NAME
Text::MiniTmpl - Compile and render templates
VERSION
This document describes Text::MiniTmpl version v2.0.1
SYNOPSIS
use Text::MiniTmpl qw( render );
$html1 = render('template.html', %params1);
$html2 = render('template.html', %params2);
DESCRIPTION
Compile templates with embedded perl code into anonymous subroutines. These subroutines can be (optionally) cached, and executed to render these templates with (optional) parameters.
Perl code in templates will be executed with:
package PACKAGE_WHERE_render_OR_tmpl2code_WAS_CALLED;
use warnings;
use strict;
Recursion in templates is supported (you can call render() or tmpl2code() inside template to "include" another template inside current one). Path name to included template can be set in several ways:
path relative to current template's directory:
'file', 'dir/file', '../file'
path relative to current working directory (where your script executed):
'./file', './dir/file'
absolute path:
'/dir/file'
When you render top-level template (i.e. call render() from your script, not inside some template) paths 'file'
and './file'
, 'dir/file'
and './dir/file'
are same.
Correctly report compile errors in templates, with template name and line number.
Unicode support
Files with templates should be in UTF8. Parameters for templates should be perl Unicode scalars. Rendered template (returned by render() or by function returned by tmpl2code()) will be in UTF8.
You can disable it using raw(1) (see below) to get more speed.
Source Filters support
Probably not all filters will work inside templates - keep in mind filter will see auto-generated (by tmpl2code()) perl function's code instead of plain template text. See `perldoc perlfilter` for more details.
Example:
&~ use Filter::CommaEquals; ~&
&~ @{ $_{users} } ,= 'GHOST' ~&
&~ for (@{ $_{users} }) { ~&
<p>Hello, @~ $_ ~@!</p>
&~ } ~&
Template syntax
Any template become perl function after parsing. This function will accept it parameters in %_
(it start with local %_ = @_;
). Of course, you can use my() and local() variables in template (their scope will be full template, not only placeholder's block where they was defined).
- &~ PERL CODE ~&
- <!--& PERL CODE -->
-
Execute PERL CODE but don't output anything.
- @~ PERL CODE ~@
-
Execute PERL CODE and output it result (last calculated expression) escaped using HTML::Entities::encode_entities().
- ^~ PERL CODE ~^
-
Execute PERL CODE and output it result (last calculated expression) escaped using URI::Escape::uri_escape_utf8().
- #~ PERL CODE ~#
-
Execute PERL CODE and output it result (last calculated expression) AS IS, without any escaping.
- any other text ...
-
... will be output AS IS
Example templates:
To: #~ $_{email} ~#
Hello, #~ $username ~#. Here is items your asked for:
&~ for (@{ $_{items} }) { ~&
#~ $_ ~#
&~ } ~&
---cut header.html---
<html>
<head><title>@~ $_{pagetitle} ~@</title></head>
<body>
---cut index.html---
#~ render('header.html', pagetitle=>'Home') ~#
<p>Hello, @~ $_{username} ~@.</p>
&~ # In HTML you may prefer <!--& instead of &~ for code blocks ~&
<!--& for (@{ $_{topics} }) { -->
<a href="news.cgi?topic=^~ $_ ~^&user=^~ $_{user} ~^">
News about @~ $_ ~@
</a>
<!--& } -->
#~ render('footer.html') ~#
---cut footer.html---
</body>
</html>
EXPORTS
Nothing by default, but all documented functions can be explicitly imported.
INTERFACE
render
$html = render( $filename, %params );
Render template from $filename with %params.
This is caching wrapper around tmpl2code(), which avoid calling tmpl2code() second time for same $filename.
Example:
$html = render( 'template/index.html',
title => $title,
name => $name,
);
Return STRING with rendered template.
tmpl2code
$code = tmpl2code( $filename );
Read template from $filename (may be absolute or relative to current template's directory or current working directory - see "DESCRIPTION"), compile it into ANON function.
This function can be executed with ( %params )
parameters, it will render $filename template with given %params
and return SCALARREF to rendered text.
Example:
$code = tmpl2code( 'template/index.html' );
$html = ${ $code->( title=>$title, name=>$name ) };
Return CODEREF to that function.
raw
raw( $is_raw );
If $is_raw TRUE disable Unicode support. To enable Unicode again call raw() with $is_raw FALSE.
Disabling Unicode support will speedup this module in about 1.5 times!
When Unicode support disabled your parameters used to render template will be used in template AS IS, without attempt to encode them to UTF8. This mean you shouldn't use perl Unicode scalars in these parameters anymore.
This affect only templates processed by tmpl2code() after calling raw() (beware caching effect of render()).
encode_js
$js = encode_js( $scalar );
Encode $scalar (string or number) for inserting into JavaScript code (usually inside HTML templates).
Example:
<script>
var int_from_perl = #~ encode_js($number) ~#;
var str_from_perl = '#~ encode_js($string) ~#';
</script>
Return encoded string.
encode_js_data
$js_data = encode_js_data( $complex );
Encode $complex data structure (HASHREF, ARRAYREF, etc. - any data type supported by JSON::XS) for inserting into JavaScript code (usually inside HTML templates).
Example:
<script>
var hash_from_perl = #~ encode_js_data( \%hash ) ~#;
var array_from_perl = #~ encode_js_data( \@array ) ~#;
</script>
Return encoded string.
SPEED AND SIZE
This module implemented under 100 lines (about 3KB) of pure Perl code. And while code is terse enough, I believe it still simple and clean.
While this is pure-perl module with many enough features, it's still very fast - you can do your own benchmarking using cool Template::Benchmark module, here is results from my system:
- instance_reuse
-
This test simulate standard FastCGI which read template from HDD and compile it only once (when this template requested first time), and for next requests it just render it using cached anon subroutine.
Rate XS? TT 29/s - Template::Toolkit (2.22) TT_X 95/s Y Template::Toolkit (2.22) with Stash::XS HM 704/s - HTML::Mason (1.45) TeCS 738/s Y Text::ClearSilver (0.10.5.4) TeMT 1131/s - Text::MicroTemplate (0.18) TeMMHM 1173/s - Text::MicroMason (2.12) using Text::MicroMason::HTMLMason * TMTU 1357/s - Text::MiniTmpl (1.1.0) with enabled Unicode MoTe 1629/s - Mojo::Template () * TMT 2054/s - Text::MiniTmpl (1.1.0) TeClev 5966/s Y Text::Clevery (0.0004) in XS mode TeXs 6761/s Y Text::Xslate (0.2015)
- uncached_disk
-
This test simulate standard CGI which read template from HDD, compile and render it on each run - no caches used at all (except HDD files caching by OS).
Rate XS? TeXs 1.4/s Y Text::Xslate (0.2015) TeClev 1.5/s Y Text::Clevery (0.0004) in XS mode HM 12.6/s - HTML::Mason (1.45) MoTe 21.0/s - Mojo::Template () TeMT 32.1/s - Text::MicroTemplate (0.18) TeMMHM 36.2/s - Text::MicroMason (2.12) using Text::MicroMason::HTMLMason * TMTU 54.1/s - Text::MiniTmpl (1.1.0) with enabled Unicode * TMT 67.9/s - Text::MiniTmpl (1.1.0) TeTmpl 448/s Y Text::Tmpl (0.33) TeCS 725/s Y Text::ClearSilver (0.10.5.4) HTP 1422/s Y HTML::Template::Pro (0.9504)
SUPPORT
Bugs / Feature Requests
Please report any bugs or feature requests through the issue tracker at https://github.com/powerman/perl-Text-MiniTmpl/issues. You will be notified automatically of any progress on your issue.
Source Code
This is open source software. The code repository is available for public review and contribution under the terms of the license. Feel free to fork the repository and submit pull requests.
https://github.com/powerman/perl-Text-MiniTmpl
git clone https://github.com/powerman/perl-Text-MiniTmpl.git
Resources
MetaCPAN Search
CPAN Ratings
AnnoCPAN: Annotated CPAN documentation
CPAN Testers Matrix
CPANTS: A CPAN Testing Service (Kwalitee)
AUTHOR
Alex Efros <powerman@cpan.org>
COPYRIGHT AND LICENSE
This software is Copyright (c) 2007- by Alex Efros <powerman@cpan.org>.
This is free software, licensed under:
The MIT (X11) License