Re: UTF-8 identifiers [was: Re: Lua t-shirts]

[David Given <dg@...>]

Jerome Vuarand wrote:
[...]
> Lua lexer uses isalpha C function, therefore depending on the locale
> some of the characters in 128-255 are already valid.

It shouldn't be! It should be using the C locale for all code-related 
parsing --- languages should never, ever, under *any* circumstances, 
change their behaviour depending on locale!

That way lies madness, with a perfectly valid program on one computer 
failing to run on another simply due to some user configuration setting. 
I've been burnt by this before. It's decidedly not fun.

> You could change
> it to allow all characters in 128-255 in addition to 'alpha' class and
> underscore, but that would be a hack imho.

Of course! But it's a *useful* hack. We lose nothing by it, but gain 
functionality in other areas --- there are other ASCII-compatible 
encodings that this would work with, for example.

--
┌─── ｄｇ＠ｃｏｗｌａｒｋ．ｃｏｍ ───── http://www.cowlark.com ─────
│
│ "They laughed at Newton. They laughed at Einstein. Of course, they
│ also laughed at Bozo the Clown." --- Carl Sagan