#============================================================= -*-perl-*-
#
# Template::FAQ
#
# DESCRIPTION
#
# AUTHOR
# Andy Wardley <abw@wardley.org>
#
# COPYRIGHT
# Copyright (C) 1996-2007 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.
#========================================================================
=head1 NAME
Template::FAQ - Frequently Asked Questions about the Template Toolkit
=head1 Template Toolkit Language
=head2 Why doesn't [% a = b IF c %] work as expected?
There's a limitation in the TT2 parser which means that the following code
doesn't work as you might expect:
[% a = b IF c %]
The parser interprets it as an attempt to set C<a> to the result of
C<b IF c>, like this:
[% a = (b IF c) %]
If you want to set C<a = b> only if C<c> is true, then do this instead:
[% SET a = b IF c %]
The explicit C<SET> keyword gives the parser the clue it needs to do the
right thing.
NOTE: this will be fixed in TT3
=head2 If I'm using TT to write out a TT template, is there a good way to escape [% and %]?
You can do something like this:
[% stag = "[\%"
etag = "%\]"
%]
and then:
[% stag; 'hello'; etag %]
Or you can use the C<TAGS> directive, like so:
[% TAGS [- -] %]
[- INCLUDE foo -] # is a directive
[% INCLUDE foo %] # not a directive
=head2 How do I iterate over a hash?
This is covered in the L<Template::Manual::VMethods> section of the
manual. A list of all the keys that are in the hash can be obtained with the
C<keys> virtual method. You can then iterate over that list and by looking up
each key in turn get the value.
[% FOREACH key = product.keys %]
[% key %] => [% product.$key %]
[% END %]
=head1 Plugins
=head2 How do I get the Table plugin to order data across rather than down?
Order the data into rows:
Steve Karen Jeff
Brooklyn Nantucket Fairfax
NY MA VA
[% USE table(data, rows=3) %]
Then ask for each column
[% FOREACH column = table.cols %]
And then print each item in the column going across the output rows
[% FOREACH item = column %]
<td>[% item %]</td>
[% END %]
=head2 Accessing Cookies
Jeff Boes E<lt>jboes@nexcerpt.comE<gt> asks:
Does anyone have a quick-n-dirty approach to accessing
cookies from templates?
Jonas Liljegren answers:
[% USE CGI %]
<p>The value is [% CGI.cookie('cookie_name') | html %]
You will need to have L<Template::Plugin::CGI> installed.
=head1 Extending the Template Toolkit
=head2 Can I serve templates from a database?
Short answer: yes, Chris Nandor has done this for Slash. You need to
subclass L<Template::Provider>. See the mailing list archives for further
info.
=head2 Can I fetch templates via http?
To do the job properly, you should subclass L<Template::Provider> to
C<Template::Provider::HTTP> and use a C<PREFIX_MAP> option to bind the C<http>
template prefix to that particular provider (you may want to go digging around
in the F<Changes> file around version 2.01 for more info on C<PREFIX_MAP> - it
may not be properly documented anywhere else...yet!). e.g.
use Template::Provider::HTTP;
my $file = Template::Provider( INCLUDE_PATH => [...] );
my $http = Template::Provider::HTTP->new(...);
my $tt2 = Template->new({
LOAD_TEMPLATES => [ $file, $http ],
PREFIX_MAP => {
file => '0', # file:foo.html
http => '1', # http:foo.html
default => '0', # foo.html => file:foo.html
}
});
Now a template specified as:
[% INCLUDE foo %]
will be served by the 'file' provider (the default). Otherwise you
can explicitly add a prefix:
[% INCLUDE file:foo.html %]
[% INCLUDE http:foo.html %]
[% INCLUDE http://www.xyz.com/tt2/header.tt2 %]
This same principal can be used to create a DBI template provider. e.g.
[% INCLUDE dbi:foo.html %]
Alas, we don't yet have a DBI provider as part of the Template Toolkit. There
has been some talk on the mailing list about efforts to develop DBI and/or
HTTP providers but as yet no-one has stepped forward to take up the
challenge...
In the mean time, Craig Barrat's post from the mailing list has some useful
pointers on how to achieve this using existing modules. See
L<http://tt2.org/pipermail/templates/2001-May/000954.html>
=head1 Miscellaneous
=head2 How can I find out the name of the main template being processed?
The C<template> variable contains a reference to the
Template::Document object for the main template you're processing
(i.e. the one provided as the first argument to the Template process()
method). The C<name> method returns its name.
[% template.name %] # e.g. index.html
=head2 How can I find out the name of the current template being processed?
The C<template> variable always references the I<main> template being processed.
So even if you call [% INCLUDE header %], and that calls [% INCLUDE menu %],
the C<template> variable will be unchanged.
index.html:
[% template.name %] # index.html
[% INCLUDE header %]
header:
[% template.name %] # index.html
[% INCLUDE menu %]
menu:
[% template.name %] # index.html
In contrast, the C<component> variable always references the I<current>
template being processed.
index.html
[% component.name %] # index.html
[% INCLUDE header %]
header:
[% component.name %] # header
[% INCLUDE menu %]
menu:
[% component.name %] # menu
=head2 How do I print the modification time of the template or component?
The C<template> and C<component> variables reference the main template
and the current template being processed (see previous questions).
The C<modtime> method returns the modification time of the
corresponding template file as a number of seconds since the Unix
epoch (00:00:00 GMT 1st January 1970).
This number doesn't mean much to anyone (except perhaps serious Unix
geeks) so you'll probably want to use the Date plugin to format it for
human consumption.
[% USE Date %]
[% template.name %] last modified [% Date.format(template.modtime) %]
=head2 How can I configure variables on a per-request basis?
One easy way to achieve this is to define a single C<PRE_PROCESS> template
which loads in other configuration files based on variables defined or other
conditions.
For example, my setup usually looks something like this:
PRE_PROCESS => 'config/main'
config/main:
[% DEFAULT style = 'text'
section = template.section or 'home';
PROCESS config/site
+ config/urls
+ config/macros
+ "config/style/$style"
+ "config/section/$section"
+ ...
%]
This allows me to set a single 'style' variable to control which config
file gets pre-processed to set my various style options (colours, img paths,
etc). For example:
config/style/basic:
[% style = {
name = style # save existing 'style' var as 'style.name'
# define various other style variables....
col = {
back => '#ffffff'
text => '#000000'
# ...etc...
}
logo = {
# ...etc...
}
# ...etc...
}
%]
Each source template can declare which section it's in via a META
directive:
[% META
title = 'General Information'
section = 'info'
%]
...
This controls which section configuration file gets loaded to set various
other variables for defining the section title, menu, etc.
config/section/info:
[% section = {
name = section # save 'section' var as 'section.name'
title = 'Information'
menu = [ ... ]
# ...etc...
}
%]
This illustrates the basic principal but you can extend it to perform
pretty much any kind of per-document initialisation that you require.
=head2 Why do I get rubbish for my utf-8 templates?
First of all, make sure that your template files define a Byte Order
Mark L<http://en.wikipedia.org/wiki/Byte_Order_Mark>
If you for some reason don't want to add BOM to your templates, you can
force Template to use a particular encoding (e.g. C<utf8>) for your
templates with the C<ENCODING> option.
my $template = Template->new({
ENCODING => 'utf8'
});
=head1 Questions About This FAQ
=head2 Why is this FAQ so short?
Because we don't have anyone maintaining it.
=head2 Can I help?
Yes please :-)
=cut
# Local Variables:
# mode: perl
# perl-indent-level: 4
# indent-tabs-mode: nil
# End:
#
# vim: expandtab shiftwidth=4: