Some UTF8 characters become ? with fixUTF8 about forceutf8

[byjg]

Hi Tim,

I tried run your code from "dev-master" and added some parameters:

echo ForceUTF8::fixUTF8('hello žš', ForceUTF8::ICONV_TRANSLIT); //outputs hello žš
echo ForceUTF8::fixUTF8('привет', ForceUTF8::ICONV_TRANSLIT); //outputs ?????? 

And I got running the first line. I tried to load with the other options: WITHOUT_ICONV and ICONV_IGNORE but there is no success.

I hope this help you.

[neitanod]

Sadly, fixUTF8 won't work for characters that are not in the Latin1 (ISO 8859-1) set.
fixUTF8() is a hacky function that helps you if somebody gives you a garbled string that was utf8_encode()'d twice or more times. You can recognize those because they are full of à and  characters.

utf8_encode() was designed to work on ISO 8859-1 strings, and never gives you russian characters.

Are you sure that it's not Encoding::toUTF8() the function you are looking for?

[neitanod]

Also, I pointed out several times that fixUTF8() was slow and not intended for production environments. It's usually useful in batch processes where you need to manually fix your data set before continue working with it.