Hi, I just wanted to let the affy developers know that the devel version of affy completely bombs out if you don't have the requisite cdf environment installed. Details: OS: Win XP R: 1.7.0 Affy: 1.2.10 Biobase: 1.3.17 tkWidgets: 1.2.13 <- I think these are the only two version-level dependencies If I read in some e.g., RG-U34A chips without the rgu34acdf environment installed, and then try to do anything with the AffyBatch, I get the following error: Error in how[[i]] : subscript out of bounds Error in inherits(cdf.env, "try-error") : Object "cdf.env" not found If you are willing to wait about 10 minutes, you can then use the computer. Otherwise it requires a reboot to recover. And yeah, I know, whaddaya expect if you don't have things set up correctly? Just doing my part to make Bioconductor fool proof ;-D Jim James W. MacDonald UMCCC Microarray Core Facility 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623

On Wed, Apr 30, 2003 at 10:42:12AM -0400, James MacDonald wrote: > Hi, > > I just wanted to let the affy developers know that the devel version of affy completely bombs out if you don't have the requisite cdf environment installed. > > Details: > > OS: Win XP > R: 1.7.0 > Affy: 1.2.10 > Biobase: 1.3.17 > tkWidgets: 1.2.13 <- I think these are the only two version-level dependencies I believe there is also a dependency towards 'reposTools' (although no version number is specified). Which version do you use ? > > If I read in some e.g., RG-U34A chips without the rgu34acdf environment installed, and then try to do anything with the AffyBatch, I get the following error: > > Error in how[[i]] : subscript out of bounds > Error in inherits(cdf.env, "try-error") : Object "cdf.env" not found It looks like a silly bug in the 'autodownload' mecanism (one more). I'll look at that. > > If you are willing to wait about 10 minutes, you can then use the computer. Otherwise it requires a reboot to recover. > This is annoying, but fairly out of the control of developpers of the affy pack. > And yeah, I know, whaddaya expect if you don't have things set up correctly? Just doing my part to make Bioconductor fool proof ;-D > The mecanism involved is an attempt to free people from the trouble of taking care of (some of) the logistics. It should be fool proof (especially with Win XP ? ;) ). Thanks for the report. L.

Hi Laurent, Ben Bolstad had me try the newest version, and it smoothly downloads the required environment and goes on from there, so no need to check it out. Thanks, Jim James W. MacDonald UMCCC Microarray Core Facility 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 >>> Laurent Gautier <laurent@cbs.dtu.dk> 04/30/03 11:21AM >>> On Wed, Apr 30, 2003 at 10:42:12AM -0400, James MacDonald wrote: > Hi, > > I just wanted to let the affy developers know that the devel version of affy completely bombs out if you don't have the requisite cdf environment installed. > > Details: > > OS: Win XP > R: 1.7.0 > Affy: 1.2.10 > Biobase: 1.3.17 > tkWidgets: 1.2.13 <- I think these are the only two version-level dependencies I believe there is also a dependency towards 'reposTools' (although no version number is specified). Which version do you use ? > > If I read in some e.g., RG-U34A chips without the rgu34acdf environment installed, and then try to do anything with the AffyBatch, I get the following error: > > Error in how[[i]] : subscript out of bounds > Error in inherits(cdf.env, "try-error") : Object "cdf.env" not found It looks like a silly bug in the 'autodownload' mecanism (one more). I'll look at that. > > If you are willing to wait about 10 minutes, you can then use the computer. Otherwise it requires a reboot to recover. > This is annoying, but fairly out of the control of developpers of the affy pack. > And yeah, I know, whaddaya expect if you don't have things set up correctly? Just doing my part to make Bioconductor fool proof ;-D > The mecanism involved is an attempt to free people from the trouble of taking care of (some of) the logistics. It should be fool proof (especially with Win XP ? ;) ). Thanks for the report. L. _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor