<?php
#
# SmartyPants - Smart typography for web sites
#
# PHP SmartyPants
# Copyright (c) 2004-2016 Michel Fortin
# <https://michelf.ca/>
#
# Original SmartyPants
# Copyright (c) 2003-2004 John Gruber
# <https://daringfireball.net/>
#
namespace Michelf;
#
# SmartyPants Parser Class
#
class SmartyPants {
### Version ###
const SMARTYPANTSLIB_VERSION = "1.6.0";
### Presets
# SmartyPants does nothing at all
const ATTR_DO_NOTHING = 0;
# "--" for em-dashes; no en-dash support
const ATTR_EM_DASH = 1;
# "---" for em-dashes; "--" for en-dashes
const ATTR_LONG_EM_DASH_SHORT_EN = 2;
# "--" for em-dashes; "---" for en-dashes
const ATTR_SHORT_EM_DASH_LONG_EN = 3;
# "--" for em-dashes; "---" for en-dashes
const ATTR_STUPEFY = -1;
# The default preset: ATTR_EM_DASH
const ATTR_DEFAULT = SmartyPants::ATTR_EM_DASH;
### Standard Function Interface ###
public static function defaultTransform($text, $attr = SmartyPants::ATTR_DEFAULT) {
#
# Initialize the parser and return the result of its transform method.
# This will work fine for derived classes too.
#
# Take parser class on which this function was called.
$parser_class = \get_called_class();
# try to take parser from the static parser list
static $parser_list;
$parser =& $parser_list[$parser_class][$attr];
# create the parser if not already set
if (!$parser)
$parser = new $parser_class($attr);
# Transform text using parser.
return $parser->transform($text);
}
### Configuration Variables ###
# Partial regex for matching tags to skip
public $tags_to_skip = 'pre|code|kbd|script|style|math';
# Options to specify which transformations to make:
protected $do_nothing = 0; # disable all transforms
protected $do_quotes = 0;
protected $do_backticks = 0; # 1 => double only, 2 => double & single
protected $do_dashes = 0; # 1, 2, or 3 for the three modes described above
protected $do_ellipses = 0;
protected $do_stupefy = 0;
protected $convert_quot = 0; # should we translate " entities into normal quotes?
### Parser Implementation ###
public function __construct($attr = SmartyPants::ATTR_DEFAULT) {
#
# Initialize a parser with certain attributes.
#
# Parser 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 == "0") {
$this->do_nothing = 1;
}
else if ($attr == "1") {
# Do everything, turn all options on.
$this->do_quotes = 1;
$this->do_backticks = 1;
$this->do_dashes = 1;
$this->do_ellipses = 1;
}
else if ($attr == "2") {
# Do everything, turn all options on, use old school dash shorthand.
$this->do_quotes = 1;
$this->do_backticks = 1;
$this->do_dashes = 2;
$this->do_ellipses = 1;
}
else if ($attr == "3") {
# Do everything, turn all options on, use inverted old school dash shorthand.
$this->do_quotes = 1;
$this->do_backticks = 1;
$this->do_dashes = 3;
$this->do_ellipses = 1;
}
else if ($attr == "-1") {
# Special "stupefy" mode.
$this->do_stupefy = 1;
}
else {
$chars = preg_split('//', $attr);
foreach ($chars as $c){
if ($c == "q") { $this->do_quotes = 1; }
else if ($c == "b") { $this->do_backticks = 1; }
else if ($c == "B") { $this->do_backticks = 2; }
else if ($c == "d") { $this->do_dashes = 1; }
else if ($c == "D") { $this->do_dashes = 2; }
else if ($c == "i") { $this->do_dashes = 3; }
else if ($c == "e") { $this->do_ellipses = 1; }
else if ($c == "w") { $this->convert_quot = 1; }
else {
# Unknown attribute option, ignore.
}
}
}
}
public function transform($text) {
if ($this->do_nothing) {
return $text;
}
$tokens = $this->tokenizeHTML($text);
$result = '';
$in_pre = 0; # Keep track of when we're inside <pre> or <code> tags.
$prev_token_last_char = ""; # This is a cheat, used to get some context
# for one-character tokens that consist of
# just a quote char. What we do is remember
# the last character of the previous text
# token, to use as context to curl single-
# character quote tokens correctly.
foreach ($tokens as $cur_token) {
if ($cur_token[0] == "tag") {
# Don't mess with quotes inside tags.
$result .= $cur_token[1];
if (preg_match('@<(/?)(?:'.$this->tags_to_skip.')[\s>]@', $cur_token[1], $matches)) {
$in_pre = isset($matches[1]) && $matches[1] == '/' ? 0 : 1;
}
} else {
$t = $cur_token[1];
$last_char = substr($t, -1); # Remember last char of this token before processing.
if (! $in_pre) {
$t = $this->educate($t, $prev_token_last_char);
}
$prev_token_last_char = $last_char;
$result .= $t;
}
}
return $result;
}
protected function educate($t, $prev_token_last_char) {
$t = $this->processEscapes($t);
if ($this->convert_quot) {
$t = preg_replace('/"/', '"', $t);
}
if ($this->do_dashes) {
if ($this->do_dashes == 1) $t = $this->educateDashes($t);
if ($this->do_dashes == 2) $t = $this->educateDashesOldSchool($t);
if ($this->do_dashes == 3) $t = $this->educateDashesOldSchoolInverted($t);
}
if ($this->do_ellipses) $t = $this->educateEllipses($t);
# Note: backticks need to be processed before quotes.
if ($this->do_backticks) {
$t = $this->educateBackticks($t);
if ($this->do_backticks == 2) $t = $this->educateSingleBackticks($t);
}
if ($this->do_quotes) {
if ($t == "'") {
# Special case: single-character ' token
if (preg_match('/\S/', $prev_token_last_char)) {
$t = "’";
}
else {
$t = "‘";
}
}
else if ($t == '"') {
# Special case: single-character " token
if (preg_match('/\S/', $prev_token_last_char)) {
$t = "”";
}
else {
$t = "“";
}
}
else {
# Normal case:
$t = $this->educateQuotes($t);
}
}
if ($this->do_stupefy) $t = $this->stupefyEntities($t);
return $t;
}
protected function educateQuotes($_) {
#
# Parameter: String.
#
# Returns: The string, with "educated" curly quote HTML entities.
#
# Example input: "Isn't this fun?"
# Example output: “Isn’t this fun?”
#
# Make our own "punctuation" character class, because the POSIX-style
# [:PUNCT:] is only available in Perl 5.6 or later:
$punct_class = "[!\"#\\$\\%'()*+,-.\\/:;<=>?\\@\\[\\\\\]\\^_`{|}~]";
# Special case if the very first character is a quote
# followed by punctuation at a non-word-break. Close the quotes by brute force:
$_ = preg_replace(
array("/^'(?=$punct_class\\B)/", "/^\"(?=$punct_class\\B)/"),
array('’', '”'), $_);
# Special case for double sets of quotes, e.g.:
# <p>He said, "'Quoted' words in a larger quote."</p>
$_ = preg_replace(
array("/\"'(?=\w)/", "/'\"(?=\w)/"),
array('“‘', '‘“'), $_);
# Special case for decade abbreviations (the '80s):
$_ = preg_replace("/'(?=\\d{2}s)/", '’', $_);
$close_class = '[^\ \r\n\[\{\(\-]';
$dec_dashes = '&\#8211;|&\#8212;';
# Get most opening single quotes:
$_ = preg_replace("{
(
\\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
$dec_dashes | # or decimal entities
&\\#x201[34]; # or hex
)
' # the quote
(?=\\w) # followed by a word character
}x", '\1‘', $_);
# Single closing quotes:
$_ = preg_replace("{
($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:
# \"<i>Custer</i>'s Last Stand.\"
}xi", '\1’', $_);
# Any remaining single quotes should be opening ones:
$_ = str_replace("'", '‘', $_);
# Get most opening double quotes:
$_ = preg_replace("{
(
\\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
$dec_dashes | # or decimal entities
&\\#x201[34]; # or hex
)
\" # the quote
(?=\\w) # followed by a word character
}x", '\1“', $_);
# Double closing quotes:
$_ = preg_replace("{
($close_class)?
\"
(?(1)|(?=\\s)) # If $1 captured, then do nothing;
# if not, then make sure the next char is whitespace.
}x", '\1”', $_);
# Any remaining quotes should be opening ones.
$_ = str_replace('"', '“', $_);
return $_;
}
protected function educateBackticks($_) {
#
# 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?”
#
$_ = str_replace(array("``", "''",),
array('“', '”'), $_);
return $_;
}
protected function educateSingleBackticks($_) {
#
# 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?’
#
$_ = str_replace(array("`", "'",),
array('‘', '’'), $_);
return $_;
}
protected function educateDashes($_) {
#
# Parameter: String.
#
# Returns: The string, with each instance of "--" translated to
# an em-dash HTML entity.
#
$_ = str_replace('--', '—', $_);
return $_;
}
protected function educateDashesOldSchool($_) {
#
# 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.
#
# em en
$_ = str_replace(array("---", "--",),
array('—', '–'), $_);
return $_;
}
protected function educateDashesOldSchoolInverted($_) {
#
# 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.)
#
# en em
$_ = str_replace(array("---", "--",),
array('–', '—'), $_);
return $_;
}
protected function educateEllipses($_) {
#
# Parameter: String.
# Returns: The string, with each instance of "..." translated to
# an ellipsis HTML entity. Also converts the case where
# there are spaces between the dots.
#
# Example input: Huh...?
# Example output: Huh…?
#
$_ = str_replace(array("...", ". . .",), '…', $_);
return $_;
}
protected function stupefyEntities($_) {
#
# Parameter: String.
# Returns: The string, with each SmartyPants HTML entity translated to
# its ASCII counterpart.
#
# Example input: “Hello — world.”
# Example output: "Hello -- world."
#
# en-dash em-dash
$_ = str_replace(array('–', '—'),
array('-', '--'), $_);
# single quote open close
$_ = str_replace(array('‘', '’'), "'", $_);
# double quote open close
$_ = str_replace(array('“', '”'), '"', $_);
$_ = str_replace('…', '...', $_); # ellipsis
return $_;
}
protected function processEscapes($_) {
#
# 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
# ------ -----
# \\ \
# \" "
# \' '
# \. .
# \- -
# \` `
#
$_ = str_replace(
array('\\\\', '\"', "\'", '\.', '\-', '\`'),
array('\', '"', ''', '.', '-', '`'), $_);
return $_;
}
protected function tokenizeHTML($str) {
#
# Parameter: String containing HTML markup.
# Returns: An array of the tokens comprising the input
# string. Each token is either a tag (possibly with nested,
# tags contained therein, such as <a href="<MTFoo>">, 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.
#
#
# Regular expression derived from the _tokenize() subroutine in
# Brad Choate's MTRegex plugin.
# <http://www.bradchoate.com/past/mtregex.php>
#
$index = 0;
$tokens = array();
$match = '(?s:<!--.*?-->)|'. # comment
'(?s:<\?.*?\?>)|'. # processing instruction
# regular tags
'(?:<[/!$]?[-a-zA-Z0-9:]+\b(?>[^"\'>]+|"[^"]*"|\'[^\']*\')*>)';
$parts = preg_split("{($match)}", $str, -1, PREG_SPLIT_DELIM_CAPTURE);
foreach ($parts as $part) {
if (++$index % 2 && $part != '')
$tokens[] = array('text', $part);
else
$tokens[] = array('tag', $part);
}
return $tokens;
}
}