Skip to content
Back to formatted view

Raw Message

Message-ID: <18268.63882.268683.604687@stat.math.ethz.ch>
Date: 2007-12-10T08:32:10Z
From: Martin Maechler
Subject: buglet in curve?
In-Reply-To: <475B6B16.6040909@ufl.edu>

>>>>> "BB" == Ben Bolker <bolker at ufl.edu>
>>>>>     on Sat, 08 Dec 2007 23:12:06 -0500 writes:

    BB> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1


    BB>   Symptoms: curve with log x axis gets the wrong x
    BB> limits (in 2.6.1, I believe introduced in this version).

    BB>  Credit goes to Mike McCoy for detecting the problem.

    BB>   Demonstration:

    BB> x = 1:5 plot(x,5*exp(-x),xlim=c(0.1,1000),log="x") xvec
    BB> = 10^seq(-1,3,length=100) lines(xvec,5*exp(-xvec))
    BB> curve(5*exp(-x),add=TRUE,col=2,lwd=3)


    BB>    I believe the problem arises from this fix:

    BB>     o curve() with unspecified 'from', 'to' and 'xlim'
    BB> now reuses the previous x limits, and not slightly
    BB> larger ones.

    BB>  and I believe the solution is to replace this ...

   [    .....   ]

Thank you, Ben  (and Mike).
That's indeed a new buglet, and you are even right about
when it happened.
Mea culpa.

    BB>   cheers Ben Bolker

 {aaaah, it feels good to get a message from you without the
  appended Nabble spam :-)}

Regards,
Martin