Hi all.

I am trying to do a simple text replace for a paragraph of text that is HTML-encoded.

So, the first method is to simply use Replace(). That fails because the text is HTML-encoded. So in this example:

<p class=MsoNormal style='text-indent:18.0pt'><span lang=EN-US
style='font-size:9.0pt;font-family:Verdana'>Middle class families' income is
much higher than the average level in </span><span lang=EN-US style='font-size:
9.0pt;font-family:Verdana'>Sichuan</span><span lang=EN-US style='font-size:
9.0pt;font-family:Verdana'> province. In 2003, the average income of middle
class families reached 76,000 RMB, 3.65 times the average families' income of
20,850 RMB.&nbsp; </span></p>

If I try to replace the word income with <B>income</B>
it will be okay. But if I try to replace the word 'class' with <B>class</B>,
it will crash the data, replacing the HTML class too. So this method doesn't work.

2nd method. I moved the Replace function into Javascript. Then I put the article text into a TD. I tried to use obj.innerText to do the replace on, but for some reason it did not do the replace. When I switched to .innerHTML, it works, but with the same problem.

I need to distinguish what is <html> and what isn't .

OK. I guess I see what has to be done now. (Just came to me). I need to use a regular expression to split the text into <dfsd> and non <dfds>. Then I do the replace on all non <edfds> and then join it back together. Right?

Or do you have a better way?

Thanks
Wengang