您需要这样的内容(假设输入UTF-8,而忽略CJK(中文,日文,韩文)):
$chr_map = array( // Windows prepage 1252 "xC2x82" => "'", // U+0082⇒U+201A single low-9 quotation mark "xC2x84" => '"', // U+0084⇒U+201E double low-9 quotation mark "xC2x8B" => "'", // U+008B⇒U+2039 single left-pointing angle quotation mark "xC2x91" => "'", // U+0091⇒U+2018 left single quotation mark "xC2x92" => "'", // U+0092⇒U+2019 right single quotation mark "xC2x93" => '"', // U+0093⇒U+201C left double quotation mark "xC2x94" => '"', // U+0094⇒U+201D right double quotation mark "xC2x9B" => "'", // U+009B⇒U+203A single right-pointing angle quotation mark // Regular Unipre // U+0022 quotation mark (") // U+0027 apostrophe (') "xC2xAB" => '"', // U+00AB left-pointing double angle quotation mark "xC2xBB" => '"', // U+00BB right-pointing double angle quotation mark "xE2x80x98" => "'", // U+2018 left single quotation mark "xE2x80x99" => "'", // U+2019 right single quotation mark "xE2x80x9A" => "'", // U+201A single low-9 quotation mark "xE2x80x9B" => "'", // U+201B single high-reversed-9 quotation mark "xE2x80x9C" => '"', // U+201C left double quotation mark "xE2x80x9D" => '"', // U+201D right double quotation mark "xE2x80x9E" => '"', // U+201E double low-9 quotation mark "xE2x80x9F" => '"', // U+201F double high-reversed-9 quotation mark "xE2x80xB9" => "'", // U+2039 single left-pointing angle quotation mark "xE2x80xBA" => "'", // U+203A single right-pointing angle quotation mark);$chr = array_keys ($chr_map); // but: for efficiency you should$rpl = array_values($chr_map); // pre-calculate these two arrays$str = str_replace($chr, $rpl, html_entity_depre($str, ENT_QUOTES, "UTF-8"));这里是背景:
每个Unipre字符都完全属于一个“常规类别”,其中可以包含引号字符的字符如下:
Ps
“标点符号,打开”Pe
“标点符号,关闭”Pi
“标点符号,初始引号(根据用法,其行为可能类似于Ps或Pe)”Pf
“标点符号,最终报价(根据用法,其行为可能类似于Ps或Pe)”Po
“标点符号,其他”
(这些页面可方便地检查您是否没有错过任何内容-
类别索引)
有时在启用Unipre的正则表达式中匹配这些类别很有用。
此外,Unipre字符具有“属性”,您感兴趣的是
Quotation_Mark。不幸的是,这些不能在正则表达式中访问。
在Wikipedia中,您可以找到具有
Quotation_Mark属性的字符组。最终参考是unipre.org上的PropList.txt,但这是一个ASCII文本文件。
如果您还需要翻译CJK字符,则只需获取它们的代码点,确定其翻译,并找到其UTF-8编码,例如,通过在fileformat.info中查找(例如,对于U +
301E:http
://www.fileformat.info/info/unipre/char/301e/index.htm)。
关于Windows代码页1252:Unipre定义了前256个代码点,以表示与ISO-8859-1完全相同的字符,但是ISO-8859-1通常与Windows代码页1252混淆,因此所有浏览器都呈现0x80-0x9F范围,在ISO-8859-1中为“空”(更确切地说:它包含控制字符),就像Windows代码页1252一样。Wikipedia页中的表列出了Unipre等效项。
注意:
strtr()通常比慢
str_replace()。使用您的输入和PHP版本为其计时。如果速度足够快,您可以直接使用地图,例如my
$chr_map。
如果您不确定您的输入是UTF-8编码的,并且愿意假设输入不是UTF-8编码的,那么它是ISO-8859-1或Windows代码页1252,那么您可以在执行其他操作之前执行此操作:
if ( !preg_match('/^\x*$/u', $str)) { $str = utf8_enpre($str);}警告:不过,这种正则表达式在极少数情况下可能无法检测到非UTF-8编码。例如:
"Gruß…"=="GruxDFx85"看起来像此正则表达式的UTF-8(U
+ 07C5是N’ko数字5)。该正则表达式可以稍作增强,但是不幸的是,可以证明,对于编码检测问题,不存在完全可靠的解决方案。
如果要将Windows代码页1252产生的范围0x80-0x9F标准化为常规Unipre代码点,则可以执行此操作(并删除
$chr_map上面的第一部分):
$normalization_map = array( "xC2x80" => "xE2x82xAC", // U+20AC Euro sign "xC2x82" => "xE2x80x9A", // U+201A single low-9 quotation mark "xC2x83" => "xC6x92", // U+0192 latin small letter f with hook "xC2x84" => "xE2x80x9E", // U+201E double low-9 quotation mark "xC2x85" => "xE2x80xA6", // U+2026 horizontal ellipsis "xC2x86" => "xE2x80xA0", // U+2020 dagger "xC2x87" => "xE2x80xA1", // U+2021 double dagger "xC2x88" => "xCBx86", // U+02C6 modifier letter circumflex accent "xC2x89" => "xE2x80xB0", // U+2030 per mille sign "xC2x8A" => "xC5xA0", // U+0160 latin capital letter s with caron "xC2x8B" => "xE2x80xB9", // U+2039 single left-pointing angle quotation mark "xC2x8C" => "xC5x92", // U+0152 latin capital ligature oe "xC2x8E" => "xC5xBD", // U+017D latin capital letter z with caron "xC2x91" => "xE2x80x98", // U+2018 left single quotation mark "xC2x92" => "xE2x80x99", // U+2019 right single quotation mark "xC2x93" => "xE2x80x9C", // U+201C left double quotation mark "xC2x94" => "xE2x80x9D", // U+201D right double quotation mark "xC2x95" => "xE2x80xA2", // U+2022 bullet "xC2x96" => "xE2x80x93", // U+2013 en dash "xC2x97" => "xE2x80x94", // U+2014 em dash "xC2x98" => "xCBx9C", // U+02DC small tilde "xC2x99" => "xE2x84xA2", // U+2122 trade mark sign "xC2x9A" => "xC5xA1", // U+0161 latin small letter s with caron "xC2x9B" => "xE2x80xBA", // U+203A single right-pointing angle quotation mark "xC2x9C" => "xC5x93", // U+0153 latin small ligature oe "xC2x9E" => "xC5xBE", // U+017E latin small letter z with caron "xC2x9F" => "xC5xB8", // U+0178 latin capital letter y with diaeresis);$chr = array_keys ($normalization_map); // but: for efficiency you should$rpl = array_values($normalization_map); // pre-calculate these two arrays$str = str_replace($chr, $rpl, $str);



