedora 20 perl-HTML-Parser-3.71-4.fc20.x86_64.rpm
Section: User Contributed Perl Documentation (3)
HTML::HeadParser - Parse <HEAD> section of a HTML document
require HTML::HeadParser; $p = HTML::HeadParser->new; $p->parse($text) and print "not finished"; $p->header('Title') # to access <title>....</title> $p->header('Content-Base') # to access <base href="http://..."> $p->header('Foo') # to access <meta http-equiv="Foo" content="..."> $p->header('X-Meta-Author') # to access <meta name="author" content="..."> $p->header('X-Meta-Charset') # to access <meta charset="...">
is a specialized (and lightweight)"HTML::Parser"
that will only parse the <HEAD
>...</HEAD>section of an HTML
document. The parse()
methodwill return a FALSE
value as soon as some <BODY
> element or bodytext are found, and should not be called again after this.
Note that the "HTML::HeadParser" might get confused if raw undecodedUTF-8 is passed to the parse() method. Make sure the strings areproperly decoded before passing them on.
The "HTML::HeadParser" keeps a reference to a header object, and theparser will update this header object as the various elements of the<HEAD> section of the HTML document are recognized. The followingheader fields are affected:
- The Content-Base header is initialized from the <basehref=``...''> element.
- The Title header is initialized from the <title>...</title>element.
- The Isindex header will be added if there is a <isindex>element in the <head>. The header value is initialized from theprompt attribute if it is present. If no prompt attribute isgiven it will have '?' as the value.
- All <meta> elements containing a "name" attribute will result inheaders using the prefix "X-Meta-" appended with the value of the"name" attribute as the name of the header, and the value of the"content" attribute as the pushed header value.
<meta> elements containing a "http-equiv" attribute will resultin headers as in above, but without the "X-Meta-" prefix in theheader name.
<meta> elements containing a "charset" attribute will result inan "X-Meta-Charset" header, using the value of the "charset"attribute as the pushed header value.
The ':' character can't be represented in header field names, soif the meta element contains this char it's substituted with '-'before forming the field name.
The following methods (in addition to those provided by thesuperclass) are available:
- $hp = HTML::HeadParser->new
- $hp = HTML::HeadParser->new( $header )
- The object constructor. The optional $header argument should be areference to an object that implement the header() and push_header()methods as defined by the "HTTP::Headers" class. Normally it will beof some class that is a or delegates to the "HTTP::Headers" class.
If no $header is given "HTML::HeadParser" will create an"HTTP::Headers" object by itself (initially empty).
- Returns a reference to the header object.
- $hp->header( $key )
- Returns a header value. It is just a shorter way to write"$hp->header->header($key)".
$h = HTTP::Headers->new; $p = HTML::HeadParser->new($h); $p->parse(<<EOT); <title>Stupid example</title> <base href="http://www.linpro.no/lwp/"> Normal text starts here. EOT undef $p; print $h->title; # should print "Stupid example"
The "HTTP::Headers" class is distributed as part of thelibwww-perl package. If you don't have that distribution installedyou need to provide the $header argument to the "HTML::HeadParser"constructor with your own object that implements the documentedprotocol.
Copyright 1996-2001 Gisle Aas. All rights reserved.
This library is free software; you can redistribute it and/ormodify it under the same terms as Perl itself.
- SEE ALSO
This document was created byman2html,using the manual pages.