Logging Packages

There are several package options for logging in R. In no particular order:

• logger. Last CRAN update 2 years ago (as of 9/2023)
• log4r. Hadley has added a few PRs and has particular interest into it since Aaron Jacobs is the maintainer and currently working at Posit
• futile.logger. Have not fully explored. Still being maintained.
• azlogr. Extension of logger package, but conveniently sends logs to Azure Log Analytics Workspace via API, which can be useful.
• paws. Paws is a Package for Amazon Web Services in R. Convenient package for sending logs to Amazon CloudWatch

In the upcoming examples, I will be using log4r, simply because that is what I am primarily using today. When I examined different logging options, what attracted me to use log4r was the microbenchmark analysis done by the package developer of log4r, and log4r performed quite well when timing how long it took to print out messages:

print(info_bench, order = "median")
#> Unit: microseconds
#>           expr      min        lq       mean    median        uq       max
#>            cat   19.727   24.8665   29.67838   27.4510   30.4120  1059.446
#>          log4r   14.296   26.0990   38.78562   32.0350   38.8730  2111.388
#>           rlog   44.754   59.5615   67.99791   66.2240   73.0670   744.355
#>         logger  152.856  175.9850  196.45468  186.4345  200.5960  2265.367
#>        logging  195.847  222.2560  252.33351  233.1320  248.1000  5634.985
#>            lgr  371.837  402.0790  498.85653  415.8595  437.0245 25077.198
#>         loggit  514.625  551.1530  632.20753  567.2740  587.6315  3101.445
#>  futile.logger 1509.811 1593.7180 1680.18994 1626.2340 1662.5675  5869.816