Dear R users, A customer of mine has asked me to develop a R-package. The sitiuation is as follows: 1) Being a special purpose package, it does not merit to be uploaded to any CRAN server. 2) It is subject to many changes, bug fixes, and the like. 3) It has then to be distributed to a relatively large and sparse base of users. I thought that it would be convenient to have it distributed in the following way: 1) I upload new versions to a server. 2) I provide a function that points to that server and installs it. Therefore, users need only invoke this function in order to access the last package version. Suppose I upload the package to http://www.myserver.com/packages/mypackage_X.Y.Z.tar.gz Typically, several versions would be stored there. The uploading function should be invoked simply. For instance,> update.mypackage()Can anybody give me a hint as to which would be key command in this function how could the uploading function be defined so that the latest version would be updated or installed? Thank you very much for your help. Carlos J. Gil Bellosta http://www.datanalytics.com
There is an example here: http://www.braju.com/R/ Scroll down to the installation section and look at hblite.R On 5/5/06, Carlos J. Gil Bellosta <cgb at datanalytics.com> wrote:> Dear R users, > > A customer of mine has asked me to develop a R-package. The sitiuation > is as follows: > > 1) Being a special purpose package, it does not merit to be uploaded to > any CRAN server. > 2) It is subject to many changes, bug fixes, and the like. > 3) It has then to be distributed to a relatively large and sparse base > of users. > > I thought that it would be convenient to have it distributed in the > following way: > > 1) I upload new versions to a server. > 2) I provide a function that points to that server and installs it. > > Therefore, users need only invoke this function in order to access the > last package version. > > Suppose I upload the package to > > http://www.myserver.com/packages/mypackage_X.Y.Z.tar.gz > > Typically, several versions would be stored there. > > The uploading function should be invoked simply. For instance, > > > update.mypackage() > > Can anybody give me a hint as to which would be key command in this > function how could the uploading function be defined so that the latest > version would be updated or installed? > > Thank you very much for your help. > > Carlos J. Gil Bellosta > http://www.datanalytics.com > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html >
On Fri, 5 May 2006, Carlos J. Gil Bellosta wrote:> Can anybody give me a hint as to which would be key command in this > function how could the uploading function be defined so that the latest > version would be updated or installed?Read the R-admin manual about how to set up a repository. Then you can just use update.packages().> Dear R users, > > A customer of mine has asked me to develop a R-package. The sitiuation > is as follows: > > 1) Being a special purpose package, it does not merit to be uploaded to > any CRAN server. > 2) It is subject to many changes, bug fixes, and the like. > 3) It has then to be distributed to a relatively large and sparse base > of users. > > I thought that it would be convenient to have it distributed in the > following way: > > 1) I upload new versions to a server. > 2) I provide a function that points to that server and installs it. > > Therefore, users need only invoke this function in order to access the > last package version. > > Suppose I upload the package to > > http://www.myserver.com/packages/mypackage_X.Y.Z.tar.gz > > Typically, several versions would be stored there. > > The uploading function should be invoked simply. For instance, > >> update.mypackage() > > Can anybody give me a hint as to which would be key command in this > function how could the uploading function be defined so that the latest > version would be updated or installed? > > Thank you very much for your help. > > Carlos J. Gil Bellosta > http://www.datanalytics.com > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html >-- Brian D. Ripley, ripley at stats.ox.ac.uk 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, UK Fax: +44 1865 272595