Skip to content

compile error RH9 & R162

7 messages · Allan Sims, Peter Dalgaard, Thomas Lumley +1 more

#
Hi,

I tried to compile source because for RH9.0 doesn't exist binary. During
compiling I got error:
***********************
make[3]: Entering directory `/home/allans/R-1.6.2/src/library/methods'
dumping R code in package 'methods'
Error in eval(expr, envir, enclos) : couldn't find function "character"
Execution halted
make[3]: *** [../../../library/methods/R/all.rda] Error 1
***********************
What I have to do to solve this?

Allan Sims
allsi@eau.ee
#
Allan Sims <allsi@eau.ee> writes:
You might try the RH8 RPMs. Please tell us if they don't work, but I
see no obvious reason for them not to. 

I haven't seen that particular issue before, but some of us wasted a
good bit of last weekend tracking down another rather nasty issue with
the RH9 version of gcc. Have a look at

http://maths.newcastle.edu.au/~rking/R/help/03a/3769.html

and its neighbours.
#
Hi,

I tried to install RPM for RH80 to computer with RH90. In download page
some short help with some requirement. I fill those requirement and still
I have problem. Have someone experience with that:
************************************************'
[root@mekor4 allans]# locate libtcl
/usr/lib/libtcl.so
/usr/lib/libtcl8.3.so
/usr/lib/libtclstub8.3.a
[root@mekor4 allans]# rpm -i R-1.6.2-1.i386.rpm
warning: R-1.6.2-1.i386.rpm: V3 DSA signature: NOKEY, key ID 97d3544e
error: Failed dependencies:
        libtcl.so.0 is needed by R-1.6.2-1
        libtk.so.0 is needed by R-1.6.2-1
*****************************************

Allan Sims
allsi@eau.ee
On Mon, 14 Apr 2003, Peter Dalgaard BSA wrote:

            
#
Allan Sims <allsi@eau.ee> writes:
Hmmm.... The RH8.0 situation is

s -l /usr/lib/libtcl*
lrwxrwxrwx    1 root     root           11 Jan 23 10:18 /usr/lib/libtcl.so -> libtcl.so.0
-r-xr-xr-x    1 root     root       605121 Aug 20  2002 /usr/lib/libtcl.so.0
lrwxrwxrwx    1 root     root           11 Jan 23 10:18 /usr/lib/libtcl8.3.so -> libtcl.so.0

and libtk similarly. That looks a bit weird, so I can't really blame
RH for fixing it up, but I'd suspect that all you need to do is to

ln -s /usr/lib/libtcl8.3.so usr/lib/libtcl.so.0
ln -s /usr/lib/libtk8.3.so usr/lib/libtk.so.0
#
On Mon, 14 Apr 2003, Allan Sims wrote:

            
I don't know about that error, but R 1.6.2 has a problem with the version
of gcc used on RH9 -- it ends up being unable to distinguish NA and NaN.
You should either try the development snapshot or wait for R1.7.0 next
week.

	-thomas
#
On Mon, 2003-04-14 at 16:05, Thomas Lumley wrote:
??? I thought it was tomorrow. Did I miss something? The imminent
release of R 1.7.0 is the reason why I haven't created an RPM for RH 9.

Martyn
#
Martyn Plummer <plummer@iarc.fr> writes:
It *is* tomorrow (if the CVS server comes back on line!). Thomas is
probably just allowing a little time for platform maintainers to roll
out the binaries.