Re: [Bugs-r-devel] WG: R2WinBUGS, wine

[Gregor G. <gre...@bf...>]

Hi John!

John P. Burkett wrote:
> Hi Gregor,
> 
> Thank you very much for your quick and helpful response.
> 
>> With R2WinBUGS under Linux you must use WinBUGS
> 
> Switching from OpenBugs to WinBUGS14 seems to have solved my initial
> problem.  However, R still produces some messages that puzzle me.

OK.

>> Can you please issue traceback() after this error? 
> 
> After installing and registering WinBUGS14 and installing the patch for
> version 1.4.2, I modified to the R command to read as follows:
> 
> radon.1 <- bugs (radon.data, radon.inits, radon.parameters,
> "/home/john/myRfiles/radon.1.bug", n.chains=3, n.iter=10, debug=TRUE,
> program="WinBugs", bugs.directory="/home/john/WinBUGS14", useWINE=TRUE,
> newWINE=TRUE, WINEPATH=WINEPATH, WINE="/usr/bin/wine")
> 
> That command got this response from R:
> fixme:keyboard:RegisterHotKey (0x10024,13,0x00000002,3): stub
> fixme:ole:GetHGlobalFromILockBytes cbSize is 13824
> err:ole:CoGetClassObject class {0003000a-0000-0000-c000-000000000046}
> not registered
> err:ole:CoGetClassObject no class object
> {0003000a-0000-0000-c000-000000000046} could be created for context 0x2

Ah, this error messages come from wine part and can be ignored. I do not 
know where exactly they come from. As you said, bugs returns its output 
as it should.

> Perhaps the messages from R are nothing to worry about. Anyway, WinBUGS
> produces output that R can plot and print, giving results similar to
> those Gelman and Hill report on p. 351.  Do you think anything needs to
> be done about the R messages?

I am glad there is one more Linux&R2WinBUGS user around!

-- 
Lep pozdrav / With regards,
     Gregor Gorjanc