[R] how to document user-defined packages for windows

Griffith Feeney feeney at un.org
Sat Mar 11 20:42:10 CET 2000

On Thu, 9 Mar 2000, Griffith Feeney wrote:

>> After several hours of trial and error I've figured out how to create
>> packages for use by rw1000 ("Writing R Extensions" seems not to cover
>> this).

[Prof Brian D Ripley wrote]
>You write documentation axactly as described in `Writing R Extensions',
>and you then install the package as descirbed in the rw-FAQ and in the
>`readme.packages' in rw1000sp.zip.

Thanks, this clarifies my confusion. All I want to do is put a few dozen
simple R programs in a package that I can attach to the search path with
library(). This to keep them from cluttering up the current workspace.
Having always downloaded and installed binaries of R, and never having used
gcc or make, it didn't occur to me that such a package should be created
with these tools. I succeeded in getting my package recognized by manually
creating the necessary directories and files, but I couldn't get the help
files recognized.

After appropriate preparation (reading the FAQ and readme files, installing
the various tools, no problem here, I think) I tried out the procedure
described in the Simple Ports section of readme.packages by downloading and
attempting to install the chron package. 'tar zxvf chron_2_2-2_tar.gz'
creates c:\rw1000\src\library\chron okay, but 'make pkg-chron' gives

C:\rw1000\src\gnuwin32>make pkg-chron
Too many parameters - /..
Makefile:49: *** missing separator.  Stop.

At this point, I'm lost. The Simple Ports section of readme.packages
doesn't say anything about editing (any of the various) make files, and
in any case the other references to such editing in the R documentation
are too terse (e.g., from INSTALL for R-1.0.0, "edit MkRules to set the
appropriate paths as needed") to guide someone who has never used make.

If it is the only way to get R to see my help files, I will eventually
find time to learn about make, but if there's a more direct route
I could move forward more quickly. Thanks in advance for any help.

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list