Greg,

More news, most of it good.

Two pieces of bad news:

1)  The build system is fragile.  I am building Debian package in a chroot
that runs unstable.  However, I 'launch' the build on a testing system, and
there I still run R 2.6.2.  Now, it turns out that when setup.py runs

    # get the Python version
    if sys.version[:3] >= '2.2':
        DEFINE = []
        UNDEF = ['PRE_2_2']
    else:
        DEFINE = [('PRE_2_2', None)]
        UNDEF = []

    # configure the R paths
    RVERSION = rpy_tools.get_R_VERSION(RHOME, force_exec=True)
    RVER     = rpy_tools.get_R_VERSION_CODE(RVERSION)

    print "### Using R verion %s installed at %s ###" % (RVERSION, RHOME)

it finds 2.6.2 on the 'outer host' system.  Would it be possible to run that
check later?  Maybe this 'just' gets in via 'setup.py clean' at the
beginning.  Unclear.

2)  I needed two patches.  The first is the one I mentioned in the earlier
email.  We no longer have Rlapack.so under Debian as we now use Lapack and
Blas from the system (having finally upgraded to current Atlas libraries
etc).  The second is that KillAllDevices() doesn't seem to be exported, so I
made it Rf_KillAllDevices().  That seems to work.

I'll send you a full diff in a short bit.

Dirk

On 14 April 2008 at 21:15, Dirk Eddelbuettel wrote:
| 
| Hi Greg,
| 
| The link options in setup.py need work -- in both 1.0.1 and in 1.0.2 which I
| tried too.  While setting
| 
|         #libraries=['R','Rlapack']
|         libraries=['R','lapack']
| 
| helps to build, I can not load:
| 
| [EMAIL PROTECTED]:/# xvfb-run python
| Python 2.5.2 (r252:60911, Mar  9 2008, 03:52:30)
| [GCC 4.3.0 20080301 (prerelease) [gcc-4_3-branch revision 132801]] on linux2
| Type "help", "copyright", "credits" or "license" for more information.
| >>> from rpy import *
| Traceback (most recent call last):
|   File "<stdin>", line 1, in <module>
|   File "/usr/lib/python2.5/site-packages/rpy.py", line 134, in <module>
|     """ % RVERSION)
| RuntimeError: /usr/lib/python2.5/site-packages/_rpy2070.so: undefined symbol: 
KillAllDevices
| 
|       RPy module can not be imported. Please check if your rpy
|       installation supports R 2.7.0. If you have multiple R versions
|       installed, you may need to set RHOME before importing rpy. For
|       example:
| 
|       >>> from rpy_options import set_options
|       >>> set_options(RHOME='c:/progra~1/r/rw2011/')
|       >>> from rpy import *
| 
| 
| >>>
| 
| even if I run this inside xvfb, so it is not simply a matter of x11 devices. 
| 
| Is KillAllDevices something that changed with R 2.7.0?  I built rpy under the
| 2.7.0 snapshot from last week.
| 
| All the best, Dirk
| 
| On 14 April 2008 at 12:53, Lucas Nussbaum wrote:
| | Package: rpy
| | Version: 1.0.1-4
| | Severity: serious
| | User: [EMAIL PROTECTED]
| | Usertags: qa-ftbfs-20080413 qa-ftbfs
| | Justification: FTBFS on i386
| | 
| | Hi,
| | 
| | During a rebuild of all packages in sid, your package failed to build on 
i386.
| | 
| | This rebuild was done with gcc 4.3 instead of gcc 4.2, because gcc 4.3 is 
now
| | the default on most architectures (even if it's not the case on i386 yet).
| | Feel free to downgrade this bug to 'important' if your package is only built
| | on i386, and this bug is specific to gcc 4.3 (i.e the package builds fine 
with
| | gcc 4.2).
| | 
| | Relevant part:
| | > gcc -pthread -shared -O2 -Wall 
build/temp.linux-i686-2.4/src/rpymodule2070.o 
build/temp.linux-i686-2.4/src/R_eval2070.o 
build/temp.linux-i686-2.4/src/io2070.o build/temp.linux-i686-2.4/src/setenv.o 
-L/usr/lib/R/bin -L/usr/lib/R/lib -Wl,-R/usr/lib/R/bin -Wl,-R/usr/lib/R/lib -lR 
-lRlapack -o build/lib.linux-i686-2.4/_rpy2070.so
| | > /usr/bin/ld: cannot find -lRlapack
| | > collect2: ld returned 1 exit status
| | 
| | The full build log is available from:
| |    http://people.debian.org/~lucas/logs/2008/04/13
| | 
| | A list of current common problems and possible solutions is available at 
| | http://wiki.debian.org/qa.debian.org/FTBFS . You're welcome to contribute!
| | 
| | About the archive rebuild: The rebuild was done on about 50 AMD64 nodes
| | of the Grid'5000 platform, using a clean chroot containing a sid i386
| | environment.  Internet was not accessible from the build systems.
| | 
| | -- 
| | | Lucas Nussbaum
| | | [EMAIL PROTECTED]   http://www.lucas-nussbaum.net/ |
| | | jabber: [EMAIL PROTECTED]             GPG: 1024D/023B3F4F |
| | 
| | 
| 
| -- 
|    I was mostly off email over the last few days, and am now catching
|    up.  My apologies for delayed responses and/or brevity. 
| --
| Three out of two people have difficulties with fractions.

-- 
   I was mostly off email over the last few days, and am now catching
   up.  My apologies for delayed responses and/or brevity. 
--
Three out of two people have difficulties with fractions.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to