[Rd] grDevice in R HEAD been broken for 6 weeks.
geDevice has been failing check for 6 weeks now with --enable-strict-barrier , bisected to: r69049 | murrell | 2015-08-14 00:03:12 +0100 (Fri, 14 Aug 2015) | 2 lines first hack at adding grid display list to recorded plot objects, so can add further grid drawing following a replayPlot() It seems to be broken by the GE_SaveSnapshotState addition in src/library/grid/src/state.c: case GE_SaveSnapshotState: +/* + * Save the current 'grid' DL. + */ +PROTECT(result = allocVector(VECSXP, 3)); +SET_VECTOR_ELT(result, 0, gridStateElement(dd, GSS_DL)); +SET_VECTOR_ELT(result, 1, gridStateElement(dd, GSS_DLINDEX)); +UNPROTECT(1); break; > showCols2() Loading required package: grid Error in grDevices:::recordPalette() : LOGICAL() can only be applied to a 'logical', not a 'list' Calls: demo ... eval -> eval -> showCols2 -> grid.newpage -> Execution halted somewhere else it seems to expect the outcome of GE_SaveSnapshotState to be a logical array rather than a list, though I cannot find where it is. Any chance of that getting fixed any time soon? __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] issues with dev.new avoiding RStudio plot device on unix?
Same problem here on Mac OS X 10.10.5 with R 3.2.2 and RStudio 0.99.473. I think dev.new() tries to find a suitable device in an interactive session with this code dsp <- Sys.getenv("DISPLAY") if (.Platform$OS.type == "windows") windows else if (.Platform$GUI == "AQUA" || ((!nzchar(dsp) || grepl("^/tmp/launch-", dsp)) && .Call(C_makeQuartzDefault))) quartz else if (nzchar(dsp) && .Platform$GUI %in% c("X11", "Tk")) X11 else defdev which fails for me because RStudio sets .Platform$GUI to "RStudio" (thanks, RStudio!) and my DISPLAY variable looks like this /private/tmp/com.apple.launchd.2wKas4wzPe/org.macosforge.xquartz:0 so the grepl() on dsp doesn't match. Best, Stefan > On 26 Sep 2015, at 06:42, Skye Bender-deMoll wrote: > > Sorry, should have given more background. x11 works fine on all my systems > when called by x11(). I'm the maintainer of a package that uses the > animation library, which has performance issues when used with the RStudio > plot device. But if you call plot.new() when using RStudio, you get an > RStudio device, not the standard device for the platform because it overrides > the device option. So I've had to have the library do platform detection and > platform-specific device calls, which R CMD check doesn't like. I believe > that noRStudioGD argument was avoided to give users a way around this, but it > doesn't seem to be behaving correctly in the unix interactive case. __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] Hidden files problem in R CMD check
Dear all, When running R CMD check on my Mac (Yosemite 10.10.5) I get suddenly NOTEs and WARNINGs that my tar.gz file does contain hidden files. However, when checking the corresponding directories with 'ls -al' (or with mc) none of these files does exist!! (with exception .BBSoptions). Here is a partial output of 00check.log: * using log directory '/Volumes/LumiData/CRAN/xps.Rcheck' * using R version 3.2.0 (2015-04-16) * using platform: x86_64-apple-darwin13.4.0 (64-bit) * using session charset: ASCII * checking for file 'xps/DESCRIPTION' ... OK * this is package 'xps' version '1.29.2' * checking package namespace information ... OK * checking package dependencies ... OK * checking if this is a source package ... OK * checking if there is a namespace ... OK * checking for executable files ... OK * checking for hidden files and directories ... NOTE Found the following hidden files and directories: .BBSoptions ._.BBSoptions ._DESCRIPTION ._NAMESPACE ._R ._README ._build ._inst ._man ._src ._vignettes R/._AffyRNAdeg.R R/._Constructors.R R/._TreeSetClasses.R R/._bgcorrect.R R/._dabg.call.R ... man/._AffyRNAdeg.Rd man/._AnalysisTreeSet-class.Rd man/._CallTreeSet-class.Rd man/._DataTreeSet-class.Rd man/._ExprTreeSet-class.Rd ... src/._rwrapper.cxx src/._rwrapper.h src/._xps.def src/._xpsLinkDef.h vignettes/._APTvsXPS.Rnw vignettes/._BoxPlot_BgrdGC_bgrd.png vignettes/._BoxPlot_BgrdMAS5_bgrd.png vignettes/._BoxPlot_BgrdRMA.png ... * checking package subdirectories ... WARNING Subdirectory 'R' contains invalid file names: '._AffyRNAdeg.R' '._Constructors.R' '._TreeSetClasses.R' '._bgcorrect.R' '._dabg.call.R' '._dfw.R' '._export.R' '._express.R' ... Subdirectory 'man' contains invalid file names: '._AffyRNAdeg.Rd' '._AnalysisTreeSet-class.Rd' '._CallTreeSet-class.Rd' '._DataTreeSet-class.Rd' '._ExprTreeSet-class.Rd' '._Filter-class.Rd' ... Does anyone know why these files are listed? For some reason all these hidden files start with '._' My suspicion was that maybe there is something wrong with Spotlight (i.e. mdutil) so I excluded the whole volume from Spotlight. This did not help. Adding the volume again and thus forcing a rebuild for Spotlight did not help either. What could I do to remove these apparently non-existing hidden files? Best regards Christian _._._._._._._._._._._._._._._._._._ C.h.r.i.s.t.i.a.n S.t.r.a.t.o.w.a V.i.e.n.n.a A.u.s.t.r.i.a e.m.a.i.l:cstrato at aon.at _._._._._._._._._._._._._._._._._._ __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
Christian, those are resource forks from your files - possibly maintained by your (likey very old?) editor. First, is seems that you are not creating the tar ball correctly - the correct way is to use R CMD build which should not include resource forks nor files on the ignore list. That said, if for some reason you intend to create is by hand (not recommended), you may want to set COPYFILE_DISABLE=1 so it doesn't include resource forks. Cheers, Simon On Sep 26, 2015, at 3:30 PM, cstrato wrote: > Dear all, > > When running R CMD check on my Mac (Yosemite 10.10.5) I get suddenly NOTEs > and WARNINGs > that my tar.gz file does contain hidden files. > > However, when checking the corresponding directories with 'ls -al' (or with > mc) > none of these files does exist!! (with exception .BBSoptions). > > Here is a partial output of 00check.log: > > * using log directory '/Volumes/LumiData/CRAN/xps.Rcheck' > * using R version 3.2.0 (2015-04-16) > * using platform: x86_64-apple-darwin13.4.0 (64-bit) > * using session charset: ASCII > * checking for file 'xps/DESCRIPTION' ... OK > * this is package 'xps' version '1.29.2' > * checking package namespace information ... OK > * checking package dependencies ... OK > * checking if this is a source package ... OK > * checking if there is a namespace ... OK > * checking for executable files ... OK > * checking for hidden files and directories ... NOTE > Found the following hidden files and directories: > .BBSoptions > ._.BBSoptions > ._DESCRIPTION > ._NAMESPACE > ._R > ._README > ._build > ._inst > ._man > ._src > ._vignettes > R/._AffyRNAdeg.R > R/._Constructors.R > R/._TreeSetClasses.R > R/._bgcorrect.R > R/._dabg.call.R > ... > man/._AffyRNAdeg.Rd > man/._AnalysisTreeSet-class.Rd > man/._CallTreeSet-class.Rd > man/._DataTreeSet-class.Rd > man/._ExprTreeSet-class.Rd > ... > src/._rwrapper.cxx > src/._rwrapper.h > src/._xps.def > src/._xpsLinkDef.h > vignettes/._APTvsXPS.Rnw > vignettes/._BoxPlot_BgrdGC_bgrd.png > vignettes/._BoxPlot_BgrdMAS5_bgrd.png > vignettes/._BoxPlot_BgrdRMA.png > ... > * checking package subdirectories ... WARNING > Subdirectory 'R' contains invalid file names: > '._AffyRNAdeg.R' '._Constructors.R' '._TreeSetClasses.R' > '._bgcorrect.R' '._dabg.call.R' '._dfw.R' '._export.R' '._express.R' > ... > Subdirectory 'man' contains invalid file names: > '._AffyRNAdeg.Rd' '._AnalysisTreeSet-class.Rd' > '._CallTreeSet-class.Rd' '._DataTreeSet-class.Rd' > '._ExprTreeSet-class.Rd' '._Filter-class.Rd' > ... > > > Does anyone know why these files are listed? > For some reason all these hidden files start with '._' > > My suspicion was that maybe there is something wrong with Spotlight (i.e. > mdutil) > so I excluded the whole volume from Spotlight. This did not help. Adding the > volume > again and thus forcing a rebuild for Spotlight did not help either. > > What could I do to remove these apparently non-existing hidden files? > > Best regards > Christian > _._._._._._._._._._._._._._._._._._ > C.h.r.i.s.t.i.a.n S.t.r.a.t.o.w.a > V.i.e.n.n.a A.u.s.t.r.i.a > e.m.a.i.l:cstrato at aon.at > _._._._._._._._._._._._._._._._._._ > > __ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel > __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
Dear Simon, Thank you very much for your help, it did solve my problems!! Great! I have googled COPYFILE_DISABLE and found the following site which does explain the issue with tar on Mac OS X, see: http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files Instead of doing: $tar czf xps_1.29.2.tar.gz xps I did now: $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps Running: $R CMD check xps_1.29.2.tar.gz now leaves only '.BBSoptions' as hidden file. However, I must admit that I do not understand it. Since more than ten years I am running the 'tar' command on my Macs (using Leopard, Snow Leopard, Lion and now Yosemite) and running R CMD check did never give these warning messages. Even two weeks ago everything was ok. Could it be that upgrading from OS X 10.10.3 to 10.10.5 did change the behaviour? (However, the link above is already from 2011) Maybe one word regarding the editor I am using: Yes, I am using 'nedit-5.5-Darwin-i386/nedit' which is pretty old, but in my opinion the best and fastest editor I know: - When I open a C++ file containing more than 10,000 lines it opens it immediately with all C++ instructions highlighted immediately (in contrast to emacs) - I can open the huge (>500 MB) Affymetrix annotation files containing more than one million lines in few seconds and scroll fast from beginning to end - It does even have a Language Mode for R I do not believe that 'nedit' does create these hidden files, since it never did. Thank you once again. Best regards, Christian On 09/26/15 21:44, Simon Urbanek wrote: Christian, those are resource forks from your files - possibly maintained by your (likey very old?) editor. First, is seems that you are not creating the tar ball correctly - the correct way is to use R CMD build which should not include resource forks nor files on the ignore list. That said, if for some reason you intend to create is by hand (not recommended), you may want to set COPYFILE_DISABLE=1 so it doesn't include resource forks. Cheers, Simon On Sep 26, 2015, at 3:30 PM, cstrato wrote: Dear all, When running R CMD check on my Mac (Yosemite 10.10.5) I get suddenly NOTEs and WARNINGs that my tar.gz file does contain hidden files. However, when checking the corresponding directories with 'ls -al' (or with mc) none of these files does exist!! (with exception .BBSoptions). Here is a partial output of 00check.log: * using log directory '/Volumes/LumiData/CRAN/xps.Rcheck' * using R version 3.2.0 (2015-04-16) * using platform: x86_64-apple-darwin13.4.0 (64-bit) * using session charset: ASCII * checking for file 'xps/DESCRIPTION' ... OK * this is package 'xps' version '1.29.2' * checking package namespace information ... OK * checking package dependencies ... OK * checking if this is a source package ... OK * checking if there is a namespace ... OK * checking for executable files ... OK * checking for hidden files and directories ... NOTE Found the following hidden files and directories: .BBSoptions ._.BBSoptions ._DESCRIPTION ._NAMESPACE ._R ._README ._build ._inst ._man ._src ._vignettes R/._AffyRNAdeg.R R/._Constructors.R R/._TreeSetClasses.R R/._bgcorrect.R R/._dabg.call.R ... man/._AffyRNAdeg.Rd man/._AnalysisTreeSet-class.Rd man/._CallTreeSet-class.Rd man/._DataTreeSet-class.Rd man/._ExprTreeSet-class.Rd ... src/._rwrapper.cxx src/._rwrapper.h src/._xps.def src/._xpsLinkDef.h vignettes/._APTvsXPS.Rnw vignettes/._BoxPlot_BgrdGC_bgrd.png vignettes/._BoxPlot_BgrdMAS5_bgrd.png vignettes/._BoxPlot_BgrdRMA.png ... * checking package subdirectories ... WARNING Subdirectory 'R' contains invalid file names: '._AffyRNAdeg.R' '._Constructors.R' '._TreeSetClasses.R' '._bgcorrect.R' '._dabg.call.R' '._dfw.R' '._export.R' '._express.R' ... Subdirectory 'man' contains invalid file names: '._AffyRNAdeg.Rd' '._AnalysisTreeSet-class.Rd' '._CallTreeSet-class.Rd' '._DataTreeSet-class.Rd' '._ExprTreeSet-class.Rd' '._Filter-class.Rd' ... Does anyone know why these files are listed? For some reason all these hidden files start with '._' My suspicion was that maybe there is something wrong with Spotlight (i.e. mdutil) so I excluded the whole volume from Spotlight. This did not help. Adding the volume again and thus forcing a rebuild for Spotlight did not help either. What could I do to remove these apparently non-existing hidden files? Best regards Christian _._._._._._._._._._._._._._._._._._ C.h.r.i.s.t.i.a.n S.t.r.a.t.o.w.a V.i.e.n.n.a A.u.s.t.r.i.a e.m.a.i.l:cstrato at aon.at _._._._._._._._._._._._._._._._._._ __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
On 26 September 2015 at 22:41, cstrato wrote: | Dear Simon, | | Thank you very much for your help, it did solve my problems!! Great! | | I have googled COPYFILE_DISABLE and found the following site which does | explain the issue with tar on Mac OS X, see: | http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files | | Instead of doing: | $tar czf xps_1.29.2.tar.gz xps | | I did now: | $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps | | Running: | $R CMD check xps_1.29.2.tar.gz | now leaves only '.BBSoptions' as hidden file. No, still wrong. As Simon said, we all are supposed to use 'R CMD build xps' to create the tarball. "Back in the day ..." straight tar cfz ... worked, it more or less stopped _many_ years ago. Cf TheOneManualThatMatters: 1.3.1 Checking packages --- Using 'R CMD check', the R package checker, one can test whether _source_ R packages work correctly. It can be run on one or more directories, or compressed package 'tar' archives with extension '.tar.gz', '.tgz', '.tar.bz2' or '.tar.xz'. It is strongly recommended that the final checks are run on a 'tar' archive prepared by 'R CMD build'. Ie "It is strongly recommended ... 'tar' archive prepared by 'R CMD build'. Dirk -- http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
Dear Dirk, Yes, I know, however forget for one moment R. If I use tar independent of R it still should not create these hidden files. BTW, do you know where these hidden files are stored on the Mac? Best regards, Christian On 09/26/15 23:01, Dirk Eddelbuettel wrote: On 26 September 2015 at 22:41, cstrato wrote: | Dear Simon, | | Thank you very much for your help, it did solve my problems!! Great! | | I have googled COPYFILE_DISABLE and found the following site which does | explain the issue with tar on Mac OS X, see: | http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files | | Instead of doing: | $tar czf xps_1.29.2.tar.gz xps | | I did now: | $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps | | Running: | $R CMD check xps_1.29.2.tar.gz | now leaves only '.BBSoptions' as hidden file. No, still wrong. As Simon said, we all are supposed to use 'R CMD build xps' to create the tarball. "Back in the day ..." straight tar cfz ... worked, it more or less stopped _many_ years ago. Cf TheOneManualThatMatters: 1.3.1 Checking packages --- Using 'R CMD check', the R package checker, one can test whether _source_ R packages work correctly. It can be run on one or more directories, or compressed package 'tar' archives with extension '.tar.gz', '.tgz', '.tar.bz2' or '.tar.xz'. It is strongly recommended that the final checks are run on a 'tar' archive prepared by 'R CMD build'. Ie "It is strongly recommended ... 'tar' archive prepared by 'R CMD build'. Dirk __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
On 26 September 2015 at 23:06, cstrato wrote: | Dear Dirk, | | Yes, I know, however forget for one moment R. No we can't. Your question was about to make R CMD check happy and 'quiet'. And one answer is to feed it a properly constructed file. | | If I use tar independent of R it still should not create these hidden | files. Hm, I doubt that. We have to tell R explicitly to exclude dotfiles, directories from version control etc pp. Recall the discussion years when .git was added to the already existing .svn? In any event, R is open source so you can check what 'build' does and follow its code into the tools package. | BTW, do you know where these hidden files are stored on the Mac? Pass -- I am too pedestrian for owing one of those machines ;-) Dirk -- http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
Dear Dirk, Please do not get me wrong. Yes, I am doing it the wrong way and I am doing it since about eight years. Nevertheless I am still allowed to be confused when something suddenly happens which did not happen before during all these years. Doing it correctly will solve the problem, but it will not explain this behavior which suddenly popped up. Best regards, Christian On 09/26/15 23:30, Dirk Eddelbuettel wrote: On 26 September 2015 at 23:06, cstrato wrote: | Dear Dirk, | | Yes, I know, however forget for one moment R. No we can't. Your question was about to make R CMD check happy and 'quiet'. And one answer is to feed it a properly constructed file. | | If I use tar independent of R it still should not create these hidden | files. Hm, I doubt that. We have to tell R explicitly to exclude dotfiles, directories from version control etc pp. Recall the discussion years when .git was added to the already existing .svn? In any event, R is open source so you can check what 'build' does and follow its code into the tools package. | BTW, do you know where these hidden files are stored on the Mac? Pass -- I am too pedestrian for owing one of those machines ;-) Dirk __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
On Sep 26, 2015, at 2:06 PM, cstrato wrote: > Dear Dirk, > > Yes, I know, however forget for one moment R. > > If I use tar independent of R it still should not create these hidden files. > > BTW, do you know where these hidden files are stored on the Mac? Your first posting showed which of several different directories they were in. Do you understand that any file whose name starts with a is called a "hidden file"? It is "hidden", i.e not displayed in a Finder window, from people who are using Finder.app unless you change the default settings. It's easy to look up the code that is needed to be pasted into a Terminal session. I never remember it. I just leave Finder set up to display these 'dotfiles' as they are also called. defaults write com.apple.finder AppleShowAllFiles YES killall Finder #The second command restarts Finder.app or you could try to restart the Finder by option (=alt) + rightclicking the Finder icon in the Dock and selecting Relaunch. -- David. > > Best regards, > Christian > > > On 09/26/15 23:01, Dirk Eddelbuettel wrote: >> >> On 26 September 2015 at 22:41, cstrato wrote: >> | Dear Simon, >> | >> | Thank you very much for your help, it did solve my problems!! Great! >> | >> | I have googled COPYFILE_DISABLE and found the following site which does >> | explain the issue with tar on Mac OS X, see: >> | >> http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files >> | >> | Instead of doing: >> | $tar czf xps_1.29.2.tar.gz xps >> | >> | I did now: >> | $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps >> | >> | Running: >> | $R CMD check xps_1.29.2.tar.gz >> | now leaves only '.BBSoptions' as hidden file. >> >> No, still wrong. As Simon said, we all are supposed to use 'R CMD build xps' >> to create the tarball. "Back in the day ..." straight tar cfz ... worked, it >> more or less stopped _many_ years ago. Cf TheOneManualThatMatters: >> >>1.3.1 Checking packages >>--- >> >>Using 'R CMD check', the R package checker, one can test whether >>_source_ R packages work correctly. It can be run on one or more >>directories, or compressed package 'tar' archives with extension >>'.tar.gz', '.tgz', '.tar.bz2' or '.tar.xz'. >> >> It is strongly recommended that the final checks are run on a 'tar' >>archive prepared by 'R CMD build'. >> >> Ie "It is strongly recommended ... 'tar' archive prepared by 'R CMD build'. >> >> Dirk >> > > __ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel David Winsemius Alameda, CA, USA __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
Hi Christian, This seems like a question about OSX rather than R. You will probably have more luck asking on an apple forum. Or just google: http://bfy.tw/1zhP Best, Ista On Sep 26, 2015 8:39 PM, "David Winsemius" wrote: > > On Sep 26, 2015, at 2:06 PM, cstrato wrote: > > > Dear Dirk, > > > > Yes, I know, however forget for one moment R. > > > > If I use tar independent of R it still should not create these hidden > files. > > > > BTW, do you know where these hidden files are stored on the Mac? > > Your first posting showed which of several different directories they were > in. Do you understand that any file whose name starts with a is > called a "hidden file"? It is "hidden", i.e not displayed in a Finder > window, from people who are using Finder.app unless you change the default > settings. It's easy to look up the code that is needed to be pasted into a > Terminal session. I never remember it. I just leave Finder set up to > display these 'dotfiles' as they are also called. > > defaults write com.apple.finder AppleShowAllFiles YES > > killall Finder > > #The second command restarts Finder.app or you could try to restart the > Finder by option (=alt) + rightclicking the Finder icon in the Dock and > selecting Relaunch. > > -- > David. > > > > > Best regards, > > Christian > > > > > > On 09/26/15 23:01, Dirk Eddelbuettel wrote: > >> > >> On 26 September 2015 at 22:41, cstrato wrote: > >> | Dear Simon, > >> | > >> | Thank you very much for your help, it did solve my problems!! Great! > >> | > >> | I have googled COPYFILE_DISABLE and found the following site which > does > >> | explain the issue with tar on Mac OS X, see: > >> | > http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files > >> | > >> | Instead of doing: > >> | $tar czf xps_1.29.2.tar.gz xps > >> | > >> | I did now: > >> | $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps > >> | > >> | Running: > >> | $R CMD check xps_1.29.2.tar.gz > >> | now leaves only '.BBSoptions' as hidden file. > >> > >> No, still wrong. As Simon said, we all are supposed to use 'R CMD build > xps' > >> to create the tarball. "Back in the day ..." straight tar cfz ... > worked, it > >> more or less stopped _many_ years ago. Cf TheOneManualThatMatters: > >> > >>1.3.1 Checking packages > >>--- > >> > >>Using 'R CMD check', the R package checker, one can test whether > >>_source_ R packages work correctly. It can be run on one or more > >>directories, or compressed package 'tar' archives with extension > >>'.tar.gz', '.tgz', '.tar.bz2' or '.tar.xz'. > >> > >> It is strongly recommended that the final checks are run on a > 'tar' > >>archive prepared by 'R CMD build'. > >> > >> Ie "It is strongly recommended ... 'tar' archive prepared by 'R CMD > build'. > >> > >> Dirk > >> > > > > __ > > R-devel@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-devel > > David Winsemius > Alameda, CA, USA > > __ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel > [[alternative HTML version deleted]] __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Hidden files problem in R CMD check
On Sep 26, 2015, at 5:06 PM, cstrato wrote: > Dear Dirk, > > Yes, I know, however forget for one moment R. > > If I use tar independent of R it still should not create these hidden files. > > BTW, do you know where these hidden files are stored on the Mac? > Please consider reading my original reply - those are not actual files (all but .BBSoptions which is an actual file you can see with ls -a), those are just resource forks of files [or extended attributes in general] -- and Apple tar if instructed to preserve resource forks encodes them as ._* in tar balls since tar has no native way of storing resource forks. And, again, as I said in my reply those likely come from some software you may have used (possibly the editor). For example, some editors store the file encoding in xattr. To list them you can use ls -l@ Cheers, Simon > Best regards, > Christian > > > On 09/26/15 23:01, Dirk Eddelbuettel wrote: >> >> On 26 September 2015 at 22:41, cstrato wrote: >> | Dear Simon, >> | >> | Thank you very much for your help, it did solve my problems!! Great! >> | >> | I have googled COPYFILE_DISABLE and found the following site which does >> | explain the issue with tar on Mac OS X, see: >> | >> http://unix.stackexchange.com/questions/9665/create-tar-archive-of-a-directory-except-for-hidden-files >> | >> | Instead of doing: >> | $tar czf xps_1.29.2.tar.gz xps >> | >> | I did now: >> | $COPYFILE_DISABLE=1 tar czf xps_1.29.2.tar.gz xps >> | >> | Running: >> | $R CMD check xps_1.29.2.tar.gz >> | now leaves only '.BBSoptions' as hidden file. >> >> No, still wrong. As Simon said, we all are supposed to use 'R CMD build xps' >> to create the tarball. "Back in the day ..." straight tar cfz ... worked, it >> more or less stopped _many_ years ago. Cf TheOneManualThatMatters: >> >>1.3.1 Checking packages >>--- >> >>Using 'R CMD check', the R package checker, one can test whether >>_source_ R packages work correctly. It can be run on one or more >>directories, or compressed package 'tar' archives with extension >>'.tar.gz', '.tgz', '.tar.bz2' or '.tar.xz'. >> >> It is strongly recommended that the final checks are run on a 'tar' >>archive prepared by 'R CMD build'. >> >> Ie "It is strongly recommended ... 'tar' archive prepared by 'R CMD build'. >> >> Dirk >> > __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel