[Groff] Re: grap-1.36 clip and logscale patch

[Ted Faber]

On Sun, Jun 12, 2005 at 03:26:04PM +0200, Hartmut Henkel wrote:
> Hi,
> 
> there is a tiny patch to grap-1.36 (i beamed it over to the base
> actually over 2 years ago, but it somehow got lost in space), that 1.
> corrects for wrong clipping when in logscale, and 2. extends grap's
> lower display limit to quantities less than 1e-6. This increases grap's
> use, e. g. for electrical engineering when one calculates in picoAmperes
> and such. I don't know whether the traditional original grap had this
> weird 1e-6 limitation.
> 
> The patch with test files and PDF output (orig. + patched) can be found
> here:
> 
> http://www.circuitwizard.de/misc/grap-1.36-clip-patch.tgz
> 
> Have fun!

I bet you thought I've forgotten about this, but no, I haven't.  I had a
bug report this week, and while I had the code open, I went back to your
patch.

I've incorporated the functionality you want, though I didn't exactly
use your patch.  You had a few places where you'd stop grap altogether
on an error, and I try to avoid stopping if at all possible.  We also
differed on what to do if there was a frame where the min and max
calculated values were the same - I added one you wanted to multiply.
The cases I've seen where the values are the same, it's also often the
case that they're both zero, so multiplying doesn't help at all.  Adding
one is certainly also a hack, but I think it falls back to some
reasonable behavior.

But, I have made changes that produce the same output as your test cases
- or very near.

If you get a chance, have a look at
http://www.lunabase.org/~faber/Vault/software/grap/grap-1.38pre.tar.gz
unless I hear very bad news from you or my bug reporter, I'm going to
roll this out in the next few days.

Thanks for your enormous patience.

-- 
http://www.lunabase.org/~faber
Unexpected attachment? http://www.lunabase.org/~faber/FAQ.html#SIG

pgpeAs36nyBif.pgp
Description: PGP signature