Re: [R] dvips function gives Documents not found error
On Thu, 27 Nov 2003, Dirk Eddelbuettel wrote: On Thu, Nov 27, 2003 at 04:21:04PM -0800, David Kelly wrote: Frank Harrell wrote: Did you make the changes I suggested (surrounding two items by dQuote())? Have you tried Linux? No, I didn't make those changes because I've just been working with a binary distribution. I may go ahead and pull down sources and try; I was But that's the beauty of it -- even in Windoze, the $R_HOME/library/$FOO/R/ directory for a package $FOO contains simple text code which you can edit. If you know how to read and write S source code, I may take you only about one minute to locate the file, and insert the suggested dQuote(). trying to avoid that. I wish I were working with Linux, but I'm doing this work for someone with many installed PCs that want to use R, and they're all running Win2K and that isn't going to change. If you organise your work environment wel, operating systems matter less and less. R, Perl, Python, ... are pretty much completely cross-platform (unless you insist on using OS-specific features). Well, it is sort of the latter. Almost all current OSes allow spaces in paths, but only Windows users seem to make use of them. Which is why early on I suggested that it would be good if Frank could ensure they are allowed in Hmisc. However (and I don't know if this is the issue here) there is the concept of `quoting hell' and quoting is certainly not even cross-sh-clones on Unix-alike. And since Dirk mentions Perl, the system() call in Perl does not even work the same on Windows 98 and 2000/XP, something R has always managed. I will try over the weekend to track down exactly where the problem lies, but only because it may help us to be offer some general guidelines. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
This is primarily an FYI to the Hmisc author, though any brilliant suggestions from him or anyone else are always welcome. Re this problem reported earlier in this thread: getting an error from Hmisc when trying dvips(latex(describe(mtcars)),file=/kellytest/kelly.ps) that says Error in system(cmd, intern = intern, wait = wait | intern, show.output.on.console = wait, : C:/Documents not found which appears to be a path-parsing problem of Documents and Settings The problem is still present. I've tried every suggestion that was sent to me. Most recently, I created a .Renviron which said TMP=c:/kellytest/rtemp and I then ran R (from within emacs/ess). I verified that TMP had been set correctly as follows: tempdir() [1] c:/kellytest/rtemp\\Rtmp5417 but I still got the C:/Documents not found error. P.S. I found that a TeX file was created when the command was executed, but I could find no evidence of a dvi or ps file. -- David Kelly __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
On Thu, 27 Nov 2003 10:04:51 -0800 David Kelly [EMAIL PROTECTED] wrote: This is primarily an FYI to the Hmisc author, though any brilliant suggestions from him or anyone else are always welcome. Re this problem reported earlier in this thread: getting an error from Hmisc when trying dvips(latex(describe(mtcars)),file=/kellytest/kelly.ps) that says Error in system(cmd, intern = intern, wait = wait | intern, show.output.on.console = wait, : C:/Documents not found which appears to be a path-parsing problem of Documents and Settings The problem is still present. I've tried every suggestion that was sent to me. Most recently, I created a .Renviron which said TMP=c:/kellytest/rtemp and I then ran R (from within emacs/ess). I verified that TMP had been set correctly as follows: tempdir() [1] c:/kellytest/rtemp\\Rtmp5417 but I still got the C:/Documents not found error. P.S. I found that a TeX file was created when the command was executed, but I could find no evidence of a dvi or ps file. -- David Kelly __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help Did you make the changes I suggested (surrounding two items by dQuote( ))? Have you tried Linux? :) --- Frank E Harrell JrProfessor and ChairSchool of Medicine Department of BiostatisticsVanderbilt University __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
Frank Harrell wrote: Did you make the changes I suggested (surrounding two items by dQuote())? Have you tried Linux? No, I didn't make those changes because I've just been working with a binary distribution. I may go ahead and pull down sources and try; I was trying to avoid that. I wish I were working with Linux, but I'm doing this work for someone with many installed PCs that want to use R, and they're all running Win2K and that isn't going to change. Thanks for the reply - David Kelly __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
On Thu, Nov 27, 2003 at 04:21:04PM -0800, David Kelly wrote: Frank Harrell wrote: Did you make the changes I suggested (surrounding two items by dQuote())? Have you tried Linux? No, I didn't make those changes because I've just been working with a binary distribution. I may go ahead and pull down sources and try; I was But that's the beauty of it -- even in Windoze, the $R_HOME/library/$FOO/R/ directory for a package $FOO contains simple text code which you can edit. If you know how to read and write S source code, I may take you only about one minute to locate the file, and insert the suggested dQuote(). trying to avoid that. I wish I were working with Linux, but I'm doing this work for someone with many installed PCs that want to use R, and they're all running Win2K and that isn't going to change. If you organise your work environment wel, operating systems matter less and less. R, Perl, Python, ... are pretty much completely cross-platform (unless you insist on using OS-specific features). Dirk -- Those are my principles, and if you don't like them... well, I have others. -- Groucho Marx __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
On Mon, 24 Nov 2003 10:53:48 + (GMT) Prof Brian Ripley [EMAIL PROTECTED] wrote: This is a question about Hmisc, I believe, although you certainly did not mention it. Looking at the source code I see sys(paste(cd, tempdir(), sc, optionsCmds(latex), tmp)) do my guess is that it is writing .dvi files in tempdir(). If you look up help on tempdir() you will see how to avoid this, Moral: don't guess, and don't make your readers guess either. And to Frank Harrell as author of Hmisc: spaces are valid in file names even on Unix/Linux, so it would be nice if this could be corrected. I've resisted this until now. The fix appears to be to surround tempdir() and tmp by dQuote( ). This will be in the next release. I'm assuming that superfluous quotes never hurt. -Frank --- Frank E Harrell JrProfessor and ChairSchool of Medicine Department of BiostatisticsVanderbilt University __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
Thanks to you both for your suggestions, and great to hear from the author himself! I may be thick, but I'm still having the problem. I did look at help for tempdir(), but tempdir() just *returns* the directory that's used; it doesn't allow me to change it. As a slightly different tack, I tried using tempfile(dsk,\\kellytest) but I still got the same error. From looking at the source code line mentioned below, I wouldn't expect the tempfile() call that I tried to make a difference anyway. So: Is there someway I can change the value of tempdir() so I can influence Hmisc? Also, in the source line quoted below, what is tmp and do I have some way to change that to give it a path without spaces? I'm looking for a workaround here to allow to do some further testing/prototyping, not a permanent fix. Thanks again -- David Kelly -- On Mon, 24 Nov 2003 10:53:48 + (GMT) Prof Brian Ripley [EMAIL PROTECTED] wrote: This is a question about Hmisc, I believe, although you certainly did not mention it. Looking at the source code I see sys(paste(cd, tempdir(), sc, optionsCmds(latex), tmp)) do my guess is that it is writing .dvi files in tempdir(). If you look up help on tempdir() you will see how to avoid this, Moral: don't guess, and don't make your readers guess either. And to Frank Harrell as author of Hmisc: spaces are valid in file names even on Unix/Linux, so it would be nice if this could be corrected. I've resisted this until now. The fix appears to be to surround tempdir() and tmp by dQuote( ). This will be in the next release. I'm assuming that superfluous quotes never hurt. -Frank --- Frank E Harrell JrProfessor and ChairSchool of Medicine Department of BiostatisticsVanderbilt University __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
Re: [R] dvips function gives Documents not found error
On Mon, 24 Nov 2003 11:09:17 -0800, David Kelly [EMAIL PROTECTED] wrote: So: Is there someway I can change the value of tempdir() so I can influence Hmisc? Not after R has started, but if you set the TMP or TEMP environment variable to the name of an existing directory, that will be used as the base of temp directories. You can do this on the R command line, e.g. set the Target field in the R shortcut to C:\R\rw1081\bin\Rgui.exe TMP=c:/junk More on this is in ?tempdir and ?Startup. Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
[R] dvips function gives Documents not found error
I'm new to R, and using version 1.8.0 on Windows XP, under emacs/ess. I'm trying to test some simple output flow using functions in the Hmisc package and hit an error that may be a more general problem than just my specific situation. If I issue the command latex(describe(mtcars), file = ) I get what looks like good TeX code output to the console. However, if I issue the command dvips(latex(describe(mtcars)),file=friday.ps) I get the following error: Error in system(cmd, intern = intern, wait = wait | intern, show.output.on.console = wait, : C:/Documents not found and I don't find any postscript output file. I'm guessing the C:/Documents reference is actually a problem with something dealing with paths with embedded spaces and that it's failing to deal with a path like C:/Documents and Settings/ What I've been unable to figure out is what file reference it's hiccuping on, so that I can hopefully find a workaround. I tried several things without any change in behavior: - insuring that the working directory for R didn't have any spaces in the pathname. - changing my TEMP and TMP environment variables to paths without any spaces. - in the possibility that some piece of software (emacs, R, LaTex) was looking for the TMPDIR environment variable, tried setting that to a path without any spaces. Any ideas, suggestions for further experiments, or solutions would be welcome! A copy of any reply to my email address (as well as just posting to the list) would be appreciated. Thank you - David Kelly __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help