/**
* SmartyPants.
*
* @param string $text
* Text to be parsed.
* @param string $attr
* Value of the smart_quotes="" attribute.
* @param string $ctx
* MT context object (unused).
*/
function SmartyPants($text, $attr = NULL, $ctx = NULL)
{
if ($attr == NULL) {
global $_typogrify_smartypants_attr;
$attr = $_typogrify_smartypants_attr;
}
// Options to specify which transformations to make.
$do_stupefy = FALSE;
// Should we translate " entities into normal quotes?
$convert_quot = 0;
// Parse attributes:
// 0 : do nothing
// 1 : set all
// 2 : set all, using old school en- and em- dash shortcuts
// 3 : set all, using inverted old school en and em- dash shortcuts
//
// q : quotes
// b : backtick quotes (``double'' and ,,double`` only)
// B : backtick quotes (``double'', ,,double``, ,single` and `single')
// d : dashes
// D : old school dashes
// i : inverted old school dashes
// e : ellipses
// w : convert " entities to " for Dreamweaver users.
if ($attr == "0") {
// Do nothing.
return $text;
} elseif ($attr == "1") {
// Do everything, turn all options on.
$do_quotes = 2;
$do_backticks = 1;
$do_dashes = 1;
$do_ellipses = 1;
} elseif ($attr == "2") {
// Do everything, turn all options on, use old school dash shorthand.
$do_quotes = 2;
$do_backticks = 1;
$do_dashes = 2;
$do_ellipses = 1;
} elseif ($attr == "3") {
// Do everything, turn all options on,
// use inverted old school dash shorthand.
$do_quotes = 2;
$do_backticks = 1;
$do_dashes = 3;
$do_ellipses = 1;
} elseif ($attr == "-1") {
// Special "stupefy" mode.
$do_stupefy = 1;
} else {
$chars = preg_split('//', $attr);
foreach ($chars as $c) {
if ($c == "q") {
$do_quotes = 1;
} elseif ($c == "Q") {
$do_quotes = 2;
} elseif ($c == "b") {
$do_backticks = 1;
} elseif ($c == "B") {
$do_backticks = 2;
} elseif ($c == "d") {
$do_dashes = 1;
} elseif ($c == "D") {
$do_dashes = 2;
} elseif ($c == "i") {
$do_dashes = 3;
} elseif ($c == "e") {
$do_ellipses = 1;
} elseif ($c == "w") {
$convert_quot = 1;
}
}
}
if ($do_quotes == 2) {
$doc_lang = $ctx['langcode'];
} else {
$doc_lang = 'en';
}
$tokens = _TokenizeHTML($text);
$result = '';
// Keep track of when we're inside <pre> or <code> tags.
$in_pre = 0;
// 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.
$prev_token_last_char = '';
foreach ($tokens as $cur_token) {
if ($cur_token[0] == 'tag') {
// Don't mess with quotes inside tags.
$result .= $cur_token[1];
if (preg_match(SMARTYPANTS_TAGS_TO_SKIP, $cur_token[1], $matches)) {
$in_pre = isset($matches[1]) && $matches[1] == '/' ? 0 : 1;
} else {
// Reading language from span.
if (preg_match('/<span .*(xml:)?lang="(..)"/', $cur_token[1], $matches)) {
$span_lang = $matches[2];
} elseif ($cur_token[1] == '</span>') {
unset($span_lang);
}
}
} else {
$t = $cur_token[1];
// Remember last char of this token before processing.
$last_char = mb_substr($t, -1);
if (!$in_pre) {
$quotes = typogrify_i18n_quotes(isset($span_lang) ? $span_lang : $doc_lang);
$t = ProcessEscapes($t);
if ($convert_quot) {
$t = preg_replace('/"/', '"', $t);
}
if ($do_dashes) {
if ($do_dashes == 1) {
$t = EducateDashes($t);
} elseif ($do_dashes == 2) {
$t = EducateDashesOldSchool($t);
} elseif ($do_dashes == 3) {
$t = EducateDashesOldSchoolInverted($t);
}
}
if ($do_ellipses) {
$t = EducateEllipses($t);
}
// Note: backticks need to be processed before quotes.
if ($do_backticks) {
$t = EducateBackticks($t);
if ($do_backticks == 2) {
$t = EducateSingleBackticks($t);
}
}
if ($do_quotes) {
$t = EducateBackticks($t);
if ($t == "'") {
// Special case: single-character ' token.
if (preg_match('/\\S/', $prev_token_last_char)) {
$t = $quotes[3];
} else {
$t = $quotes[2];
}
} elseif ($t == '"') {
// Special case: single-character " token.
if (preg_match('/\\S/', $prev_token_last_char)) {
$t = $quotes[1];
} else {
$t = $quotes[0];
}
} else {
// Normal case:
$t = EducateQuotes($t, $quotes);
}
}
if ($do_stupefy) {
$t = StupefyEntities($t);
}
}
$prev_token_last_char = $last_char;
$result .= $t;
}
}
return $result;
}
function SmartyPants($text, $attr = NULL, $ctx = NULL)
{
global $smartypants_attr, $sp_tags_to_skip;
# Paramaters:
$text;
# text to be parsed
$attr;
# value of the smart_quotes="" attribute
$ctx;
# MT context object (unused)
if ($attr == NULL) {
$attr = $smartypants_attr;
}
# Options to specify which transformations to make:
$do_stupefy = FALSE;
$convert_quot = 0;
# should we translate " entities into normal quotes?
# Parse attributes:
# 0 : do nothing
# 1 : set all
# 2 : set all, using old school en- and em- dash shortcuts
# 3 : set all, using inverted old school en and em- dash shortcuts
#
# q : quotes
# b : backtick quotes (``double'' only)
# B : backtick quotes (``double'' and `single')
# d : dashes
# D : old school dashes
# i : inverted old school dashes
# e : ellipses
# w : convert " entities to " for Dreamweaver users
if ($attr == "0") {
# Do nothing.
return $text;
} else {
if ($attr == "1") {
# Do everything, turn all options on.
$do_quotes = 1;
$do_backticks = 1;
$do_dashes = 1;
$do_ellipses = 1;
} else {
if ($attr == "2") {
# Do everything, turn all options on, use old school dash shorthand.
$do_quotes = 1;
$do_backticks = 1;
$do_dashes = 2;
$do_ellipses = 1;
} else {
if ($attr == "3") {
# Do everything, turn all options on, use inverted old school dash shorthand.
$do_quotes = 1;
$do_backticks = 1;
$do_dashes = 3;
$do_ellipses = 1;
} else {
if ($attr == "-1") {
# Special "stupefy" mode.
$do_stupefy = 1;
} else {
$chars = preg_split('//', $attr);
foreach ($chars as $c) {
if ($c == "q") {
$do_quotes = 1;
} else {
if ($c == "b") {
$do_backticks = 1;
} else {
if ($c == "B") {
$do_backticks = 2;
} else {
if ($c == "d") {
$do_dashes = 1;
} else {
if ($c == "D") {
$do_dashes = 2;
} else {
if ($c == "i") {
$do_dashes = 3;
} else {
if ($c == "e") {
$do_ellipses = 1;
} else {
if ($c == "w") {
$convert_quot = 1;
} else {
# Unknown attribute option, ignore.
}
}
}
}
}
}
}
}
}
}
}
}
}
}
$tokens = _TokenizeHTML($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 .= $cur_token[1];
if (preg_match("@{$sp_tags_to_skip}@", $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) {
$t = ProcessEscapes($t);
if ($convert_quot) {
$t = preg_replace('/"/', '"', $t);
}
if ($do_dashes) {
if ($do_dashes == 1) {
$t = EducateDashes($t);
}
if ($do_dashes == 2) {
$t = EducateDashesOldSchool($t);
}
if ($do_dashes == 3) {
$t = EducateDashesOldSchoolInverted($t);
}
}
if ($do_ellipses) {
$t = EducateEllipses($t);
}
# Note: backticks need to be processed before quotes.
if ($do_backticks) {
$t = EducateBackticks($t);
if ($do_backticks == 2) {
$t = EducateSingleBackticks($t);
}
}
if ($do_quotes) {
if ($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:
$t = EducateQuotes($t);
}
}
}
if ($do_stupefy) {
$t = StupefyEntities($t);
}
}
$prev_token_last_char = $last_char;
$result .= $t;
}
}
return $result;
}
function SmartyPants($text = '', $mode = DEFAULT_OPERATION_MODE)
{
// quick return for empty string
if ($text == '') {
return '';
}
// default all commands to FALSE, then reset according to $mode
$do_quotes = $do_backticks_double = $do_backticks_all = FALSE;
$do_dashes = $do_oldschool_dashes = $do_inverted_oldschool_dashes = FALSE;
$do_ellipses = $do_stupefy = $convert_quot = FALSE;
// setting these flags introduces a layer of abstraction that turned out
// to be unnecessary. Such is porting.
if ($mode & DO_QUOTES) {
$do_quotes = TRUE;
}
if ($mode & DO_BACKTICKS_DOUBLE) {
$do_backticks_double = TRUE;
}
if ($mode & DO_BACKTICKS_ALL) {
$do_backticks_all = TRUE;
}
if ($mode & DO_DASHES) {
$do_dashes = TRUE;
}
if ($mode & DO_OLDSCHOOL_DASHES) {
$do_oldschool_dashes = TRUE;
}
if ($mode & DO_INVERTED_OLDSCHOOL_DASHES) {
$do_inverted_oldschool_dashes = TRUE;
}
if ($mode & DO_ELLIPSES) {
$do_ellipses = TRUE;
}
if ($mode & DO_QUOT_CONV) {
$convert_quot = TRUE;
}
// tokenize input string -- break it into HTML tags and the text between them.
$tokens = array();
_tokenize($text, $tokens);
$result = '';
$in_pre = FALSE;
# 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 $data) {
if ($data['type'] == TOKENS_TYPE_TAG) {
# Don't mess with quotes inside tags.
$result .= $data['body'];
// if the current tag contains text that should not be
// modified, set $in_pre to TRUE
if (preg_match(TAGS_TO_SKIP, $data['body'], $hits)) {
$in_pre = $hits[1] == '' ? TRUE : FALSE;
}
} else {
$t = $data['body'];
$last_char = substr($t, -1);
# Remember last char of this token before processing.
if (!$in_pre) {
$t = ProcessEscapes($t);
if ($convert_quot) {
$t = str_replace('"', '"', $t);
}
if ($do_dashes) {
$t = EducateDashes($t);
} elseif ($do_oldschool_dashes) {
$t = EducateDashesOldSchool($t);
} elseif ($do_inverted_oldschool_dashes) {
$t = EducateDashesOldSchoolInverted($t);
}
if ($do_ellipses) {
$t = EducateEllipses($t);
}
# Note: backticks need to be processed before quotes.
if ($do_backticks_double || $do_backticks_all) {
$t = EducateBackticks($t);
}
if ($do_backticks_all) {
$t = EducateSingleBackticks($t);
}
if ($do_quotes) {
if ($t == "'") {
# Special case: single-character ' token
if (preg_match("/\\S/", $prev_token_last_char)) {
$t = "’";
} else {
$t = "‘";
}
} elseif ($t == '"') {
# Special case: single-character " token
if (preg_match("/\\S/", $prev_token_last_char)) {
$t = "”";
} else {
$t = "“";
}
} else {
$t = EducateQuotes($t);
}
}
if ($do_stupefy) {
$t = StupefyEntities($t);
}
}
$prev_token_last_char = $last_char;
$result .= $t;
}
}
return $result;
}