add check_folder()

NAMESPACE

@@ -10,6 +10,7 @@ export(check_description)
 export(check_documentation)
 export(check_environment)
 export(check_filename)
+export(check_folder)
 export(check_license)
 export(check_lintr)
 export(check_package)

NEWS.md

@@ -28,6 +28,7 @@
   `Config/checklist/communities` and `Config/checklist/keywords`.
 * Store author information to reuse when running `create_package()` or
   `create_project()`.
+* Add `check_folder()`.
 
 # checklist 0.2.6
 

R/check_folder.R

@@ -0,0 +1,246 @@
+#' Check the folder structure
+#'
+#' For the time being, this function only checks projects.
+#' Keep in mind that R packages have requirements for the folder structure.
+#' `check_cran()` checks those requirements.
+#'
+#' # Recommended folder structure
+#'
+#' - `source`: contains all `R` scripts and `Rmd` files.
+#' - `data`: contains all data files.
+#'
+#' # `source`
+#'
+#' A simple project with only `R` scripts or only `Rmd` files can place all the
+#' files directly in the `source` folder.
+#'
+#' More elaborate projects should place in the files in several folders under
+#' `source`.
+#' Place every `bookdown` document in a dedicated folder.
+#' And create an RStudio project for that folder.
+#'
+#' # `data`
+#'
+#' Simple projects in which `source` has no sub folders, place `data` at the
+#' root of the project.
+#' For more elaborate project you must choose between either `data` at the root
+#' of the project or `data` as sub folder of the sub folders of `source`.
+#' E.g. `source/report/data`.
+#'
+#' Place the data in an open file format.
+#' E.g. `csv`, `txt` or `tsv` for tabular data.
+#' We strongly recommend to use `git2rdata::write_vc()` to store such data.
+#' Use the [`geopackage`](https://www.geopackage.org/) format for spatial data.
+#' Optionally add description of the data as markdown files.
+#'
+#' @inheritParams read_checklist
+#' @family project
+#' @export
+#' @importFrom fs dir_ls path path_rel
+check_folder <- function(x = ".") {
+  x <- read_checklist(x = x)
+  if (x$package) {
+    return(x)
+  }
+
+  dir_ls(x$get_path, type = "directory") |>
+    path_rel(x$get_path) -> root_dir
+  suppressWarnings(
+    path(x$get_path, "source") |>
+      dir_ls(type = "directory", fail = FALSE) -> source_1
+  )
+
+  paste(open_data_ext, collapse = "|") |>
+    sprintf(fmt = "\\.(%s)$") -> data_regexp
+  dir_ls(x$get_path, type = "file", recurse = TRUE, regexp = data_regexp) |>
+    path_rel(x$get_path) -> data_files
+  suppressWarnings(
+    path(x$get_path, "output") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = data_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_data_files
+  )
+  data_files <- data_files[!data_files %in% ignore_data_files]
+  suppressWarnings(
+    path(x$get_path, "data") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = data_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_data_files
+  )
+  data_files <- data_files[!data_files %in% ignore_data_files]
+  suppressWarnings(
+    path(x$get_path, "renv") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = data_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_data_files
+  )
+  data_files <- data_files[!data_files %in% ignore_data_files]
+  suppressWarnings(
+    path(source_1, "data") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = data_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_data_files
+  )
+  data_files <- data_files[!data_files %in% ignore_data_files]
+
+  paste(graphics_ext, collapse = "|") |>
+    sprintf(fmt = "\\.(%s)$") -> graphics_regexp
+  dir_ls(x$get_path, type = "file", recurse = TRUE, regexp = graphics_regexp) |>
+    path_rel(x$get_path) -> graphics_files
+  paste0("_files.*", graphics_regexp) |>
+    dir_ls(x$get_path, type = "file", recurse = TRUE, regexp = _) |>
+    path_rel(x$get_path) -> ignore_graphics_files
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+  suppressWarnings(
+    path(x$get_path, "output") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = graphics_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_graphics_files
+  )
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+  suppressWarnings(
+    path(x$get_path, "media") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = graphics_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_graphics_files
+  )
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+  suppressWarnings(
+    path(source_1, "media") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = graphics_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_graphics_files
+  )
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+  suppressWarnings(
+    path(x$get_path, "renv") |>
+      dir_ls(
+        type = "file", recurse = TRUE, regexp = graphics_regexp, fail = FALSE
+      ) |>
+      path_rel(x$get_path) -> ignore_graphics_files
+  )
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+
+  dir_ls(x$get_path, recurse = TRUE, regexp = "_(bookdown|quarto)\\.yml$") |>
+    dirname() |>
+    c(source_1) |>
+    vapply(check_data_media, vector(mode = "list", length = 1)) -> data_media_ok
+  vapply(data_media_ok, "[[", vector(mode = "list", length = 1), "data") |>
+    unlist() |>
+    path_rel(x$get_path) -> ignore_data_files
+  data_files <- data_files[!data_files %in% ignore_data_files]
+  vapply(data_media_ok, "[[", vector(mode = "list", length = 1), "cover") |>
+    unlist() |>
+    path_rel(x$get_path) -> ignore_cover_files
+  data_files <- data_files[!data_files %in% ignore_cover_files]
+  vapply(
+    data_media_ok, "[[", vector(mode = "list", length = 1), "extra_media"
+  ) |>
+    unlist() |>
+    path_rel(x$get_path) -> ignore_graphics_files
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+  vapply(data_media_ok, "[[", vector(mode = "list", length = 1), "media") |>
+    unlist() |>
+    path_rel(x$get_path) -> ignore_graphics_files
+  graphics_files <- graphics_files[!graphics_files %in% ignore_graphics_files]
+
+  dir_ls(x$get_path, recurse = TRUE, regexp = "_bookdown\\.yml$") |>
+    dirname() |>
+    vapply(check_bookdown, vector(mode = "list", length = 1)) |>
+    unlist() |>
+    unname() |>
+    c(
+      paste(
+        "A project should only have `data`, `inst`, `output`, `renv` and ",
+        "`source` as main folder."
+      )[
+        !all(root_dir %in% c("data", "inst", "output", "renv", "source"))
+      ],
+      sprintf(
+        "Data files found outside of a `data` folder:\n  %s",
+        paste(data_files, collapse = "\n  ")
+      )[length(data_files) > 1],
+      sprintf(
+        "Media files found outside of a `media` folder:\n  %s",
+        paste(graphics_files, collapse = "\n  ")
+      )[length(graphics_files) > 1]
+    ) -> warn
+
+  c(
+    "No `source` main folder found"[!"source" %in% root_dir],
+    "`src` main folder is not allowed"["src" %in% root_dir],
+    "`source` cannot have `data` as a subfolder"[
+      "data" %in% basename(source_1)
+    ],
+    paste(
+      "Use either a common `data` folder at the root or a `data` folder within",
+      "the subfolders of `source`."
+    )["data" %in% root_dir && length(ignore_data_files) > 0],
+    paste(
+      "Use either a common `media` folder at the root or a `media` folder",
+      "within the subfolders of `source`."
+    )["media" %in% root_dir && length(ignore_graphics_files) > 0]
+  ) -> problems
+
+  x$add_error(problems, item = "folder conventions")
+  x$add_warnings(warn, item = "folder conventions")
+  return(x)
+}
+
+#' @importFrom fs dir_ls
+check_bookdown <- function(path) {
+  rstudio <- dir_ls(path, regexp = "\\.Rproj$")
+  c(
+    paste("No Rstudio project found for bookdown", path)[length(rstudio) == 0],
+    paste(
+      "Multiple Rstudio projects found for bookdown", path
+    )[length(rstudio) > 1]
+  ) -> warn
+  list(warn)
+}
+
+#' @importFrom fs dir_ls is_dir path
+check_data_media <- function(path) {
+  cover_ok <- dir_ls(path, type = "file")
+  cover_ok <- cover_ok[basename(cover_ok) == "cover.txt"]
+
+  suppressWarnings(
+    paste(open_data_ext, collapse = "|") |>
+      sprintf(fmt = "\\.(%s)$") |>
+      dir_ls(
+        path = path(path, "data"), type = "file", recurse = TRUE, fail = FALSE,
+        regexp = _
+      ) -> data_ok
+  )
+
+  suppressWarnings(
+    paste(graphics_ext, collapse = "|") |>
+      sprintf(fmt = "\\.(%s)$") |>
+      dir_ls(
+        path = path(path, "media"), type = "file", recurse = TRUE, fail = FALSE,
+        regexp = _
+      ) -> media_ok
+  )
+
+  suppressWarnings(
+    paste(graphics_ext, collapse = "|") |>
+      sprintf(fmt = "\\.(%s)$") |>
+      dir_ls(
+        path = path(path, c("_book", "_extensions", "_freeze", "libs")),
+        type = "file", recurse = TRUE, regexp = _, fail = FALSE
+      ) -> extra_media_ok
+  )
+
+  list(
+    cover = list(cover_ok), data = list(data_ok),
+    extra_media = list(extra_media_ok), media = list(media_ok)
+  ) |>
+    list()
+}

R/check_project.R

@@ -6,7 +6,7 @@
 #' @export
 #' @importFrom assertthat assert_that
 #' @importFrom fs is_file
-#' @family source
+#' @family project
 check_project <- function(x = ".", fail = !interactive(), quiet = FALSE) {
   assert_that(
     inherits(x, "checklist") || is_file(path(x, "checklist.yml")),

R/check_source.R

@@ -23,7 +23,7 @@
 #' @importFrom fs dir_exists file_exists path
 #' @importFrom utils file_test
 #' @export
-#' @family source
+#' @family project
 check_source <- function(x = ".", fail = !interactive()) {
   # nocov start
   .Deprecated("check_project", package = "checklist")

R/checklist_class.R

@@ -285,7 +285,7 @@ Please contact the maintainer of the `checklist` package."
     available_checks = c(
       "checklist", "CITATION", "DESCRIPTION", "documentation",
       "R CMD check", "codemeta", "license", "repository secret",
-      "filename conventions", "lintr", "spelling"
+      "filename conventions", "folder conventions", "lintr", "spelling"
     ),
     # stores a named logical vector of checked items.
     # names must match the available checks

_pkgdown.yml

@@ -31,10 +31,10 @@ reference:
   contents:
   - check_package
   - has_concept("package")
-- title: Functions relevant for checking scripts
+- title: Functions relevant for checking projects with R and Rmd scripts
   contents:
-  - has_concept("source")
-- title: Functions relevant for checking packages and scripts
+  - has_concept("project")
+- title: Functions relevant for checking packages and projects
   contents:
   - has_concept("both")
 - title: Setting up a project use the checklist package

data-raw/iso_639_3.R

@@ -42,4 +42,7 @@ sprintf("`%s`", graphics_ext) |>
 #' These are fonts files which often require their own file name scheme.
 ") |>
   writeLines("man-roxygen/graphics.R")
-save(iso_639_3, email_regexp, graphics_ext, file = "R/sysdata.rda")
+open_data_ext <- c("csv", "gpkg", "tsv", "txt")
+save(
+  email_regexp, graphics_ext, iso_639_3, open_data_ext, file = "R/sysdata.rda"
+)

docker/.gitignore

@@ -0,0 +1 @@
+build_push.sh