This function prepares the computer for cmsy_run(): it installs the missing R packages from
CRAN, downloads the JAGS installer that matches this system and R version (see
cmsy_jags_versions(); with cmsy_download_jags(), which checks its size and checksums) and
opens it, and checks the bundled CMSY++ files. It asks before installing anything when R is
used interactively.
Usage
cmsy_setup(
jags = "recommended",
os = "auto",
ask = TRUE,
data_dir = fao_data_dir(),
silent = fao_setting("silent")
)Arguments
- jags
The JAGS version to recommend: "recommended" (default; the newest JAGS 4 release built for this R version, which works with rjags from CRAN), "latest" (JAGS 5.0.0, which needs rjags 5-x from SourceForge) or a version number such as "4.3.2".
- os
The operating system whose settings are used: "auto" (default, detect it), "windows", "macos" or "linux".
- ask
A logical value. If TRUE (default), ask before installing anything. In a non-interactive session nothing is installed when
askis TRUE.- data_dir
The directory where downloads are kept. Default is
fao_data_dir().- silent
A logical value. If TRUE, no summary is printed. Default is FALSE, or the value set with
fao_settings(silent = ).
Value
Invisibly, the result of cmsy_system() after the installation. With silent = TRUE
no summary or download messages are printed (questions are still asked when ask
is TRUE).
Details
JAGS is a separate program, so it cannot be installed from R. On Windows and macOS the
installer is opened and you click through it; on Linux the commands to install it are shown.
Restart R after installing JAGS and run cmsy_setup() again to check.
The JAGS installer must match the R version (see cmsy_jags_versions()): on Windows JAGS
4.3.2 for R 4.5 and 4.6 and JAGS 4.3.1 for R 4.2 to 4.4; on macOS JAGS 4.3.2 for R 4.3 and
later and 4.3.1 for R 4.2; on Linux the JAGS of the package manager. JAGS 4 is used by
default because rjags on CRAN (4-17) needs JAGS 4. JAGS 5.0.0 (jags = "latest") also
installs rjags 5-3 from SourceForge.
References
Plummer, M. (2003). JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003), Vienna. https://mcmc-jags.sourceforge.io/
