Template::Perlish - Yet Another Templating system for Perl
This document describes Template::Perlish version 1.06. Most likely, this version number here is outdate, and you should peek the source.
use Template::Perlish; my $tp = Template::Perlish->new(); # A complex template, including some logic as Perl code my $tmpl = <<END_OF_TEMPLATE Dear [% name %], we are pleased to present you the following items: [% my $items = $variables{items}; # Available %variables my $counter = 0; for my $item (@$items) { %] [%= ++$counter %]. [% $item %] [% } %] Please consult our complete catalog at [% uris.2.catalog %]. Yours, [% director.name %] [% director.surname %]. END_OF_TEMPLATE my $processed = $tt->process($template, { name => 'Ciccio Riccio', items => [ qw( ciao a tutti quanti ) ], uris => [ 'http://whatever/', undef, { catalog => 'http://whateeeeever/', } ], director => { surname => 'Poletti' }, });
The above prints:
Dear Ciccio Riccio, we are pleased to present you the following items: 1. ciao 2. a 3. tutti 4. quanti Please consult our complete catalog at http://whateeeeever/. Yours, Poletti.
There is also a convenience function for one-shot templates:
use Template::Perlish qw( render ); my $rendered = render($template, \%variables);
You're probably looking at the tons and tons of templating systems available around - should you use this?
This system is quite basic and useful for simple situations. Say you have a bunch of templates in which you want to put some variables - then it's ok. On top of this, say that you want to add some simple logic - like a couple of IF's or iterating over an array - then it's ok again. For everything more complicated you should probably look elsewhere.
As a summary:
lightweight, a single-file module with minimal requirements that you can easily embed in your script;
simple approach to variable substitution, following Template::Toolkit to cope with scalars, hashes and arrays;
Perl code to handle all logic. This can be regarded as a PRO if you're a Perl programmer, because you already know the syntax; this is probably (definitively?) a CON in all other cases;
you have to explicitly code everything that goes beyond simple variable stuffing into a template.
If you think that this module does not fit your requirements, my personal suggestion for a templating system is Template::Toolkit: it's complete, easy to use and extensible, has excellent documentation (including a book and a quick reference guide) and support. Do you need anything more?
But don't trust me! Take a look at Choosing a Templating System at http://perl.apache.org/docs/tutorials/tmpl/comparison/comparison.html, where you can find a fairly complete comparison about the streamline templating systems in Perl, and decide by yourself!
You bet, this is another templating system for Perl. Yes, because it's the dream of every Perl programmer, me included. I needed something that's easily portable, with no dependencies apart a recent Perl version (but with some tweaking this should be solved), much in the spirit of the ::Tiny modules. And yes, my dream is to fill that ::Tiny gap some time in the future, but with another module.
Wherever possible I try to mimic Template::Toolkit, but I stop quite early. If you only have to fill a template with a bunch of variables, chances are that TT2 templates are good for Template::Perlish as well. If you need even the slightest bit of logic, you'll have to part from TT2 - and get full Perl power.
A template is simply a text (even if not necessarily) with some particular markup to embed commands. In particular, all the stuff included between [% and %] is considered as some sort of command, and treated specially. All the rest is treated as simple text. Of course, you can modify the start and stop delimiter for a command.
[%
%]
Commands can be of four different types:
that are expanded with the particular value for a given variable, where variables are passed as a hash reference. A variable can be defined as a sequence of alphanumeric (actually \w) tokens, separated by dots. The variables hash is visited considering each token as a subkey, in order to let you visit complex data structures. You can also put arrays in, but remember to use numbers ;)
variable
\w
that is expanded with the value of the given scalar variable;
this MUST have a = equal sign immediately after the opener, and contain a valid Perl expression. This expression is evaluated in scalar context and the result is printed;
=
good old Perl code, in order to provide you with control structures, modules, etc etc. This the most lazy approach I could think about, and it's also why this module is called Perlish.
Perlish
Take a look at the example in the "SYNOPSIS", it actually contains all that this module provides.
To start, you'll need a Template::Perlish object and, of course, a template. Templates are provided as text strings; if you have them into files, you are in charge of loading them first.
Template::Perlish
# get a Template::Perlish object my $tp = Template::Perlish->new(); # get the template (yes, it's your duty) my $tmpl = do { open my $fh, '<', 'filename'; local $/; <$fh> };
The basic operation mode is via the "process" method, which works much like in TT2. Anyway, this method will always give you back the generated stuff, and won't print anything. This can probably be less memory efficient when big templates are involved, but in this case you should probably head somewhere else.
# print out the template filled with some variables print $tp->process($tmpl, { key => 'value' });
Each template is transformed into Pure Perl code, then the code is evaluated in order to get the output. Thus, if you want to operate on the same template many times, a typical usage is:
# compile the template with something like: my $compiled = $tp->compile($template); # use the compiled template multiple times with different data for my $dataset (@available_data) { print {*STDOUT} "DATASET\n", $tp->evaluate($compiled, $dataset), "\n\n"; }
Anyway, there's a drawback in this approach: each time "evaluate" is called, the code in the string $compiled is exec'ed as a string, thus involving Perl parsing etc. In this case, the "compile_as_sub" method can come handy:
$compiled
exec
my $sub = $tp->compile_as_sub($template) or die "template did not compile: $EVAL_ERROR"; for my $dataset (@available_data) { print {*STDOUT} "DATASET\n", $sub->($dataset), "\n\n"; }
This has the advantage that $sub is compiled only once, and is a Perl subroutine that can be invoked directly. As an added bonus, you get immediate feedback about the generated code, because you'll get back undef if something goes wrong during parsing. On the other hand, this means that the Perl code is always executed (to generate the sub), thus you have to endure this penalty even if the compiled form isn't actually needed.
$sub
undef
The following convenience function can be used to quickly render a template:
use Template::Perlish qw( render ); my $rendered = render($template); # OR my $rendered = render($template, %variables); # OR my $rendered = render($template, \%variables);
if you already have a template and the variables to fill it in, this is probably the quickest thing to do.
You can pass the template alone, or you can pass the variables as well, either as a flat list (that will be converted back to a hash) or as a single hash reference.
Returns the rendered template, i.e. the same output as "process".
$tp = Template::Perlish->new(%opts); # OR $tp = Template::Perlish->new(\%opts);
constructor, does exactly what you think. You can provide any parameter, but only the following will make sense:
delimiter for the start of a command (as opposed to plain text/data);
delimiter for the end of a command;
variables that will be passed to all invocations of "process" and/or "evaluate".
Parameters can be given directly or via a hash reference.
By default, the delimiters are the same as TT2, i.e. [% and %], and the variables hash is empty.
The return value is a reference to an anonymous hash, whose three elements are the ones described above. You can modify them at will.
$code_text = $tp->compile($template);
compile a template generating the relevant Perl code. Using this method is useful when the same template has to be used multiple times, so the compilation can be done one time only.
Please note that the generated Perl code will be parsed each time you use it, of course.
Returns a text containing Perl code.
$sub_reference = $tp->compile_as_sub($template);
Much like "compile", this method does exactly the same compilation, but returns a reference to an anonymous subroutine that can be used each time you want to "explode" the template. This has the advantage that you immediately know if your template compiles good (otherwise you will get undef) and that each following template explosion will not require any Perl code compilation. On the other hand, you'll have to endure the code compilation immediately, which could be bad for you if you don't end up using it.
The anonymous sub that is returned accepts a single, optional parameter, namely a reference to a hash of variables to be used in addition to the "streamline" ones.
Note that if you add/change/remove values using the variables member of the Template::Perlish object, these changes will reflect on the anonymous sub, so you end up using different values in two subsequent invocations of the sub. This is consistent with the behaviuor of the "evaluate" method.
variables
$final_text = $tp->evaluate($compiled); # OR $final_text = $tp->evaluate($compiled, \%variables);
evaluate a template (in its compiled for, see "compile") with the available variables. In the former form, only the already configured variables are used; in the latter, the given $variables (which is a hash reference) are added, overriding any matching key.
$variables
Returns the processed text as a string.
$final_text = $tp->process($template); # OR $final_text = $tp->process($template, $variables);
this method included "compile" and "evaluate" into a single step.
There's really very little to say: write your document/text/whatever, and embed special parts with the delimiters of your choice (or stick to the defaults). If you have to print stuff, just print to STDOUT, it will be automatically catpured (unless you're calling the generated code by yourself).
Anything inside these "special" parts matching the regular expression /^\s*\w+(?:\.\w+)*\s*$/, i.e. consisting only of a sequence of alphanumeric tokens separated by dots, are considered to be variables and processed accordingly. Thus, available variables can be accessed in two ways: using the dotted notation, as in
[% some.value.3.lastkey %]
or explicitly using the %variables hash:
%variables
[% print $variables{some}{value}[3]{lastkey} %]
The former is cleaner, but the latter is more powerful of course.
If you happen to have a value you want to print inside a simple scalar variable, instead of:
[% print $variable; %]
you can also you the short form:
[% $variable %]
Note: only the scalar variable name, nothing else apart optional spaces. If you have something fancier, i.e. a Perl expression, you can use a shortcut to evaluate it and print all in one single command:
[%= my $value = 100; "*** $variable -> $value ***" %]
Note that there is an equal sign (") immediately after the command opener [%. The Perl expression is evaluated in scalar context, and the result is printed (if defined, otherwise it's skipped). This sort of makes the previous short form for simple scalars a bit outdated, but you spare a character in any case and it's just DWIM.
"
If you know Perl, you should not have problems using the control structures. Just intersperse the code with the templates as you would normally do in any other templating system:
[% if ($variables{this}) { %] blah blah [% this %], foo bar! [% } else { %] yak yak that! [% } %]
Take care to always terminate your commands with a ; each time you would do it in actual code.
;
There's no escaping mechanism, so if you want to include literal [% or %] you either have to change delimiters, or you have to resort to tricks. In particular, a stray closing inside a textual part won't be a problem, e.g.:
[% print "variable"; %] %] [% print "another"; %]
prints:
variable %] another
The tricky part is including the closing in the Perl code, but there can be many tricks:
[% print '>>>%'.']<<<' %]
prints
>>>%]<<<
To include a starter in the text just print it inside a Perl block:
here it comes [% print '[%'; %] the delimiter
here it comes [% the delimiter
Another trick is to separate the two chars with an empty block:
here it comes [[%%]% the delimiter
Including the starter in the Perl code is not a problem, of course.
So the bottom line is: who needs escaping?
Unfortunately, the diagnostic is still quite poor.
open(): %s
the only perlfunc/open is done to print stuff to a string. If you get this error, you're probably using a version of Perl that's too old.
perlfunc/open
unclosed %s at position %d
a Perl block was opened but not closed.
Template::Perlish requires no configuration files or environment variables.
None, apart a fairly recent version of Perl.
None reported.
No bugs have been reported.
Please report any bugs or feature requests through http://rt.cpan.org/
Due to the fact that Perl code is embedded directly into the template, you have to take into consideration all the possible security implications. In particular, you should avoid taking templates from outside, because in this case you'll be evaluating Perl code that you haven't checked. CAVEAT EMPTOR.
Flavio Poletti <flavio [at] polettix [dot] it>
<flavio [at] polettix [dot] it>
Copyright (c) 2008, Flavio Poletti <flavio [at] polettix [dot] it>. All rights reserved.
This module is free software; you can redistribute it and/or modify it under the same terms as Perl 5.8.x itself. See perlartistic and perlgpl.
Questo modulo è software libero: potete ridistribuirlo e/o modificarlo negli stessi termini di Perl 5.8.x stesso. Vedete anche perlartistic e perlgpl.
BECAUSE THIS SOFTWARE IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE SOFTWARE, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE SOFTWARE "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE SOFTWARE IS WITH YOU. SHOULD THE SOFTWARE PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR, OR CORRECTION.
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE SOFTWARE AS PERMITTED BY THE ABOVE LICENCE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE SOFTWARE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE SOFTWARE TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
Poiché questo software viene dato con una licenza gratuita, non c'è alcuna garanzia associata ad esso, ai fini e per quanto permesso dalle leggi applicabili. A meno di quanto possa essere specificato altrove, il proprietario e detentore del copyright fornisce questo software "così com'è" senza garanzia di alcun tipo, sia essa espressa o implicita, includendo fra l'altro (senza però limitarsi a questo) eventuali garanzie implicite di commerciabilità e adeguatezza per uno scopo particolare. L'intero rischio riguardo alla qualità ed alle prestazioni di questo software rimane a voi. Se il software dovesse dimostrarsi difettoso, vi assumete tutte le responsabilità ed i costi per tutti i necessari servizi, riparazioni o correzioni.
In nessun caso, a meno che ciò non sia richiesto dalle leggi vigenti o sia regolato da un accordo scritto, alcuno dei detentori del diritto di copyright, o qualunque altra parte che possa modificare, o redistribuire questo software così come consentito dalla licenza di cui sopra, potrà essere considerato responsabile nei vostri confronti per danni, ivi inclusi danni generali, speciali, incidentali o conseguenziali, derivanti dall'utilizzo o dall'incapacità di utilizzo di questo software. Ciò include, a puro titolo di esempio e senza limitarsi ad essi, la perdita di dati, l'alterazione involontaria o indesiderata di dati, le perdite sostenute da voi o da terze parti o un fallimento del software ad operare con un qualsivoglia altro software. Tale negazione di garanzia rimane in essere anche se i dententori del copyright, o qualsiasi altra parte, è stata avvisata della possibilità di tali danneggiamenti.
Se decidete di utilizzare questo software, lo fate a vostro rischio e pericolo. Se pensate che i termini di questa negazione di garanzia non si confacciano alle vostre esigenze, o al vostro modo di considerare un software, o ancora al modo in cui avete sempre trattato software di terze parti, non usatelo. Se lo usate, accettate espressamente questa negazione di garanzia e la piena responsabilità per qualsiasi tipo di danno, di qualsiasi natura, possa derivarne.
The best templating system in the world is undoubtfully Template::Toolkit.
See http://perl.apache.org/docs/tutorials/tmpl/comparison/comparison.html for a comparison (and a fairly complete list) of different templating modules.
To install Template::Perlish, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Template::Perlish
CPAN shell
perl -MCPAN -e shell install Template::Perlish
For more information on module installation, please visit the detailed CPAN module installation guide.