=encoding UTF-8

=head1 NAME

JSON::Tokenize - Tokenize JSON

=head1 SYNOPSIS

    
    use JSON::Tokenize ':all';
    
    my $input = '{"tuttie":["fruity", true, 100]}';
    my $token = tokenize_json ($input);
    print_tokens ($token, 0);
    
    sub print_tokens
    {
        my ($token, $depth) = @_;
        while ($token) {
            my $start = tokenize_start ($token);
            my $end = tokenize_end ($token);
            my $type = tokenize_type ($token);
            print "   " x $depth;
            my $value = substr ($input, $start, $end - $start);
            print "'$value' has type '$type'.\n";
            my $child = tokenize_child ($token);
            if ($child) {
                print_tokens ($child, $depth+1);
            }
            my $next = tokenize_next ($token);
            $token = $next;
        }
    }


This outputs

    '{"tuttie":["fruity", true, 100]}' has type 'object'.
       '"tuttie"' has type 'string'.
       ':' has type 'colon'.
       '["fruity", true, 100]' has type 'array'.
          '"fruity"' has type 'string'.
          ',' has type 'comma'.
          'true' has type 'literal'.
          ',' has type 'comma'.
          '100' has type 'number'.


=head1 VERSION

This documents version 0.61 of JSON::Tokenize corresponding to
L<git commit 033269fa8972fdce8626aa65cd11a5394ab50492|https://github.com/benkasminbullock/JSON-Parse/commit/033269fa8972fdce8626aa65cd11a5394ab50492> released on Thu Feb 11 09:14:04 2021 +0900.



=head1 DESCRIPTION

This is a module for tokenizing a JSON string. "Tokenizing" means
breaking the string into individual tokens, without creating any Perl
structures. It uses the same underlying code as
L<JSON::Parse>. Tokenizing can be used for tasks such as picking out
or searching through parts of a large JSON structure without storing
each part of the entire structure in memory.

This module is an experimental part of L<JSON::Parse> and its
interface is likely to change. The tokenizing functions are currently
written in a very primitive way.

=head1 FUNCTIONS

=head2 tokenize_child

    my $child = tokenize_child ($child);

Walk the tree of tokens.

=head2 tokenize_end

    my $end = tokenize_end ($token);

Get the end of the token as a byte offset from the start of the
string. Note this is a byte offset not a character offset.

=head2 tokenize_json

    my $token = tokenize_json ($json);

=head2 tokenize_next

    my $next = tokenize_next ($token);

Walk the tree of tokens.

=head2 tokenize_start

    my $start = tokenize_start ($token);

Get the start of the token as a byte offset from the start of the
string. Note this is a byte offset not a character offset.

=head2 tokenize_text

    my $text = tokenize_text ($json, $token);

Given a token C<$token> from this parsing and the JSON in C<$json>,
return the text which corresponds to the token. This is a convenience
function written in Perl which uses L</tokenize_start> and
L</tokenize_end> and C<substr> to get the string from C<$json>.

=head2 tokenize_type

    my $type = tokenize_type ($token);

Get the type of the token as a string. The possible return values are

    "array",
    "initial state",
    "invalid",
    "literal",
    "number",
    "object",
    "string",
    "unicode escape"



=head1 AUTHOR

Ben Bullock, <bkb@cpan.org>

=head1 COPYRIGHT & LICENCE

This package and associated files are copyright (C) 
2016-2021
Ben Bullock.

You can use, copy, modify and redistribute this package and associated
files under the Perl Artistic Licence or the GNU General Public
Licence.