Re: [gentoo-user] dev-lang/R installation : a hard nut

2011-07-01 Thread pk 
On 2011-06-30 11:31, Helmut Jarausch wrote:

> Many thanks for your suggestions.

Bitte schön! :-)

> Finally it turned that some old files (lzma) in /usr/local/include were 
> in the way and caused those problem.

It's always the little things... :-/

MfG / Best regards

Peter K

Re: [gentoo-user] dev-lang/R installation : a hard nut

2011-06-30 Thread Helmut Jarausch 
On 06/29/2011 07:23:19 PM, pk wrote:
> On 2011-06-29 11:32, Helmut Jarausch wrote:
> 
> > I'm looking for advice on the following hard nut.
> > On one of my machines emerging dev-lang/R fails while on
> > a very similar one it succeeds!
> 
> I don't have much to offer, unfortunately, but what are the
> differences
> between your machines (you say it's similar but...)? x86 or x64?
> Compiler settings? USE flags? Also, have you tried another version?
> 

Many thanks for your suggestions.
Finally it turned that some old files (lzma) in /usr/local/include were 
in the way and caused those problem.

Helmut.

Re: [gentoo-user] dev-lang/R installation : a hard nut

2011-06-29 Thread pk 
On 2011-06-29 11:32, Helmut Jarausch wrote:

> I'm looking for advice on the following hard nut.
> On one of my machines emerging dev-lang/R fails while on
> a very similar one it succeeds!

I don't have much to offer, unfortunately, but what are the differences
between your machines (you say it's similar but...)? x86 or x64?
Compiler settings? USE flags? Also, have you tried another version?

MfG / Best regards

Peter K

[gentoo-user] dev-lang/R installation : a hard nut

2011-06-29 Thread Helmut Jarausch 
Hi,

I'm looking for advice on the following hard nut.
On one of my machines emerging dev-lang/R fails while on
a very similar one it succeeds!

On the 'bad one' I get lots of

Warning in lazyLoadDBinsertVariable(vars[i], from, datafile, ascii, 
compress,  :
  internal error 8 in R_compress3

and finally

** R
** data
**  moving datasets to lazyload DB
Warning: lzma decoding result 8
Error in if (!grepl("RD[AX]2\n", magic)) { : argument is of length zero
ERROR: lazydata failed for package 'survival'

On both machines I have xz-utils-5.0.3 installed and the 'bad machine'
has even been reinstalled by an emerge --emtpytree some weeks, ago.
The dev-lang/R problem happened before that and is still happening.
I have re-emerged xz-utils but still no luck.
It must have be related to some settings which are not changed by an
emerge --emtpytree

(As a work around, copying the binary package from that other machine 
and doing an emerge -k works just, but I'd like to know why)

Many thanks for any hints,
Helmut.