Unicode, UTF-8, UTF-16, UCS-2, UCS-4 and URIs
Unicode can be confusing !
For a start there are a number of different encodings such as :
- UTF-8 (for example € in UTF-8 is 0xE2 0x82 0xAC)
-
UTF-16 (which uses surrogate pairs to represent "characters" outside the Basic Multilingual Plane (BMP)
-
UCS-2 (a predecessor of UTF-16)
- UCS-4
A RFC-2396 URI must be encoded / escaped using UTF-8 (and %hex-values) so if you want to acccess a web page called
- http://my.host/€.php
the URI will be
- http://my.host/%E2%82%AC.php
and different browsers seem to work with Unicode URIs in different ways !
- Safari works with both (€.php and %E2%82%AC.php) and helpfully (?) redisplays %E2%82%AC.php as €.php in the address bar
- Firefox converts €.php (sometimes incorrectly to %80.php) so you can only use / see %E2%82%AC.php
- IE works with both (€.php and %E2%82%AC.php) but leaves both versions unchanged in the address bar