R-beta: base not loading

Peter Dalgaard BSA p.dalgaard@biostat.ku.dk
16 Aug 1998 12:36:50 +0200

Martin Maechler <maechler@stat.math.ethz.ch> writes:

>     PaulG> (PS. There also seem to be some problems making the f2c with gcc
>     PaulG> 2.8.1. I've been using g77 so that isn't the problem.)
> NOTE: Looking at  RHOME/library/base/R/base (which is the R file that is
> 	NOT correctly loaded by the internal C code),
>       I now bet, that it is  .Alias(.)  which is failing;  more exactly,
>       line 64
> 	------------------------------------------------------
> 	colours <- .Alias(colors)
> 	------------------------------------------------------
>       That's (the only way?) how the error message
> 	------------------------------------------------------
> 	Error: Object "colors" not found 
> 	------------------------------------------------------
>       can emerge...
> [[this may help find gcc's problem, maybe...]]

While messing around with a Windows cross-compile, I realized that
this can also happen if colorstuff.R occurs before New-Internal.R in
the base file. The former has:

## nice to the English
colours <- colors

And the latter has

colors <- function().Internal(colors())
colours <- .Alias(colors)

Now, if ls gets the idea of sorting differently on some platforms,
which is not at all unlikely - Ripley reported something about locale
settings resulting in the collating sequence AaBbCc..., and he *is*
using Sol2.6! - then all hell breaks loose.

The thing in colorstuff.R is simply wrong, so I'll remove it.

PS: No, the 0.62.3 crosscompile doesn't work (yet?).

   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch