This commit and previous should address remaining ropensci reviewer i…

…ssues.

#21 #22 #23 #24 #25 #26 #27 #28 #29 #30 #31 #32

DESCRIPTION

@@ -4,7 +4,7 @@ Title: Construct Reproducible Analytic Data Sets as R Packages
 Authors@R:
     c(person(given = "Greg Finak", role=c("aut","cre","cph"), email="[email protected]"),
       person(given = "Paul Obrecht", role=c("ctb")))
-Version: 0.13.6
+Version: 0.14.0
 Description: Construct reproducible analytic data sets as R packages. 
 License: MIT + file LICENSE
 Depends: R (>= 3.5.0)
@@ -28,6 +28,7 @@ RoxygenNote: 6.0.1
 Suggests:
     testthat,
     covr,
-    data.tree
+    data.tree,
+    mtcars20
 URL: https://github.com/RGLab/DataPackageR
 BugReports: https://github.com/RGLab/DataPackageR/issues

R/build.R

@@ -7,6 +7,7 @@
 #' @param packageName \code{character} path to package source directory. Defaults to the current path when NULL.
 #' @param vignettes \code{logical} specify whether to build vignettes. Default FALSE.
 #' @param log log level \code{INFO,WARN,DEBUG,FATAL}
+#' @param deps \code{logical} should we pass data objects into subsequent scripts? Default TRUE
 #' @importFrom roxygen2 roxygenise roxygenize
 #' @importFrom devtools build_vignettes build parse_deps
 #' @importFrom usethis use_build_ignore use_rstudio proj_set use_directory
@@ -32,7 +33,8 @@
 #' package_build(file.path(tempdir(),pname))
 package_build <- function(packageName = NULL,
                           vignettes = FALSE,
-                          log=INFO) {
+                          log = INFO,
+                          deps = TRUE) {
   flog.threshold(log)
   flog.appender(appender.console())
   requireNamespace("rprojroot")
@@ -71,7 +73,7 @@ package_build <- function(packageName = NULL,
 
   # Return success if we've processed everything
   success <-
-    DataPackageR(arg = package_path)
+    DataPackageR(arg = package_path, deps = deps)
   ifelse(success,
     flog.info("DataPackageR succeeded"),
     flog.warn("DataPackageR failed")

R/processData.R

@@ -108,12 +108,13 @@ NULL
 #' Meant to be called before R CMD build.
 #' @name DataPackageR
 #' @param arg \code{character} name of the package to build.
+#' @param deps \code{logical} should scripts pass data objects to each other (default=TRUE) 
 #' @return logical TRUE if succesful, FALSE, if not.
 #' @importFrom desc desc
 #' @importFrom rmarkdown render
 #' @importFrom utils getSrcref modifyList
 #' @importFrom usethis proj_set proj_get
-DataPackageR <- function(arg = NULL) {
+DataPackageR <- function(arg = NULL, deps = TRUE) {
   requireNamespace("futile.logger")
   requireNamespace("yaml")
   pkg_dir <- arg
@@ -132,7 +133,7 @@ DataPackageR <- function(arg = NULL) {
       stop("exiting", call. = FALSE)
     }
   } else {
-    logpath <-
+    logpath <- 
       normalizePath(
         file.path(pkg_dir, "inst/extdata"),
         winslash = "/"
@@ -232,20 +233,9 @@ DataPackageR <- function(arg = NULL) {
     pkg_description <- try(read.description(file = description_file),
       silent = TRUE
     )
-    if (inherits(pkg_description, "try-error")) {
-      flog.fatal("No valid DESCRIPTION file")
-      {
-        stop(
-          paste0(
-            "You need a valid package DESCRIPTION file.",
-            "Please see Writing R Extensions",
-            "(http://cran.r-project.org/doc/manuals/",
-            "r-release/R-exts.html#The-DESCRIPTION-file).\n"
-          ),
-          pkg_description
-        )
-      }
-    }
+    # The test for a valid DESCRIPTION here is no longer needed since
+    # we use proj_set().
+
     # check that we have at least one file
     # This is caught elsewhere
 
@@ -264,8 +254,9 @@ DataPackageR <- function(arg = NULL) {
     for (i in seq_along(r_files)) {
       dataenv <- new.env(hash = TRUE, parent = .GlobalEnv)
       # assign ENVS into dataenv.
-      # provide functions in the package to read from it.
-      assign(x = "ENVS", value = ENVS, dataenv)
+      # provide functions in the package to read from it (if deps = TRUE)
+      if(deps)
+        assign(x = "ENVS", value = ENVS, dataenv)
       flog.info(paste0(
         "Processing ", i, " of ",
         length(r_files), ": ", r_files[i],

tests/testthat/test-skeleton.R

@@ -37,7 +37,8 @@ test_that("package can be built from different locations", {
         file.path(tmp, "subsetCars"))),
     "subsetCars_1.0.tar.gz")
 
-  old <- setwd(file.path(tmp, "subsetCars"))
+  old <- 
+    setwd(file.path(tmp, "subsetCars"))
   on.exit(setwd(old))
   expect_equal(basename(package_build(".")), "subsetCars_1.0.tar.gz")
   expect_error(package_build("subsetCars"))
@@ -474,6 +475,9 @@ test_that("package built in different edge cases", {
          force = TRUE,
          recursive = TRUE)
   package.skeleton("foo", path = tmp)
+  suppressWarnings(expect_error(
+    DataPackageR:::DataPackageR(
+      file.path(tmp, "foo"))))
   dir.create(file.path(tmp, "foo", "data-raw"))
   suppressWarnings(expect_error(
     DataPackageR:::DataPackageR(
@@ -492,8 +496,8 @@ test_that("package built in different edge cases", {
          recursive = TRUE)
 
 
-  package.skeleton("foo", path = tmp)
-  expect_error(yml_find(file.path(tmp, "foo")))
+  package.skeleton("foo", path = tempdir(),force=TRUE)
+  expect_error(yml_find(file.path(tempdir(), "foo")))
   dir.create(file.path(tmp, "foo", "data-raw"))
   unlink(file.path(tmp, "foo", "DESCRIPTION"))
   yml <- DataPackageR:::construct_yml_config("foo.Rmd")

vignettes/usingDataPackageR.R

@@ -5,58 +5,104 @@ knitr::opts_chunk$set(
   eval = TRUE
 )
 
-## ------------------------------------------------------------------------
-library(data.tree)
+## ----minimal_example, results='hide'-------------------------------------
 library(DataPackageR)
-tmp = normalizePath(tempdir())
-processing_code = system.file("extdata","tests","subsetCars.Rmd",package="DataPackageR")
-print(processing_code)
-setwd(tmp)
-DataPackageR::datapackage.skeleton("Test", 
-                                   force=TRUE, 
-                                   code_files = processing_code, 
-                                   r_object_names = "cars_over_20") # cars_over_20 is an R object 
-                                                                    # created in the Rmd file.
+
+# Let's reproducibly package up
+# the cars in the mtcars dataset
+# with speed > 20.
+# Our dataset will be called cars_over_20.
+
+# Get the code file that turns the raw data
+# to our packaged and processed analysis-ready dataset.
+processing_code <-
+  system.file("extdata", 
+              "tests",
+              "subsetCars.Rmd",
+              package = "DataPackageR")
+
+# Create the package framework.
+DataPackageR::datapackage_skeleton(
+  "mtcars20",
+  force = TRUE,
+  code_files = processing_code,
+  r_object_names = "cars_over_20",
+  path = tempdir()
+  ) 
 
 ## ----dirstructure,echo=FALSE---------------------------------------------
-df = data.frame(pathString=file.path("Test",(list.files(tmp,recursive=TRUE))))
+library(data.tree)
+df = data.frame(pathString = file.path(
+  "mtcars20",
+  list.files(
+  file.path(tempdir(), "mtcars20"),
+  include.dirs = TRUE,
+  recursive = TRUE
+  )
+  ))
 as.Node(df)
 
 ## ---- echo=FALSE---------------------------------------------------------
-library(yaml)
-setwd(tmp)
-cat(as.yaml(yaml.load_file("Test/datapackager.yml")))
+cat(yaml::as.yaml(yaml::yaml.load_file(file.path(tempdir(),"mtcars20","datapackager.yml"))))
 
-## ------------------------------------------------------------------------
-# Within the package directory
-setwd(tmp)
-DataPackageR:::package_build("Test") 
+## ----eval=TRUE-----------------------------------------------------------
+# Run the preprocessing code to build cars_over_20
+# and reproducibly enclose it in a package.
+DataPackageR:::package_build(file.path(tempdir(),"mtcars20"))
 
 ## ---- echo=FALSE---------------------------------------------------------
-library(yaml)
-setwd(tmp)
-df = data.frame(pathString=file.path("Test",(list.files("Test",recursive=TRUE))))
-as.Node(df)
+df = data.frame(pathString = file.path(
+  "mtcars20",
+  list.files(
+  file.path(tempdir(), "mtcars20"),
+  include.dirs = TRUE,
+  recursive = TRUE
+  )
+  ))
+  as.Node(df)
 
-## ---- echo=FALSE---------------------------------------------------------
-setwd(tmp)
-cat(readLines("Test/DATADIGEST"),sep="\n")
+## ------------------------------------------------------------------------
+# Let's use the package we just created.
+install.packages(file.path(tempdir(),"mtcars20_1.0.tar.gz"), type = "source", repos = NULL)
+library(mtcars20)
+data("cars_over_20") # load the data
+cars_over_20  # Now we can use it.
+?cars_over_20 # See the documentation you wrote in data-raw/documentation.R.
 
-## ----echo=FALSE----------------------------------------------------------
-setwd(tmp)
-cat(readLines("Test/DESCRIPTION"),sep="\n")
+vignettes = vignette(package="mtcars20")
+vignettes$results
 
-## ----construct_config----------------------------------------------------
-#assume I have file1.Rmd and file2.R located in /data-raw, and these create 'object1' and 'object2' respectively.
+## ------------------------------------------------------------------------
+# We can easily check the version of the data
+DataPackageR::data_version("mtcars20")
 
-config = construct_yml_config(code = c("file1.Rmd","file2.R"), data = c("object1","object2"))
-cat(as.yaml(config))
+# You can use an assert to check the data version in  reports and
+# analyses that use the packaged data.
+assert_data_version(data_package_name = "mtcars20",
+                    version_string = "0.1.0",
+                    acceptable = "equal")  #If this fails, execution stops
+                                           #and provides an informative error.
 
-## ------------------------------------------------------------------------
-path_to_package = tempdir() #pretend this is the root of our package
-yml_write(config,path = path_to_package)
+## ----construct_config, echo=1:2------------------------------------------
+# assume I have file1.Rmd and file2.R located in /data-raw, 
+# and these create 'object1' and 'object2' respectively.
+
+config = construct_yml_config(code = c("file1.Rmd", "file2.R"),
+                              data = c("object1", "object2"))
+cat(yaml::as.yaml(config))
 
 ## ------------------------------------------------------------------------
+path_to_package = tempdir() #e.g., if tempdir() was the root of our package.
+yml_write(config, path = path_to_package)
+
+## ----echo=1:2------------------------------------------------------------
 config = yml_disable_compile(config,filenames = "file2.R")
-cat(as.yaml(config))
+yml_write(config, path = path_to_package) # write modified yml to the package.
+cat(yaml::as.yaml(config))
+
+## ---- echo=FALSE---------------------------------------------------------
+cat(readLines(file.path(tempdir(),"mtcars20","DATADIGEST")),sep="\n")
+
+## ----echo=FALSE----------------------------------------------------------
+cat(readLines(file.path(tempdir(),"mtcars20","DESCRIPTION")),sep="\n")
 

vignettes/usingDataPackageR.Rmd

@@ -290,6 +290,11 @@ A script (e.g., `script2.Rmd`) running after `script1.Rmd` can access a stored d
 
 `DataPackageR::datapackager_object_read("script1_dataset")`. 
 
+Passing of data objects amongst scripts can be turned off via:
+
+`package_build(deps = FALSE)`
+
+
 # Additional Details
 
 We provide some additional details for the interested.