How to create vignette.pdf for R-2.13.0?
Dear Herve, Since 'R CMD check' has already options '--no-vignettes' and '--no-rebuild-vignettes', so why not option '--no-clean-vignettes'? But, as you admitted, it is only a suggestion. Maybe it is time to close this thread. (If you want we can discuss this issue further privately.) Best regards Christian
On 5/9/11 7:20 AM, Herv? Pag?s wrote:
Christian, On 11-05-08 01:19 PM, cstrato wrote:
Dear Herve, In my initial mail: https://stat.ethz.ch/pipermail/r-devel/2011-April/060688.html I wrote: "While R CMD check and R CMD INSTALL have always created the vignettes on R-2.12.1 or any earlier versions of R, I am no longer able to build the vignettes on R-2.13.0." Although I made the mistake to include R CMD INSTALL, I wanted simply to know why R CMD check suddenly (after all these years) does no longer build the pdf-versions of the vignettes. After many replies and discussions Prof. Ripley gave finally the answer: https://stat.ethz.ch/pipermail/r-devel/2011-May/060782.html I understand the reason for this and made the suggestion to include an option such as "--no-clean-vignettes" to R CMD check, since this would be a convenient option for me: https://stat.ethz.ch/pipermail/r-devel/2011-May/060797.html
and then someone tried to explain you that you didn't need that, you just need to use the standard development cycle. And in your last 5 emails you're still trying to convince the people around that you have a better/faster development cycle that doesn't use 'R CMD build' at all. No matter this better development cycle is broken, it should be supported anyway by adding a "--no-clean-vignettes" option to 'R CMD check' just because that would be a "convenient option for you"? I know, you are just suggesting this option, but still... H.
Best regards Christian On 5/8/11 9:29 PM, Herv? Pag?s wrote:
Hi Christian, On 11-05-04 01:02 PM, cstrato wrote:
Dear Simon, I did not complain about the R doing something wrong. I only wanted to know why, after all these years, R CMD check does suddenly no longer build the pdf-files of the vignettes. I also think that this is a legal question. Let me compare the times spent: 1, the "official" development cycle: - R CMD build: 5 minutes - R CMD INSTALL: 3 minutes - R CMD check: 6 minutes 2, my own development cycle: - manual tarring: 2 seconds - R CMD check: 6 minutes This means spending 14 minutes vs 6 minutes.
My understanding is that R CMD INSTALL is not strictly necessary in the official devel cycle. However, to be fair you should either include that step in both cycles (devel and yours) or not. Then the numbers are 14 minutes vs 9 minutes (or 11 minutes vs 6 minutes).
If I assume that in one evening I have to make 10 corrections this would mean 140 minutes vs 60 minutes, i.e. the official development cycle would take 1 hr and 20 min longer than my own development cycle. This is time I do not have.
No matter how you put it and how much benefits you get from this, manual tarring is *wrong* and "after all these years" you should really know this. Furthermore, here is what you claimed in a previous email: This is my development cycle: First, I run R CMD check until there are no more warnings/errors. Since years it was very convenient that R CMD check builds the pdf-files of the vignettes, too, since this allowed me to correct errors in the manual files and the vignette files at the same time! Afterwards I run R CMD INSTALL to install my package and do more tests until everything works. As you see I do not use R CMD build, since every run takes about 5 minutes, it overwrites my zipped source code, and I would need to unzip it to get access to the vignette pdf-files. You didn't mention the manual tarring. Then later: Thank you, however since I use "R CMD INSTALL xps.tar.gz" my source code is not polluted. So you are contradicting yourself. No, I simply do "tar czf xps_1.13.1.tar.gz xps". Whaooo, finally! Note that, among other problems, this manual tarring is very likely to pollute your source tree (this is one of the things R CMD build will take care of). You are free to do manual tarring for your own internal business if you want, nobody cares, but you should keep this kind of dirty little secrets for you. For your 10 corrections, you can do manual tarring 10 times, that's your problem, but that doesn't prevent you from doing the right thing, i.e. R CMD build + R CMD check, as a *final* pass (e.g. just before you commit your changes to svn). Now that means 71 minutes vs 60 minutes, but you did the *right* thing. Cheers, H.
Best regards Christian On 5/4/11 12:13 AM, Simon Urbanek wrote:
On May 3, 2011, at 5:15 PM, cstrato wrote:
No, I simply do "tar czf xps_1.13.1.tar.gz xps".
Well, that's your problem then, not R's. Source packages are created using R CMD build, not by manual tarring (you can do the latter if you know what you're doing, but then you can't complain about the R doing something wrong). Cheers, S
On 5/3/11 11:11 PM, Simon Urbanek wrote:
On May 3, 2011, at 4:48 PM, cstrato<cstrato at aon.at> wrote:
Dear Uwe, Thank you, however since I use "R CMD INSTALL xps.tar.gz" my source code is not polluted.
But then you already used build to create the tar ball so the vignette has been built. So what is your point? Cheers, S
Furthermore, I forgot to mention that finally I upload the source code only to the BioC svn repository. The rest is done by the BioC servers, including building the pdf-files for the vignettes. Best regards Christian On 5/3/11 10:13 PM, Uwe Ligges wrote:
On 03.05.2011 21:14, cstrato wrote:
Dear Uwe, This is my development cycle: First, I run R CMD check until there are no more warnings/errors. Since years it was very convenient that R CMD check builds the pdf-files of the vignettes, too, since this allowed me to correct errors in the manual files and the vignette files at the same time! Afterwards I run R CMD INSTALL to install my package and do more tests until everything works. As you see I do not use R CMD build, since every run takes about 5 minutes, it overwrites my zipped source code, and I would need to unzip it to get access to the vignette pdf-files.
Then this is the main problem here. The *recommended* development cycle from the manuals is to run 1. R CMD build in order to get a valid source tarball and clean the sources 2. R CMD INSTALL to check if your package can be installed 3. R CMD check in order to finally check your package Running R CMD INSTALL on your source directory may pollute it, hence this is not recommended at all. Best, UWe
Best regards Christian On 5/3/11 1:07 PM, Uwe Ligges wrote:
On 02.05.2011 21:24, cstrato wrote:
Dear Prof. Ripley, Thank you for your confirmation and explanation, I understand the reason for cleaning things up to save memory. However, it was very convenient to have this feature in earlier versions of R. It would be really helpful to have an additional option to R CMD check, e.g. "--no-clean-vignettes". FYI, I did not claim "..create the vignettes *in<pkg>inst/doc*", instead my words were: One interesting observation is that xps.Rcheck from R-2.12.2 contains the subdirectory "inst/doc" with the vignettes while xps.Rcheck from R-2.13.0 does not contain "inst".
But you do not need it. I do not know how often I have to mention that vignettes are produced by R CMD build! They are already build when running R CMD check. And please do not tell us about tzhe PDF version oif manuals which are *unrelated* to vignettes, because they are not built in advance and need to be checked, since they should be produced at user level while vignettes are built at developer level already. Uwe Ligges
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 _._._._._._._._._._._._._._._._._._ On 5/2/11 7:08 AM, Prof Brian Ripley wrote:
On Sun, 1 May 2011, Duncan Murdoch wrote:
On 11-05-01 4:10 PM, cstrato wrote:
Dear Duncan, dear Uwe, Since I have installed both WinXP and OpenSUSE 11.3 on my Mac in a virtual machine, I have now done the following tests on all three architectures: 1, R CMD build xps: This creates "xps_1.13.1.tar.gz" which DOES contain all vignettes as pdf-file. Thus R CMD build is ok. 2, R CMD check xps: This does NOT build the vignettes as pdf-files on all three architectures. Or to be more precise, R-2.13.0 does no longer build the vignettes since with R-2.12.2 and earlier versions R did create the vignettes as pdf-files. Thus the question is: Why does R CMD check no longer create the vignettes?
Probably the answer is simply "because it doesn't". For a truly reliable check, you should build the package, then check the tar.gz file. Anything else is, and always has been, an approximation.
Actually, it does. What earlier versions never did (despite 'cstrato's repeated delusional claims earlier) was to create the vignettes *in <pkg>inst/doc*. All of them re-created (by default) vignettes in a working directory. The difference is that 2.13.0 deletes that working directory if the test was successful, whereas earlier versions left the results somewhere in<pkg>.Rcheck (the 'somewhere' has varied). However, earier versions of R CMD check sometimes failed when R CMD build succeeded Using Animal (a small CRAN package with one vignette). R 2.12.2 gave * checking package vignettes in ?inst/doc? ... WARNING Package vignettes without corresponding PDF: /tmp/Animal/inst/doc/Animal.Rnw and the vignette was re-created in Animal.Rcheck/inst/doc. R 2.13.0 gives * checking package vignettes in ?inst/doc? ... WARNING Package vignette(s) without corresponding PDF: Animal.Rnw Non-ASCII package vignette(s) without specified encoding: Animal.Rnw * checking running R code from vignettes ... OK * checking re-building of vignettes ... OK and the working directory was Animal.Rcheck/vign_test . The main reason for cleaning up is that to mimic R CMD build the test has to make a complete copy of the package sources, and that adds up: checking CRAN already takes 17GB for each flavour.
Duncan Murdoch
Best regards Christian On 4/27/11 10:16 AM, Uwe Ligges wrote:
On 26.04.2011 21:58, cstrato wrote:
Dear Duncan, dear Uwe,
Just now I have re-run everything, and today xps.Rnw
can be
converted to
a vignette w/o any problems using:
a, buildVignettes("xps", dir="/Volumes/CoreData/CRAN/xps",
quiet=F)
b, R CMD Sweave xps.Rnw
In both cases the vignette xps.pdf is created (maybe my
Mac did
not
like
to work during eastern holidays).
However, one issue remains:
"R64 CMD check xps_1.13.1.tar.gz" no longer creates any
pdf files
for
the vignettes.
Dioes it give an error or warning? It should check the code. R CMD build creates the pdf files. Uwe Ligges
Best regards Christian On 4/25/11 9:31 PM, Duncan Murdoch wrote:
On 25/04/2011 3:16 PM, cstrato wrote:
Thank you. My problem seems to be that at the moment the problem can be seen only on my Mac, since e.g. the Bioconductor servers have no problems creating the vignettes.
Then you are definitely the one in the best position to diagnose the problem. Use the usual approach: simplify it by cutting out everything that looks unrelated. Verify that the problem still exists, then cut some more. Eventually you'll have isolated the error to a particular small snippet of code, and then you can add print() statements, or use trace(), or do whatever is necessary to see what's so special about your system. I suspect it will turn out to be an assumption in the code that is not true on your system. If the assumption is being made by code you wrote, then fix it. If it's being made by R, let us know. Duncan Murdoch
Best regards Christian On 4/25/11 8:55 PM, Duncan Murdoch wrote:
cstrato wrote:
Dear Duncan, Thank you for your example, however it is different since it does not use x and y. What about print(x+y)?
Try it.
Sorry, I do not believe that there is a bug in my code, since: 1, it worked in all versions from R starting with R-2.6.0 till R-2.12.2. 2, the identical code works in the examples 3, this code (or a similar code) is the starting code which all users of xps have to use, and there was never a problem.
This might be a problem in R, or might be a problem in your code. As far as I know, it has only shown up in your code, so I'd guess that's where the problem is. In any case, you're the one in the best position to isolate it and debug it. If it turns out to be a problem in R, put together an example illustrating the problem that doesn't involve your code, and I'll take a look. Duncan Murdoch
Maybe the reason could be that my code has to import - the CEL-files from the package dir - the file SchemeTest3.root from the package dir ?? Best regards Christian On 4/25/11 8:00 PM, Duncan Murdoch wrote:
cstrato wrote:
Dear Uwe, Your suggestion to look at the Sweave manual helped me to solve the problem. It seems that in R-2.13.0 every chunk can use the code from the chunk before but not from an earlier chunk.
I'm either misreading what you wrote, or it's wrong. If I have this in a Sweave file: <<>>= x<- 1 @ <<>>= y<- 2 @ <<>>= print(x) @ I will see the value of x getting printed, even though it came from two chunks earlier. I think Uwe is right: there is some bug in the code you're running. Sweave isn't the problem. Duncan Murdoch
Concretely, the following does not work since chunk
5 needs
the
code
from chunk 3 and 4:
###################################################
### chunk number 3:
###################################################
#line 126 "xps.Rnw"
celdir<- file.path(.path.package("xps"), "raw")
###################################################
### chunk number 4:
###################################################
#line 132 "xps.Rnw"
scheme.test3<-
root.scheme(file.path(.path.package("xps"),
"schemes",
"SchemeTest3.root"))
###################################################
### chunk number 5:
###################################################
#line 137 "xps.Rnw"
celfiles<- c("TestA1.CEL","TestA2.CEL")
data.test3<- import.data(scheme.test3,
"tmpdt_DataTest3",
celdir=celdir, celfiles=celfiles, verbose=FALSE)
However, when I add "celdir" to chunk 5 then
everything
works
since
now chunk 5 needs only the code from chunk 4 but not
from
chunk 3:
###################################################
### chunk number 5:
###################################################
#line 137 "xps.Rnw"
celdir<- file.path(.path.package("xps"), "raw")
celfiles<- c("TestA1.CEL","TestA2.CEL")
data.test3<- import.data(scheme.test3,
"tmpdt_DataTest3",
celdir=celdir, celfiles=celfiles, verbose=FALSE)
Now buildVignettes() is able to create the
vignettes,
however
R CMD
check still does not build the vignettes.
Yes, I get a Warning in both cases:
* checking package vignettes in 'inst/doc' ...
WARNING
Package vignettes without corresponding PDF:
........
However, with R-2.12.2 the following lines are
added:
/Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/APTvsXPS.Rnw
/Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xps.Rnw
/Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xpsClasses.Rnw
/Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xpsPreprocess.Rnw
and in xps.Rcheck the subdirectory "inst/doc"
will be
created
which
contains the vignette data such as xps.Rnw, but also
xps.tex
and
xps.pdf.
In contrast, R-2.13.0 does not create the
subdirectory
"inst/doc"
and
no vignettes are built.
One more issue:
In contrast to my former believe R CMD INSTALL does
not
build
the
vignettes, neither in R-2.12.2 nor in R-2.13.0. I
have to
run
buildVignettes() after installation. Is this the
usual case?
Best regards
Christian
On 4/25/11 4:00 PM, Uwe Ligges wrote:
On 24.04.2011 23:10, cstrato wrote:
Dear Uwe, On 4/24/11 10:37 PM, Uwe Ligges wrote:
On 24.04.2011 20:59, cstrato wrote:
Dear Uwe, Thank you for your reply. ad 2, Yes, i know that "xps-manual.pdf" is the collection of help pages, I have mentioned it only to show that creating pdf-files does work for R-2.13.0. ad 1, Could it be that this is a Mac-specific problem since I see it on both my old MacBook Pro and my new Mac Mini.
Have you tried on any other OS? I did not since installing root is a bit too much effort.
No, until now I did not try another OS, but I will, since xps has to work on all three OSes.
Using R CMD check with R-2.12.2 I get:
I thought we are talking about R-2.13.0?
I showed you the output of R-2.12.2 first and then the output of R-2.13.0, so that you can see that on the same machine with the same file it works with one version but not the other.
But you got a Warning in both cases: * checking package vignettes in 'inst/doc' ... WARNING Package vignettes without corresponding PDF: ........
Where is the R CMD build output (since R CMD build is supposed to prepare the vignette).
$ R64 CMD check xps_1.13.1.tar.gz * using log directory '/Volumes/CoreData/CRAN/xps.Rcheck' * using R version 2.12.2 (2011-02-25) * using platform: x86_64-apple-darwin9.8.0 (64-bit) * using session charset: ASCII * checking for file 'xps/DESCRIPTION' ... OK * this is package 'xps' version '1.13.1' ... ... * checking package vignettes in 'inst/doc' ... WARNING Package vignettes without corresponding PDF: /Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/APTvsXPS.Rnw /Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xps.Rnw /Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xpsClasses.Rnw /Volumes/CoreData/CRAN/xps.Rcheck/00_pkg_src/xps/inst/doc/xpsPreprocess.Rnw * checking PDF version of manual ... OK Using "RSwitch.app" I switch to R-2.13.0, but now I get: $ R64 CMD check xps_1.13.1.tar.gz * using log directory '/Volumes/CoreData/CRAN/xps.Rcheck' * using R version 2.13.0 (2011-04-13) * using platform: x86_64-apple-darwin9.8.0 (64-bit) * using session charset: ASCII * checking for file 'xps/DESCRIPTION' ... OK * this is package 'xps' version '1.13.1' ... ... * checking package vignettes in 'inst/doc' ... WARNING Package vignette(s) without corresponding PDF: APTvsXPS.Rnw xps.Rnw xpsClasses.Rnw xpsPreprocess.Rnw * checking running R code from vignettes ... OK * checking re-building of vignettes ... OK * checking PDF version of manual ... OK I must admit that I have never built the vignettes manually, and I cannot find a hint how I can do it from the command line. Is this possible?
Is it possible to build the vignettes from the command line?
R --help suggests there is R CMD Sweave .....
However, building the vignettes within R-2.13.0 I get:
library(tools)
buildVignettes("xps",
dir="/Volumes/CoreData/CRAN/xps",
quiet=F)
Overfull ?vbox (21.68121pt too high) has
occurred while
?output is
Writing to file xps.tex
Processing code chunks with options ...
1 : term verbatim
2 : echo term hide
3 : echo term verbatim
4 : echo term verbatim
5 : echo term verbatim
SysError in<TFile::TFile>: file
/tmpdt_DataTest3_cel.root/tmpdt_DataTest3_cel_20110424_201301.root
can
not be opened (No such file or directory)
Error: Could not create
file</tmpdt_DataTest3_cel.root>
Here are the two important chunks 4 and 5, which
work
fine with
R-2.12.2
and earlier versions:
###################################################
### chunk number 4:
###################################################
#line 132 "xps.Rnw"
scheme.test3<-
root.scheme(file.path(.path.package("xps"),
"schemes",
"SchemeTest3.root"))
###################################################
### chunk number 5:
###################################################
#line 137 "xps.Rnw"
celfiles<- c("TestA1.CEL","TestA2.CEL")
data.test3<- import.data(scheme.test3,
"tmpdt_DataTest3",
celdir=celdir, celfiles=celfiles, verbose=FALSE)
However, in R-2.13.0 chunk 5 crashes!!!
It works only when replacing chunk 5 with:
###################################################
### chunk number 5:
###################################################
#line 137 "xps.Rnw"
celfiles<- c("TestA1.CEL","TestA2.CEL")
scheme.test3<-
root.scheme(file.path(.path.package("xps"),
"schemes",
"SchemeTest3.root"))
data.test3<- import.data(scheme.test3,
"tmpdt_DataTest3",
celdir=celdir, celfiles=celfiles, verbose=FALSE)
As you see R-2.13.0 does no longer remember the
result of
chunk 4,
i.e.
"scheme.test3". Now I have to include the line
defining
"scheme.test3"
also in chunk 5.
How objects generated in one chunk can be reused later on is explained in the Sweave manual.
Do you have any idea for this behavior?
An example, which does work in R-2.13.0 are the
following
two
chunks:
#################################################
### chunk number 20:
###################################################
#line 300 "xps.Rnw"
library(xps)
scheme.test3<-
root.scheme(file.path(.path.package("xps"),
"schemes",
"SchemeTest3.root"))
data.test3<- root.data(scheme.test3,
file.path(.path.package("xps"),"rootdata",
"DataTest3_cel.root"))
###################################################
### chunk number 21:
###################################################
#line 318 "xps.Rnw"
data.rma<- rma(data.test3, "tmpdt_Test3RMA",
verbose=FALSE)
In summary it is not quite clear to me what has
changed in
R-2.13.0 so
that chunk 5 no longer works.
BTW, the error:
SysError in<TFile::TFile>: file
/tmpdt_DataTest3_cel.root/tmpdt_DataTest3_cel_20110424_201301.root
can
not be opened
is not clear to me since
"/tmpdt_DataTest3_cel.root/" is
definitively
NOT a directory.
Time to debug what root.scheme is doing with the supplied path.
Sorry, I do not understand what you mean. root.scheme does not change the path, otherwise it would not work on all older versions of R.
Right, but since something changed obviously, and you tell us the path is not a path, it might be a function used by root.scheme .... That's why I said time to debug the code in your vignette!
Do you have any ideas? Please note that the vignette "xps.Rnw" did work for the last two years w/o problem. Furthermore, the Bioconductor servers are able to build the vignette, see: http://www.bioconductor.org/packages/release/bioc/html/xps.html
See http://bioconductor.org/checkResults/2.8/bioc-20110225/xps/liverpool-checksrc.html and find that their Mac Servers are also fine with it. So maybe it is your setup that is corrupted?
Maybe, you are right that my setup is corrupted, but on two independent machines? Maybe the download of "R-2.13.0.pkg" is corrupted, but then nothing should work, or am I wrong? How can I check if my setup is corrupted?
By debugging the code in your package's vignette. Uwe Ligges
Best regards Christian
Best wishes, Uwe
Best regards Christian On 4/24/11 4:12 PM, Uwe Ligges wrote:
On 23.04.2011 21:50, cstrato wrote:
Dear all, While R CMD check and R CMD INSTALL have always created the vignettes on R-2.12.1 or any earlier versions of R, I am no longer able to build the vignettes on R-2.13.0. Instead R CMD check gives me the following output: * checking for unstated dependencies in vignettes ... OK * checking package vignettes in 'inst/doc' ... WARNING Package vignette(s) without corresponding PDF: APTvsXPS.Rnw xps.Rnw xpsClasses.Rnw xpsPreprocess.Rnw * checking running R code from vignettes ... OK * checking re-building of vignettes ... OK * checking PDF version of manual ... OK Does someone know what the reason might be?
No, it does for me for other packages. Perhaps an error when processing the vignettes? Have you tried to build them manually?
(R64 CMD check --help says that be default rebuild-vignettes is turned on.) Interestingly, R CMD check still creates the file "xps-manual.pdf".
That is the collection of help pages, unrelated to the vignette. Uwe Ligges
Here is my sessionInfo:
sessionInfo()
R version 2.13.0 (2011-04-13) Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit) locale: [1] C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] xps_1.13.1 loaded via a namespace (and not attached): [1] Biobase_2.12.1 Biostrings_2.20.0 IRanges_1.10.0 [4] affy_1.30.0 affyPLM_1.28.5 affyio_1.20.0 [7] preprocessCore_1.14.0 Thank you in advance. 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 at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel