[R] difftime / RBloomberg

[Prof Brian Ripley]

What you need to do is to isolate the package giving the problem and 
send a report to its maintainer.  It is not RODBC nor chron, and 
RBloomberg is not available pre-compiled from CRAN.

On Thu, 22 Mar 2007, Judith Schiller wrote:

> The error is "evaluation nested too deeply: infinite recursion / 
> options(expressions=)?", and sessionInfo() results in
>
> R version 2.4.1 (2006-12-18)
> i386-pc-mingw32
>
> locale:
> LC_COLLATE=German_Austria.1252;LC_CTYPE=German_Austria.1252;LC_MONETARY=German_Austria.1252;LC_NUMERIC=C;LC_TIME=German_Austria.1252
>
> attached base packages:
> [1] "splines"   "stats"     "graphics"  "grDevices" "datasets"  "tcltk" 
> "utils"     "methods"   "base" 
> other attached packages:
>     RODBC  RBloomberg       chron         zoo RDCOMClient        MASS 
> xtable    survival    svSocket        svIO      R2HTML      svMisc
>   "1.1-7"    "0.1-10"     "2.3-8"     "1.2-1"    "0.91-0"    "7.2-30" 
> "1.4-2"      "2.30"     "0.9-5"     "0.9-5"      "1.58"     "0.9-5"
>     svIDE
>   "0.9-5"
>
> The workaround is a nice idea, but as the problem arisis in 
> "as.COMDate.chron", other users of this function might be affected as well.
>
> Prof Brian Ripley schrieb:
>> On Thu, 22 Mar 2007, Schiller Judith 1541 EB wrote:
>>
>> 
>>> hi,
>>> 
>>> I've troubles with some difftime objects. e.g.
>>> 
>>> ISOdate(2001, 4, 26) - ISOdate(2001, 2, 26) - 2
>>> 
>>> works, telling me "Time difference of 57 days". But when I'd like to add
>>> days, such as
>>> 
>>> ISOdate(2001, 4, 26) - ISOdate(2001, 2, 26) + 2
>>> 
>>> the function gives me an error.
>>> 
>> 
>> The error being?  (It does not give me an error.)
>>
>>
>>>  Function "as.COMDate.chron" of the
>>> Rbloomberg package doesn't work for that reason.
>>> I'm running R on a Windows machine, R version 2.4.1.
>>> 
>> 
>> I cross-checked on that system.
>> 
>> It is quite possible the error is due to some other package that you are 
>> using: that's why we ask for the result of sessionInfo().
>> 
>> I would expect
>> 
>> (ISOdate(2001, 4, 26) - ISOdate(2001, 2, 26)) - (-2)
>> 
>> to be a workaround, BTW.
>>
>>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595