Having trouble with netCDF since I upgraded to 1.2.1 (did directly from 1.1.1). Package was recompiled after the upgrade. Symptoms: open.netCDF returns cleanly, but read.netCDF causes R to segfault and dump core. What else do you need me to write to help? --please do not edit the information below-- Version: platform = i586-pc-linux-gnu arch = i586 os = linux-gnu system = i586, linux-gnu status = major = 1 minor = 2.1 year = 2001 month = 01 day = 15 language = R Search Path: .GlobalEnv, package:netCDF, package:RODBC, package:ts, package:local, package:ctest, Autoloads, package:base -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Thu, 25 Jan 2001 jasont@indigoindustrial.co.nz wrote:> Having trouble with netCDF since I upgraded to 1.2.1 (did directly > from 1.1.1). Package was recompiled after the upgrade.You did use netCDF_1.1 for 1.2.1 did you. I wonder? netCDF_1.0-1 will not work.> Symptoms: > open.netCDF returns cleanly, but read.netCDF causes R to segfault > and dump core. What else do you need me to write to help?The first item is to report that it had passed R CMD check netCDF? That uses read.netCDF. If so the problem is data-specific, if not it is with your installation. So we either need the data or a lot more about your installation of netCDF. -- Brian D. Ripley, ripley@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
First and foremost, regarding:> Please do not send bug reports to R-bugs and R-help: the first > set goes to R-devel too which means that the active developers see it > twice (and are probably twice as likely to ignore it, as I will next time).My appologies. I'm well aware that R is a labour of love for those involved, and I truly appreciate what they've given the rest of us. The last thing I want to do is irritate any of you. Now trying to pull foot from mouth. And making every note to not do this sort of thing in future. Thanks for the quick reply with information.> You did use netCDF_1.1 for 1.2.1 did you.Yes.> The first item is to report that it had passed R CMD check netCDF?Output from that is.... ---Begin ~/R/netCDF.Rcheck/00check.log --- * using log directory /home/jasont/R/netCDF.Rcheck * checking for file `netCDF/DESCRIPTION' ... OK * checking package directory ... OK * checking DESCRIPTION Package field ... OK * checking DESCRIPTION Version field ... OK * checking DESCRIPTION Title field ... OK * checking DESCRIPTION License field ... OK * checking DESCRIPTION Author field ... OK * checking DESCRIPTION Maintainer field ... WARNING * no DESCRIPTION Maintainer field found * checking Rd files ... OK * checking for undocumented objects ... OK * creating netCDF-Ex.R ... OK * checking examples ... OK * creating netCDF-manual.tex ... OK * checking netCDF-manual.tex ... OK ---End ~/R/netCDF.Rcheck/00check.log --- Looks good to me. As you've probably guessed, I'm new to R, so take the following with the requisite number of grains of salt: in the R-ex directory, the read.netCDF.R file contains either blank lines or commented lines, and nothing else. There might be some keywords there for the example software? ---Begin ~/R/netCDF.Rcheck/netCDF/R-ex/read.netCDF.R --- ###--- >>> `read.netCDF' <<<----- Read subsets of a netCDF file ## alias help(read.netCDF) ##___ Examples ___: ## Keywords: 'file'. ---End ~/R/netCDF.Rcheck/netCDF/R-ex/read.netCDF.R --- That's it. NetCDF installation: version 3.4, built from source. Other NetCDF utilities continue to work nicely, which is how I'm working around the issue. Now, I'm taking existing netCDF files, converting them to csv files, and either reading them directly, or loading them to PostgreSQL, and using RPgSQL or RODBC. Conversion to csv is done using home-brewed c programs which were dynamically linked against libnetcdf.so. As such, I'm pretty sure the library is still working fine. I have a nice sample netCDF file, but I doubt it's appropriate for the bugs list (waste of bandwidth). It's just under 2K. -- Indigo Industrial Controls Ltd. 64-21-343-545 jasont@indigoindustrial.co.nz -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
First and foremost, regarding:> Please do not send bug reports to R-bugs and R-help: the first > set goes to R-devel too which means that the active developers see it > twice (and are probably twice as likely to ignore it, as I will next time).My appologies. I'm well aware that R is a labour of love for those involved, and I truly appreciate what they've given the rest of us. The last thing I want to do is irritate any of you. Now trying to pull foot from mouth. And making every note to not do this sort of thing in future. Thanks for the quick reply with information.> You did use netCDF_1.1 for 1.2.1 did you.Yes.> The first item is to report that it had passed R CMD check netCDF?Output from that is.... ---Begin ~/R/netCDF.Rcheck/00check.log --- * using log directory /home/jasont/R/netCDF.Rcheck * checking for file `netCDF/DESCRIPTION' ... OK * checking package directory ... OK * checking DESCRIPTION Package field ... OK * checking DESCRIPTION Version field ... OK * checking DESCRIPTION Title field ... OK * checking DESCRIPTION License field ... OK * checking DESCRIPTION Author field ... OK * checking DESCRIPTION Maintainer field ... WARNING * no DESCRIPTION Maintainer field found * checking Rd files ... OK * checking for undocumented objects ... OK * creating netCDF-Ex.R ... OK * checking examples ... OK * creating netCDF-manual.tex ... OK * checking netCDF-manual.tex ... OK ---End ~/R/netCDF.Rcheck/00check.log --- Looks good to me. As you've probably guessed, I'm new to R, so take the following with the requisite number of grains of salt: in the R-ex directory, the read.netCDF.R file contains either blank lines or commented lines, and nothing else. There might be some keywords there for the example software? ---Begin ~/R/netCDF.Rcheck/netCDF/R-ex/read.netCDF.R --- ###--- >>> `read.netCDF' <<<----- Read subsets of a netCDF file ## alias help(read.netCDF) ##___ Examples ___: ## Keywords: 'file'. ---End ~/R/netCDF.Rcheck/netCDF/R-ex/read.netCDF.R --- That's it. NetCDF installation: version 3.4, built from source. Other NetCDF utilities continue to work nicely, which is how I'm working around the issue. Now, I'm taking existing netCDF files, converting them to csv files, and either reading them directly, or loading them to PostgreSQL, and using RPgSQL or RODBC. Conversion to csv is done using home-brewed c programs which were dynamically linked against libnetcdf.so. As such, I'm pretty sure the library is still working fine. I have a nice sample netCDF file, but I doubt it's appropriate for the bugs list (waste of bandwidth). It's just under 2K. -- Indigo Industrial Controls Ltd. 64-21-343-545 jasont@indigoindustrial.co.nz -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Another interesting point - read.netCDF does *not* cause R to dump core if debug(read.netCDF) is on. To be clear:> rm(list=ls(all=T)) > library(netCDF)# --- works ---> debug(read.netCDF) > fd<-open.netCDF("data.nc") > z<-read.netCDF(fd,"foo")--- lots of stuff deleted ---> length(z)[1] 143 --- what it should be> close.netCDF(fd)[1] TRUE #--- end works --- That works fine. Carrying on in the same session: # --- doesn't work ---> rm(z,fd) > undebug(read.netCDF) > fd<-open.netCDF("data.nc") > z<-read.netCDF(fd,"foo")Segmentation fault (core dumped) # --- end doesn't work --- Now more puzzled. Jason Turner -- Indigo Industrial Controls Ltd. 64-21-343-545 jasont@indigoindustrial.co.nz -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Thu, 25 Jan 2001 jasont@indigoindustrial.co.nz wrote:> NetCDF installation: version 3.4, built from source. > Other NetCDF utilities continue to work nicely, which > is how I'm working around the issue. Now, I'm taking > existing netCDF files, converting them to csv files, > and either reading them directly, or loading them > to PostgreSQL, and using RPgSQL or RODBC. Conversion > to csv is done using home-brewed c programs which > were dynamically linked against libnetcdf.so. As > such, I'm pretty sure the library is still working fine. > > I have a nice sample netCDF file, but I doubt it's appropriate > for the bugs list (waste of bandwidth). It's just under 2K.I think the thing to do now is to contact the netCDF package maintainer (Thomas Lumley) with that example. As I am sure you appreciate we need to eliminate the simple causes first, and they are common. -- Brian D. Ripley, ripley@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._