#============================================================= -*-perl-*-
# Template
#   Module implementing a simple, user-oriented front-end to the Template
#   Toolkit.
#   Andy Wardley   <abw@wardley.org>
#   Copyright (C) 1996-2020 Andy Wardley.  All Rights Reserved.
#   This module is free software; you can redistribute it and/or
#   modify it under the same terms as Perl itself.

package Template;

use strict;
use warnings;
use 5.006;
use base 'Template::Base';

use Template::Config;
use Template::Constants;
use Template::Provider;
use Template::Service;
use File::Basename;
use File::Path;
use Scalar::Util qw(blessed);

our $VERSION = '3.009';
our $ERROR   = '';
our $DEBUG   = 0;
our $BINMODE = 0 unless defined $BINMODE;

# preload all modules if we're running under mod_perl
Template::Config->preload() if $ENV{ MOD_PERL };

# process($input, \%replace, $output)
# Main entry point for the Template Toolkit.  The Template module
# delegates most of the processing effort to the underlying SERVICE
# object, an instance of the Template::Service class.

sub process {
    my ($self, $template, $vars, $outstream, @opts) = @_;
    my ($output, $error);
    my $options = (@opts == 1) && ref($opts[0]) eq 'HASH'
        ? shift(@opts) : { @opts };

    $options->{ binmode } = $BINMODE
        unless defined $options->{ binmode };

    # we're using this for testing in t/output.t and t/filter.t so
    # don't remove it if you don't want tests to fail...
    $self->DEBUG("set binmode\n") if $DEBUG && $options->{ binmode };

    $output = $self->{ SERVICE }->process($template, $vars);

    if (defined $output) {
        $outstream ||= $self->{ OUTPUT };
        unless (ref $outstream) {
            my $outpath = $self->{ OUTPUT_PATH };
            $outstream = "$outpath/$outstream" if $outpath;

        # send processed template to output stream, checking for error
        return ($self->error($error))
            if ($error = &_output($outstream, \$output, $options));

        return 1;
    else {
        return $self->error($self->{ SERVICE }->error);

# service()
# Returns a reference to the internal SERVICE object which handles
# all requests for this Template object

sub service {
    my $self = shift;
    return $self->{ SERVICE };

# context()
# Returns a reference to the CONTEXT object within the SERVICE
# object.

sub context {
    my $self = shift;
    return $self->{ SERVICE }->{ CONTEXT };

sub template {

#                     -- PRIVATE METHODS --

# _init(\%config)
sub _init {
    my ($self, $config) = @_;

    # convert any textual DEBUG args to numerical form
    my $debug = $config->{ DEBUG };
    $config->{ DEBUG } = Template::Constants::debug_flags($self, $debug)
        || return if defined $debug && $debug !~ /^\d+$/;

    # prepare a namespace handler for any CONSTANTS definition
    if (my $constants = $config->{ CONSTANTS }) {
        my $ns  = $config->{ NAMESPACE } ||= { };
        my $cns = $config->{ CONSTANTS_NAMESPACE } || 'constants';
        $constants = Template::Config->constants($constants)
            || return $self->error(Template::Config->error);
        $ns->{ $cns } = $constants;

    $self->{ SERVICE } = $config->{ SERVICE }
        || Template::Config->service($config)
        || return $self->error(Template::Config->error);

    $self->{ OUTPUT      } = $config->{ OUTPUT } || \*STDOUT;
    $self->{ OUTPUT_PATH } = $config->{ OUTPUT_PATH };

    return $self;

# _output($where, $text)

sub _output {
    my ($where, $textref, $options) = @_;
    my $reftype;
    my $error = 0;

    # call a CODE reference
    if (($reftype = ref($where)) eq 'CODE') {
    # print to a glob (such as \*STDOUT)
    elsif ($reftype eq 'GLOB') {
        print $where $$textref;
    # append output to a SCALAR ref
    elsif ($reftype eq 'SCALAR') {
        $$where .= $$textref;
    # push onto ARRAY ref
    elsif ($reftype eq 'ARRAY') {
        push @$where, $$textref;
    # call the print() method on an object that implements the method
    # (e.g. IO::Handle, Apache::Request, etc)
    elsif (blessed($where) && $where->can('print')) {
    # a simple string is taken as a filename
    elsif (! $reftype) {
        local *FP;
        # make destination directory if it doesn't exist
        my $dir = dirname($where);
        eval { mkpath($dir) unless -d $dir; };
        if ($@) {
            # strip file name and line number from error raised by die()
            ($error = $@) =~ s/ at \S+ line \d+\n?$//;
        elsif (open(FP, '>', $where)) {
            # binmode option can be 1 or a specific layer, e.g. :utf8
            my $bm = $options->{ binmode  };
            if ($bm && $bm eq 1) {
                binmode FP;
            elsif ($bm){
                binmode FP, $bm;
            print FP $$textref;
            close FP;
        else {
            $error  = "$where: $!";
    # give up, we've done our best
    else {
        $error = "output_handler() cannot determine target type ($where)\n";

    return $error;



=head1 NAME

Template - Front-end module to the Template Toolkit


    use Template;

    # some useful options (see below for full list)
    my $config = {
        INCLUDE_PATH => '/search/path',  # or list ref
        INTERPOLATE  => 1,               # expand "$var" in plain text
        POST_CHOMP   => 1,               # cleanup whitespace
        PRE_PROCESS  => 'header',        # prefix each template
        EVAL_PERL    => 1,               # evaluate Perl code blocks

    # create Template object
    my $template = Template->new($config);

    # define template variables for replacement
    my $vars = {
        var1  => $value,
        var2  => \%hash,
        var3  => \@list,
        var4  => \&code,
        var5  => $object,

    # specify input filename, or file handle, text reference, etc.
    my $input = 'myfile.html';

    # process input template, substituting variables
    $template->process($input, $vars)
        || die $template->error();


This documentation describes the Template module which is the direct
Perl interface into the Template Toolkit.  It covers the use of the
module and gives a brief summary of configuration options and template
directives.  Please see L<Template::Manual> for the complete reference
manual which goes into much greater depth about the features and use
of the Template Toolkit.  The L<Template::Tutorial> is also available
as an introductory guide to using the Template Toolkit.

=head1 METHODS

=head2 new(\%config)

The C<new()> constructor method (implemented by the
L<Template::Base|Template::Base#new()> base class) instantiates a new
C<Template> object. A reference to a hash array of configuration items may be
passed as a parameter.

    my $tt = Template->new({
        INCLUDE_PATH => '/usr/local/templates',
        EVAL_PERL    => 1,
    }) || die $Template::ERROR, "\n";

A reference to a new C<Template> object is returned, or undef on error. In the
latter case, the error message can be retrieved by calling L<error()> as a
class method or by examining the C<$Template::ERROR> package variable

    my $tt = Template->new(\%config)
        || die Template->error(), "\n";

    my $tt = Template->new(\%config)
        || die $Template::ERROR, "\n";

For convenience, configuration items may also be specified as a list
of items instead of a hash array reference.  These are automatically
folded into a hash array by the constructor.

    my $tt = Template->new(INCLUDE_PATH => '/tmp', POST_CHOMP => 1)
        || die $Template::ERROR, "\n";

=head2 process($template, \%vars, $output, %options)

The C<process()> method is called to process a template. The first parameter
indicates the input template as one of: a filename relative to
C<INCLUDE_PATH>, if defined; a reference to a text string containing the
template text; or a file handle reference (e.g. C<IO::Handle> or sub-class) or
C<GLOB> (e.g. C<\*STDIN>), from which the template can be read. A reference to
a hash array may be passed as the second parameter, containing definitions of
template variables.

    # filename
        || die $tt->error(), "\n";

    # text reference
    $text = "[% INCLUDE header %]\nHello world!\n[% INCLUDE footer %]";
        || die $tt->error(), "\n";

    # file handle (GLOB)
        || die $tt->error(), "\n";

    [% INCLUDE header %]
    This is a template defined in the __END__ section which is
    accessible via the DATA "file handle".
    [% INCLUDE footer %]

By default, the processed template output is printed to C<STDOUT>. The
C<process()> method then returns C<1> to indicate success. A third parameter
may be passed to the C<process()> method to specify a different output location.
This value may be one of: a plain string indicating a filename which will be
opened (relative to C<OUTPUT_PATH>, if defined) and the output written to; a file
GLOB opened ready for output; a reference to a scalar (e.g. a text string) to
which output/error is appended; a reference to a subroutine which is called,
passing the output as a parameter; or any object reference which implements a
C<print()> method (e.g. C<IO::Handle>, C<Apache::Request>, etc.) which will be called,
passing the generated output as a parameter.


    # output filename
    $tt->process('welcome.tt2', $vars, 'welcome.html')
        || die $tt->error(), "\n";

    # reference to output subroutine
    sub myout {
        my $output = shift;
    $tt->process('welcome.tt2', $vars, \&myout)
        || die $tt->error(), "\n";

    # reference to output text string
    my $output = '';
    $tt->process('welcome.tt2', $vars, \$output)
        || die $tt->error(), "\n";

    print "output: $output\n";

In an Apache/mod_perl handler:

    sub handler {
        my $req = shift;

        # ...your code here...

        # direct output to Apache::Request via $req->print($output)
        $tt->process($file, $vars, $req) || do {
            return SERVER_ERROR;
        return OK;

After the optional third output argument can come an optional
reference to a hash or a list of C<(name, value)> pairs providing further
options for the output.  The only option currently supported is
C<binmode> which, when set to any true value will ensure that files
created (but not any existing file handles passed) will be set to
binary mode.

    # either: hash reference of options
    $tt->process($infile, $vars, $outfile, { binmode => 1 })
        || die $tt->error(), "\n";

    # or: list of name, value pairs
    $tt->process($infile, $vars, $outfile, binmode => 1)
        || die $tt->error(), "\n";

Alternately, the C<binmode> argument can specify a particular IO layer such
as C<:utf8>.

    $tt->process($infile, $vars, $outfile, binmode => ':utf8')
        || die $tt->error(), "\n";

The C<OUTPUT> configuration item can be used to specify a default output
location other than C<\*STDOUT>.  The C<OUTPUT_PATH> specifies a directory
which should be prefixed to all output locations specified as filenames.

    my $tt = Template->new({
        OUTPUT      => sub { ... },       # default
        OUTPUT_PATH => '/tmp',
    }) || die Template->error(), "\n";

    # use default OUTPUT (sub is called)
    $tt->process('welcome.tt2', $vars)
        || die $tt->error(), "\n";

    # write file to '/tmp/welcome.html'
    $tt->process('welcome.tt2', $vars, 'welcome.html')
        || die $tt->error(), "\n";

The C<process()> method returns C<1> on success or C<undef> on error. The
error message generated in the latter case can be retrieved by calling the
L<error()> method. See also L<CONFIGURATION SUMMARY> which describes how error
handling may be further customised.

=head2 error()

When called as a class method, it returns the value of the C<$ERROR> package
variable.  Thus, the following are equivalent.

    my $tt = Template->new()
        || die Template->error(), "\n";

    my $tt = Template->new()
        || die $Template::ERROR, "\n";

When called as an object method, it returns the value of the internal
C<_ERROR> variable, as set by an error condition in a previous call to

        || die $tt->error(), "\n";

Errors are represented in the Template Toolkit by objects of the
L<Template::Exception> class. If the L<process()> method returns a false value
then the C<error()> method can be called to return an object of this class.
The L<type()|Template::Exception#type()> and
L<info()|Template::Exception#info()> methods can called on the object to
retrieve the error type and information string, respectively. The
method can be called to return a string of the form C<$type - $info>. This
method is also overloaded onto the stringification operator allowing the
object reference itself to be printed to return the formatted error string.

    $tt->process('somefile') || do {
        my $error = $tt->error();
        print "error type: ", $error->type(), "\n";
        print "error info: ", $error->info(), "\n";
        print $error, "\n";

=head2 service()

The C<Template> module delegates most of the effort of processing templates
to an underlying L<Template::Service> object.  This method returns a reference
to that object.

=head2 context()

The L<Template::Service> module uses a core L<Template::Context> object for
runtime processing of templates.  This method returns a reference to
that object and is equivalent to C<< $template-E<gt>service-E<gt>context() >>.

=head2 template($name)

This method is a simple wrapper around the L<Template::Context> method of the
same name.  It returns a compiled template for the source provided as an


The following list gives a short summary of each Template Toolkit
configuration option.  See L<Template::Manual::Config> for full details.

=head2 Template Style and Parsing Options


Specifies the character encoding.


Define tokens that indicate start and end of directives
(default: 'C<[%>' and 'C<%]>').

=head3 TAG_STYLE

Set C<START_TAG> and C<END_TAG> according to a pre-defined style (default:
'C<template>', as above).


Removes whitespace before/after directives (default: 0/0).

=head3 TRIM

Remove leading and trailing whitespace from template output (default: 0).


Interpolate variables embedded like C<$this> or C<${this}> (default: 0).

=head3 ANYCASE

Allow directive keywords in lower case (default: 0 - UPPER only).

=head2 Template Files and Blocks


One or more directories to search for templates.


Delimiter for separating paths in C<INCLUDE_PATH> (default: 'C<:>').


Allow absolute file names, e.g. C</foo/bar.html> (default: 0).


Allow relative filenames, e.g. C<../foo/bar.html> (default: 0).

=head3 DEFAULT

Default template to use when another not found.

=head3 BLOCKS

Hash array pre-defining template blocks.


Enabled by default causing C<BLOCK> definitions to be reset each time a
template is processed.  Disable to allow C<BLOCK> definitions to persist.


Flag to permit recursion into templates (default: 0).

=head2 Template Variables


Hash array of variables and values to pre-define in the stash.

=head2 Runtime Processing Options

=head3 EVAL_PERL

Flag to indicate if C<PERL>/C<RAWPERL> blocks should be processed (default: 0).


Name of template(s) to process before/after main template.

=head3 PROCESS

Name of template(s) to process instead of main template.

=head3 ERROR

Name of error template or reference to hash array mapping error types to

=head3 OUTPUT

Default output location or handler.


Directory into which output files can be written.

=head3 DEBUG

Enable debugging messages.

=head2 Caching and Compiling Options


Maximum number of compiled templates to cache in memory (default:
undef - cache all)


Filename extension for compiled template files (default: undef - don't


Root of directory in which compiled template files should be written
(default: undef - don't compile).

=head2 Plugins and Filters

=head3 PLUGINS

Reference to a hash array mapping plugin names to Perl packages.


One or more base classes under which plugins may be found.

=head3 LOAD_PERL

Flag to indicate regular Perl modules should be loaded if a named plugin
can't be found  (default: 0).

=head3 FILTERS

Hash array mapping filter names to filter subroutines or factories.

=head2 Customisation and Extension


List of template providers.


List of plugin providers.


List of filter providers.


Set providers to tolerate errors as declinations (default: 0).

=head3 SERVICE

Reference to a custom service object (default: L<Template::Service>).

=head3 CONTEXT

Reference to a custom context object (default: L<Template::Context>).

=head3 STASH

Reference to a custom stash object (default: L<Template::Stash>).

=head3 PARSER

Reference to a custom parser object (default: L<Template::Parser>).

=head3 GRAMMAR

Reference to a custom grammar object (default: L<Template::Grammar>).


The following list gives a short summary of each Template Toolkit directive.
See L<Template::Manual::Directives> for full details.

=head2 GET

Evaluate and print a variable or value.

    [%   GET variable %]    # 'GET' keyword is optional
    [%       variable %]
    [%       hash.key %]
    [%         list.n %]
    [%     code(args) %]
    [% obj.meth(args) %]
    [%  "value: $var" %]

=head2 CALL

As per L<GET> but without printing result (e.g. call code)

    [%  CALL variable %]

=head2 SET

Assign a values to variables.

    [% SET variable = value %]    # 'SET' also optional
    [%     variable = other_variable
           variable = 'literal text @ $100'
           variable = "interpolated text: $var"
           list     = [ val, val, val, val, ... ]
           list     = [ val..val ]
           hash     = { var => val, var => val, ... }

=head2 DEFAULT

Like L<SET>, but variables are only set if currently unset (i.e. have no
true value).

    [% DEFAULT variable = value %]

=head2 INSERT

Insert a file without any processing performed on the contents.

    [% INSERT legalese.txt %]

=head2 PROCESS

Process another template file or block and insert the generated output.
Any template L<BLOCK>s or variables defined or updated in the C<PROCESS>ed
template will thereafter be defined in the calling template.

    [% PROCESS template %]
    [% PROCESS template  var = val, ... %]

=head2 INCLUDE

Similar to C<PROCESS>, but using a local copy of the current variables.
Any template C<BLOCK>s or variables defined in the C<INCLUDE>d template
remain local to it.

    [% INCLUDE template %]
    [% INCLUDE template  var = val, ... %]

=head2 WRAPPER

The content between the C<WRAPPER> and corresponding C<END> directives is first
evaluated, with the output generated being stored in the C<content> variable.
The named template is then process as per C<INCLUDE>.

    [% WRAPPER layout %]
       Some template markup [% blah %]...
    [% END %]

A simple F<layout> template might look something like this:

    Your header here...
    [% content %]
    Your footer here...

=head2 BLOCK

Define a named template block for L<INCLUDE>, L<PROCESS> and L<WRAPPER>
to use.

    [% BLOCK hello %]
       Hello World
    [% END %]

    [% INCLUDE hello %]

=head2 FOREACH

Repeat the enclosed C<FOREACH> ... C<END> block for each value in the list.

    [% FOREACH variable IN [ val, val, val ] %]    # either
    [% FOREACH variable IN list %]                 # or
       The variable is set to [% variable %]
    [% END %]

=head2 WHILE

The block enclosed between C<WHILE> and C<END> block is processed while
the specified condition is true.

    [% WHILE condition %]
    [% END %]


The enclosed block is processed if the condition is true / false.

    [% IF condition %]
    [% ELSIF condition %]
    [% ELSE %]
    [% END %]

    [% UNLESS condition %]
    [% # ELSIF/ELSE as per IF, above %]
    [% END %]

=head2 SWITCH / CASE

Multi-way switch/case statement.

    [% SWITCH variable %]
    [%   CASE val1 %]
    [%   CASE [ val2, val3 ] %]
    [%   CASE %]         # or [% CASE DEFAULT %]
    [% END %]

=head2 MACRO

Define a named macro.

    [% MACRO name <directive> %]
    [% MACRO name(arg1, arg2) <directive> %]
    [% name %]
    [% name(val1, val2) %]

=head2 FILTER

Process enclosed C<FILTER> ... C<END> block then pipe through a filter.

    [% FILTER name %]                       # either
    [% FILTER name( params ) %]             # or
    [% FILTER alias = name( params ) %]     # or
    [% END %]

=head2 USE

Load a plugin module (see C<Template::<Manual::Plugins>), or any regular Perl
module when the C<LOAD_PERL> option is set.

    [% USE name %]                      # either
    [% USE name( params ) %]            # or
    [% USE var = name( params ) %]      # or
    [% name.method %]
    [% var.method %]


Evaluate enclosed blocks as Perl code (requires the C<EVAL_PERL> option to be

    [% PERL %]
     # perl code goes here
     $stash->set('foo', 10);
     print "set 'foo' to ", $stash->get('foo'), "\n";
     print $context->include('footer', { var => $val });
    [% END %]

    [% RAWPERL %]
       # raw perl code goes here, no magic but fast.
       $output .= 'some output';
    [% END %]


Exception handling.

    [% TRY %]
       [% THROW type info %]
    [% CATCH type %]
     catch content
       [% error.type %] [% error.info %]
    [% CATCH %] # or [% CATCH DEFAULT %]
    [% FINAL %]
       this block is always processed
    [% END %]

=head2 NEXT

Jump straight to the next item in a C<FOREACH> or C<WHILE> loop.

    [% NEXT %]

=head2 LAST

Break out of C<FOREACH> or C<WHILE> loop.

    [% LAST %]

=head2 RETURN

Stop processing current template and return to including templates.

    [% RETURN %]

=head2 STOP

Stop processing all templates and return to caller.

    [% STOP %]

=head2 TAGS

Define new tag style or characters (default: C<[%> C<%]>).

    [% TAGS html %]
    [% TAGS <!-- --> %]


Ignored and deleted.

    [% # this is a comment to the end of line
       foo = 'bar'

    [%# placing the '#' immediately inside the directive
        tag comments out the entire directive


The source code for the Template Toolkit is held in a public git repository
on Github: L<https://github.com/abw/Template2>

=head1 AUTHOR

Andy Wardley E<lt>abw@wardley.orgE<gt> L<http://wardley.org/>

=head1 VERSION

Template Toolkit version 3.009, released on July 13 2020.


Copyright (C) 1996-2020 Andy Wardley.  All Rights Reserved.

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


# Local Variables:
# mode: perl
# perl-indent-level: 4
# indent-tabs-mode: nil
# End:
# vim: expandtab shiftwidth=4: