# Copyright Tamil 2011
#
# This file is part of Koha.
#
# Koha is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# Koha is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Koha; if not, see <https://www.gnu.org/licenses>.

#simple parser for HTML with Template Toolkit directives. Tokens are put into @tokens and are accessible via next_token and peep_token
package C4::TTParser;
use base qw(HTML::Parser);
use C4::TmplToken;
use strict;
use warnings;

#seems to be handled post tokenizer
##hash where key is tag we are interested in and the value is a hash of the attributes we want
#my %interesting_tags = (
#    img => { alt => 1 },
#);

#tokens found so far (used like a stack)
my (@tokens);

#shiftnext token or undef
sub next_token {
    return shift @tokens;
}

#unshift token back on @tokens
sub unshift_token {
    my $self = shift;
    unshift @tokens, shift;
}

#have a peep at next token
sub peep_token {
    return $tokens[0];
}

#wrapper for parse
#please use this method INSTEAD of the HTML::Parser->parse_file method (and HTML::Parser->parse)
#signature build_tokens( self, filename)
sub build_tokens {
    my ( $self, $filename ) = @_;
    $self->{filename} = $filename;
    $self->handler( start => "start", "self, line, tagname, attr, text" )
        ;    #signature is start( self, linenumber, tagname, hash of attributes, original text )
    $self->handler( text => "text", "self, line, text, is_cdata" )
        ;    #signature is text( self, linenumber, original text, is_cdata )
    $self->handler( end => "end", "self, line, tag, attr, text" )
        ;    #signature is end( self, linenumber, tagename, original text )
    $self->handler( declaration => "declaration", "self, line, text, is_cdata" );    # declaration
    $self->handler( comment     => "comment",     "self, line, text, is_cdata" );    # comments
    $self->handler( process     => "process",     "self, line, text, is_cdata" );    # processing statement <?...?>

    #    $self->handler(default => "default", "self, line, text, is_cdata"); # anything else
    $self->marked_sections(1)
        ;    #treat anything inside CDATA tags as text, should really make it a C4::TmplTokenType::CDATA
    $self->unbroken_text(1);    #make contiguous whitespace into a single token (can span multiple lines)
    open( my $fh, "<:encoding(utf8)", $filename ) || die "Cannot open $filename ($!)";
    $self->parse_file($fh);
    return $self;
}

#handle parsing of text
sub text {
    my $self     = shift;
    my $line     = shift;
    my $work     = shift;       # original text
    my $is_cdata = shift;

    # If there is a split template toolkit tag
    if ( $work =~ m/^(?:(?!\[%).)*?%\]/s ) {
        my @strings = ($&);
        my $token;
        $work = $';
        while ( $token = pop @tokens ) {
            if ( $token->string =~ m/\[%.*?$/s ) {
                push @tokens,  C4::TmplToken->new( $`, $token->type, $token->line_number, $token->pathname );
                push @strings, $&;
                last;
            } else {
                push @strings, $token->string;
            }
        }
        push @tokens,
            C4::TmplToken->new(
            join( '', reverse @strings ),         C4::TmplTokenType::DIRECTIVE,
            $token ? $token->line_number : $line, $self->{filename}
            );
    }
    while ($work) {

        # if there is a template_toolkit tag
        if ( $work =~ m/\[%.*?%\]/ ) {

            #everything before this tag is text (or possibly CDATA), add a text token to tokens if $`
            if ($`) {
                my $t = C4::TmplToken->new(
                    $`, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
                    $self->{filename}
                );
                push @tokens, $t;
            }

            #the match itself is a DIRECTIVE $&
            my $t = C4::TmplToken->new( $&, C4::TmplTokenType::DIRECTIVE, $line, $self->{filename} );
            push @tokens, $t;

            # put work still to do back into work
            $work = $' ? $' : 0;
        } else {

            # If there is some left over work, treat it as text token
            my $t = C4::TmplToken->new(
                $work, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
                $self->{filename}
            );

            push @tokens, $t;
            last;
        }
    }
}

sub declaration {
    my $self     = shift;
    my $line     = shift;
    my $work     = shift;                #original text
    my $is_cdata = shift;
    my $t        = C4::TmplToken->new(
        $work, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
        $self->{filename}
    );
    push @tokens, $t;
}

sub comment {
    my $self     = shift;
    my $line     = shift;
    my $work     = shift;                #original text
    my $is_cdata = shift;
    my $t        = C4::TmplToken->new(
        $work, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
        $self->{filename}
    );
    push @tokens, $t;
}

sub process {
    my $self     = shift;
    my $line     = shift;
    my $work     = shift;                #original text
    my $is_cdata = shift;
    my $t        = C4::TmplToken->new(
        $work, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
        $self->{filename}
    );
    push @tokens, $t;
}

sub default {
    my $self     = shift;
    my $line     = shift;
    my $work     = shift;                #original text
    my $is_cdata = shift;
    my $t        = C4::TmplToken->new(
        $work, ( $is_cdata ? C4::TmplTokenType::CDATA : C4::TmplTokenType::TEXT ), $line,
        $self->{filename}
    );
    push @tokens, $t;
}

#handle opening html tags
sub start {
    my $self = shift;
    my $line = shift;
    my $tag  = shift;
    my $hash = shift;                                                                          #hash of attr/value pairs
    my $text = shift;                                                                          #original text
    my $t    = C4::TmplToken->new( $text, C4::TmplTokenType::TAG, $line, $self->{filename} );
    my %attr;

    # tags seem to be uses in an 'interesting' way elsewhere..
    for my $key (%$hash) {
        next unless defined $hash->{$key};
        if ( $key eq "/" ) {
            $attr{ +lc($key) } = [ $key, $hash->{$key}, $key . "=" . $hash->{$key}, 1 ];
        } else {
            $attr{ +lc($key) } = [ $key, $hash->{$key}, $key . "=" . $hash->{$key}, 0 ];
        }
    }
    $t->set_attributes( \%attr );
    push @tokens, $t;
}

#handle closing html tags
sub end {
    my $self = shift;
    my $line = shift;
    my $tag  = shift;
    my $hash = shift;
    my $text = shift;

    # what format should this be in?
    my $t = C4::TmplToken->new( $text, C4::TmplTokenType::TAG, $line, $self->{filename} );
    my %attr;

    # tags seem to be uses in an 'interesting' way elsewhere..
    for my $key (%$hash) {
        next unless defined $hash->{$key};
        $attr{ +lc($key) } = [ $key, $hash->{$key}, $key . "=" . $hash->{$key}, 0 ];
    }
    $t->set_attributes( \%attr );
    push @tokens, $t;
}

1;