Re: EGG10 - Proposed Automatic Validation Checks

Hello Stephen:

The ibischk2 parser already does some numerical checking of some table
end-points values and correctly handles numbers/units/scientific notation
combinations. For example, .5, 500m, 500e3u, and 500e-3 are all correctly
interpreted as .5 So the warning tests should be on actual selected numerical
values, not on the format.

Bob Ross
Interconnectix, Inc.

> To: ibis@vhdl.org
> Cc: crokusek@qdt.com
> Subject: Re: EGG10 - Proposed Automatic Validation Checks
> Date: Fri, 23 Feb 1996 08:28:27 -0800
> From: Stephen Peters <speters@ichips.intel.com>
> Status: R

> Hello Arpad:

> Mmmmm.....that's OK if a resistance is expressed as '0.2' instead
> of '200m' - it just results in an additional warning message. We could
> handle scientific notation by simply checking the exponent value, but
> the real point of this bird is to detect the missing suffix (the most
> common error).

> Thanks for your comments!

> Regards,
> Stephen

> Text item:

> Stephen,

> Sounds like a good idea, but what do you do if the value is expressed in the
> base unit? This might happen often with the resistance values, or if one uses
> engineering or scientific notation.

> Arpad
> ==============================================================================

> Hello Chris, and Fellow IBISans:

> Reading Chris's EEG 10 got me thinking about another test
> I would like to see added to this golden parser EGG. How about a 'units'
> check? I would like to see a warning if the values listed with
> [C_comp], [Package], R_pin, L_pin, C_pin, [Ramp] and the waveform
> fixture variables do not have a suffix attached. For example, if
> a model maker entered '5' instead of '5p' as the value of C_comp,
> the golden parser would flag it with a warning. I have seen this
> error several times, especially with the ramp rate number (leaving
> the 'n' off the Dt portion of the ratio).

> Regards,
> Stephen Peters
> Intel Corp.
Received on Fri Feb 23 09:47:45 1996