#3 could be assigned from within the user's .Rprofile via options("dbuser") or whatever this is how I have some REST services set up since it's a drag to write examples otherwise On Mon, Dec 20, 2010 at 8:35 AM, Sean Davis wrote: > On Mon, Dec 20, 2010 at 11:24 AM, Jeff Allen wrote: > > > Hello. > > > > I'm looking to create an R package which requires a user-specific > parameter > > in order to work. To be more descriptive: I'm making a database API > > available via an R package, but I want to control access to this API by > > using keys so that I can monitor activity at the user-level and make sure > no > > one is abusing the database. So the R package is complete, but I need a > way > > for users to specify their key so that it can be included in each call to > > the API. I'm looking for a way to avoid having the include the API key in > > every function call, but instead be able to include the key when you > first > > attach the package. > > > > The functionality I'm looking for would essentially be to be able to pass > a > > parameter to a package through the library call. That doesn't seem to be > > possible, so I'm looking for alternatives that would be user-friendly. > > > > I'd appreciate any thoughts! > > > > > Hi, Jeff. > > I think the standard way this is usually handled (in multiple languages, > not > just R) is to create a connection object (where connection credentials > would > be specified) and then use that object in further calls to the db API. You > might consider the DBI interface ( > http://cran.r-project.org/web/packages/DBI/index.html) as the basis for > your > API. > > That said, if you don't choose to use the standard interface, there are a > number of ways that this type of thing could be done. > > 1) Read a file from the system that contains the information (like .psqlrc > of .my.cnf file) > 2) Read an environment variable or two--these could be set before starting > R or using R commands to set the environment variable > 3) Setting an R option > 4) Other, more creative solutions such as reading the username, etc., from > environment > 5) Combinations of the above.... > 6) There may be others. > > All of these except (4) will require the user to specify once per R session > the information that you need. > > Sean > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioc-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/bioc-devel > -- If people do not believe that mathematics is simple, it is only because they do not realize how complicated life is. John von Neumann [[alternative HTML version deleted]]