On 19/08/2010 7:55 AM, Marius Hofert wrote:
Dear Tim,

if I run ...

## Example 2: texi2dvi also deletes the generated .pdf
require(tools)
filepath <- paste(getwd(),"/test/myLaTeXFile2.tex",sep="")
sink(file=filepath)
          cat("\\documentclass{article}\n")
          cat("\\usepackage[T1]{fontenc}\n")
          cat("\\usepackage[english]{babel}\n")
          cat("\\usepackage{blindtext}\n")
          cat("\\begin{document}\n")
          cat("\\blindtext[10]\n")
          cat("\\end{document}\n")
sink()
texi2dvi(filepath,pdf=TRUE,clean=TRUE,quiet=TRUE)
... on a "fresh-and-clean" working directory (and also no files in the 
subdirectory ./test), I only get the file myLaTeXFile2.tex in ./test, so *everything* 
(including the .pdf) is deleted :-(

The texi2dvi function has two possible execution paths: in one, it basically constructs a call to the system texi2dvi command and calls that, and in the other it emulates the behaviour of texi2dvi. I haven't checked recently, but I think the second method doesn't do anything with clean=TRUE, so it looks as though this is a problem with the texi2dvi command you have installed on your system.

To confirm this, run debug(tools:::.shell_with_capture), and run texi2dvi. When R breaks in that .shell_with_capture, it will show you what it is trying to execute. If it is using your system texi2dvi command, it will call it once with option --help to find out what options are supported, and then a second call will show you the main command line. You can use this to report a bug to Apple or whoever supplied your texi2dvi.

Duncan Murdoch


I work with R version 2.11.1 (2010-05-31) on a MacBook Pro running 10.6.4.

Cheers,

Marius

On 2010-08-19, at 13:38 , Tim Gruene wrote:

> Dear Marius,
> > yes, that's what I had meant. > > I am sorry I had not actually tested my suggestion nor your problem (bad
> practice). Now I have and actually in case 2 you describe I only end up
> with test/myLaTeXFile2.tex and myLaTeXFile2.pdf.
> > If, however, I run texi2dvi at least once with option 'clean=F', the auxiliary files
> are generated and left in the path. Also upon a subsequent run with the option
> 'clean=T', they are not being deleted. > > Could this also be the case for you, i.e., that you once used the option
> 'clean=F' (or not set at all) and now are made believe 'clean=T' does not work
> because they are still there but actually left-overs from the old run?
> > I run "R version 2.7.1 (2008-06-23)" on Debian stable. > > Kind regards, Tim > > On Thu, Aug 19, 2010 at 12:41:57PM +0200, Marius Hofert wrote:
>> Dear Tim,
>> >> you mean >> >> require(tools)
>> filepath <- paste(getwd(),"/test/myLaTeXFile2.tex",sep="")
>> sink(file=filepath)
>>          cat("\\documentclass{article}\n")
>>          cat("\\usepackage[T1]{fontenc}\n")
>>          cat("\\usepackage[english]{babel}\n")
>>          cat("\\usepackage{blindtext}\n")
>>          cat("\\begin{document}\n")
>>          cat("\\blindtext[10]\n")
>>          cat("\\end{document}\n")
>> sink()
>> texi2dvi(filepath,pdf=TRUE,quiet=TRUE,texi2dvi=getOption("texi2dvi 
--mostly-clean"))
>> >> ? >> The problem is that this does not work. It produces the same output as my "Example 2" without the "clean"-option. So it leaves all files in the working directory. >> >> The reason why I am looking for a "clean-up" is that I have to compile several hundred LaTeX files in different directories... >> >> Cheers, >> >> Marius >> >> >> On 2010-08-19, at 12:33 , Tim Gruene wrote: >> >>> Dear Marius, >>> >>> from reading the texi2dvi man-page and R-help on texi2dvi I suggest to set
>>> 'clean=F' and use 'texi2dvi = getOption("texi2dvi --mostly-clean")' instead.
>>> >>> Tim >>> >>> On Thu, Aug 19, 2010 at 10:36:30AM +0200, Marius Hofert wrote:
>>>> Dear expeRts,
>>>> >>>> below are two examples of using texi2dvi. The problem becomes clear from the >>>> second example: on cleaning, texi2dvi also deletes the generated .pdf file.
>>>> Is there a solution?
>>>> >>>> Cheers, >>>> >>>> Marius >>>> >>>> ## Example 1: texi2dvi cleans fine, without deleting the generated .pdf
>>>> require(tools)
>>>> sink(file="myLaTeXFile1.tex")
>>>>         cat("\\documentclass{article}\n")
>>>>         cat("\\usepackage[T1]{fontenc}\n")
>>>>         cat("\\usepackage[english]{babel}\n")
>>>>         cat("\\usepackage{blindtext}\n")
>>>>         cat("\\begin{document}\n")
>>>>         cat("\\blindtext[10]\n")
>>>>         cat("\\end{document}\n")
>>>> sink()
>>>> texi2dvi("myLaTeXFile1.tex",pdf=TRUE,clean=TRUE,quiet=TRUE) >>>> >>>> ## Example 2: texi2dvi also deletes the generated .pdf
>>>> require(tools)
>>>> filepath <- paste(getwd(),"/test/myLaTeXFile2.tex",sep="")
>>>> sink(file=filepath)
>>>>         cat("\\documentclass{article}\n")
>>>>         cat("\\usepackage[T1]{fontenc}\n")
>>>>         cat("\\usepackage[english]{babel}\n")
>>>>         cat("\\usepackage{blindtext}\n")
>>>>         cat("\\begin{document}\n")
>>>>         cat("\\blindtext[10]\n")
>>>>         cat("\\end{document}\n")
>>>> sink()
>>>> texi2dvi(filepath,pdf=TRUE,clean=TRUE,quiet=TRUE)
>>>> >>>> ______________________________________________
>>>> R-help@r-project.org mailing list
>>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>>> PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
>>>> and provide commented, minimal, self-contained, reproducible code.
>>> >>> -- >>> --
>>> Tim Gruene
>>> Institut fuer anorganische Chemie
>>> Tammannstr. 4
>>> D-37077 Goettingen
>>> >>> GPG Key ID = A46BEE1A >>> > > -- > --
> Tim Gruene
> Institut fuer anorganische Chemie
> Tammannstr. 4
> D-37077 Goettingen
> > GPG Key ID = A46BEE1A >
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to