RMark, r code, github and Sweave

[jlaake]

I put a new release of version 2.1.1 of RMark on my github site (https://github.com/jlaake) and will release to Cran soon. It now checks to make sure individual covariate names are not longer than 10 characters which is a limitation in MARK. Oddly enough I had two separate reports about this problem in 2 days and I had not run into it personally because I use short variable names. If you are using time-varying covariates remember that the time value is tacked onto the end of the variable name and the entire name has to be less than 10 characters.

I mentioned before that I put new releases of RMark on my github site but I've received a number of questions about access to source so I thought I'd address that here. Almost all of the R code that I've developed is on my github site and you can access the source as well as the R package binary for some of the packages. github is a website intended to provide open and private access to code repositories. All of my code is public and open access. The 2 primary packages of interest will be RMark and mrds which is used in Distance for analysis of distance sampling data. Both are available on CRAN and you can access the source of any package on CRAN although most folks only install the package binary. The most recent source and binaries will always be on my github site. The source can be downloaded by clicking on the Download button which brings up a new screen. Press on download as zip to get the source. Note that it will download a zip file within a zip. Open the downloaded zip and then the zip file it contains and then you should be able to copy the directory (eg mrds or RMark) to somewhere on your disk. The sub-directory R contains the source and man contains the help files. The package binaries are listed on the same download page and can be individually selected. I keep the current and any archived versions there.

You do not need a github account to access public repositories. You do need an account to create your own repositories but it is free if all of your repositories are public. Feel free to explore R code (not necessarily just mine) to help with your learning process. If you want to learn how to develop your own packages downloading a package source is a good way to learn. Most of my packages use roxygen which embeds the help file material at the top of each source file. If you are interested in learning about package development there is documentation on the web and I suggest that you use the devtools R package to make it easier.

You can also find some documentation, R examples and information about using LyX and Sweave to create reproducible research documents at https://github.com/NMML/R-User-Meetings. I'll be updating my LyX/Sweave tutorial on that site within the next week.

regards --jeff