Perl::ToPerl6 - Transmogrify Perl5 code into Perl6
perlmogrify source.pl
Perl::ToPerl6 is an extensible framework for transforming Perl5 code into Perl6. The framework owes a great deal to Jeffrey Thalmer and Perl::Critic, inasmuch as the source was almost completely cribbed from there. Any remaining bugs are my fault, however.
This is meant to be extensible so that others can create their own Perl5 to Perl6 custom transformation modules. Feel free to add them to the appropriate Perl::ToPerl6::Transformerx namespace.
As the original Perl::Critic module still remains under the hood, you should still be able to use the original configuration and .rc system.
The core transformations were simmply chosen by taking an existing perl5 file and editing it until it compiled cleanly under Perl6. The author does not guarantee that the code semantically matches the old Perl5 source, merely that it is now compilable Perl6 code.
The documentation is still basically a s/// of the original Perl::Critic documentation, with the exception of this README. If you're brave or foolish, feel free to use Test::Perl::ToPerl6 to translate your code as part of your test suite or source code control platform, but caveat emptor.
As of the initial release, it contains transformers for:
Basic data types (integers, floats, strings)
Binary, Octal and Hexadecimal integer
Floating-point trailing decimal
Here-docs
Interpolated values ("${x}", "$x-foo" etc.)
Interpolated case modifiers ("\lFoo", "\Ufoo\E" etc.)
Builtins
'print $fh "text hee"'
Q types
qw(), qr(), qx()
Conditional constructs ('if', 'elsif', 'unless', 'when')
Looping constructs ('for', 'foreach', 'while', 'until')
Transformation of C-style loops
First-order operators
'map {} @foo', 'grep !2 @a'
Core operators ('->', '.', '<<', '>>', '!' etc.)
Package declarations ('package My::Package;', 'package My::Package {}')
Package usage ('use Foo::Bar')
Pragmas ('utf8', 'warnings', 'overload' etc.)
Dereferencing ('%{ $foo }', '%$foo' etc.)
Hashes ('$foo{a}', "$foo{'a'}" etc.)
Sigils ('$x', '$a[1]', '@a', '%a{a,b}' etc.)
Special variables ('@ARGV', '@+', '%ENV' etc.)
Special literals ('__END__', '__PACKAGE__', '__FILE__' etc.)
Please read Perl::Critic>s documentation for more information. Quite a bit of this will be invalid soon as while the infrastructure it provides is wonderful, some of the indiviual flags and data types Perl::Critic uses aren't really appropriate for this.
Please see the Perl::ToPerl6::Transformer:: namespace for a full listing of the core modules that come with Perl::ToPerl6. The core modules are all documented, albeit in a minimal sense. 'perlmogrify -doc PATTERN' might still work and should bring up documentation for the appropriate module.
If other people do write transformer modules, I might be tempted to add support for a LPerl::ToPerl6::TransformerX namespace, but we'll discuss that issue if it ever comes up.
This is a completely unused feature for the moment, it may come into use once I figure out how to properly integrate it.
I don't really see a point to this at the moment unless you need to retain a specific module or some tricky bit of syntax that fails to convert, but the '## no mogrify' line marker may work for your needs, as it's a holdover from this module's genesis as Perl::Critic.
THE Perl::ToPerl6 PHILOSOPHY
For the moment, there is no facility for defining the order that transformers are run i, so each individual module cannot rely on other transformations being done. So far during development this hasn't been a problem, but I do make allowances at certain points, for instance there are two separate transformation modules for the C loop, and the last one to run assumes that nothing has been run.
There is a Perl::ToPerl6::Utils namespace where utilities for transformers will reside, such as determining the style of a 'for' loop. This was a pain and probably already exists on CPAN but I'm already pulling in quite a few dependencies for this module as it is.
The simplest way to go about this is find a module in the list that performs a task like what you want, copy that, and start walking its PPI tree. Each module is presumed to act on one node of the tree at a time, in other words C<$elem> will always be a single element of the type you're modifying. This keeps code simple, and lets the main body collect statistics about what it's modifying.
For instance, when running it you'll get an awful lot of output about what the modules are doing, complete with line and column numbers of where the modifications are happening. This is more or less so that you can trace back to the point of origin when a module does something you don't expect.
Your module receives the original document in C<$doc> and the element to process in C<$elem>. If you make no modifications to the element, just return. Otherwise, calling C<transformation()> tells the main application that your module has changed source.
Just to keep the source tree clean and reasonably Perlish, I try to create new tokens for whitespace and such where it's practical. Please also note that at some points I'm forced to violate PPI encapsulation, for instance changing brace styles or a heredoc's marker.
Something else to keep in mind as you're creating tests is that the expression you're looking for won't always begin at the start of a LPPI::Statement. As a trivial example, C<$x++> may occur at the end of a long statement, such as C<1 if $x++>. So, when creating your test suites be sure that at least a few of your test cases don't begin precisely at the statement boundary.
Feel free to send me a pull request on GitHub if you've developed a module and want it integrated.
Perl::ToPerl6 requires the following modules:
Currently the development team is just me, [mailto:drforr@pobox.com] or send me a pull request for the appropriate module on GitHUb. I'll keep an eye out for requests and integrate them as they come in, or within a reasonable time frame.
You can also catch me on IRC at either [irc://irc.perl.org/#perl] or [irc://irc.freenode.net/#perl], and follow me on Twitter at [https://twitter.com/drforr]
Please note this is mostly based on Perl::Critic code, so please see that module if you have questions about why something was implemented. Also most of the documentation still is from that module, and will be replaced on a purely ad-hoc basis.
Feel free to submit bugs via either RT or GitHub. GitHub and personal email gets checked more frequently, or just bounce me a note on IRC if I happen to be active.
Jeffrey Thalhammer - For creating the framework I'm shamelessly ripping off, so I don't have to create an entire plugin architecture.
Adam Kennedy - For creating PPI, the heart and soul of Perl::ToPerl6.
Damian Conway - For writing Perl Best Practices, finally :)
Chris Dolan - For contributing the best features and Transformer modules.
Andy Lester - Wise sage and master of all-things-testing.
Elliot Shank - The self-proclaimed quality freak.
Giuseppe Maxia - For all the great ideas and positive encouragement.
Jeffrey Goff drforr@pobox.com
Jeffrey Ryan Thalhammer jeff@imaginative-software.com
Copyright (c) 2015 Jeffrey Goff drforr@pobox.com. All rights reserved.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The full text of this license can be found in the LICENSE file included with this module.