On Wed, Aug 6, 2008 at 7:51 AM, Peter TB Brett <[EMAIL PROTECTED]> wrote:
> It's usual (as far as I know) in GLib-based software to use the GLib basic
> types, as far as I can tell. I've used them almost exclusively in the gEDA
> code I've written, anyway. *shrug*

Regardless of current practice, I dislike the wholesale
s/TYPENAME/gTYPENAME/ replacement.  I just don't understand the
supposed benefits, while I do understand the costs: you now have many
more keyword-like tokens in your source code that don't get
syntax-highlighted.  You now *have to* #include most or all of glib
just to compile your program.  Bubble gum wrapper message: "Did you
know?  The compiler has to chow through more than a megabyte of input
for each gEDA source file it compiles."

I have more sympathy for G_MAXINT and friends, as I can more easily
imagine <limits.h> being missing or broken on some systems (any
specific examples?) on which we want gaf to be able to build.

It's not like there's a difference between "int" and "gint" - unless
GLib starts doing "typedef int gint __attribute__((mode (TI)));"


_______________________________________________
geda-dev mailing list
[email protected]
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev

Reply via email to