modules/typogrify/unicode-conversion.php

<?php 
// $Id: unicode-conversion.php,v 1.5 2009/10/20 09:19:03 mikl Exp $

/**
 * Return the unicode conversion maps.
 *
 * @param string $type
 *    The map type we're looking for, one of 'ligature', 'punctuation', 
 *    'arrow' 'nested' or 'all'.
 * @return array
 *    Array of conversions, keyed by the original string.
 */
function unicode_conversion_map($type = 'all') {
  $map = array(
    // See http://www.unicode.org/charts/PDF/UFB00.pdf
    'ligature' => array(
      'ffi' => '&#xfb03;',
      'ffl' => '&#xfb04;',
      'ff'  => '&#xfb00;',
      'fi'  => '&#xfb01;',
      'fl'  => '&#xfb02;',
      'ij'  => '&#x0133;',
      'IJ'  => '&#x0132;',
      'st'  => '&#xfb06;',
      'ss'  => '&szlig;',
    ),
    // See http:#www.unicode.org/charts/PDF/U2000.pdf
    'punctuation' => array(
      '...'   => '&#x2026;',
      '..'    => '&#x2025;',
      '. . .' => '&#x2026;',
      '---'   => '&mdash;',
      '--'    => '&ndash;',
    ),
    // See http:#www.unicode.org/charts/PDF/U2190.pdf
    'arrow' => array(
      '->>' => '&#x21a0;',
      '<<-' => '&#x219e;',
      '->|' => '&#x21e5;',
      '|<-' => '&#x21e4;',
      '<->' => '&#x2194;',
      '->'  => '&#x2192;',
      '<-'  => '&#x2190;',
      '<=>' => '&#x21d4;',
      '=>'  => '&#x21d2;',
      '<='  => '&#x21d0;',
    ),
  );

  if ($type == 'all') {
    return array_merge($map['ligature'], $map['arrow'], $map['punctuation']);
  }
  elseif ($type == 'nested') {
    return $map;
  }
  else {
    return $map[$type];
  }
}

/**
 * Perform character conversion.
 *
 * @param string $test
 *    Text to be parsed.
 * @param array $characters_to_convert
 *    Array of ASCII characters to convert.
 * @return string
 *    The result of the conversion.
 */
function convert_characters($text, $characters_to_convert) {
  if (($characters_to_convert == NULL) || (count($characters_to_convert) < 1)) {
    // do nothing
    return $text;
  }

  // get ascii to unicode mappings
  $unicode_map = unicode_conversion_map();
  
  foreach ($characters_to_convert as $ascii_string) {
    $unicode_strings[] = $unicode_map[$ascii_string];
  }
  
  $tokens = _TokenizeHTML($text);
  $result = '';
  $in_pre = 0;  // Keep track of when we're inside <pre> or <code> tags
  foreach ($tokens as $cur_token) {
    if ($cur_token[0] == "tag") {
      // Don't mess with text inside tags, <pre> blocks, or <code> blocks
      $result .= $cur_token[1];
      // Get the tags to skip regex from SmartyPants
      if (preg_match(SMARTYPANTS_TAGS_TO_SKIP, $cur_token[1], $matches)) {
        $in_pre = isset($matches[1]) && $matches[1] == '/' ? 0 : 1;
      }
    } else {
      $t = $cur_token[1];
      if ($in_pre == 0) {
        $t = ProcessEscapes($t);
        $t = str_replace($characters_to_convert, $unicode_strings, $t);
      }
      $result .= $t;
    }
  }
  return $result;
}


// _TokenizeHTML is shared between PHP SmartyPants and PHP Markdown.
// We're borrowing it for Typogrify.module, too
// We only define it if it is not already defined.
if (!function_exists('_TokenizeHTML')) {
	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();
	
		$match = '(?s:<!(?:--.*?--\s*)+>)|'.	# comment
				 '(?s:<\?.*?\?>)|'.				# processing instruction
												# regular tags
				 '(?:<[/!$]?[-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;
	}
}

1	<?php
2	// $Id: unicode-conversion.php,v 1.5 2009/10/20 09:19:03 mikl Exp $
3
4	/**
5	* Return the unicode conversion maps.
6	*
7	* @param string $type
8	* The map type we're looking for, one of 'ligature', 'punctuation',
9	* 'arrow' 'nested' or 'all'.
10	* @return array
11	* Array of conversions, keyed by the original string.
12	*/
13	function unicode_conversion_map($type = 'all') {
14	$map = array(
15	// See http://www.unicode.org/charts/PDF/UFB00.pdf
16	'ligature' => array(
17	'ffi' => 'ﬃ',
18	'ffl' => 'ﬄ',
19	'ff' => 'ﬀ',
20	'fi' => 'ﬁ',
21	'fl' => 'ﬂ',
22	'ij' => 'ĳ',
23	'IJ' => 'Ĳ',
24	'st' => 'ﬆ',
25	'ss' => 'ß',
26	),
27	// See http:#www.unicode.org/charts/PDF/U2000.pdf
28	'punctuation' => array(
29	'...' => '…',
30	'..' => '‥',
31	'. . .' => '…',
32	'---' => '—',
33	'--' => '–',
34	),
35	// See http:#www.unicode.org/charts/PDF/U2190.pdf
36	'arrow' => array(
37	'->>' => '↠',
38	'<<-' => '↞',
39	'->\|' => '⇥',
40	'\|<-' => '⇤',
41	'<->' => '↔',
42	'->' => '→',
43	'<-' => '←',
44	'<=>' => '⇔',
45	'=>' => '⇒',
46	'<=' => '⇐',
47	),
48	);
49
50	if ($type == 'all') {
51	return array_merge($map['ligature'], $map['arrow'], $map['punctuation']);
52	}
53	elseif ($type == 'nested') {
54	return $map;
55	}
56	else {
57	return $map[$type];
58	}
59	}
60
61	/**
62	* Perform character conversion.
63	*
64	* @param string $test
65	* Text to be parsed.
66	* @param array $characters_to_convert
67	* Array of ASCII characters to convert.
68	* @return string
69	* The result of the conversion.
70	*/
71	function convert_characters($text, $characters_to_convert) {
72	if (($characters_to_convert == NULL) \|\| (count($characters_to_convert) < 1)) {
73	// do nothing
74	return $text;
75	}
76
77	// get ascii to unicode mappings
78	$unicode_map = unicode_conversion_map();
79
80	foreach ($characters_to_convert as $ascii_string) {
81	$unicode_strings[] = $unicode_map[$ascii_string];
82	}
83
84	$tokens = _TokenizeHTML($text);
85	$result = '';
86	$in_pre = 0; // Keep track of when we're inside <pre> or <code> tags
87	foreach ($tokens as $cur_token) {
88	if ($cur_token[0] == "tag") {
89	// Don't mess with text inside tags, <pre> blocks, or <code> blocks
90	$result .= $cur_token[1];
91	// Get the tags to skip regex from SmartyPants
92	if (preg_match(SMARTYPANTS_TAGS_TO_SKIP, $cur_token[1], $matches)) {
93	$in_pre = isset($matches[1]) && $matches[1] == '/' ? 0 : 1;
94	}
95	} else {
96	$t = $cur_token[1];
97	if ($in_pre == 0) {
98	$t = ProcessEscapes($t);
99	$t = str_replace($characters_to_convert, $unicode_strings, $t);
100	}
101	$result .= $t;
102	}
103	}
104	return $result;
105	}
106
107
108	// _TokenizeHTML is shared between PHP SmartyPants and PHP Markdown.
109	// We're borrowing it for Typogrify.module, too
110	// We only define it if it is not already defined.
111	if (!function_exists('_TokenizeHTML')) {
112	function _TokenizeHTML($str) {
113	//
114	// Parameter: String containing HTML markup.
115	// Returns: An array of the tokens comprising the input
116	// string. Each token is either a tag (possibly with nested,
117	// tags contained therein, such as <a href="<MTFoo>">, or a
118	// run of text between tags. Each element of the array is a
119	// two-element array; the first is either 'tag' or 'text';
120	// the second is the actual value.
121	//
122	//
123	// Regular expression derived from the _tokenize() subroutine in
124	// Brad Choate's MTRegex plugin.
125	// <http://www.bradchoate.com/past/mtregex.php>
126	//
127	$index = 0;
128	$tokens = array();
129
130	$match = '(?s:<!(?:--.?--\s)+>)\|'. # comment
131	'(?s:<\?.*?\?>)\|'. # processing instruction
132	# regular tags
133	'(?:<[/!$]?[-a-zA-Z0-9:]+\b(?>[^"\'>]+\|"[^"]"\|\'[^\']\')*>)';
134
135	$parts = preg_split("{($match)}", $str, -1, PREG_SPLIT_DELIM_CAPTURE);
136
137	foreach ($parts as $part) {
138	if (++$index % 2 && $part != '')
139	$tokens[] = array('text', $part);
140	else
141	$tokens[] = array('tag', $part);
142	}
143	return $tokens;
144	}
145	}
146