Re: ucwords() bug -- Newbie trying to fix it

Ernesto wrote:

Hi, I'm using PHP (4.1.0) for Windows and found a bug with the ucwords() function. The problem is with certain international characters. Example: the "tamaños", is converted as "TamañOs". As you can see, the "ñ" is interpreted as a word separator. What looks stranger is the fact that it will interpret correctly some international characters (i.e. á, é and ë; ASC 160, 161, 130) and will interpret incorrectly some others (i.e. ñ and ü; ASCII 164 and 129). I was looking at the PHP source code hoping I can do something to fix it, but I found that the isspace() function is used to determine what characters are word separators. Unfortunately, isspace() is a function from the C libraries, which I can not fix (I'm no expert with C) :( Anyway, I'm not supossed to fix it (the isspace function), because all Windows string functions are supossed to have international character support and Visual C++ is supossed to call the Windows API to acomplish this kind of tasks. I'm I right? Is there any way I can fix this?

If you submit bug report, I guess someone will fix it. (it may be fixed already) It should be easy one to fix :) -- Yasuo Ohgaki