Re: Wish-List for GTOP

[Martin Baulig <baulig math38 uni-trier de>]

I think I'll write this kind of library:

On Tue, 31 Mar 1998, Sebastian Wilhelmi wrote:

> 1. Define a set of 'top'-relevant information. a good orientation surly
> is gtop. (this set should be an union in the sense of the set theory for
> different systems). 

Well, under Linux nearly all those information can be accessed via the
proc filesystem. I think this one - and the ipc header files from GNU libc
- can be a good starting point.

> 
> 2. Define a function to return which of these infos are available in
> this particular implementation.
> 
> 3. Define a set of functions to return the available data. (e.g.
> memory-usage; maybe net-usage; of course process-info and so on.)

This could be designed similar to how GNU libc does it: returning -1 and
setting errno to ENOSYS when a particular information/function is not
present.

> 
> 4. write a program, that reads a short command and raw data on stdin and
> writes the result to the stdout. (you will see later, why)
>

Maybe it's standard output should dump the proc filesystem under Linux?

> 5. reprogram the in point 2. and 3. mentioned functions using only the
> in point 4. mentioned program.
> 
> On top of this libtop you could then easily make a truly
> system-independent gtop.

I'll design a first interface the next days that can then be discussed by
a wider public.

Martin

-----------------------------------------------------------------
   Martin Baulig - Angewandte Mathematik - Universitaet Trier

   baulig@castor.uni-trier.de, http://www.home-of-linux.com/
   Key: 1024-bit key with ID C8178435 created 1997/01/24 
   ID:  67 C1 84 A0 47 F5 11 C5  5F 68 4C 84 99 05 C3 92
   Finger me for public key or fetch finger.txt from the url above
------------------------------------------------------------------