UTF8 locales

[AliceWonderMiscreations]

Hello,

 

I am writing a program that makes use of strftime() to produce localized results related to date and time.

 

The output is XML sent as UTF-8 so I need to make sure the locale is a UTF-8 locale.

 

setlocale(LC_TIME, 'ru_RU.UTF8')

 

for example works beautifully and does exactly what I need it to do, but detecting that is the locale to use is where I am having a problem.

 

For example, there are two Russian locales that FireFox in CentOS 7 can report - one is just ru and the second is ru-md

 

Appending a .UTF8 to either of those doesn't work.

 

I guess what I'm looking for, and I hope there is already a pear or PECL library, is some way to take the locale the browser reports and get a UTF8 safe locale I can use with setlocale for the purpose of strftime.

 

Any suggestions?

[AliceWonderMiscreations]

Oh and yes UTF8 is a must.

 

The php script creates an XML file that is then translated via XSLT into another XML format. UTF8 is the only way to do that with sanity.

[AliceWonderMiscreations]

okay I think I found a solution

 

On the server -

 

locale -a |grep "\.utf8$" > whatever.txt

 

Then from a php script I can read whatever.txt and parse it into an two dimensional array, e.g. $foo[ru] is an array that contains ru_RU and ru_UA

 

When the language array only has one entry, that's the default for any browser reported locale for the language,

 

When the language array has two or more entries, I am doing some manual work on the script that turns it into an array so that the proper default is the first listed - e.g. en_GB should be the first in $foo[en]

 

I wish there was a cleaner way to do it, but this will work to solve my problem.