package Text::SmartyPants; use strict; use vars qw($VERSION); $VERSION = "1.3"; # Configurable variables: my $smartypants_attr = "1"; # 1 => "--" for em-dashes; no en-dash support # 2 => "---" for em-dashes; "--" for en-dashes # 3 => "--" for em-dashes; "---" for en-dashes # See docs for more configuration options. # Globals: my $tags_to_skip = qr!<(/?)(?:pre|code|kbd|script)[\s>]!; sub process { shift if ( $_[0] eq __PACKAGE__ ); # oops, called in OOP fashion. # Paramaters: my $text = shift; # text to be parsed # value of the smart_quotes="" attribute. Default to 'everything on' my $attr = shift || '1'; # Options to specify which transformations to make: my ( $do_quotes, $do_backticks, $do_dashes, $do_ellipses, $do_stupefy ); # should we translate " entities into normal quotes? my $convert_quot = 0; # Parse attributes: # 0 : do nothing # 1 : set all # 2 : set all, using old school en- and em- dash shortcuts # 3 : set all, using inverted old school en and em- dash shortcuts # # q : quotes # b : backtick quotes (``double'' only) # B : backtick quotes (``double'' and `single') # d : dashes # D : old school dashes # i : inverted old school dashes # e : ellipses # w : convert " entities to " for Dreamweaver users if ( $attr eq "0" ) { # Do nothing. return $text; } elsif ( $attr eq "1" ) { # Do everything, turn all options on. $do_quotes = 1; $do_backticks = 1; $do_dashes = 1; $do_ellipses = 1; } elsif ( $attr eq "2" ) { # Do everything, turn all options on, use old school dash shorthand. $do_quotes = 1; $do_backticks = 1; $do_dashes = 2; $do_ellipses = 1; } elsif ( $attr eq "3" ) { # Do everything, turn all options on, use inverted old school dash shorthand. $do_quotes = 1; $do_backticks = 1; $do_dashes = 3; $do_ellipses = 1; } elsif ( $attr eq "-1" ) { # Special "stupefy" mode. $do_stupefy = 1; } else { my @chars = split( //, $attr ); foreach my $c (@chars) { if ( $c eq "q" ) { $do_quotes = 1; } elsif ( $c eq "b" ) { $do_backticks = 1; } elsif ( $c eq "B" ) { $do_backticks = 2; } elsif ( $c eq "d" ) { $do_dashes = 1; } elsif ( $c eq "D" ) { $do_dashes = 2; } elsif ( $c eq "i" ) { $do_dashes = 3; } elsif ( $c eq "e" ) { $do_ellipses = 1; } elsif ( $c eq "w" ) { $convert_quot = 1; } else { # Unknown attribute option, ignore. } } } my $tokens ||= _tokenize($text); my $result = ''; my $in_pre = 0; # Keep track of when we're inside
 or  tags.

    my $prev_token_last_char = "";    # This is a cheat, used to get some context
                                      # for one-character tokens that consist of
                                      # just a quote char. What we do is remember
                                      # the last character of the previous text
                                      # token, to use as context to curl single-
                                      # character quote tokens correctly.

    foreach my $cur_token (@$tokens) {
        if ( $cur_token->[0] eq "tag" ) {

            # Don't mess with quotes inside tags.
            $result .= $cur_token->[1];
            if ( $cur_token->[1] =~ m/$tags_to_skip/ ) {
                $in_pre = defined $1 && $1 eq '/' ? 0 : 1;
            }
        }
        else {
            my $t = $cur_token->[1];
            my $last_char =
                substr( $t, -1 );    # Remember last char of this token before processing.
            if ( !$in_pre ) {
                $t = ProcessEscapes($t);

                if ($convert_quot) {
                    $t =~ s/"/"/g;
                }

                if ($do_dashes) {
                    $t = EducateDashes($t)                  if ( $do_dashes == 1 );
                    $t = EducateDashesOldSchool($t)         if ( $do_dashes == 2 );
                    $t = EducateDashesOldSchoolInverted($t) if ( $do_dashes == 3 );
                }

                $t = EducateEllipses($t) if $do_ellipses;

                # Notes: backticks need to be processed before quotes.
                if ($do_backticks) {
                    $t = EducateBackticks($t);
                    $t = EducateSingleBackticks($t) if ( $do_backticks == 2 );
                }

                if ($do_quotes) {
                    if ( $t eq q/'/ ) {

                        # Special case: single-character ' token
                        if ( $prev_token_last_char =~ m/\S/ ) {
                            $t = "’";
                        }
                        else {
                            $t = "‘";
                        }
                    }
                    elsif ( $t eq q/"/ ) {

                        # Special case: single-character " token
                        if ( $prev_token_last_char =~ m/\S/ ) {
                            $t = "”";
                        }
                        else {
                            $t = "“";
                        }
                    }
                    else {

                        # Normal case:
                        $t = EducateQuotes($t);
                    }
                }

                $t = StupefyEntities($t) if $do_stupefy;
            }
            $prev_token_last_char = $last_char;
            $result .= $t;
        }
    }

    return $result;
}

sub SmartQuotes {

    # Paramaters:
    my $text = shift;    # text to be parsed
    my $attr = shift;    # value of the smart_quotes="" attribute

    my $do_backticks;    # should we educate ``backticks'' -style quotes?

    if ( $attr == 0 ) {

        # do nothing;
        return $text;
    }
    elsif ( $attr == 2 ) {

        # smarten ``backticks'' -style quotes
        $do_backticks = 1;
    }
    else {
        $do_backticks = 0;
    }

    # Special case to handle quotes at the very end of $text when preceded by
    # an HTML tag. Add a space to give the quote education algorithm a bit of
    # context, so that it can guess correctly that it's a closing quote:
    my $add_extra_space = 0;
    if ( $text =~ m/>['"]\z/ ) {
        $add_extra_space = 1;    # Remember, so we can trim the extra space later.
        $text .= " ";
    }

    my $tokens ||= _tokenize($text);
    my $result = '';
    my $in_pre = 0;              # Keep track of when we're inside 
 or  tags

    my $prev_token_last_char = "";    # This is a cheat, used to get some context
                                      # for one-character tokens that consist of
                                      # just a quote char. What we do is remember
                                      # the last character of the previous text
                                      # token, to use as context to curl single-
                                      # character quote tokens correctly.

    foreach my $cur_token (@$tokens) {
        if ( $cur_token->[0] eq "tag" ) {

            # Don't mess with quotes inside tags
            $result .= $cur_token->[1];
            if ( $cur_token->[1] =~ m/$tags_to_skip/ ) {
                $in_pre = defined $1 && $1 eq '/' ? 0 : 1;
            }
        }
        else {
            my $t = $cur_token->[1];
            my $last_char =
                substr( $t, -1 );    # Remember last char of this token before processing.
            if ( !$in_pre ) {
                $t = ProcessEscapes($t);
                if ($do_backticks) {
                    $t = EducateBackticks($t);
                }

                if ( $t eq q/'/ ) {

                    # Special case: single-character ' token
                    if ( $prev_token_last_char =~ m/\S/ ) {
                        $t = "’";
                    }
                    else {
                        $t = "‘";
                    }
                }
                elsif ( $t eq q/"/ ) {

                    # Special case: single-character " token
                    if ( $prev_token_last_char =~ m/\S/ ) {
                        $t = "”";
                    }
                    else {
                        $t = "“";
                    }
                }
                else {

                    # Normal case:
                    $t = EducateQuotes($t);
                }

            }
            $prev_token_last_char = $last_char;
            $result .= $t;
        }
    }

    if ($add_extra_space) {
        $result =~ s/ \z//;    # Trim trailing space if we added one earlier.
    }
    return $result;
}

sub SmartDashes {

    # Paramaters:
    my $text = shift;          # text to be parsed
    my $attr = shift;          # value of the smart_dashes="" attribute

    # reference to the subroutine to use for dash education, default to EducateDashes:
    my $dash_sub_ref = \&EducateDashes;

    if ( $attr == 0 ) {

        # do nothing;
        return $text;
    }
    elsif ( $attr == 2 ) {

        # use old smart dash shortcuts, "--" for en, "---" for em
        $dash_sub_ref = \&EducateDashesOldSchool;
    }
    elsif ( $attr == 3 ) {

        # inverse of 2, "--" for em, "---" for en
        $dash_sub_ref = \&EducateDashesOldSchoolInverted;
    }

    my $tokens;
    $tokens ||= _tokenize($text);

    my $result = '';
    my $in_pre = 0;    # Keep track of when we're inside 
 or  tags
    foreach my $cur_token (@$tokens) {
        if ( $cur_token->[0] eq "tag" ) {

            # Don't mess with quotes inside tags
            $result .= $cur_token->[1];
            if ( $cur_token->[1] =~ m/$tags_to_skip/ ) {
                $in_pre = defined $1 && $1 eq '/' ? 0 : 1;
            }
        }
        else {
            my $t = $cur_token->[1];
            if ( !$in_pre ) {
                $t = ProcessEscapes($t);
                $t = $dash_sub_ref->($t);
            }
            $result .= $t;
        }
    }
    return $result;
}

sub SmartEllipses {

    # Paramaters:
    my $text = shift;    # text to be parsed
    my $attr = shift;    # value of the smart_ellipses="" attribute

    if ( $attr == 0 ) {

        # do nothing;
        return $text;
    }

    my $tokens;
    $tokens ||= _tokenize($text);

    my $result = '';
    my $in_pre = 0;      # Keep track of when we're inside 
 or  tags
    foreach my $cur_token (@$tokens) {
        if ( $cur_token->[0] eq "tag" ) {

            # Don't mess with quotes inside tags
            $result .= $cur_token->[1];
            if ( $cur_token->[1] =~ m/$tags_to_skip/ ) {
                $in_pre = defined $1 && $1 eq '/' ? 0 : 1;
            }
        }
        else {
            my $t = $cur_token->[1];
            if ( !$in_pre ) {
                $t = ProcessEscapes($t);
                $t = EducateEllipses($t);
            }
            $result .= $t;
        }
    }
    return $result;
}

sub EducateQuotes {

    #
    #   Parameter:  String.
    #
    #   Returns:    The string, with "educated" curly quote HTML entities.
    #
    #   Example input:  "Isn't this fun?"
    #   Example output: “Isn’t this fun?”
    #

    local $_ = shift;

    # Tell perl not to gripe when we use $1 in substitutions,
    # even when it's undefined. Use $^W instead of "no warnings"
    # for compatibility with Perl 5.005:
    local $^W = 0;

    # Make our own "punctuation" character class, because the POSIX-style
    # [:PUNCT:] is only available in Perl 5.6 or later:
    my $punct_class = qr/[!"#\$\%'()*+,-.\/:;<=>?\@\[\\\]\^_`{|}~]/;

    # Special case if the very first character is a quote
    # followed by punctuation at a non-word-break. Close the quotes by brute force:
    s/^'(?=$punct_class\B)/’/;
    s/^"(?=$punct_class\B)/”/;

    # Special case for double sets of quotes, e.g.:
    #   

He said, "'Quoted' words in a larger quote."

s/"'(?=\w)/“‘/g; s/'"(?=\w)/‘“/g; my $close_class = qr![^\ \t\r\n\[\{\(]!; # Single closing quotes: s { ($close_class)? ' (?(1)| # If $1 captured, then do nothing; (?=\s | s\b) # otherwise, positive lookahead for a whitespace ) # char or an 's' at a word ending position. This # is a special case to handle something like: # "Custer's Last Stand." } {$1’}xgi; # Single opening quotes: s/'/‘/g; # Double closing quotes: s { ($close_class)? " (?(1)|(?=\s)) # If $1 captured, then do nothing; # if not, then make sure the next char is whitespace. } {$1”}xg; # Double opening quotes: s/"/“/g; return $_; } sub EducateBackticks { # # Parameter: String. # Returns: The string, with ``backticks'' -style double quotes # translated into HTML curly quote entities. # # Example input: ``Isn't this fun?'' # Example output: “Isn't this fun?” # local $_ = shift; s/``/“/g; s/''/”/g; return $_; } sub EducateSingleBackticks { # # Parameter: String. # Returns: The string, with `backticks' -style single quotes # translated into HTML curly quote entities. # # Example input: `Isn't this fun?' # Example output: ‘Isn’t this fun?’ # local $_ = shift; s/`/‘/g; s/'/’/g; return $_; } sub EducateDashes { # # Parameter: String. # # Returns: The string, with each instance of "--" translated to # an em-dash HTML entity. # local $_ = shift; s/--/—/g; return $_; } sub EducateDashesOldSchool { # # Parameter: String. # # Returns: The string, with each instance of "--" translated to # an en-dash HTML entity, and each "---" translated to # an em-dash HTML entity. # local $_ = shift; s/---/—/g; # em s/--/–/g; # en return $_; } sub EducateDashesOldSchoolInverted { # # Parameter: String. # # Returns: The string, with each instance of "--" translated to # an em-dash HTML entity, and each "---" translated to # an en-dash HTML entity. Two reasons why: First, unlike the # en- and em-dash syntax supported by # EducateDashesOldSchool(), it's compatible with existing # entries written before SmartyPants 1.1, back when "--" was # only used for em-dashes. Second, em-dashes are more # common than en-dashes, and so it sort of makes sense that # the shortcut should be shorter to type. (Thanks to Aaron # Swartz for the idea.) # local $_ = shift; s/---/–/g; # en s/--/—/g; # em return $_; } sub EducateEllipses { # # Parameter: String. # Returns: The string, with each instance of "..." translated to # an ellipsis HTML entity. # # Example input: Huh...? # Example output: Huh…? # local $_ = shift; s/\.\.\./…/g; return $_; } sub StupefyEntities { # # Parameter: String. # Returns: The string, with each SmartyPants HTML entity translated to # its ASCII counterpart. # # Example input: “Hello — world.” # Example output: "Hello -- world." # local $_ = shift; s/–/-/g; # en-dash s/—/--/g; # em-dash s/‘/'/g; # open single quote s/’/'/g; # close single quote s/“/"/g; # open double quote s/”/"/g; # close double quote s/…/.../g; # ellipsis return $_; } sub SmartyPantsVersion { return $VERSION; } sub ProcessEscapes { # # Parameter: String. # Returns: The string, with after processing the following backslash # escape sequences. This is useful if you want to force a "dumb" # quote or other character to appear. # # Escape Value # ------ ----- # \\ \ # \" " # \' ' # \. . # \- - # \` ` # local $_ = shift; s! \\\\ !\!gx; s! \\" !"!gx; s! \\' !'!gx; s! \\\. !.!gx; s! \\- !-!gx; s! \\` !`!gx; return $_; } sub _tokenize { # # Parameter: String containing HTML markup. # Returns: Reference to an array of the tokens comprising the input # string. Each token is either a tag (possibly with nested, # tags contained therein, such as , or a # run of text between tags. Each element of the array is a # two-element array; the first is either 'tag' or 'text'; # the second is the actual value. # # # Based on the _tokenize() subroutine from Brad Choate's MTRegex plugin. # # my $str = shift; my $pos = 0; my $len = length $str; my @tokens; # pattern to match balanced nested <> pairs, up to two levels deep: my $nested_angles = qr/<(?:[^<>]|<[^<>]*>)*>/; while ( $str =~ m/($nested_angles)/gs ) { my $whole_tag = $1; my $sec_start = pos $str; my $tag_start = $sec_start - length $whole_tag; if ( $pos < $tag_start ) { push @tokens, [ 'text', substr( $str, $pos, $tag_start - $pos ) ]; } push @tokens, [ 'tag', $whole_tag ]; $pos = pos $str; } push @tokens, [ 'text', substr( $str, $pos, $len - $pos ) ] if $pos < $len; \@tokens; } 1; __END__ =pod =head1 Name B =head1 Synopsis SmartyPants is a free web publishing plug-in for Movable Type, Blosxom, and BBEdit that easily translates plain ASCII punctuation characters into "smart" typographic punctuation HTML entities. =head1 Description SmartyPants can perform the following transformations: =over 4 =item * Straight quotes ( " and ' ) into "curly" quote HTML entities =item * Backticks-style quotes (``like this'') into "curly" quote HTML entities =item * Dashes (C<--> and C<--->) into en- and em-dash entities =item * Three consecutive dots (C<...>) into an ellipsis entity =back This means you can write, edit, and save your posts using plain old ASCII straight quotes, plain dashes, and plain dots, but your published posts (and final HTML output) will appear with smart quotes, em-dashes, and proper ellipses. SmartyPants is a combination plug-in -- the same file works with Movable Type, Blosxom, and BBEdit. It can also be used from a Unix-style command-line. Version requirements and installation instructions for each of these tools can be found in the corresponding sub-section under "Installation", below. SmartyPants does not modify characters within C<<
 >>, C<< 
>>, C<<  >>, or C<<