Johannes,
On Mon, Sep 26, 2005 at 12:08:05PM -0700, Johannes Graumann wrote:
> Package: python-rpy
> Version: 0.4.6-1
> Severity: important
>
> After upgrading R on my unsteble box to 2.2.0 beta (from 2.1.1), rpy
> ceases to work. Following a 'from rpy import *' in a python script, the
> following error is thrown:
>
> Traceback (most recent call last):
> File "./DistAn", line 3, in ?
> from rpy import *
> File "/usr/lib/python2.3/site-packages/rpy.py", line 51, in ?
> exec("import _rpy%s as _rpy" % RVER)
> File "<string>", line 1, in ?
> ImportError: No module named _rpy2020
>
> Downgrading R fixes the issue. While ultimately I can wait for an upstream
> release to accomodate this R version, there is clearly a dependency missing:
> rpy 0.4.6 should require R 2.1.1 and conflict with R 2.2.0.
"Unstable" is not guaranteed to work at all times.
Prior to a release of R, I tend to make several snapshots available -- in a
way that precludes them from entering testing. So R and rpy in testing work,
and will work -- until R 2.2.0 enters testing AND if and only if rpy will
not have updated. So your bug report is appreciated as a heads-up, but is
also entirely "upstream".
So would you agree with me leaving this open as a reminder, CCing it to Greg
and tagging it as both 'unstable' and 'upstream' ?
Lastly, I don;t think the strongly versioned help here as it really is a
one-to-one relationship. We simply need to update rpy; I wouldn't want it to
block R from entering testing as there are many R users who do not use rpy.
> Thanks for your work in making this great package available to me,
My pleasure. Thanks for the feedback and your helping it keeping the
quality level up.
Gruesse, Dirk
> Joh
>
> -- System Information:
> Debian Release: testing/unstable
> APT prefers unstable
> APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
> Architecture: i386 (i686)
> Shell: /bin/sh linked to /bin/bash
> Kernel: Linux 2.6.12-1-k7
> Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
>
> Versions of packages python-rpy depends on:
> ii python 2.3.5-3 An interactive high-level
> object-o
> ii python2.3-rpy 0.4.6-1 Python interface to the GNU R
> lang
>
> python-rpy recommends no packages.
>
> -- no debconf information
--
Statistics: The (futile) attempt to offer certainty about uncertainty.
-- Roger Koenker, 'Dictionary of Received Ideas of Statistics'
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]