#
#   Interface Definition Language (OMG IDL CORBA v3.0)
#
#   Lexer module
#

package CORBA::IDL::Lexer;

use strict;
use warnings;

our $VERSION = '2.63';

use Math::BigInt;
use Math::BigFloat;

sub _StringLexer {
    my ($parser, $token) = @_;
    my $str = q{};

    while ($parser->YYData->{line}) {

        for ($parser->YYData->{line}) {

            s/^\"//
                and return ($token, $str);

            s/^([^"\\]+)//
                and $str .= $1,     # any character except single quote or backslash
                    last;

            s/^\\a//
                and $str .= "\a",   # alert
                    last;
            s/^\\b//
                and $str .= "\b",   # backspace
                    last;
            s/^\\f//
                and $str .= "\f",   # form feed
                    last;
            s/^\\n//
                and $str .= "\n",   # new line
                    last;
            s/^\\r//
                and $str .= "\r",   # carriage return
                    last;
            s/^\\t//
                and $str .= "\t",   # horizontal tab
                    last;
            s/^\\v//
                and $str .= "\013", # vertical tab
                    last;
            s/^\\([\?'"])//
                and $str .= $1,     # backslash, question mark, single quote, double quote
                    last;

            s/^\\([0-7]{1,3})//
                and $str .= chr oct $1,
                    (oct $1) ? 1 : $parser->Error("null character in a string.\n"),
                    last;
            s/^\\x([0-9A-Fa-f]{1,2})//
                and $str .= chr hex $1,
                    (hex $1) ? 1 : $parser->Error("null character in a string.\n"),
                    last;
            if ($token eq 'WIDE_STRING_LITERAL') {
                s/^\\u([0-9A-Fa-f]{1,4})//
                    and $str .= chr hex $1,
                        (hex $1) ? 1 : $parser->Error("null character in a string.\n"),
                        last;
            }
            s/^\\//
                and $parser->Error("invalid escape sequence.\n"),
                    last
        }
    }

    $parser->Error("untermined string.\n");
    $parser->YYData->{lineno} ++;
    return ($token, $str);
}

sub _CharLexer {
    my ($parser, $token) = @_;

    $_ = $parser->YYData->{line};
    s/^([^'\\])\'//
        and return ($token, $1);        # any character except single quote or backslash

    s/^\\n\'//
        and return ($token, "\n");  # new line
    s/^\\t\'//
        and return ($token, "\t");  # horizontal tab
    s/^\\v\'//
        and return ($token, "\013");    # vertical tab
    s/^\\b\'//
        and return ($token, "\b");  # backspace
    s/^\\r\'//
        and return ($token, "\r");  # carriage return
    s/^\\f\'//
        and return ($token, "\f");  # form feed
    s/^\\a\'//
        and return ($token, "\a");  # alert
    s/^\\([\?'"])\'//
        and return ($token, $1);        # backslash, question mark, single quote, double quote
    s/^\\([0-7]{1,3})\'//
        and return ($token, chr oct $1);
    s/^\\x([0-9A-Fa-f]{1,2})\'//
        and return ($token, chr hex $1);
    if ($token eq 'WIDE_STRING_LITERAL') {
        s/^\\u([0-9A-Fa-f]{1,4})\'//
            and return ($token, chr hex $1);
    }

    s/^\\([^\s\(\)\[\]\{\}<>,;:="]*)//
        and $parser->Error("invalid escape sequence $1.\n"),
            return ($token, ' ');

    s/^([^\s\(\)\[\]\{\}<>,;:="]*)//
        and $parser->Error("invalid character $1.\n"),
            return ($token, q{ });

    print "INTERNAL_ERROR:_CharLexer $_\n";
    return ($token, q{ });
}

sub _Identifier {
    my ($parser, $ident) = @_;

    my $key = uc $ident;
    if (exists $parser->YYData->{keyword}{$key}) {
        my ($keywd, $version, $lang) = @{$parser->YYData->{keyword}{$key}};
        if ($CORBA::IDL::Parser::IDL_VERSION ge $version) {
            if ($ident eq $keywd) {
                return ($key, $ident);
            }
            else {
                $parser->Error("'$ident' collides with keyword '$keywd'.\n");
                return ('IDENTIFIER', $ident);
            }
        }
        else {
            if (defined $lang) {
                if ($ident eq $keywd) {
                    $parser->Info("'$ident' is a keyword of $lang.\n");
                }
                else {
                    $parser->Info("'$ident' collides with keyword '$keywd' of $lang.\n");
                }
            }
            else {
                if ($ident eq $keywd) {
                    $parser->Info("'$ident' is a future keyword.\n");
                }
                else {
                    $parser->Info("'$ident' collides with future keyword '$keywd'.\n");
                }
            }
        }
    }
    return ('IDENTIFIER', $ident);
}

sub _EscIdentifier {
    my ($parser, $ident) = @_;

    if ($CORBA::IDL::Parser::IDL_VERSION ge '2.3') {
        my $key = uc $ident;
        unless (exists $parser->YYData->{keyword}{$key}) {
            Info("Unnecessary escaped identifier '$ident'.\n");
        }
    }
    else {
        $parser->Warning("Escaped identifier is not allowed.\n");
    }
    return ('IDENTIFIER', $ident);
}

sub _OctInteger {
    my ($parser, $str) = @_;

    my $val = new Math::BigInt(0);
    foreach (split //, $str) {
        $val = $val * new Math::BigInt(8) + new Math::BigInt(oct $_);
    }
    return ('INTEGER_LITERAL', $val);
}

sub _HexInteger {
    my ($parser, $str) = @_;

    my $val = new Math::BigInt(0);
    foreach (split //, $str) {
        $val = $val * new Math::BigInt(16) + new Math::BigInt(hex $_);
    }
    return ('INTEGER_LITERAL', $val);
}

sub _CommentLexer {
    my ($parser) = @_;

    while (1) {
            $parser->YYData->{line}
        or  $parser->YYData->{line} = readline $parser->YYData->{fh}
        or  return;

        for ($parser->YYData->{line}) {
            s/^\n//
                    and $parser->YYData->{lineno} ++,
                    last;
            s/^\*\///
                    and return;
            s/^.[^*\n]*//
                    and last;
        }
    }
}

sub _DocLexer {
    my ($parser) = @_;

    $parser->YYData->{doc} = q{};
    my $flag = 1;
    while (1) {
            $parser->YYData->{line}
        or  $parser->YYData->{line} = readline $parser->YYData->{fh}
        or  return;

        for ($parser->YYData->{line}) {
            s/^(\n)//
                    and $parser->YYData->{lineno} ++,
                        $parser->YYData->{doc} .= $1,
                        $flag = 0,
                        last;
            s/^\r//
                    and last;
            s/^\*\///
                    and return;
            unless ($flag) {
                s/^\*//
                        and $flag = 1,
                        last;
            }
            s/^([ \t\f\013]+)//
                    and $parser->YYData->{doc} .= $1,
                    last;
            s/^(.[\w \t]*)//
                    and $parser->YYData->{doc} .= $1,
                    $flag = 1,
                    last;
        }
    }
}

sub _DocAfterLexer {
    my ($parser) = @_;

    unless (defined $parser->YYData->{curr_node}) {
        $parser->_DocLexer();
        return;
    }

    unless (exists $parser->YYData->{curr_node}->{doc}) {
        $parser->YYData->{curr_node}->{doc} = q{};
    }
    my $flag = 1;
    while (1) {
            $parser->YYData->{line}
        or  $parser->YYData->{line} = readline $parser->YYData->{fh}
        or  return;

        for ($parser->YYData->{line}) {
            s/^(\n)//
                    and $parser->YYData->{lineno} ++,
                        $parser->YYData->{curr_node}->{doc} .= $1,
                        $flag = 0,
                        last;
            s/^\r//
                    and last;
            s/^\*\///
                    and return;
            unless ($flag) {
                s/^\*//
                        and $flag = 1,
                        last;
            }
            s/^([ \t\f\013]+)//
                    and $parser->YYData->{curr_node}->{doc} .= $1,
                    last;
            s/^(.[\w \t]*)//
                    and $parser->YYData->{curr_node}->{doc} .= $1,
                    $flag = 1,
                    last;
        }
    }
}

sub _CodeLexer {
    my ($parser) = @_;
    my $frag = q{};

    while (1) {
            $parser->YYData->{line}
        or  $parser->YYData->{line} = readline $parser->YYData->{fh}
        or  return;

        for ($parser->YYData->{line}) {
            s/^(\n)//
                    and $parser->YYData->{lineno} ++,
                        $frag .= $1,
                        last;
            s/^%\}.*//
                    and return ('CODE_FRAGMENT', $frag);
            s/^(.[^%\n]*)//
                    and $frag .= $1,
                        last;
        }
    }
}

sub _PragmaLexer {                      #   10.6.5  Pragma Directives for RepositoryId
    my ($parser, $line) = @_;

    for ($line) {
        s/^ID[ \t]+([0-9A-Za-z_:]+)[ \t]+\"([^\s">]+)\"//
                and $parser->YYData->{symbtab}->PragmaID($1,$2),
                    return;
        s/^prefix[ \t]+\"([0-9A-Za-z_:\.\/\-]*)\"//
                and $parser->YYData->{symbtab}->PragmaPrefix($1),
                    return;
        s/^version[ \t]+([0-9A-Za-z_:]+)[ \t]+([0-9]+)\.([0-9]+)//
                and $parser->YYData->{symbtab}->PragmaVersion($1,$2,$3),
                    return;

        $parser->Info("Non standard pragma.\n");
        return;
    }
}

sub _AttachDoc {
    my ($parser, $comment) = @_;

    if (defined $parser->YYData->{curr_node}) {
        if (exists $parser->YYData->{curr_node}->{doc}) {
            $parser->YYData->{curr_node}->{doc} .= $comment;
        }
        else {
            $parser->YYData->{curr_node}->{doc} = $comment;
        }
    }
}

sub Lexer {
    my ($parser) = @_;

    while (1) {
            $parser->YYData->{line}
        or  $parser->YYData->{line} = readline $parser->YYData->{fh}
        or  return (q{}, undef);

        unless (exists $parser->YYData->{srcname}) {
            if ($parser->YYData->{line} =~ /^#\s*(line\s+)?\d+\s+["<]([^\s">]+)[">]\s*\n/ ) {
                $parser->YYData->{srcname} = $2;
            }
            else {
                print "INTERNAL_ERROR:_Lexer\n";
            }
            if (defined $parser->YYData->{srcname}) {
                my @st = stat($parser->YYData->{srcname});
                $parser->YYData->{srcname_size} = $st[7];
                $parser->YYData->{srcname_mtime} = $st[9];
            }
        }

        for ($parser->YYData->{line}) {
            s/^#\s+[\d]+\s+"<[^>]+>"\s*\n//                         # cpp 3.2.3 ("<build-in>", "<command line>")
                    and last;

            s/^#\s+([\d]+)\s+["<]([^\s">]+)[">]\s+([\d]+)\s*\n//    # cpp
                    and $parser->YYData->{lineno} = $1,
                        $parser->YYData->{filename} = $2,
                        $parser->YYData->{doc} = q{},
                        $parser->YYData->{curr_node} = undef,
                        last;

            s/^#\s+([\d]+)\s+["<]([^\s">]+)[">]\s*\n//              # cpp
                    and $parser->YYData->{lineno} = $1,
                        $parser->YYData->{filename} = $2,
                        $parser->YYData->{doc} = q{},
                        $parser->YYData->{curr_node} = undef,
                        last;
            s/^#\s*line\s+([\d]+)\s+["<]([^\s">]+)[">]\s*\n//       # CL.EXE Microsoft VC
                    and $parser->YYData->{lineno} = $1,
                        $parser->YYData->{filename} = $2,
                        $parser->YYData->{doc} = q{},
                        $parser->YYData->{curr_node} = undef,
                        last;

            s/^[ \r\t\f\013]+//;                            # whitespaces
            s/^\n//
                    and $parser->YYData->{lineno} ++,
                        $parser->YYData->{curr_node} = undef,
                        last;

            s/^#pragma\s+(.*)\n//
                    and _PragmaLexer($parser, $1),
                        $parser->YYData->{lineno} ++,
                        $parser->YYData->{curr_node} = undef,
                        last;

            s/^\/\*\*<//                                    # documentation after
                    and _DocAfterLexer($parser),
                        last;
            s/^\/\*\*//                                     # documentation
                    and _DocLexer($parser),
                        last;
            s/^\/\/\/(.*)\n//                               # single line documentation
                    and _AttachDoc($parser, $1),
                    and $parser->YYData->{lineno} ++,
                        last;

            s/^\/\*//                                       # multiple line comment
                    and _CommentLexer($parser),
                        last;
            s/^\/\/.*\n//                                   # single line comment
                    and $parser->YYData->{lineno} ++,
                        $parser->YYData->{curr_node} = undef,
                        last;

            s/^%\{//                                        # code fragment
                    and return _CodeLexer($parser);

            if ($parser->YYData->{prop}) {
                s/^([A-Za-z][0-9A-Za-z_]*)//
                        and return ('PROP_KEY', $1);

                s/^\(([^\)]+)\)//
                        and return ('PROP_VALUE', $1);
            }

            if ($parser->YYData->{native}) {
                s/^([^\)]+)\)//
                        and return ('NATIVE_TYPE', $1);
            }

            s/^__declspec\s*\(\s*([A-Za-z]*)\s*\)//
                    and return ('DECLSPEC', $1);

            s/^([0-9]+)([Dd])//
                    and $parser->YYData->{lexeme} = $1 . $2,
                        return ('FIXED_PT_LITERAL', new Math::BigFloat($1));
            s/^([0-9]+\.)([Dd])//
                    and $parser->YYData->{lexeme} = $1 . $2,
                        return ('FIXED_PT_LITERAL', new Math::BigFloat($1));
            s/^(\.[0-9]+)([Dd])//
                    and $parser->YYData->{lexeme} = $1 . $2,
                        return ('FIXED_PT_LITERAL', new Math::BigFloat($1));
            s/^([0-9]+\.[0-9]+)([Dd])//
                    and $parser->YYData->{lexeme} = $1 . $2,
                        return ('FIXED_PT_LITERAL', new Math::BigFloat($1));

            s/^([0-9]+\.[0-9]+[Ee][+\-]?[0-9]+)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));
            s/^([0-9]+[Ee][+\-]?[0-9]+)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));
            s/^(\.[0-9]+[Ee][+\-]?[0-9]+)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));
            s/^([0-9]+\.[0-9]+)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));
            s/^([0-9]+\.)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));
            s/^(\.[0-9]+)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('FLOATING_PT_LITERAL', new Math::BigFloat($1));

            s/^0([0-7]+)//
                    and $parser->YYData->{lexeme} = '0' . $1,
                        return _OctInteger($parser, $1);
            s/^(0[Xx])([A-Fa-f0-9]+)//
                    and $parser->YYData->{lexeme} = $1 . $2,
                        return _HexInteger($parser, $2);
            s/^(0)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('INTEGER_LITERAL', new Math::BigInt($1));
            s/^([1-9][0-9]*)//
                    and $parser->YYData->{lexeme} = $1,
                        return ('INTEGER_LITERAL', new Math::BigInt($1));

            s/^\"//
                    and return _StringLexer($parser, 'STRING_LITERAL');

            if ($CORBA::IDL::Parser::IDL_VERSION ge '2.3') {
                s/^L\"//
                        and return _StringLexer($parser, 'WIDE_STRING_LITERAL');
            }
            else {
                s/^L\"//
                        and $parser->Warning("literal 'wstring' is not allowed.\n")
                        and return _StringLexer($parser, 'STRING_LITERAL');
            }

            s/^\'//
                    and return _CharLexer($parser, 'CHARACTER_LITERAL');

            if ($CORBA::IDL::Parser::IDL_VERSION ge '2.3') {
                s/^L\'//
                        and return _CharLexer($parser, 'WIDE_CHARACTER_LITERAL');
            }
            else {
                s/^L\'//
                        and $parser->Warning("literal 'wchar' is not allowed.\n")
                        and return _CharLexer($parser, 'CHARACTER_LITERAL');
            }

            s/^([A-Za-z][0-9A-Za-z_]*)//
                    and return _Identifier($parser, $1);
            s/^_([A-Za-z][0-9A-Za-z_]*)//
                    and return _EscIdentifier($parser, $1);

            s/^(<<)//
                    and return ($1, $1);
            s/^(>>)//
                    and return ($1, $1);
            s/^(::)//
                    and return ($1, $1);
            s/^(\.\.\.)//
                    and return ($1, $1);

            s/^([\+&\/%\*~\|\-\^\(\)\[\]\{\}<>,;:=])//
                    and return ($1, $1);                    # punctuators

            s/^([\S]+)//
                    and $parser->Error("lexer error $1.\n"),
                        last;
        }
    }
}

sub InitLexico {
    my ($parser) = @_;

    # 3.2.4 Keywords
    my %keywords = (
        'ABSTRACT'      => [ 'abstract',    '2.3' ],
        'ANY'           => [ 'any',         '2.0' ],
        'ATTRIBUTE'     => [ 'attribute',   '2.0' ],
        'BOOLEAN'       => [ 'boolean',     '2.0' ],
        'BYTE'          => [ 'byte',        '9.9',  'MIDL/MODL' ],
        'CASE'          => [ 'case',        '2.0' ],
        'CHAR'          => [ 'char',        '2.0' ],
        'COMPONENT'     => [ 'component',   '3.0' ],
        'CONST'         => [ 'const',       '2.0' ],
        'CONSUMES'      => [ 'consumes',    '3.0' ],
        'CONTEXT'       => [ 'context',     '2.0' ],
        'CUSTOM'        => [ 'custom',      '2.3' ],
        'DEFAULT'       => [ 'default',     '2.0' ],
        'DOUBLE'        => [ 'double',      '2.0' ],
        'EMITS'         => [ 'emits',       '3.0' ],
        'ENUM'          => [ 'enum',        '2.0' ],
        'EVENTTYPE'     => [ 'eventtype',   '3.0' ],
        'EXCEPTION'     => [ 'exception',   '2.0' ],
        'FACTORY'       => [ 'factory',     '2.3' ],
        'FALSE'         => [ 'FALSE',       '2.0' ],
        'FINDER'        => [ 'finder',      '3.0' ],
        'FIXED'         => [ 'fixed',       '2.1' ],
        'FLOAT'         => [ 'float',       '2.0' ],
        'GETRAISES'     => [ 'getraises',   '3.0' ],
        'HOME'          => [ 'home',        '3.0' ],
        'IMPORT'        => [ 'import',      '3.0' ],
        'IN'            => [ 'in',          '2.0' ],
        'INOUT'         => [ 'inout',       '2.0' ],
        'INTERFACE'     => [ 'interface',   '2.0' ],
        'LOCAL'         => [ 'local',       '2.4' ],
        'LONG'          => [ 'long',        '2.0' ],
        'MODULE'        => [ 'module',      '2.0' ],
        'MULTIPLE'      => [ 'multiple',    '3.0' ],
        'NATIVE'        => [ 'native',      '2.2' ],
        'OBJECT'        => [ 'Object',      '2.0' ],
        'OCTET'         => [ 'octet',       '2.0' ],
        'ONEWAY'        => [ 'oneway',      '2.0' ],
        'OUT'           => [ 'out',         '2.0' ],
        'PRIMARYKEY'    => [ 'primarykey',  '3.0' ],
        'PRIVATE'       => [ 'private',     '2.3' ],
        'PROVIDES'      => [ 'provides',    '3.0' ],
        'PUBLIC'        => [ 'public',      '2.3' ],
        'PUBLISHES'     => [ 'publishes',   '3.0' ],
        'RAISES'        => [ 'raises',      '2.0' ],
        'READONLY'      => [ 'readonly',    '2.0' ],
        'SEQUENCE'      => [ 'sequence',    '2.0' ],
        'SETRAISES'     => [ 'setraises',   '3.0' ],
        'SHORT'         => [ 'short',       '2.0' ],
        'STRING'        => [ 'string',      '2.0' ],
        'STRUCT'        => [ 'struct',      '2.0' ],
        'SUPPORTS'      => [ 'supports',    '2.3' ],
        'SWITCH'        => [ 'switch',      '2.0' ],
        'TRUE'          => [ 'TRUE',        '2.0' ],
        'TRUNCATABLE'   => [ 'truncatable', '2.3' ],
        'TYPEDEF'       => [ 'typedef',     '2.0' ],
        'TYPEID'        => [ 'typeid',      '3.0' ],
        'TYPEPREFIX'    => [ 'typeprefix',  '3.0' ],
        'UNION'         => [ 'union',       '2.0' ],
        'UNSIGNED'      => [ 'unsigned',    '2.0' ],
        'USES'          => [ 'uses',        '3.0' ],
        'VALUEBASE'     => [ 'ValueBase',   '2.3' ],
        'VALUETYPE'     => [ 'valuetype',   '2.3' ],
        'VOID'          => [ 'void',        '2.0' ],
        'WCHAR'         => [ 'wchar',       '2.1' ],
        'WSTRING'       => [ 'wstring',     '2.1' ]
    );

    $parser->YYData->{keyword} = \%keywords;
}

1;