Re: [patch/rfc] "tui"

[Bob Rossi <bob@brasko.net>]

On Wed, Feb 11, 2004 at 09:47:34AM -0500, Andrew Cagney wrote:
> >On Tue, Feb 10, 2004 at 12:53:26PM -0700, Kevin Buettner wrote:
> >
> >>On Tue, 10 Feb 2004 13:40:10 -0500
> >>Daniel Jacobowitz <drow@mvista.com> wrote:
> >>
> >
> >>> On Tue, Feb 10, 2004 at 01:32:13PM -0500, Andrew Cagney wrote:
> >
> >>> > 
> >>> > This modifies the build machinery so that the program "tui" is built. 
> >>> > "tui" starts up GDB's TUI (or Text User Interface).  Alternative 
> >>program > > names include "tgdb", "gdbtui", ...
> >>> > 
> >>> > Comments?  Otherwize look to commit it round 16th Feb.
> >
> >>> 
> >>> Any of those names sound great to me.  I think that one with "gdb" in it
> >>> would be clearer; unlike Insight, the TUI does not have a separate life
> >>> from GDB.
> >
> >>
> >>I too would prefer one of the names with "gdb" in it somewhere.
> >
> >
> >I would also, I actually like the name gdbtui. It is clear that it is
> >GDB, and it is clear that it is the text user interface.
> 
> Then:
> 	tui<tab>
> or
> 	tg<tab>
> wouldn't work.
> 
> Speaking of which, if that tcl/tk code ever integrated into GDB's code 
> base (assuming it's first contributed), it to will need a name: tclgdb, 
> tkgdb, and even gdbtk come to mind (ah, the ultimate irony).

Well thats pretty funny, I thought the uninformed user would type 
gd<tab> and see gdb,gdbtui. That would be some pretty nice eye candy for
those that didn't know that the tui was on there system.

I like the idea of thinking that everything begins with gdb, to keep the
names uniformed and easy to find. If the convention was gdbX, they would
certainly be easy to find. For example 

gdb,
gdbtui
gdbtk

Also, putting the name on the end gives the feel that the tui is part of
GDB, and not just another front end. For example, it seems that most 
developers that write front ends add the abbreviation up front. 
A little proof would be,

xxgdb
tgdb
kgdb
cgdb

I am sure there are others ...

In all, I think the extension on the end would be a little proof that
the exe is part of GNU GDB. Also, it would be easy to find with gd<tab>.

Either way, I'll be interested to see what it ends up being.

Bob Rossi