How to create vignette.pdf for R-2.13.0?

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
_._._._._._._._._._._._._._._._._._