File Coverage

blib/lib/MarpaX/Demo/JSONParser.pm

Criterion	Covered	Total	%
statement	112	115	97.3
branch	54	64	84.3
condition			n/a
subroutine	15	15	100.0
pod	1	6	16.6
total	182	200	91.0

package MarpaX::Demo::JSONParser;

use strict;
use warnings;

use File::Basename; # For basename.
use File::Slurper 'read_text';

use Marpa::R2;

use MarpaX::Demo::JSONParser::Actions;
use MarpaX::Simple qw(gen_parser);

use Moo;

use Types::Standard qw/Any Str/;

has base_name =>
(
	default		=> sub {return ''},
	is			=> 'rw',
	isa			=> Str,
	required	=> 0,
);

has bnf_file =>
(
	default		=> sub {return ''},
	is			=> 'rw',
	isa			=> Str,
	required	=> 1,
);

has grammar =>
(
	default		=> sub {return ''},
	is			=> 'rw',
	isa			=> Any,
	required	=> 0,
);

has parser =>
(
	default		=> sub {return ''},
	is			=> 'rw',
	isa			=> Any,
	required	=> 0,
);

has scanner =>
(
	default		=> sub {return ''},
	is			=> 'rw',
	isa			=> Any,
	required	=> 0,
);

our $VERSION = '1.07';

# ------------------------------------------------

sub BUILD
{
	my($self) = @_;
	my $bnf   = read_text $self -> bnf_file;

	$self -> base_name(basename($self -> bnf_file) );

	if ($self -> base_name eq 'json.1.bnf')
	{
		$self-> grammar
		(
			Marpa::R2::Scanless::G -> new
			({
				default_action => 'do_first_arg',
				source         => \$bnf,
			})
		)
	}
	elsif ($self -> base_name eq 'json.2.bnf')
	{
		$self-> grammar
		(
			Marpa::R2::Scanless::G -> new
			({
				bless_package => 'MarpaX::Demo::JSONParser::Actions',
				source        => \$bnf,
			})
		)
	}
	elsif ($self -> base_name eq 'json.3.bnf')
	{
		$self-> parser
		(
			gen_parser
			(
				grammar => $bnf,
			)
		);
	}
	else
	{
		die "Unknown BNF. Use either 'json.[123].bnf'\n";
	}

	if ($self -> base_name ne 'json.3.bnf')
	{
		$self -> scanner
		(
			Marpa::R2::Scanless::R -> new
			({
				grammar           => $self -> grammar,
				semantics_package => 'MarpaX::Demo::JSONParser::Actions',
			})
		);
	}

} # End of BUILD.

# ------------------------------------------------

sub decode_string
{
	my ($self, $s) = @_;

	$s =~ s/\\u([0-9A-Fa-f]{4})/chr(hex($1))/eg;
	$s =~ s/\\n/\n/g;
	$s =~ s/\\r/\r/g;
	$s =~ s/\\b/\b/g;
	$s =~ s/\\f/\f/g;
	$s =~ s/\\t/\t/g;
	$s =~ s/\\\\/\\/g;
	$s =~ s{\\/}{/}g;
	$s =~ s{\\"}{"}g;

	return $s;

} # End of decode_string.

# ------------------------------------------------

sub eval_json
{
	my($self, $thing) = @_;
	my($type) = ref $thing;

	if ($type eq 'REF')
	{
		return \$self -> eval_json( ${$thing} );
	}
	elsif ($type eq 'ARRAY')
	{
		return [ map { $self -> eval_json($_) } @{$thing} ];
	}
	elsif ($type eq 'MarpaX::Demo::JSONParser::Actions::string')
	{
		my($string) = substr $thing->[0], 1, -1;

		return $self -> decode_string($string) if ( index $string, '\\' ) >= 0;
		return $string;
	}
	elsif ($type eq 'MarpaX::Demo::JSONParser::Actions::hash')
	{
		return { map { $self -> eval_json( $_->[0] ), $self -> eval_json( $_->[1] ) } @{ $thing->[0] } };
	}

	return 1  if $type eq 'MarpaX::Demo::JSONParser::Actions::true';
	return '' if $type eq 'MarpaX::Demo::JSONParser::Actions::false';
	return $thing;

} # End of eval_json.

# ------------------------------------------------

sub parse
{
	my($self, $string) = @_;

	if ($self -> base_name eq 'json.3.bnf')
	{
		my $parse_value = $self -> parser -> ($string);

		return $self -> post_process(@{$parse_value});
	}
	else
	{
		$self -> scanner -> read(\$string);

		my($value_ref) = $self -> scanner -> value;

		die "Parse failed\n" if (! defined $value_ref);

		$value_ref = $self -> eval_json($value_ref) if ($self -> base_name eq 'json.2.bnf');

		return $$value_ref;
	}

} # End of parse.

# ------------------------------------------------

sub post_process
{
	my ($self, $type, @value) = @_;

	return $value[0] if $type eq 'number';
	return undef if $type eq 'null';
	return $value[0] if $type eq 'easy string';
	return $self -> unescape($value[0]) if $type eq 'any char';
	return chr(hex(substr($value[0],2))) if $type eq 'hex char';
	return 1 if $type eq 'true';
	return q{} if $type eq 'false';

	if ($type eq 'array')
	{
		my @result = ();
		push @result, $self -> post_process(@{$_}) for @{$value[0]};

		return \@result;
	}

	if ($type eq 'hash')
	{
		my %result = ();

		for my $pair (@{$value[0]})
		{
			my $key = $self -> post_process(@{$pair->[0]});
			$result{$key} = $self -> post_process(@{$pair->[1]});
		}

		return \%result;
	}

	if ($type eq 'string')
	{
		return join q{}, map { $self -> post_process( @{$_} ) } @{$value[0]};
	}

	die join q{ }, 'post process failed:', $type, @value;

} # End of post_process.

# ------------------------------------------------

sub unescape
{
	my($self, $char) = @_;

	return "\b" if $char eq 'b';
	return "\f" if $char eq 'f';
	return "\n" if $char eq 'n';
	return "\r" if $char eq 'r';
	return "\t" if $char eq 't';
	return '/'  if $char eq '/';
	return '\\' if $char eq '\\';
	return '"'  if $char eq '"';

	# If the character is not legal, return it anyway
	# As an alternative, we could fail here.

	return $char;

} # End of unescape.

# ------------------------------------------------

1;

=pod

=head1 NAME

C - A JSON parser with a choice of grammars

=head1 Synopsis

	#!/usr/bin/env perl

	use strict;
	use warnings;

	use MarpaX::Demo::JSONParser;

	use Try::Tiny;

	my($app_name) = 'MarpaX-Demo-JSONParser';
	my($bnf_name) = 'json.1.bnf'; # Or 'json.2.bnf'. See scripts/find.grammars.pl below.
	my($bnf_file) = "data/$bnf_name";
	my($string)   = '{"test":"1.25e4"}';

	my($message);
	my($result);

	# Use try to catch die.

	try
	{
		$message = '';
		$result  = MarpaX::Demo::JSONParser -> new(bnf_file => $bnf_file) -> parse($string);
	}
	catch
	{
		$message = $_;
		$result  = 0;
	};

	print $result ? "Result: test => $$result{test}. Expect: 1.25e4. \n" : "Parse failed. $message";

This script ships as scripts/demo.pl.

You can test failure by deleting the '{' character in line 17 of demo.pl and re-running it.

See also t/basic.tests.t for more sample code.

=head1 Description

C demonstrates 2 grammars for parsing JSON.

Only 1 grammar is loaded per run, as specified by the C option to C<< new() >>.

See t/basic.tests.t for sample code.

=head1 Installation

Install C as you would for any C module:

Run:

	cpanm MarpaX::Demo::JSONParser

or run:

	sudo cpan MarpaX::Demo::JSONParser

or unpack the distro, and then either:

	perl Build.PL
	./Build
	./Build test
	sudo ./Build install

or:

	perl Makefile.PL
	make (or dmake or nmake)
	make test
	make install

=head1 Constructor and Initialization

C is called as C<< my($parser) = MarpaX::Demo::JSONParser -> new(k1 => v1, k2 => v2, ...) >>.

It returns a new object of type C.

Key-value pairs accepted in the parameter list (see corresponding methods for details
[e.g. bnf_file([$string])]):

=over 4

=item o bnf_file aUserGrammarFileName

Specify the name of the file containing your Marpa::R2-style grammar.

See data/json.1.bnf, data/json.2.bnf and data/json.3.bnf for the cases handled by the code.

This option is mandatory.

Default: ''.

=back

=head1 Methods

=head2 parse($string)

Parses the given $string using the grammar whose file name was provided by the C option to
C<< new() >>.

Dies if the parse fails, or returns the result of the parse if it succeeded.

=head1 Files Shipped with this Module

=head2 Data Files

These JSON grammars are discussed in the L below.

=over 4

=item o data/json.1.bnf

This JSON grammar was devised by Peter Stuifzand.

=item o data/json.2.bnf

This JSON grammar was devised by Jeffrey Kegler.

=item o data/json.3.bnf

This JSON grammar was devised by Jeffrey Kegler.

=back

=head2 Scripts

=over 4

=item o scripts/demo.pl

This program is exactly what is displayed in the L above.

Before installation of this module, run it with:

	shell> perl -Ilib scripts/demo.pl

And after installation, just use:

	shell> perl scripts/demo.pl

=item o scripts/find.grammars.pl

After installation of the module, run it with:

	shell> perl scripts/find.grammars.pl (Defaults to json.1.bnf)
	shell> perl scripts/find.grammars.pl json.1.bnf

Or use json.2.bnf or json.2.bnf.

It will print the name of the path to given grammar file.

=back

=head1 FAQ

=head2 Where are the grammar files actually installed?

They are not installed (when the source code is). They are shipped in the data/ dir.

I used to use L and L to install them, but Module::Install is now
unusable. See Changes for details.

=head2 Which JSON BNF is best?

This is not really a fair question. They were developed under different circumstances.

=over 4

=item o json.1.bnf is by Peter Stuifzand.

json.1.bnf is the first attempt, when the Marpa SLIF still did not handle utf8. And it's meant to be a practical
grammar. The sophisticated test suite is his, too.

=item o json.2.bnf is by Jeffrey Kegler, the author of L.

json.2.bnf was written later, after Jeffey had a chance to study json.1.bnf. He used it to help optimise Marpa,
but with a minimal test suite, so it had a different purpose.

I (Ron) converted their code into forms suitable for building this module.

=item o json.3.bnf is by Jeffrey Kegler.

He developed this in August, 2014, after recent significant progress in the writing of Marpa.

=back

=head2 Where is Marpa's Homepage?

L.

=head2 Are there any articles discussing Marpa?

Yes, many by its author, and several others. See Marpa's homepage, just above, and:

L, (in progress, by Peter Stuifzand and Ron Savage).

L, by Peter Stuifzand.

L, by Peter Stuifzand.

L, by Ron Savage.

=head1 See Also

L.

L.

L.

=head1 Machine-Readable Change Log

The file Changes was converted into Changelog.ini by L.

=head1 Repository

L

=head1 Version Numbers

Version numbers < 1.00 represent development versions. From 1.00 up, they are production versions.

=head1 Support

Email the author, or log a bug on RT:

L.

=head1 Author

L was written by Ron Savage [e-mail]> in 2013.

Home page: L.

=head1 Copyright

Australian copyright (c) 2013, Ron Savage.

	All Programs of mine are 'OSI Certified Open Source Software';
	you can redistribute them and/or modify them under the terms of
	The Perl License, a copy of which is available at:
	http://www.opensource.org/licenses/index.html

=cut