Message-ID: <494A6297.8010201@biostat.ku.dk>
Date: 2008-12-18T14:47:51Z
From: Peter Dalgaard
Subject: "Error: bad value" problem
In-Reply-To: <1229594690.2948.98.camel@localhost.localdomain>
Martyn Plummer wrote:
> On Thu, 2008-12-18 at 10:57 +0100, Peter Dalgaard wrote:
>> Martyn Plummer wrote:
>>> This has all the hallmarks of a bug I found and fixed in R-devel
>>> (r46998). I did not port the patch over to the R release branch because
>>> I could not reproduce the bug.
[....]
>> Yes, this does appear to fix the issue. (Did you forget a NEWS file
>> entry, though?)
>
> You mean this?
>
> * Fixed obscure, poorly reproducible bug that nobody else has reported.
>
> Anyway, it's the same bug. There is a call to parse() in plot.mmfit()
> which generates a parse error. This has been wrapped in a silent call
> to try, so you never see the error message:
>
> Error in parse(text = colnames(z)[seq(1, 2 * np, 2)]) :
> unexpected numeric constant in "Parameter 1"
>
> This error leaves the parser in a bad state.
>
>> It's a two-line change fixing a live bug, so it could go in 2.8.1 even
>> in code freeze. The problem is that it is inside gram.y which will
>> trigger some maintainer-mode activity, and did trip up my Fedora laptop
>> slightly. Hmmmm....
>>
>> -p
>
> Fedora 10 certainly won't let me do a maintainer mode build. I leave it
> in your hands.
>
> M.
Now committed for 2.8.1.
--
O__ ---- Peter Dalgaard ?ster Farimagsgade 5, Entr.B
c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
(*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907