-1
我试图用它来删除字符串中的所有注释。RegEx在ie8中替换
str = str.replace(/<!--(.*?)-->/g, "");
它完美的铬,但是当涉及到IE8运行,惊喜惊讶的是,它不会删除文本的任何部分。
我注意到,当铬会甩掉评论,他们会变成什么aliong这些行:
<!--[if !supportLists]-->· <!--[endif]-->
这是基于我的全部功能,在This Link
function CleanWordHTML(str) {
str = str.replace(/<o:p>\s*<\/o:p>/g, "");
str = str.replace(/<o:p>.*?<\/o:p>/g, " ");
str = str.replace(/\s*mso-[^:]+:[^;"]+;?/gi, "");
str = str.replace(/\s*MARGIN: 0cm 0cm 0pt\s*;/gi, "");
str = str.replace(/\s*MARGIN: 0cm 0cm 0pt\s*"/gi, "\"");
str = str.replace(/\s*TEXT-INDENT: 0cm\s*;/gi, "");
str = str.replace(/\s*TEXT-INDENT: 0cm\s*"/gi, "\"");
str = str.replace(/\s*TEXT-ALIGN: [^\s;]+;?"/gi, "\"");
str = str.replace(/\s*PAGE-BREAK-BEFORE: [^\s;]+;?"/gi, "\"");
str = str.replace(/\s*FONT-VARIANT: [^\s;]+;?"/gi, "\"");
str = str.replace(/\s*tab-stops:[^;"]*;?/gi, "");
str = str.replace(/\s*tab-stops:[^"]*/gi, "");
str = str.replace(/\s*face="[^"]*"/gi, "");
str = str.replace(/\s*face=[^ >]*/gi, "");
str = str.replace(/\s*FONT-FAMILY:[^;"]*;?/gi, "");
str = str.replace(/<(\w[^>]*) class=([^ |>]*)([^>]*)/gi, "<$1$3");
str = str.replace(/<(\w[^>]*) style="([^\"]*)"([^>]*)/gi, "<$1$3");
str = str.replace(/\s*style="\s*"/gi, '');
str = str.replace(/<SPAN\s*[^>]*>\s* \s*<\/SPAN>/gi, ' ');
str = str.replace(/<SPAN\s*[^>]*><\/SPAN>/gi, '');
str = str.replace(/<(\w[^>]*) lang=([^ |>]*)([^>]*)/gi, "<$1$3");
str = str.replace(/<SPAN\s*>(.*?)<\/SPAN>/gi, '$1');
str = str.replace(/<FONT\s*>(.*?)<\/FONT>/gi, '$1');
str = str.replace(/<\\?\?xml[^>]*>/gi, "");
str = str.replace(/<\/?\w+:[^>]*>/gi, "");
str = str.replace(/<H\d>\s*<\/H\d>/gi, '');
str = str.replace(/<H1([^>]*)>/gi, '');
str = str.replace(/<H2([^>]*)>/gi, '');
str = str.replace(/<H3([^>]*)>/gi, '');
str = str.replace(/<H4([^>]*)>/gi, '');
str = str.replace(/<H5([^>]*)>/gi, '');
str = str.replace(/<H6([^>]*)>/gi, '');
str = str.replace(/<\/H\d>/gi, '<br>'); //remove this to take out breaks where Heading tags were
str = str.replace(/<(U|I|STRIKE)> <\/\1>/g, ' ');
str = str.replace(/<(B|b)> <\/\b|B>/g, '');
str = str.replace(/<([^\s>]+)[^>]*>\s*<\/\1>/g, '');
str = str.replace(/<([^\s>]+)[^>]*>\s*<\/\1>/g, '');
str = str.replace(/<([^\s>]+)[^>]*>\s*<\/\1>/g, '');
//comments
str = str.replace(/<!--(.*?)-->/g, "");
//spaces
str = str.replace(/ /g, '');
//empty tags
str = str.replace(/<[^\/>][^>]*><\/[^>]+>/g, '');
//some RegEx code for the picky browsers
var re = new RegExp("(<P)([^>]*>.*?)(<\/P>)", "gi");
str = str.replace(re, "<div$2</div>");
var re2 = new RegExp("(<font|<FONT)([^*>]*>.*?)(<\/FONT>|<\/font>)", "gi");
str = str.replace(re2, "<div$2</div>");
str = str.replace(/size|SIZE = ([\d]{1})/g, '');
return str;
}
谢谢
强制性备注:http://stackoverflow.com/questions/590747/using-regular-expressions-to-parse-html-why-not – epascarello 2014-12-02 15:24:35
@epascarello无论如何,不应该仍然有办法抓住一个<! - 使用RegEx? – starvator 2014-12-02 15:27:45
你试过了吗?<! - (。*?) - >'?也许我走了,但我怀疑IE8可能会在解析脚本文本之前删除HTML注释。 (即它读作'',然后运行中间的'...'部分)。然后,这似乎会把你的正则表达式变成一个''''JS的评论,所以我可能是错的。 – apsillers 2014-12-02 15:35:21