Components of ecosystem assessment with BiocBuildTools

Overview

In this vignette we illustrate the use of software to capture information from R CMD check and BiocCheck::BiocCheck in concise formats.

Harvesting check outputs

Acquire some source repos

td = tempdir()
od = getwd()
setwd(td)
library(BiocBuildTools)
ps = PackageSet(c("parody", "BiocFileCache"))

## 'getOption("repos")' replaces Bioconductor standard repositories, see
## '?repositories' for details
## 
## replacement repositories:
##     CRAN: https://cloud.r-project.org

getwd()

## [1] "/tmp/RtmpIao2O2"

system("mkdir gits")
populate_local_gits(ps, "gits")

Some helper software

date_string = function(x) as.character(as.Date(x, origin="1970-01-01"))
last_commit_date = function(repo) {
  ans = Sys.time() + NA # if gert not available, produce a classed NA
  if (requireNamespace("gert")) ans = gert::git_log(repo=repo, max=1)$time
  ans
}

Run R CMD check

setwd(td)
rc = lapply(dir("gits", full=TRUE), function(x) {
   ans = rcmdcheck::rcmdcheck(x,quiet=TRUE)
   attr(ans, "last_commit_date") <- last_commit_date(x)
   attr(ans, "check_date") <- Sys.time()
   ans
   })
rdfs = rcclist_to_dataframes(rc)

Run BiocCheck

In this chunk we use BiocParallel to get separate processes for BiocCheck runs. Without this we can see namespace detachment problems.

setwd(td)
library(BiocParallel)
register(SnowParam(2))  # gets around namespace detachment issue
ini = dir("gits")
bdfs = bplapply(dir("gits", full=TRUE), function(x) {
   ans = BiocCheck::BiocCheck(x)
   attr(ans, "last_commit_date") <- last_commit_date(x)
   attr(ans, "check_date") <- Sys.time()
   bco2df(ans)
   })
names(bdfs) = ini

Results

names(rdfs)

## [1] "basic"    "notes"    "warnings" "errors"   "inst"     "desc"

rdfs[1:4]

## $basic
##         package version date_commit date_check
## 1 BiocFileCache   2.6.0  2022-11-01 2022-11-16
## 2        parody  1.56.0  2022-11-01 2022-11-16
## 
## $notes
##         package
## 1 BiocFileCache
## 2        parody
##                                                                                                                                                                                                                                                                                                                                                                                                 notes
## 1 checking R code for possible problems ... NOTE\n.sql_clean_cache: no visible binding for global variable ‘rid’\n.sql_clean_cache: no visible binding for global variable ‘access_time’\n.sql_get_resource_table: no visible binding for global variable ‘rid’\n.sql_get_resource_table: no visible binding for global variable ‘id’\nUndefined global functions or variables:\n  access_time id rid
## 2                                                                                                                                                                                                                                                                                                                                                                                            NO NOTES
##   date_commit date_check
## 1  2022-11-01 2022-11-16
## 2  2022-11-01 2022-11-16
## 
## $warnings
##         package    warnings date_commit date_check
## 1 BiocFileCache NO WARNINGS  2022-11-01 2022-11-16
## 2        parody NO WARNINGS  2022-11-01 2022-11-16
## 
## $errors
##         package    errors date_commit date_check
## 1 BiocFileCache NO ERRORS  2022-11-01 2022-11-16
## 2        parody NO ERRORS  2022-11-01 2022-11-16

bdfs

## $BiocFileCache
## $BiocFileCache$errors
##                        type
## 1 checkExportsAreDocumented
## 2 checkExportsAreDocumented
## 3 checkExportsAreDocumented
##                                                                                        message
## 1 * ERROR: At least 80% of man pages documenting exported objects must have runnable examples.
## 2                                                                  The following pages do not:
## 3                                                            makeBiocFileCacheFromDataFrame.Rd
##   commit_date check_date
## 1  2022-11-01 2022-11-16
## 2  2022-11-01 2022-11-16
## 3  2022-11-01 2022-11-16
## 
## $BiocFileCache$warnings
##                                   type
## 1                   checkVersionNumber
## 2 checkDescriptionNamespaceConsistency
## 3                 checkForValueSection
## 4                 checkForValueSection
##                                                                  message
## 1                   * WARNING: y of x.y.z version should be odd in devel
## 2           * WARNING: Import tools in DESCRIPTION as well as NAMESPACE.
## 3 * WARNING: Add non-empty \\value sections to the following man pages: 
## 4                                         man/makeCachedActiveBinding.Rd
##   commit_date check_date
## 1  2022-11-01 2022-11-16
## 2  2022-11-01 2022-11-16
## 3  2022-11-01 2022-11-16
## 4  2022-11-01 2022-11-16
## 
## 
## $parody
## $parody$errors
##    type message commit_date check_date
## 1 error    none  2022-11-01 2022-11-16
## 
## $parody$warnings
##                                   type
## 1                   checkVersionNumber
## 2 checkDescriptionNamespaceConsistency
## 3                 checkForValueSection
## 4                 checkForValueSection
##                                                                  message
## 1                   * WARNING: y of x.y.z version should be odd in devel
## 2           * WARNING: Import stats in DESCRIPTION as well as NAMESPACE.
## 3 * WARNING: Add non-empty \\value sections to the following man pages: 
## 4                                       man/box.scale.Rd, man/tukeyor.Rd
##   commit_date check_date
## 1  2022-11-01 2022-11-16
## 2  2022-11-01 2022-11-16
## 3  2022-11-01 2022-11-16
## 4  2022-11-01 2022-11-16

Vincent J. Carey, stvjc at channing.harvard.edu