[Rd] R 3.6.2 is released
The build system rolled up R-3.6.2.tar.gz (codename "Dark and Stormy Night") this morning. The list below details the changes in this release. You can get the source code from http://cran.r-project.org/src/base/R-3/R-3.6.2.tar.gz or wait for it to be mirrored at a CRAN site nearer to you. Binaries for various platforms will appear in due course. For the R Core Team, Peter Dalgaard These are the checksums (md5 and SHA-256) for the freshly created files, in case you wish to check that they are uncorrupted: MD5 (AUTHORS) = b9c44f9f78cab3184ad9898bebc854b4 MD5 (COPYING) = eb723b61539feef013de476e68b5c50a MD5 (COPYING.LIB) = a6f89e2100d9b6cdffcea4f398e37343 MD5 (FAQ) = 28a3942a7129877e9af1d5ea16202052 MD5 (INSTALL) = 7893f754308ca31f1ccf62055090ad7b MD5 (NEWS) = 45437b38c75e0248b527c00e6d42ee6a MD5 (NEWS.0) = bfcd7c147251b5474d96848c6f57e5a8 MD5 (NEWS.1) = eb78c4d053ec9c32b815cf0c2ebea801 MD5 (NEWS.2) = 591dcf615162127f904e4e461f330ce9 MD5 (R-latest.tar.gz) = 90d23d138cee26d275da14b58296e521 MD5 (README) = f468f281c919665e276a1b691decbbe6 MD5 (RESOURCES) = 529223fd3ffef95731d0a87353108435 MD5 (THANKS) = bb45f89c01d509721c47fd41f147da60 MD5 (VERSION-INFO.dcf) = 9c33701e25092aefc1d16beb5858f20f MD5 (R-3/R-3.6.2.tar.gz) = 90d23d138cee26d275da14b58296e521 2cde824a7b18958e5f06b391c801c8288be0f84fa8934b7ddefef23c67e60c09 AUTHORS e6d6a009505e345fe949e1310334fcb0747f28dae2856759de102ab66b722cb4 COPYING 6095e9ffa777dd22839f7801aa845b31c9ed07f3d6bf8a26dc5d2dec8ccc0ef3 COPYING.LIB 38219d9c6221ccfbf075ef03711b420a1aa8731f890c8f2337148b602a217c2d FAQ f87461be6cbaecc4dce44ac58e5bd52364b0491ccdadaf846cb9b452e9550f31 INSTALL 0ceb6fbab3e0e29bc374683fd5c2ccd0c9c62ce8eca2a394a4603775b3ef129c NEWS 4e21b62f515b749f80997063fceab626d7258c7d650e81a662ba8e0640f12f62 NEWS.0 12b30c724117b1b2b11484673906a6dcd48a361f69fc420b36194f9218692d01 NEWS.1 ca04f78ffe54afa326fe3ed40e7e1411acaed2fa5ead97ddf51c6aa5b7bc NEWS.2 bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954 R-latest.tar.gz 2fdd3e90f23f32692d4b3a0c0452f2c219a10882033d1774f8cadf25886c3ddc README 408737572ecc6e1135fdb2cf7a9dbb1a6cb27967c757f1771b8c39d1fd2f1ab9 RESOURCES 2a8dca916cd92229ef9e328f3610ca204809c262823b860252b42072dac2473a THANKS 40cc7cea5f0e67cf8f2f7b25a534ae6bc53f38eae2ab2c2649a952ed37f0654a VERSION-INFO.dcf bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954 R-3/R-3.6.2.tar.gz This is the relevant part of the NEWS file CHANGES IN R 3.6.2: NEW FEATURES: * runmed(x, *) gains a new option na.action determining _how_ to handle NaN or NA in x. * dotchart() gains new options ann, xaxt, frame.plot and log. INSTALLATION on a UNIX-ALIKE: * Detection of the C stack direction has been moved from run-time to configure: this is safer with LTO builds and allows the detection to be overridden - see file config.site. * Source-code changes enable installation on platforms using gcc -fno-common (the expected default for gcc 10.x). C-LEVEL FACILITIES: * installTrChar (which is nowadays is wrapped by installChar) is defined in Rinternals.h. (Neither are part of the API.) PACKAGE INSTALLATION: * Header Rconfig.h contains the value of FC_LEN_T deduced at installation which is used by the prototypes in headers R_ext/BLAS.h and R_ext/Lapack.h but to avoid extensive breakage this is only exposed when USE_FC_LEN_T is defined. If a package's C/C++ calls to BLAS/LAPACK allow for the 'hidden' arguments used by most Fortran compilers to pass the lengths of Fortran character arguments, define USE_FC_LEN_T and include Rconfig.h (possibly _via_ R.h) before including R_ext/BLAS.h or R_ext/Lapack.h. * A package with Fortran source code and perhaps C (but not C++) sources can request for its shared object/DLL to be linked by the Fortran compiler by including a line USE_FC_TO_LINK= in src/Makevars[.win] and using $(SHLIB_OPENMP_FFLAGS) as part of PKG_LIBS. The known reason for doing so is a package which uses Fortran (only) OpenMP on a platform where the Fortran OpenMP runtime is incompatible with the C one (e.g. gfortran 9.x with clang). UTILITIES: * R CMD check has a new option to mitigate checks leaving files/directories in /tmp. See the 'R Internals' manual - this is part of --as-cran. Windows: * The default standard for C++ in package installation is C++11 (as it has been on other platforms where available since R 3.6.0: the default toolchain on Windows was defaulting to C++98). DEPRECATED AND DEFUNCT: * Support for specifying C++98 in package installation is deprecated. * Support in R CMD config for F77, FCPIFCPLAGS, CPP, CXXCPP and CXX98 and similar is deprecated. (CPP is found from the system make and may well not be set.) Use $CC -E and $CXX -E instead of CPP and CXXCPP. BUG FIXES:
Re: [Rd] R 3.6.2 is released
Hi. Under R-news there is an entry for 3.6.2 patched regarding LAPACK. However, when uncompresding the current R-patched, it creates R-Rc directories. Is this a naming oversight or is the patched version actually the unadjusted release candidate? Thank you, Avi On Thu, Dec 12, 2019 at 4:58 AM Peter Dalgaard via R-devel < r-devel@r-project.org> wrote: > The build system rolled up R-3.6.2.tar.gz (codename "Dark and Stormy > Night") this morning. > > The list below details the changes in this release. > > You can get the source code from > > http://cran.r-project.org/src/base/R-3/R-3.6.2.tar.gz > > or wait for it to be mirrored at a CRAN site nearer to you. > > Binaries for various platforms will appear in due course. > > > For the R Core Team, > > Peter Dalgaard > > These are the checksums (md5 and SHA-256) for the freshly created files, > in case you wish > to check that they are uncorrupted: > > MD5 (AUTHORS) = b9c44f9f78cab3184ad9898bebc854b4 > MD5 (COPYING) = eb723b61539feef013de476e68b5c50a > MD5 (COPYING.LIB) = a6f89e2100d9b6cdffcea4f398e37343 > MD5 (FAQ) = 28a3942a7129877e9af1d5ea16202052 > MD5 (INSTALL) = 7893f754308ca31f1ccf62055090ad7b > MD5 (NEWS) = 45437b38c75e0248b527c00e6d42ee6a > MD5 (NEWS.0) = bfcd7c147251b5474d96848c6f57e5a8 > MD5 (NEWS.1) = eb78c4d053ec9c32b815cf0c2ebea801 > MD5 (NEWS.2) = 591dcf615162127f904e4e461f330ce9 > MD5 (R-latest.tar.gz) = 90d23d138cee26d275da14b58296e521 > MD5 (README) = f468f281c919665e276a1b691decbbe6 > MD5 (RESOURCES) = 529223fd3ffef95731d0a87353108435 > MD5 (THANKS) = bb45f89c01d509721c47fd41f147da60 > MD5 (VERSION-INFO.dcf) = 9c33701e25092aefc1d16beb5858f20f > MD5 (R-3/R-3.6.2.tar.gz) = 90d23d138cee26d275da14b58296e521 > > > 2cde824a7b18958e5f06b391c801c8288be0f84fa8934b7ddefef23c67e60c09 AUTHORS > e6d6a009505e345fe949e1310334fcb0747f28dae2856759de102ab66b722cb4 COPYING > 6095e9ffa777dd22839f7801aa845b31c9ed07f3d6bf8a26dc5d2dec8ccc0ef3 > COPYING.LIB > 38219d9c6221ccfbf075ef03711b420a1aa8731f890c8f2337148b602a217c2d FAQ > f87461be6cbaecc4dce44ac58e5bd52364b0491ccdadaf846cb9b452e9550f31 INSTALL > 0ceb6fbab3e0e29bc374683fd5c2ccd0c9c62ce8eca2a394a4603775b3ef129c NEWS > 4e21b62f515b749f80997063fceab626d7258c7d650e81a662ba8e0640f12f62 NEWS.0 > 12b30c724117b1b2b11484673906a6dcd48a361f69fc420b36194f9218692d01 NEWS.1 > ca04f78ffe54afa326fe3ed40e7e1411acaed2fa5ead97ddf51c6aa5b7bc NEWS.2 > bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954 > R-latest.tar.gz > 2fdd3e90f23f32692d4b3a0c0452f2c219a10882033d1774f8cadf25886c3ddc README > 408737572ecc6e1135fdb2cf7a9dbb1a6cb27967c757f1771b8c39d1fd2f1ab9 RESOURCES > 2a8dca916cd92229ef9e328f3610ca204809c262823b860252b42072dac2473a THANKS > 40cc7cea5f0e67cf8f2f7b25a534ae6bc53f38eae2ab2c2649a952ed37f0654a > VERSION-INFO.dcf > bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954 > R-3/R-3.6.2.tar.gz > > This is the relevant part of the NEWS file > > CHANGES IN R 3.6.2: > > NEW FEATURES: > > * runmed(x, *) gains a new option na.action determining _how_ to > handle NaN or NA in x. > > * dotchart() gains new options ann, xaxt, frame.plot and log. > > INSTALLATION on a UNIX-ALIKE: > > * Detection of the C stack direction has been moved from run-time > to configure: this is safer with LTO builds and allows the > detection to be overridden - see file config.site. > > * Source-code changes enable installation on platforms using gcc > -fno-common (the expected default for gcc 10.x). > > C-LEVEL FACILITIES: > > * installTrChar (which is nowadays is wrapped by installChar) is > defined in Rinternals.h. (Neither are part of the API.) > > PACKAGE INSTALLATION: > > * Header Rconfig.h contains the value of FC_LEN_T deduced at > installation which is used by the prototypes in headers > R_ext/BLAS.h and R_ext/Lapack.h but to avoid extensive breakage > this is only exposed when USE_FC_LEN_T is defined. > > If a package's C/C++ calls to BLAS/LAPACK allow for the 'hidden' > arguments used by most Fortran compilers to pass the lengths of > Fortran character arguments, define USE_FC_LEN_T and include > Rconfig.h (possibly _via_ R.h) before including R_ext/BLAS.h or > R_ext/Lapack.h. > > * A package with Fortran source code and perhaps C (but not C++) > sources can request for its shared object/DLL to be linked by the > Fortran compiler by including a line USE_FC_TO_LINK= in > src/Makevars[.win] and using $(SHLIB_OPENMP_FFLAGS) as part of > PKG_LIBS. > > The known reason for doing so is a package which uses Fortran > (only) OpenMP on a platform where the Fortran OpenMP runtime is > incompatible with the C one (e.g. gfortran 9.x with clang). > > UTILITIES: > > * R CMD check has a new option to mitigate checks leaving > files/directories in /tmp. See the 'R Internals' manual - this > is part of --as-cran. > > Windows: > > * The defau
Re: [Rd] R 3.6.2 is released
It is not obvious what it is that you are calling "R-patched", nor where there
could be an entry for "3.6.2 patched".
The prerelease/patched versions are snapshots of R-3-6-branch made at 00:20 so
the current one will have been made before the release version run started at
09:00 this morning, and hence the nightly tarball will be of the release
candidate. However it will not be called R-patched:
lrwxr-xr-x 1 pd staff29 Dec 12 00:20 R-latest.tar.gz ->
R-rc_2019-12-06_r77555.tar.gz
The current version in SVN differs only by the VERSION file. Its NEWS.Rd starts
Peters-MacBook-Air:R pd$ more VERSION
3.6.2 Patched
Peters-MacBook-Air:R pd$ more doc/NEWS.Rd
% -*- coding: utf-8 -*-
\newcommand{\Rlogo}{\if{html}{\figure{../../html/Rlogo.svg}{options:
class="toplogo" alt="[R logo]"}}\if{latex}{\figure{Rlogo.pdf}{options:
width=0.5in}}}
\name{NEWS}
\title{R News}
\encoding{UTF-8}
\section{\Rlogo CHANGES IN R 3.6.2}{
\subsection{NEW FEATURES}{
\itemize{
and any changes for 3.6.2 patched should go above the entries for 3.6.2.
- pd
> On 12 Dec 2019, at 16:34 , Avraham Adler wrote:
>
> Hi.
>
> Under R-news there is an entry for 3.6.2 patched regarding LAPACK. However,
> when uncompresding the current R-patched, it creates R-Rc directories. Is
> this a naming oversight or is the patched version actually the unadjusted
> release candidate?
>
> Thank you,
>
> Avi
>
> On Thu, Dec 12, 2019 at 4:58 AM Peter Dalgaard via R-devel
> wrote:
> The build system rolled up R-3.6.2.tar.gz (codename "Dark and Stormy Night")
> this morning.
>
> The list below details the changes in this release.
>
> You can get the source code from
>
> http://cran.r-project.org/src/base/R-3/R-3.6.2.tar.gz
>
> or wait for it to be mirrored at a CRAN site nearer to you.
>
> Binaries for various platforms will appear in due course.
>
>
> For the R Core Team,
>
> Peter Dalgaard
>
> These are the checksums (md5 and SHA-256) for the freshly created files, in
> case you wish
> to check that they are uncorrupted:
>
> MD5 (AUTHORS) = b9c44f9f78cab3184ad9898bebc854b4
> MD5 (COPYING) = eb723b61539feef013de476e68b5c50a
> MD5 (COPYING.LIB) = a6f89e2100d9b6cdffcea4f398e37343
> MD5 (FAQ) = 28a3942a7129877e9af1d5ea16202052
> MD5 (INSTALL) = 7893f754308ca31f1ccf62055090ad7b
> MD5 (NEWS) = 45437b38c75e0248b527c00e6d42ee6a
> MD5 (NEWS.0) = bfcd7c147251b5474d96848c6f57e5a8
> MD5 (NEWS.1) = eb78c4d053ec9c32b815cf0c2ebea801
> MD5 (NEWS.2) = 591dcf615162127f904e4e461f330ce9
> MD5 (R-latest.tar.gz) = 90d23d138cee26d275da14b58296e521
> MD5 (README) = f468f281c919665e276a1b691decbbe6
> MD5 (RESOURCES) = 529223fd3ffef95731d0a87353108435
> MD5 (THANKS) = bb45f89c01d509721c47fd41f147da60
> MD5 (VERSION-INFO.dcf) = 9c33701e25092aefc1d16beb5858f20f
> MD5 (R-3/R-3.6.2.tar.gz) = 90d23d138cee26d275da14b58296e521
>
>
> 2cde824a7b18958e5f06b391c801c8288be0f84fa8934b7ddefef23c67e60c09 AUTHORS
> e6d6a009505e345fe949e1310334fcb0747f28dae2856759de102ab66b722cb4 COPYING
> 6095e9ffa777dd22839f7801aa845b31c9ed07f3d6bf8a26dc5d2dec8ccc0ef3 COPYING.LIB
> 38219d9c6221ccfbf075ef03711b420a1aa8731f890c8f2337148b602a217c2d FAQ
> f87461be6cbaecc4dce44ac58e5bd52364b0491ccdadaf846cb9b452e9550f31 INSTALL
> 0ceb6fbab3e0e29bc374683fd5c2ccd0c9c62ce8eca2a394a4603775b3ef129c NEWS
> 4e21b62f515b749f80997063fceab626d7258c7d650e81a662ba8e0640f12f62 NEWS.0
> 12b30c724117b1b2b11484673906a6dcd48a361f69fc420b36194f9218692d01 NEWS.1
> ca04f78ffe54afa326fe3ed40e7e1411acaed2fa5ead97ddf51c6aa5b7bc NEWS.2
> bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954
> R-latest.tar.gz
> 2fdd3e90f23f32692d4b3a0c0452f2c219a10882033d1774f8cadf25886c3ddc README
> 408737572ecc6e1135fdb2cf7a9dbb1a6cb27967c757f1771b8c39d1fd2f1ab9 RESOURCES
> 2a8dca916cd92229ef9e328f3610ca204809c262823b860252b42072dac2473a THANKS
> 40cc7cea5f0e67cf8f2f7b25a534ae6bc53f38eae2ab2c2649a952ed37f0654a
> VERSION-INFO.dcf
> bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954
> R-3/R-3.6.2.tar.gz
>
> This is the relevant part of the NEWS file
>
> CHANGES IN R 3.6.2:
>
> NEW FEATURES:
>
> * runmed(x, *) gains a new option na.action determining _how_ to
> handle NaN or NA in x.
>
> * dotchart() gains new options ann, xaxt, frame.plot and log.
>
> INSTALLATION on a UNIX-ALIKE:
>
> * Detection of the C stack direction has been moved from run-time
> to configure: this is safer with LTO builds and allows the
> detection to be overridden - see file config.site.
>
> * Source-code changes enable installation on platforms using gcc
> -fno-common (the expected default for gcc 10.x).
>
> C-LEVEL FACILITIES:
>
> * installTrChar (which is nowadays is wrapped by installChar) is
> defined in Rinternals.h. (Neither are part of the API.)
>
> PACKAGE INSTALLATION:
>
> * Header Rconfig.h contains the value of FC_LEN_T deduced at
> installation which is used by the prototypes in headers
>
Re: [Rd] Inconsistencies in wilcox.test
> Karolis Koncevičius > on Mon, 9 Dec 2019 23:43:36 +0200 writes: > So I tried adding Infinity support for all cases. > And it is (as could be expected) more complicated than I thought. "Of course !" Thank you, Karolis, in any case! > It is easy to add Inf support for the test. The problems start with conf.int=TRUE. > Currently confidence intervals are computed via `uniroot()` and, in the > case of infinities, we are computationally looking for roots over > infinite interval which results in an error. I suspect this is the > reason Inf values were removed in the first place. Maybe. It's still wrong to be done "up front". I'm sure 98% (or so ;-) of all calls to wilcox.test() do *not* use conf.int = TRUE > Just a note, I found a few more errors/inconsistencies when requesting > confidence intervals with paired=TRUE (due to Infinities being left in). > Current error in Inf-Inf scenario: > wilcox.test(c(1,2,Inf), c(4,8,Inf), paired=TRUE, conf.int=TRUE) > Error in if (ZEROES) x <- x[x != 0] : > missing value where TRUE/FALSE needed Good catch .. notably as it also happens when conf.int=FALSE as by default. My version of wilcox.test() now does give the same as when the to 'Inf' are left away. > NaN confidence intervals: > wilcox.test(c(1:9,Inf), c(21:28,Inf,30), paired=TRUE, conf.int=TRUE) > Wilcoxon signed rank test with continuity correction > data: c(1:9, Inf) and c(21:28, Inf, 30) > V = 9.5, p-value = 0.0586 > alternative hypothesis: true location shift is not equal to 0 > 0 percent confidence interval: > NaN NaN > sample estimates: > midrange > NaN I don't see a big problem here. The NaN's in some sense show the best that can be computed with this data. Statistic and P-value, but no conf.int. > The easiest "fix" for consistency would be to simply remove Infinity > support from the paired=TRUE case. I strongly disagree. We are not pruning good functionality just for some definition of consistency. > But going with the more desirable approach of adding Infinity support > for non-paired cases - it is currently not clear to me what confidence > intervals and pseudomedian should be. Specially when Infinities are on > both sides. I deem that not to be a big deal. They are not defined given the default formulas and that is reflected by NA / NaN in those parts of the result. > Regards, > Karolis Koncevičius. But I have also spent a few hours now on wilcox.test.default() behavior notably also looking at the "rounding" / "machine precision" situation, and also on your remark that the 'method: ...' does not indicate well enough what was computed. In my (not yet committed) but hereby proposed enhancement of wilcox.test(), I have a new argument, 'digits.rank = Inf' (the default 'Inf' corresponding to the current behavior) with help page documentation: digits.rank: a number; if finite, ‘rank(signif(r, digits.rank))’ will be used to compute ranks for the test statistic instead of (the default) ‘rank(r)’. and then in 'Details :' For stability reasons, it may be advisable to use rounded data or to set ‘digits.rank = 7’, say, such that determination of ties does not depend on very small numeric differences (see the example). and then in 'Examples: ' ## accuracy in ties determination via 'digits.rank': wilcox.test( 4:2, 3:1, paired=TRUE) # Warning: cannot compute exact p-value with ties wilcox.test((4:2)/10, (3:1)/10, paired=TRUE) # no ties => *no* warning wilcox.test((4:2)/10, (3:1)/10, paired=TRUE, digits.rank = 9) # same ties as (4:2, 3:1) -- Lastly, I propose to replace "test" by "exact test" in the 'method' component (and print out) in case exact computations were used. This information should be part of the returned "htest" object, and not only visible from the arguments and warnings that are printed during the computations. This last change is in some sense the "most back-incompatible" change of these, because many wilcox.test() printouts would slightly change, e.g., > w0 <- wilcox.test( 1:5, 4*(0:4), paired=TRUE) Wilcoxon signed rank exact test data: 1:5 and 4 * (0:4) V = 1, p-value = 0.125 alternative hypothesis: true location shift is not equal to 0 where before (in R <= 3.6.x) it is just Wilcoxon signed rank test data: . ... ... but I think R 4.0.0 is a good occasion for such a change. Constructive feedback on all this is very welcome! Martin > On 2019-12-07 23:18, Karolis Koncevičius wrote: >> Thank you for a fast response. Nice to see this mailing list being so >> alive. >> >> Regarding Inf issue: I agree with your assessment that Inf should not >> be removed. The code gave me an impression that Inf values were >>
Re: [Rd] R 3.6.2 is released
Thank you.
I apologize for not providing the link. [1] Under the news for R-revel
there is a single entry for R 3.6.2-patched.
The file I downloaded was [2] with a date of 2019-12-12 01:50.
Is it safe to say that 3.6.2 has the LAPACK upgrades and fixes?
Apologies in advance if iOS links the URL below. I cannot access gmail
desktop from behind my corporate firewall.
Thank you again,
Avi
[1]
https://cran.r-project.org/doc/manuals/r-devel/NEWS.html
[2]
https://stat.ethz.ch/R/daily/R-patched.tar.gz
On Thu, Dec 12, 2019 at 11:10 AM Peter Dalgaard wrote:
> It is not obvious what it is that you are calling "R-patched", nor where
> there could be an entry for "3.6.2 patched".
>
> The prerelease/patched versions are snapshots of R-3-6-branch made at
> 00:20 so the current one will have been made before the release version run
> started at 09:00 this morning, and hence the nightly tarball will be of the
> release candidate. However it will not be called R-patched:
>
> lrwxr-xr-x 1 pd staff29 Dec 12 00:20 R-latest.tar.gz ->
> R-rc_2019-12-06_r77555.tar.gz
>
> The current version in SVN differs only by the VERSION file. Its NEWS.Rd
> starts
>
> Peters-MacBook-Air:R pd$ more VERSION
> 3.6.2 Patched
> Peters-MacBook-Air:R pd$ more doc/NEWS.Rd
> % -*- coding: utf-8 -*-
> \newcommand{\Rlogo}{\if{html}{\figure{../../html/Rlogo.svg}{options:
> class="toplogo" alt="[R logo]"}}\if{latex}{\figure{Rlogo.pdf}{options:
> width=0.5in}}}
>
> \name{NEWS}
> \title{R News}
> \encoding{UTF-8}
>
> \section{\Rlogo CHANGES IN R 3.6.2}{
>
> \subsection{NEW FEATURES}{
> \itemize{
>
>
> and any changes for 3.6.2 patched should go above the entries for 3.6.2.
>
> - pd
>
>
> > On 12 Dec 2019, at 16:34 , Avraham Adler
> wrote:
> >
> > Hi.
> >
> > Under R-news there is an entry for 3.6.2 patched regarding LAPACK.
> However, when uncompresding the current R-patched, it creates R-Rc
> directories. Is this a naming oversight or is the patched version actually
> the unadjusted release candidate?
> >
> > Thank you,
> >
> > Avi
> >
> > On Thu, Dec 12, 2019 at 4:58 AM Peter Dalgaard via R-devel <
> r-devel@r-project.org> wrote:
> > The build system rolled up R-3.6.2.tar.gz (codename "Dark and Stormy
> Night") this morning.
> >
> > The list below details the changes in this release.
> >
> > You can get the source code from
> >
> > http://cran.r-project.org/src/base/R-3/R-3.6.2.tar.gz
> >
> > or wait for it to be mirrored at a CRAN site nearer to you.
> >
> > Binaries for various platforms will appear in due course.
> >
> >
> > For the R Core Team,
> >
> > Peter Dalgaard
> >
> > These are the checksums (md5 and SHA-256) for the freshly created files,
> in case you wish
> > to check that they are uncorrupted:
> >
> > MD5 (AUTHORS) = b9c44f9f78cab3184ad9898bebc854b4
> > MD5 (COPYING) = eb723b61539feef013de476e68b5c50a
> > MD5 (COPYING.LIB) = a6f89e2100d9b6cdffcea4f398e37343
> > MD5 (FAQ) = 28a3942a7129877e9af1d5ea16202052
> > MD5 (INSTALL) = 7893f754308ca31f1ccf62055090ad7b
> > MD5 (NEWS) = 45437b38c75e0248b527c00e6d42ee6a
> > MD5 (NEWS.0) = bfcd7c147251b5474d96848c6f57e5a8
> > MD5 (NEWS.1) = eb78c4d053ec9c32b815cf0c2ebea801
> > MD5 (NEWS.2) = 591dcf615162127f904e4e461f330ce9
> > MD5 (R-latest.tar.gz) = 90d23d138cee26d275da14b58296e521
> > MD5 (README) = f468f281c919665e276a1b691decbbe6
> > MD5 (RESOURCES) = 529223fd3ffef95731d0a87353108435
> > MD5 (THANKS) = bb45f89c01d509721c47fd41f147da60
> > MD5 (VERSION-INFO.dcf) = 9c33701e25092aefc1d16beb5858f20f
> > MD5 (R-3/R-3.6.2.tar.gz) = 90d23d138cee26d275da14b58296e521
> >
> >
> > 2cde824a7b18958e5f06b391c801c8288be0f84fa8934b7ddefef23c67e60c09 AUTHORS
> > e6d6a009505e345fe949e1310334fcb0747f28dae2856759de102ab66b722cb4 COPYING
> > 6095e9ffa777dd22839f7801aa845b31c9ed07f3d6bf8a26dc5d2dec8ccc0ef3
> COPYING.LIB
> > 38219d9c6221ccfbf075ef03711b420a1aa8731f890c8f2337148b602a217c2d FAQ
> > f87461be6cbaecc4dce44ac58e5bd52364b0491ccdadaf846cb9b452e9550f31 INSTALL
> > 0ceb6fbab3e0e29bc374683fd5c2ccd0c9c62ce8eca2a394a4603775b3ef129c NEWS
> > 4e21b62f515b749f80997063fceab626d7258c7d650e81a662ba8e0640f12f62 NEWS.0
> > 12b30c724117b1b2b11484673906a6dcd48a361f69fc420b36194f9218692d01 NEWS.1
> > ca04f78ffe54afa326fe3ed40e7e1411acaed2fa5ead97ddf51c6aa5b7bc NEWS.2
> > bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954
> R-latest.tar.gz
> > 2fdd3e90f23f32692d4b3a0c0452f2c219a10882033d1774f8cadf25886c3ddc README
> > 408737572ecc6e1135fdb2cf7a9dbb1a6cb27967c757f1771b8c39d1fd2f1ab9
> RESOURCES
> > 2a8dca916cd92229ef9e328f3610ca204809c262823b860252b42072dac2473a THANKS
> > 40cc7cea5f0e67cf8f2f7b25a534ae6bc53f38eae2ab2c2649a952ed37f0654a
> VERSION-INFO.dcf
> > bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954
> R-3/R-3.6.2.tar.gz
> >
> > This is the relevant part of the NEWS file
> >
> > CHANGES IN R 3.6.2:
> >
> > NEW FEATURES:
> >
> > * runmed(x, *) gains a new option na.action determining _how_ to
> > handle NaN or NA in x.
> >
> > * dot
Re: [Rd] Dynamic load library problem using R 4.0.0
Case closed. My error was that I had not included an "export" option to
the linker and the symbols were not being exported (I was out of practice
building Window's dlls). The nm command output didn't help me. Thanks to
Tomas Kalibera and Jeroen Ooms for some helpful correspondence.
On Tue, Dec 3, 2019 at 1:41 PM Nicholas Crookston
wrote:
> I'm having problems getting code that runs under 3.6.1 on MacOSX (and
> Ubuntu)
> to run the R version 4.0.0 on Windows. I have not tested R 4.0.0 on mac
> (yet).
>
> Key question: Has anything changed in R 4.0.0/Windows regarding how
> dynamically loaded
> libraries are accessed? Can someone please point me in the right
> direction?
>
> WORKS On MacOSX Under:
> R version 3.6.1 (2019-07-05) -- "Action of the Toes"
> after dynamically loading ../../bin/qFVSie.so:
>
> > getLoadedDLLs()$qFVSie
> DLL name: qFVSie
> Filename:
>
> /Users/ncrookston/open-fvs/branches/NickDev/tests/APIviaR/../../bin/qFVSie.so
> Dynamic lookup: TRUE
>
> # expected output is correct:
> > i=1
> > .C("CfvsSpeciesCode",fvs_code="",fia_code="",plant_code="",
> indx=as.integer(i))
> $fvs_code
> [1] "WP"
> $fia_code
> [1] "119"
> $plant_code
> [1] "PIMO3"
> $indx
> [1] 1
>
> - from the nm command:
> $ nm ../../bin/qFVSie.so | grep -i cfvsspeciescode
> 000a7980 T _CfvsSpeciesCode
>
>
>
> Fails On Win10 Under:
> R version 4.0.0 Under development (Rtools 4.0) (2019-11-23 r77455) --
> "Blame Jeroen"
> NOTE: The .dll is built using the most recent Rtools as advised by Jeroen.
>
> > getLoadedDLLs()$qFVSie
> DLL name: qFVSie
> Filename:
>
> C:/Users/nickc/open-fvs/branches/NickDev/tests/APIviaR/../../bin/qFVSie.dll
> Dynamic lookup: TRUE
>
> > i=1
> > .C("CfvsSpeciesCode",fvs_code="",fia_code="",plant_code="",
> indx=as.integer(i))
> Error in .C("CfvsSpeciesCode", fvs_code = "", fia_code = "", plant_code =
> "", :
> C symbol name "CfvsSpeciesCode" not in load table
>
> - from the nm command on Win10:
> nm ../../bin/qFVSie.dll | grep -i cfvsspeciescode
> 707a21f0 T CfvsSpeciesCode
>
> ---
>
> If advised to do so, I'll create a simple and fully reproducible example,
> but perhaps
> I missing something simple (like compiler switch needed to create my .dll
> that has not
> been necessary until now).
>
> Thanks, Nick
>
> --
> Nicholas L. Crookston
> Forestry Research Consultant
> Moscow Idaho USA
>
--
Nicholas L. Crookston
Forestry Research Consultant
Moscow Idaho USA
[[alternative HTML version deleted]]
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] Build failure on powerpc64
Hi R folks, Went to build R 3.6.2 for Fedora/EPEL and got failures across the board. Disabling the test suite for all non-intel architectures resolves most of the failures, but powerpc64 dies in the compiler, specifically here: gcc -m64 -I../../src/extra/xdr -I. -I../../src/include -I../../src/include -I/usr/local/include -I../../src/nmath -DHAVE_CONFIG_H -fopenmp -fPIC -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mcpu=power8 -mtune=power8 -fasynchronous-unwind-tables -fstack-clash-protection -c arithmetic.c -o arithmetic.o arithmetic.c:180:26: error: initializer element is not constant 180 | static LDOUBLE q_1_eps = (1 / LDBL_EPSILON); | ^ make[3]: *** [../../Makeconf:124: arithmetic.o] Error 1 Took me a bit to figure out why, but this is happening because on powerpc64, gcc is compiled with -mlong-double-128, and the long double format used on PPC is IBM's 128bit long double which is two doubles added together. As a result, gcc can't safely do const assignments to long doubles on ppc64, so it dies there. The fix is easy enough, do not try to assign a value to a static long double on ppc64. --- ./src/main/arithmetic.c.orig2019-12-12 18:30:12.416334062 + +++ ./src/main/arithmetic.c 2019-12-12 18:30:44.966334062 + @@ -179,7 +179,10 @@ void attribute_hidden InitArithmetic() #endif } -#if HAVE_LONG_DOUBLE && (SIZEOF_LONG_DOUBLE > SIZEOF_DOUBLE) +/* PowerPC 64 (when gcc has -mlong-double-128) breaks here because + * of issues constant folding 128bit IBM long doubles. + */ +#if HAVE_LONG_DOUBLE && (SIZEOF_LONG_DOUBLE > SIZEOF_DOUBLE) && !__PPC64__ static LDOUBLE q_1_eps = 1 / LDBL_EPSILON; #else static double q_1_eps = 1 / DBL_EPSILON; Hope that helps someone else. Tom [[alternative HTML version deleted]] __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] R 3.6.2 is released
It is possible that this means that it is the intention to port the changes
from R-devel to 3.6.2 patched once 3.6.2 is out. In that case, you can be
pretty sure that the changes are NOT in 3.6.2 (and actually not in 3.6.2
patched either -- yet).
If so, you have provided a nice illustration of why I keep telling people not
to do that...
-pd
> On 12 Dec 2019, at 18:05 , Avraham Adler wrote:
>
> Thank you.
>
> I apologize for not providing the link. [1] Under the news for R-revel there
> is a single entry for R 3.6.2-patched.
>
> The file I downloaded was [2] with a date of 2019-12-12 01:50.
>
> Is it safe to say that 3.6.2 has the LAPACK upgrades and fixes?
>
> Apologies in advance if iOS links the URL below. I cannot access gmail
> desktop from behind my corporate firewall.
>
> Thank you again,
>
> Avi
>
> [1]
> https://cran.r-project.org/doc/manuals/r-devel/NEWS.html
> [2]
> https://stat.ethz.ch/R/daily/R-patched.tar.gz
>
> On Thu, Dec 12, 2019 at 11:10 AM Peter Dalgaard wrote:
> It is not obvious what it is that you are calling "R-patched", nor where
> there could be an entry for "3.6.2 patched".
>
> The prerelease/patched versions are snapshots of R-3-6-branch made at 00:20
> so the current one will have been made before the release version run started
> at 09:00 this morning, and hence the nightly tarball will be of the release
> candidate. However it will not be called R-patched:
>
> lrwxr-xr-x 1 pd staff29 Dec 12 00:20 R-latest.tar.gz ->
> R-rc_2019-12-06_r77555.tar.gz
>
> The current version in SVN differs only by the VERSION file. Its NEWS.Rd
> starts
>
> Peters-MacBook-Air:R pd$ more VERSION
> 3.6.2 Patched
> Peters-MacBook-Air:R pd$ more doc/NEWS.Rd
> % -*- coding: utf-8 -*-
> \newcommand{\Rlogo}{\if{html}{\figure{../../html/Rlogo.svg}{options:
> class="toplogo" alt="[R logo]"}}\if{latex}{\figure{Rlogo.pdf}{options:
> width=0.5in}}}
>
> \name{NEWS}
> \title{R News}
> \encoding{UTF-8}
>
> \section{\Rlogo CHANGES IN R 3.6.2}{
>
> \subsection{NEW FEATURES}{
> \itemize{
>
>
> and any changes for 3.6.2 patched should go above the entries for 3.6.2.
>
> - pd
>
>
> > On 12 Dec 2019, at 16:34 , Avraham Adler wrote:
> >
> > Hi.
> >
> > Under R-news there is an entry for 3.6.2 patched regarding LAPACK. However,
> > when uncompresding the current R-patched, it creates R-Rc directories. Is
> > this a naming oversight or is the patched version actually the unadjusted
> > release candidate?
> >
> > Thank you,
> >
> > Avi
> >
> > On Thu, Dec 12, 2019 at 4:58 AM Peter Dalgaard via R-devel
> > wrote:
> > The build system rolled up R-3.6.2.tar.gz (codename "Dark and Stormy
> > Night") this morning.
> >
> > The list below details the changes in this release.
> >
> > You can get the source code from
> >
> > http://cran.r-project.org/src/base/R-3/R-3.6.2.tar.gz
> >
> > or wait for it to be mirrored at a CRAN site nearer to you.
> >
> > Binaries for various platforms will appear in due course.
> >
> >
> > For the R Core Team,
> >
> > Peter Dalgaard
> >
> > These are the checksums (md5 and SHA-256) for the freshly created files, in
> > case you wish
> > to check that they are uncorrupted:
> >
> > MD5 (AUTHORS) = b9c44f9f78cab3184ad9898bebc854b4
> > MD5 (COPYING) = eb723b61539feef013de476e68b5c50a
> > MD5 (COPYING.LIB) = a6f89e2100d9b6cdffcea4f398e37343
> > MD5 (FAQ) = 28a3942a7129877e9af1d5ea16202052
> > MD5 (INSTALL) = 7893f754308ca31f1ccf62055090ad7b
> > MD5 (NEWS) = 45437b38c75e0248b527c00e6d42ee6a
> > MD5 (NEWS.0) = bfcd7c147251b5474d96848c6f57e5a8
> > MD5 (NEWS.1) = eb78c4d053ec9c32b815cf0c2ebea801
> > MD5 (NEWS.2) = 591dcf615162127f904e4e461f330ce9
> > MD5 (R-latest.tar.gz) = 90d23d138cee26d275da14b58296e521
> > MD5 (README) = f468f281c919665e276a1b691decbbe6
> > MD5 (RESOURCES) = 529223fd3ffef95731d0a87353108435
> > MD5 (THANKS) = bb45f89c01d509721c47fd41f147da60
> > MD5 (VERSION-INFO.dcf) = 9c33701e25092aefc1d16beb5858f20f
> > MD5 (R-3/R-3.6.2.tar.gz) = 90d23d138cee26d275da14b58296e521
> >
> >
> > 2cde824a7b18958e5f06b391c801c8288be0f84fa8934b7ddefef23c67e60c09 AUTHORS
> > e6d6a009505e345fe949e1310334fcb0747f28dae2856759de102ab66b722cb4 COPYING
> > 6095e9ffa777dd22839f7801aa845b31c9ed07f3d6bf8a26dc5d2dec8ccc0ef3
> > COPYING.LIB
> > 38219d9c6221ccfbf075ef03711b420a1aa8731f890c8f2337148b602a217c2d FAQ
> > f87461be6cbaecc4dce44ac58e5bd52364b0491ccdadaf846cb9b452e9550f31 INSTALL
> > 0ceb6fbab3e0e29bc374683fd5c2ccd0c9c62ce8eca2a394a4603775b3ef129c NEWS
> > 4e21b62f515b749f80997063fceab626d7258c7d650e81a662ba8e0640f12f62 NEWS.0
> > 12b30c724117b1b2b11484673906a6dcd48a361f69fc420b36194f9218692d01 NEWS.1
> > ca04f78ffe54afa326fe3ed40e7e1411acaed2fa5ead97ddf51c6aa5b7bc NEWS.2
> > bd65a45cddfb88f37370fbcee4ac8dd3f1aebeebe47c2f968fd9770ba2bbc954
> > R-latest.tar.gz
> > 2fdd3e90f23f32692d4b3a0c0452f2c219a10882033d1774f8cadf25886c3ddc README
> > 408737572ecc6e1135fdb2cf7a9dbb1a6cb27967c757f1771b8c39d1fd2f1ab9 RE
