Deciphering
Glyph
( )
Today while helping a friend, I was exposed to PH...

Sat 29 October 2005

Today while helping a friend, I was exposed to PHP's answer to unicode and encodings. Ahem, I'm sorry. multibyte strings.

Some days I cannot help but be amazed at how much better Python is than anything else available today. Obviously PHP is garbage, but the reason I thought to disturb you with this posting, dear reader, was this gem in the comments section of the documentation:
Setting mbstring.func_overload = 2 may break your applications that deal with binary data.
What is it, 1975? We are still writing - not even applications - systems that break when presented with data that has the high bit set??? Maybe we need a new HTTP request header, like the BINARY command in FTP.

Apparently there is something in the works for the next release - PHP6 or so - which will actually deal with unicode more sanely. I haven't looked at it but I am foolishly giving them the benefit of the doubt. So, cool. Maybe that release will also include a way to access "high memory" so I can get at more than 640K of HTML at a time!