Skip to content

Possible R CMD check problem (PR#3070)

4 messages · Friedrich Leisch, Ross Boylan

Original
Ross Boylan
# 
Using R 1.7.0 I get 
* checking parcv-manual.tex ... ERROR
Could not create DVI version.

Although there is no apparent error.  The dvi file exists.
Possibly there is some problem with my TeX setup, but the following
messages don't suggest that either.

Here's the full log, which does show some documentation issues:
sheep:~$R CMD check --library=.R/library/ src/parcv
* checking for working latex ... OK
* using log directory '/space/home/ross/parcv.Rcheck'
* checking for file 'parcv/DESCRIPTION' ... OK
* checking if this is a source package ... OK

* Installing *source* package 'parcv' ...
** R
** demo
** inst
** help
 >>> Building/Updating help pages for package 'parcv'
     Formats: text html latex example
* DONE (parcv)

* DONE (INSTALL)

* checking package directory ... OK
* checking for sufficient/correct file permissions ... OK
* checking DESCRIPTION meta-information ... OK
* checking package dependencies ... OK
* checking index information ... WARNING
Demos with missing or empty index information:
[1] "parDemo1"
See the information on INDEX files and package subdirectories in
* section
'Creating R packages' of the 'Writing R Extensions' manual.
* checking package subdirectories ... OK
* checking R files for syntax errors ... OK
* checking R files for library.dynam ... OK
* checking generic/method consistency ... OK
* checking for assignment functions with final arg not named 'value'
* ... OK
* checking Rd files ... WARNING
Rd files without 'alias':
  man/parcv-internal.Rd
These tags are required in an Rd file.
See chapter 'Writing R documentation' in manual 'Writing R
Extensions'.
* checking for undocumented objects ... WARNING
Undocumented code objects:
[1] "crossval.fit"      "crossval.outerfit" "crossval.setup"
[4] "gcv"
* checking for code/documentation mismatches ... OK
* checking for undocumented arguments in \usage ... OK
* creating parcv-Ex.R ... OK
* checking examples ... OK
* creating parcv-manual.tex ... OK
* checking parcv-manual.tex ... ERROR
Could not create DVI version.
This typically indicates Rd problems.
Friedrich Leisch
# 
Why do you think there is a bug in R rather than in your package? (I
assume that it is a package of yours, at least it's not on CRAN).

How should we be able to track this problem for you? You don't specify
your computing platform and give us no examples we could try to
reproduce.

Does running latex on parcv-manual.tex in a shell work?

fritz leisch
> Using R 1.7.0 I get 
  > * checking parcv-manual.tex ... ERROR
  > Could not create DVI version.

  > Although there is no apparent error.  The dvi file exists.
  > Possibly there is some problem with my TeX setup, but the following
  > messages don't suggest that either.

  > Here's the full log, which does show some documentation issues:
  > sheep:~$R CMD check --library=.R/library/ src/parcv
  > * checking for working latex ... OK
  > * using log directory '/space/home/ross/parcv.Rcheck'
  > * checking for file 'parcv/DESCRIPTION' ... OK
  > * checking if this is a source package ... OK

  > * Installing *source* package 'parcv' ...
  > ** R
  > ** demo
  > ** inst
  > ** help
  >>>> Building/Updating help pages for package 'parcv'
  >      Formats: text html latex example
  > * DONE (parcv)

  > * DONE (INSTALL)

  > * checking package directory ... OK
  > * checking for sufficient/correct file permissions ... OK
  > * checking DESCRIPTION meta-information ... OK
  > * checking package dependencies ... OK
  > * checking index information ... WARNING
  > Demos with missing or empty index information:
  > [1] "parDemo1"
  > See the information on INDEX files and package subdirectories in
  > * section
  > 'Creating R packages' of the 'Writing R Extensions' manual.
  > * checking package subdirectories ... OK
  > * checking R files for syntax errors ... OK
  > * checking R files for library.dynam ... OK
  > * checking generic/method consistency ... OK
  > * checking for assignment functions with final arg not named 'value'
  > * ... OK
  > * checking Rd files ... WARNING
  > Rd files without 'alias':
  >   man/parcv-internal.Rd
  > These tags are required in an Rd file.
  > See chapter 'Writing R documentation' in manual 'Writing R
  > Extensions'.
  > * checking for undocumented objects ... WARNING
  > Undocumented code objects:
  > [1] "crossval.fit"      "crossval.outerfit" "crossval.setup"
  > [4] "gcv"
  > * checking for code/documentation mismatches ... OK
  > * checking for undocumented arguments in \usage ... OK
  > * creating parcv-Ex.R ... OK
  > * checking examples ... OK
  > * creating parcv-manual.tex ... OK
  > * checking parcv-manual.tex ... ERROR
  > Could not create DVI version.
  > This typically indicates Rd problems.

  > ______________________________________________
  > R-devel@stat.math.ethz.ch mailing list
  > https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
Ross Boylan
#
On Thu, 2003-05-22 at 03:52, Friedrich.Leisch@ci.tuwien.ac.at wrote:
Well, I did say "possible" problem.  The reason I think so is that part
of the error message (Could not create DVI version) is wrong (a dvi file
was produced) and part of it (This typically indicates Rd problems) is
misleading.  There is also some chance that the problems are deeper than
that.
I'm happy to try to isolate it here.  I'm running on Solaris 2.8.  I can
send the files if that would help.
Yes (once I found out how to tell it where Rd.sty was).  However, there
is an error in the tex log, also visible (now that I look!) in the
original run.  It seems to indicate some problem with my font setup:
-----------------------------------
(/usr/local/teTeX/share/texmf/tex/latex/base/t1enc.def)kpathsea: Running
mktextfm  ecrm1000
mkdir: Failed to make directory "/usr/local/teTeXfonts/tfm"; Permission
denied
kpsestat: /usr/local/teTeXfonts/tfm/..: No such file or directory
usage:  chmod [-fR] <absolute-mode> file ...
        chmod [-fR] <symbolic-mode-list> file ...
where   <symbolic-mode-list> is a comma-separated list of
        [ugoa]{+|-|=}[rwxXlstugo]
mkdir: Failed to make directory "/usr/local/teTeXfonts/tfm/jknappen"; No
such file or directory
kpsestat: /usr/local/teTeXfonts/tfm/jknappen/..: No such file or
directory
usage:  chmod [-fR] <absolute-mode> file ...
        chmod [-fR] <symbolic-mode-list> file ...
where   <symbolic-mode-list> is a comma-separated list of
        [ugoa]{+|-|=}[rwxXlstugo]
mkdir: Failed to make directory "/usr/local/teTeXfonts/tfm/jknappen/ec";
No such file or directory
kpsestat: /usr/local/teTeXfonts/tfm/jknappen/ec/..: No such file or
directory
usage:  chmod [-fR] <absolute-mode> file ...
        chmod [-fR] <symbolic-mode-list> file ...
where   <symbolic-mode-list> is a comma-separated list of
        [ugoa]{+|-|=}[rwxXlstugo]
mktextfm: mktexdir /usr/local/teTeXfonts/tfm/jknappen/ec failed.
kpathsea: Appending font creation commands to missfont.log.

! Font T1/cmr/m/n/10=ecrm1000 at 10.0pt not loadable: Metric (TFM) file
not fou
nd.
<to be read again> 
                   relax 
l.96 \fontencoding\encodingdefault\selectfont
                                             
? 
) (/usr/local/teTeX/share/texmf/tex/latex/ae/ae.sty
(/usr/local/teTeX/share/texmf/tex/latex/base/fontenc.sty
(/usr/local ....
--------------------------------------

So here's my theory:
latex runs into the problem above, which appears to be with our local
TeX setup, and returns an error code.
R thinks the code means TeX has failed to produce the dvi file.  And
then it emits the error message.

Does this seem a plausible diagnosis?

If that's correct, the undesirable behavior on R's part just relates to
the error handling and reporting.
Friedrich Leisch
# 

        
> On Thu, 2003-05-22 at 03:52, Friedrich.Leisch@ci.tuwien.ac.at wrote:

        
>> Why do you think there is a bug in R rather than in your package? (I
  >> assume that it is a package of yours, at least it's not on CRAN).

  > Well, I did say "possible" problem.

Yes, but you also filed it into the R bug tracking system ... and
reporting "bugs" that in fact are mostly installation problems on your side
doesn't make life easier for us. 


  > The reason I think so is that part
  > of the error message (Could not create DVI version) is wrong (a dvi file
  > was produced)

Thanks for the hint, I'll fix that message.


  > and part of it (This typically indicates Rd problems) is
  > misleading. There is also some chance that the problems are deeper than
  > that.

Well, the message says that *typically* Rd problems are the reason
(which probably is true in 95% of all case), not that they are the
only possible cause.

Regards