[Rd] "R CMD check" accepts but "R CMD INSTALL" rejects a tar ball.
I have somehow managed to made a source tar ball which "R CMD check" accepts but "R CMD INSTALL" rejects with: -- Warning in untar2(tarfile, files, list, exdir) : checksum error for entry 'pax_global_header' Error in untar2(tarfile, files, list, exdir) : unsupported entry type ‘g’ -- This happens with both R 2.12.2 (x86 linux) and R svn (x86_64 linux). Since R CMD check does install as part of the check process, there is probably a bug somewhere. The tar ball is uploaded at: http://htl10.users.sourceforge.net/tmp/Matrix_0.999375-48.tar.gz and tar -xzpvf works. It is possible to do R CMD INSTALL from the untar'ed data, so I am a bit lost at where the problem is. The tar ball was generated with git archive ... | gzip > package.tar.gz similar to the example at the bottom of git-archive. It is the result of git cherry-pick trunk@2658 + Matrix-for-R-2.13@2657 . __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] Fw: Re: core Matrix package segfaulted on R CMD check --use-gct
repost from the subscribed address...
--- On Fri, 1/4/11, Hin-Tak Leung wrote:
> --- On Wed, 30/3/11, Douglas Bates
>
> wrote:
>
> > I isolated the problem and tested then committed a
> fix. I
> > am going to
> > ask Martin to upload the new release as I have gotten
> out
> > of sync with
> > some of his recent changes and he will, I hope,
> reconcile
> > the branches
> > and trunk. If you need the fixed version
> immediately,
> > say for
> > testing, the changes are in the file
> > Matrix/src/chm_common.c You can
> > visit the SVN archive for the project,
> > https://r-forge.r-project.org/scm/?group_id=61, click
> > on the link to
> > Browse Subversion Repository and go to
> > pkg/Matrix/src/chm_common.c
> >
> > I made the mistake that I chastised others for
> making,
> > performing an
> > operation on the result of a call to install and
> another
> > value that
> > may have required allocation of memory. The first
> > time install is
> > called on a particular string it allocates and
> SEXPREC,
> > which may
> > cause a garbage collection.
>
> Have a different (but similiar) problem git cherry-picking
>
> trunk@2658 on top of Matrix-for-R-2.13@2657.
> trunk@2658 + Matrix-for-R-2.13@2657 should be identical to
> Matrix-for-R-2.13@2659 which you committed earlier
> (according to git svn rebase):
>
> > pkgname <- "Matrix"
> > source(file.path(R.home("share"), "R",
> "examples-header.R"))
> > gctorture(TRUE)
> > options(warn = 1)
> > library('Matrix')
> Loading required package: lattice
> Error in regexpr("package:", envName, fixed = TRUE) :
> unprotected object (0x3a52538) encountered (was
> INTSXP)
> Error: package/namespace load failed for 'Matrix'
> Execution halted
>
> I also found a different problem, but it looks like it is a
> different part of R and possibly in utils. I'll write
> separately and CC' you (the Matrix authors) on it.
>
>
>
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] "R CMD check" accepts but "R CMD INSTALL" rejects a tar ball.
?untar: You may see warnings from the internal implementation such as unsupported entry type 'x' This often indicates an invalid archive: entry types ‘"A-Z"’ are allowed as extensions, but other types are reserved (this example is from Mac OS 10.6.3). The only thing you can do with such an archive is to find a ‘tar’ program that handles it, and look carefully at the resulting files. So the difference is whether you use external or internal tar. 'g' is the global pax header extension so the format you created is really pax and not tar (pax defines two new types 'x' and 'g'). Cheers, Simon On Apr 1, 2011, at 10:19 AM, Hin-Tak Leung wrote: > I have somehow managed to made a source tar ball which "R CMD check" accepts > but "R CMD INSTALL" rejects with: > > -- > Warning in untar2(tarfile, files, list, exdir) : > checksum error for entry 'pax_global_header' > Error in untar2(tarfile, files, list, exdir) : unsupported entry type ‘g’ > -- > > This happens with both R 2.12.2 (x86 linux) and R svn (x86_64 linux). Since R > CMD check does install as part of the check process, there is probably a bug > somewhere. The tar ball is uploaded at: > > http://htl10.users.sourceforge.net/tmp/Matrix_0.999375-48.tar.gz > > and tar -xzpvf works. It is possible to do R CMD INSTALL from the untar'ed > data, so I am a bit lost at where the problem is. > > The tar ball was generated with > git archive ... | gzip > package.tar.gz > similar to the example at the bottom of git-archive. > > It is the result of git cherry-pick trunk@2658 + Matrix-for-R-2.13@2657 . > > __ > 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] "R CMD check" accepts but "R CMD INSTALL" rejects a tar ball.
--- On Fri, 1/4/11, Simon Urbanek wrote: > ?untar: > > You may see warnings from the > internal implementation such as > > unsupported entry type 'x' > > This often indicates an invalid > archive: entry types ‘"A-Z"’ are > allowed as extensions, but other > types are reserved (this example > is from Mac OS 10.6.3). The > only thing you can do with such an > archive is to find a ‘tar’ > program that handles it, and look > carefully at the resulting files. > > > So the difference is whether you use external or internal > tar. 'g' is the global pax header extension so the format > you created is really pax and not tar (pax defines two new > types 'x' and 'g'). > > Cheers, > Simon Okay, thanks. So I guess git-archive --format=tar uses GNU tar extensions (not too surprising). So this is documented... and a documented incompatibilities between different tar/tar extensions. But this behavior of R is a bit unexpected - When R CMD check (which involves installing to a temporary location then loads it and runs various things) works and R CMD INSTALL itself does not. OTOH, should this be reported to the GIT people? > > > On Apr 1, 2011, at 10:19 AM, Hin-Tak Leung wrote: > > > I have somehow managed to made a source tar ball which > "R CMD check" accepts but "R CMD INSTALL" rejects with: > > > > -- > > Warning in untar2(tarfile, files, list, exdir) : > > checksum error for entry 'pax_global_header' > > Error in untar2(tarfile, files, list, exdir) : > unsupported entry type ‘g’ > > -- > > > > This happens with both R 2.12.2 (x86 linux) and R svn > (x86_64 linux). Since R CMD check does install as part of > the check process, there is probably a bug somewhere. The > tar ball is uploaded at: > > > > http://htl10.users.sourceforge.net/tmp/Matrix_0.999375-48.tar.gz > > > > and tar -xzpvf works. It is possible to do R CMD > INSTALL from the untar'ed data, so I am a bit lost at where > the problem is. > > > > The tar ball was generated with > > git archive ... | gzip > > package.tar.gz > > similar to the example at the bottom of git-archive. > > > > It is the result of git cherry-pick trunk@2658 + > Matrix-for-R-2.13@2657 . > > > > __ > > 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] "R CMD check" accepts but "R CMD INSTALL" rejects a tar ball.
On Apr 1, 2011, at 12:16 PM, Hin-Tak Leung wrote: > --- On Fri, 1/4/11, Simon Urbanek wrote: > >> ?untar: >> >> You may see warnings from the >> internal implementation such as >> >> unsupported entry type 'x' >> >> This often indicates an invalid >> archive: entry types ‘"A-Z"’ are >> allowed as extensions, but other >> types are reserved (this example >> is from Mac OS 10.6.3). The >> only thing you can do with such an >> archive is to find a ‘tar’ >> program that handles it, and look >> carefully at the resulting files. >> >> >> So the difference is whether you use external or internal >> tar. 'g' is the global pax header extension so the format >> you created is really pax and not tar (pax defines two new >> types 'x' and 'g'). >> >> Cheers, >> Simon > > Okay, thanks. So I guess git-archive --format=tar uses GNU tar extensions > (not too surprising). So this is documented... and a documented > incompatibilities between different tar/tar extensions. But this behavior of > R is a bit unexpected - When R CMD check (which involves installing to a > temporary location then loads it and runs various things) works and R CMD > INSTALL itself does not. > > OTOH, should this be reported to the GIT people? > I think you should re-read my e-mail more carefully -- those are not GNU extensions, those are headers used by pax, not tar. pax is a format inspired by tar and introduced in POSIX.1-2001, see http://en.wikipedia.org/wiki/Pax_(Unix) Cheers, Simon >> >> >> On Apr 1, 2011, at 10:19 AM, Hin-Tak Leung wrote: >> >>> I have somehow managed to made a source tar ball which >> "R CMD check" accepts but "R CMD INSTALL" rejects with: >>> >>> -- >>> Warning in untar2(tarfile, files, list, exdir) : >>> checksum error for entry 'pax_global_header' >>> Error in untar2(tarfile, files, list, exdir) : >> unsupported entry type ‘g’ >>> -- >>> >>> This happens with both R 2.12.2 (x86 linux) and R svn >> (x86_64 linux). Since R CMD check does install as part of >> the check process, there is probably a bug somewhere. The >> tar ball is uploaded at: >>> >>> http://htl10.users.sourceforge.net/tmp/Matrix_0.999375-48.tar.gz >>> >>> and tar -xzpvf works. It is possible to do R CMD >> INSTALL from the untar'ed data, so I am a bit lost at where >> the problem is. >>> >>> The tar ball was generated with >>>git archive ... | gzip > >> package.tar.gz >>> similar to the example at the bottom of git-archive. >>> >>> It is the result of git cherry-pick trunk@2658 + >> Matrix-for-R-2.13@2657 . >>> >>> __ >>> 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
[Rd] Phrase "package writer" in R-exts
In a conversation with a programmer new to writing R packages, he mentioned that he was very confused by phrase "package writer" used in the document, and said that he "[was] literally imagining some sort of function that writes something related to packages". I can see his point: not only is it confusing, but I think it's also bad English (one wouldn't say "the novel's writer"). Can this be changed to "package author"? Davor __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Adding a "description" meta-tag to the R homepage
On Apr 1, 2011, at 08:49 , Tal Galili wrote: > Hello all, > I hope I'm writing to the correct place. > > I believe that the R homepage will benefit from including the "description" > meta tag in it's homepage. > The reason is that google uses that tag to decide what to show when the R > homepage shows up on a search result. > The current description of the first result of searching "R" in google is: > > "*R*, also called GNU S, is a strongly functional language and environment > to statistically explore data sets, make many graphical displays of data > from custom *..."* > * > * > This defention is truncated. Interesting. What's odd is that although that phrasing is used in multiple places on the net, it is not in the actual www.r-project.org/index.html, nor in any other "official" places that I can spot. However, what gets displayed for SAS and Stata is not what is in their description tags? SAS: Stata: What gives? > > > It seems that other statistical packages (and some known open source > projects) already noticed this. Here is what they write about themselves > (when searching for their name on google): > > - SAS: 50,000 Customer Sites Use SAS for CRM, BI, Analytics & More-Get > Info > - SPSS: Analytical Software at SPSS.com. Specializing in data mining, > customer relationship management, business intelligence and data analysis. > - JMP: JMP is statistical software for expert data analysis, DOE, and Six > Sigma, from SAS. For Mac, Windows, and Linux. > - Stata: Integrated statistical package for data analysis, data > management and graphics. > - Perl: The Perl Programming Language at Perl.org. Links and other > helpful resources for new and experienced Perl programmers. > - Matematica: Wolfram Research, makers of Mathematica, the only fully > integrated technical computing software. > - It seems that Python, Ruby, and Matlab are not using any description > tag (and their google description is also truncated) > > > I hope this helps. > > Best, > Tal > > > Contact > Details:--- > Contact me: tal.gal...@gmail.com | 972-52-7275845 > Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | > www.r-statistics.com (English) > -- > > [[alternative HTML version deleted]] > > __ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel -- Peter Dalgaard Center for Statistics, Copenhagen Business School Solbjerg Plads 3, 2000 Frederiksberg, Denmark Phone: (+45)38153501 Email: pd@cbs.dk Priv: pda...@gmail.com __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Phrase "package writer" in R-exts
> From: r-devel-boun...@r-project.org [mailto:r-devel-bounces@r- > project.org] On Behalf Of Davor Cubranic > Sent: April-01-11 2:23 PM > > In a conversation with a programmer new to writing R packages, he > mentioned that he was very confused by phrase "package writer" used in > the document, and said that he "[was] literally imagining some sort of > function that writes something related to packages". > > I can see his point: not only is it confusing, but I think it's also bad English (one > wouldn't say "the novel's writer"). Can this be changed to "package author"? > No, it is not "bad English." Neither would I regard it as being confusing. To my eye, it is unambiguous. This reminds me of a conversation I had a few years ago with a well educated individual who claimed that the sentence "He was hit by the ball" was grammatically wrong, and that the correct way to express that idea was "The ball hit him." Needless to say, we ended up agreeing to disagree. Of course, obviously, BOTH are correct. This individual had an earned Ph.D. from a prominent north american university. The English langage is beautifully large and complex, having drawn both vocabulary and grammatical structures from a wide variety of languages. That makes it hard to learn to use well, but it is arguably the most expressive language on the planet (what oher language has a vocabulary as large as that in english or grammar as flexible as is english). I spent a little time working in India, and found that ordinary working people, who happen to speak as many as 5 local and regional languages in addition to english, and found that they use such a different subset if standard english vocabulary from what the same sort of people would normally use in Toronto (where I was born), that I had to listen carefully in order to understand what they meant. I even found myself translating from standard english into standard english, for two bright people with very different ethic backgrounds and who had learned english after they turned 30. The influences of their respective mother tongues so distorted their use of english that they were mutually incomprehensible even though it was obvious to me what each was saying. One sweet lady I met in Singapore (a waitress) told me she liked servicing me, but didn't like serving the Australians next to me. She said she found it easy to understand me but she couldn't understand anything the Aussies at the next table were saying. I would never have noticed a significan difference between how we spoke, but she had plenty of trouble trying to figure out what they were saying. One of my best friends spent 6 months of his Ph.D. program just watching TV. That was because even though his grammar was perfect and his written english was beautiful, he had never hear english spoken by a native english speaker. Thus his diction was so distorted by habits developed in speaking perfect Mandarin (his mother tongue), that his spoken english was incomprehensible to everyone outside China. This is actually a major problem for any author that aspires to a global market. How do you write in such a way that you will be understood by everyone in your target market? I won't presume to offer writing advice to the authors in question, in part because I don't know a good solution. However, Davor, I notice you are at one of the finest universities in Canada, so might I suggest you approach someone in the english department there for insight into the evolution of english in the context of a multicultural society, and how that can help make our writing more readily understood by a global audience? Note, while I see the author's responsibility to write perfectly (or at least try), the reader also has a responsibility to work at determining what the author meant by what he or she wrote. Cheers, Ted __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Adding a "description" meta-tag to the R homepage
On 4/1/11 1:38 PM, "peter dalgaard" wrote: > On Apr 1, 2011, at 08:49 , Tal Galili wrote: > >> I believe that the R homepage will benefit from including the "description" >> meta tag in it's homepage. >> The reason is that google uses that tag to decide what to show when the R >> homepage shows up on a search result. >> The current description of the first result of searching "R" in google is: >> >> "*R*, also called GNU S, is a strongly functional language and environment >> to statistically explore data sets, make many graphical displays of data >> from custom *..."* >> * >> * >> This defention is truncated. > > > Interesting. What's odd is that although that phrasing is used in multiple > places on the net, it is not in the actual www.r-project.org/index.html, nor > in any other "official" places that I can spot. > > However, what gets displayed for SAS and Stata is not what is in their > description tags? > > SAS: > > > Stata: > > > What gives? Preferences then? R: - or - >> It seems that other statistical packages (and some known open source >> projects) already noticed this. Here is what they write about themselves >> (when searching for their name on google): >> >> - SAS: 50,000 Customer Sites Use SAS for CRM, BI, Analytics & More-Get >> Info >> - SPSS: Analytical Software at SPSS.com. Specializing in data mining, >> customer relationship management, business intelligence and data analysis. >> - JMP: JMP is statistical software for expert data analysis, DOE, and Six >> Sigma, from SAS. For Mac, Windows, and Linux. >> - Stata: Integrated statistical package for data analysis, data >> management and graphics. >> - Perl: The Perl Programming Language at Perl.org. Links and other >> helpful resources for new and experienced Perl programmers. >> - Matematica: Wolfram Research, makers of Mathematica, the only fully >> integrated technical computing software. >> - It seems that Python, Ruby, and Matlab are not using any description >> tag (and their google description is also truncated) >> __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Adding a "description" meta-tag to the R homepage
> Interesting. What's odd is that although that phrasing is used in multiple > places on the net, it is not in the actual www.r-project.org/index.html, nor > in any other "official" places that I can spot. > > However, what gets displayed for SAS and Stata is not what is in their > description tags? Google takes the description tag as advice, but if it thinks the page is actually about something else, it will write its own summary. Hadley -- Assistant Professor / Dobelman Family Junior Chair Department of Statistics / Rice University http://had.co.nz/ __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] "R CMD check" accepts but "R CMD INSTALL" rejects a tar ball.
--- On Fri, 1/4/11, Simon Urbanek wrote: > >> So the difference is whether you use external or > internal > >> tar. 'g' is the global pax header extension so the > format > >> you created is really pax and not tar (pax defines > two new > >> types 'x' and 'g'). > >> > >> Cheers, > >> Simon > > > > Okay, thanks. So I guess git-archive --format=tar uses > GNU tar extensions (not too surprising). So this is > documented... and a documented incompatibilities between > different tar/tar extensions. But this behavior of R is a > bit unexpected - When R CMD check (which involves installing > to a temporary location then loads it and runs various > things) works and R CMD INSTALL itself does not. > > > > OTOH, should this be reported to the GIT people? > > > > I think you should re-read my e-mail more carefully -- > those are not GNU extensions, those are headers used by pax, > not tar. pax is a format inspired by tar and introduced in > POSIX.1-2001, see > http://en.wikipedia.org/wiki/Pax_(Unix) > > Cheers, > Simon Argh, got it. The more informative page is in GNU tar's man page, but even GNU tar doesn't write pax by default. I happened to have git's git repository in my hard disc, and you are correct, git-archive, according to its source code, does write pax header, and refers to pax in several places. So git-archive's documentation (and its '--format=tar' option) is misleading; although even GNU fileutils says the ungzip'ed bundle is "posix tar". Go figure... OTOH, "R CMD check" extracts the content (and does an install) and 'R CMD INSTALL' itself does not, that's a bit inconsistent. > > > >> > >> > >> On Apr 1, 2011, at 10:19 AM, Hin-Tak Leung wrote: > >> > >>> I have somehow managed to made a source tar > ball which > >> "R CMD check" accepts but "R CMD INSTALL" rejects > with: > >>> > >>> -- > >>> Warning in untar2(tarfile, files, list, exdir) > : > >>> checksum error for entry > 'pax_global_header' > >>> Error in untar2(tarfile, files, list, exdir) > : > >> unsupported entry type ‘g’ > >>> -- > >>> > >>> This happens with both R 2.12.2 (x86 linux) > and R svn > >> (x86_64 linux). Since R CMD check does install as > part of > >> the check process, there is probably a bug > somewhere. The > >> tar ball is uploaded at: > >>> > >>> http://htl10.users.sourceforge.net/tmp/Matrix_0.999375-48.tar.gz > >>> > >>> and tar -xzpvf works. It is possible to do R > CMD > >> INSTALL from the untar'ed data, so I am a bit lost > at where > >> the problem is. > >>> > >>> The tar ball was generated with > >>> git archive ... | gzip > > >> package.tar.gz > >>> similar to the example at the bottom of > git-archive. > >>> > >>> It is the result of git cherry-pick trunk@2658 > + > >> Matrix-for-R-2.13@2657 . > >>> > >>> > __ > >>> 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] Phrase "package writer" in R-exts
Considering that the package.skeleton() function really is a "package writer" of sorts, I agree that "package author" is clearer. Kevin On Fri, Apr 1, 2011 at 1:22 PM, Davor Cubranic wrote: > In a conversation with a programmer new to writing R packages, he mentioned > that he was very confused by phrase "package writer" used in the document, > and said that he "[was] literally imagining some sort of function that > writes something related to packages". > > I can see his point: not only is it confusing, but I think it's also bad > English (one wouldn't say "the novel's writer"). Can this be changed to > "package author"? > > Davor > __ > 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
[Rd] logos
Dear all, I'd like to link one of the logos in our Google Summer of Code Profile. If I remember correctly, behind the logo link at http://developer.r-project.org/ (i.e. http://developer.r-project.org/Logo) also small versions can be found. However, I'm not allowed to access that directory. I have no problem with putting a small version of the "normal" logo somewhere online, but I'm wondering whether the logo directory really isn't supposed to be accessed by the public. Thanks, Claudia __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] logos
On Apr 1, 2011, at 4:42 PM, Claudia Beleites wrote: > Dear all, > > I'd like to link one of the logos in our Google Summer of Code Profile. If I > remember correctly, behind the logo link at http://developer.r-project.org/ > (i.e. http://developer.r-project.org/Logo) also small versions can be found. > However, I'm not allowed to access that directory. > I have no problem with putting a small version of the "normal" logo somewhere > online, but I'm wondering whether the logo directory really isn't supposed to > be accessed by the public. > Thanks, yes, the developer pages had indexes disabled by default, I have enabled them now (which was the case way back before we switched servers if I remember correctly). Thanks, Simon __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Phrase "package writer" in R-exts
On 2011-04-01, at 12:17 PM, Ted Byers wrote: > Note, while I see the author's > responsibility to write perfectly (or at least try), the reader also has a > responsibility to work at determining what the author meant by what he or > she wrote. Most people don't read reference manuals to engage with the author (writer?) and analyze the subtext, but to find relevant information quickly. For the programmer I was talking to at least (a native English speaker, BTW), R-exts fell short in that respect. When I first heard about his confusion I also thought "how can you miss that?", but now believe that the manual would be improved by replacing "writer" with "author". It is already dense enough with real information, why make it more obscure ? Davor __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
