Multiple charsets in one document

zalath

Okay. This is going to be good 😉

I've a page to display product information. Including - ta-daa - the product name in different languages, a few of them which happen to include less-than-ordinary characters.

Now, it appears the russian text is in windows-1251 . I deduced this from the fact that when I changed the document charset to windows-1251, the cyrillic characters were all displayed.

That, naturally caused problems, because most of the data is to be displayed in ISO-8859-1.

No sweat, I thought, and dug out the manual. Now, there's htmlentities that's supposed to do the trick for me. So, I go and try htmlentities($string, ENT_COMPAT, 'windows-1251'); which produced - no changes. Still a mess.

I also tried to first convert_cyr_string from windows-1251 to ISO-8859-5 and then htmlentities and/or htmlspecialchars it to no avail. All I'm getting is a garbled mess, which quite frankly isn't too attractive.

I'm missing the obvious, aren't I?

herve

Hi,

may be the solution is to use Unicode charset (UTF-8 to be more precise) in you HTML page ?

(pb : how to convert your russian words from your
actual coding to UTF-8 ?)

Hervé.

zalath

As you so helpfully point out, the page encoding isn't the problem here - and I never said I didn't use UTF-8 ;-) (I don't, at least yet, but I did try that one too)

herve

Well,

here's a lot of information :

http://www.kloth.net/informations/cyr2utf8.htm

http://webcenter.ru/~kazarn/eng/fonts_tools.htm#ttconv

http://www.selectorweb.com/unicode_utf.html

http://php3.de/manual/es/html/ref.mbstring.php

but I've read also (from the last URL) :

As of PHP 4.3.0, the following character encoding support will be added experimentaly : EUC-CN, CP936, HZ, EUC-TW, CP950,
BIG-5, EUC-KR, UHC (CP949), ISO-2022-KR, Windows-1251 (CP1251), ...

So, for the time being, with PHP 4.2.3
it seems there's no solution (in regular PHP) :]

May be s.o. has crafted a special fct for
this purpose ?

Hervé.

zalath

...but I did it. I hand-crafted a translation map between windows-1251 and Unicode (well, around 90 characters). I'm planning on doing the same for at least 1250 as well (we'll see how many sets I need). I'm considering packaging the whole shebang so if anyone else needs it, feel free to mail me.