resubmission

Author: idmn

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: cloudfs
 Title: Streamlined Interface to Interact with Cloud Storage Platforms
-Version: 0.1.1
+Version: 0.1.2
 Authors@R: c(
     person("Iaroslav", "Domin", email = "[email protected]", role = c("aut", "cre")),
     person("Stefan", "Musch", email = "[email protected]", role = c("aut")),

NEWS.md

@@ -1,3 +1,3 @@
-# cloudfs 0.1.1
+# cloudfs 0.1.2
 
 * Initial version.

R/drive_transfer.R

@@ -59,9 +59,12 @@ cloud_drive_upload <- function(file, root = NULL) {
 #' @return Invisibly returns `NULL` after successfully downloading the file.
 #' 
 #' @examplesIf interactive() 
-#' # downloads data/demo.csv from project's Google Drive folder 
-#' # (provided it exists) and saves it to local 'data' folder
-#' cloud_drive_download("data/demo.csv")
+#' # downloads toy_data/demo.csv from project's Google Drive folder 
+#' # (provided it exists) and saves it to local 'toy_data' folder
+#' cloud_drive_download("toy_data/demo.csv")
+#' 
+#' # clean up
+#' unlink("toy_data", recursive = TRUE)
 #' 
 #' @export
 cloud_drive_download <- function(file, root = NULL) {

R/drive_transfer_bulk.R

@@ -167,12 +167,15 @@ cloud_drive_upload_bulk <- function(content, quiet = FALSE, root = NULL) {
 #' @return Invisibly returns the input `content` dataframe.
 #' 
 #' @examplesIf interactive() 
-#' # provided there's a folder called "data" in the root of your project's
+#' # provided there's a folder called "toy_data" in the root of your project's
 #' # Google Drive folder, and this folder contains "csv" files
-#' cloud_drive_ls("data") |> 
+#' cloud_drive_ls("toy_data") |> 
 #'   filter(type == "csv") |> 
 #'   cloud_drive_download_bulk()
 #'   
+#' # clean up
+#' unlink("toy_data", recursive = TRUE)
+#'   
 #' @export
 cloud_drive_download_bulk <- function(content, quiet = FALSE) {
   cont <- cloud_drive_prep_bulk(content, what = "download", quiet = quiet)

R/s3_transfer.R

@@ -57,9 +57,12 @@ cloud_s3_upload <- function(file, root = NULL) {
 #' @return Invisibly returns `NULL` after successfully downloading the file.
 #' 
 #' @examplesIf interactive() 
-#' # downloads data/demo.csv from project's S3 folder (provided it exists)
-#' # and saves it to local 'data' folder
-#' cloud_s3_download("data/demo.csv")
+#' # downloads toy_data/demo.csv from project's S3 folder (provided it exists)
+#' # and saves it to local 'toy_data' folder
+#' cloud_s3_download("toy_data/demo.csv")
+#' 
+#' # clean up
+#' unlink("toy_data", recursive = TRUE)
 #' 
 #' @export
 cloud_s3_download <- function(file, root = NULL) {

R/s3_transfer_bulk.R

@@ -131,12 +131,15 @@ cloud_s3_upload_bulk <- function(content, quiet = FALSE, root = NULL) {
 #' @return Invisibly returns the input `content` dataframe.
 #' 
 #' @examplesIf interactive()
-#' # provided there's a folder called "data" in the root of your project's
+#' # provided there's a folder called "toy_data" in the root of your project's
 #' # S3 folder, and this folder contains "csv" files
-#' cloud_s3_ls("data") |> 
+#' cloud_s3_ls("toy_data") |> 
 #'   filter(type == "csv") |> 
 #'   cloud_s3_download_bulk()
 #'   
+#' # clean up
+#' unlink("toy_data", recursive = TRUE)
+#'   
 #' @export
 cloud_s3_download_bulk <- function(content, quiet = FALSE, root = NULL) {
   check_string(root, alt_null = TRUE)

cran-comments.md

@@ -1,52 +1,96 @@
 ## Resubmission
-This is a resubmission. In this version I have:
-
-* **Updated the `Description` field in the `DESCRIPTION` file** to:
-  - Place software names ('Google Drive' and 'Amazon S3') in single quotes as 
-  per CRAN guidelines.
-  - Add links to the respective web services for 'Google Drive' and 'Amazon S3'
-  for clarity and compliance.
-
-* **Removed examples from unexported functions.**
-
-In response to the feedback, I have made the following changes to minimize the
-use of the `\dontrun` tag:
-
-* **Examples Updated to Execute Safely**:
-  - In `cloud_read_excel.Rd` and `cloud_get_roots.Rd`, I introduced examples 
-  that can be safely executed and removed the `\dontrun` tags.
-
-* **Limited Use of `\dontrun`**:
-  - For `cloud_local_ls.Rd`, I retained the `\dontrun` tag for only one example.
-
-* **Conditional Execution with `\dontshow`**:
-  The majority of the functions in the package require the user to associate the
-  project directory with a cloud storage, a process that involves inserting a
-  URL of a 'Google Drive' or an 'Amazon S3' folder into the console. For these
-  functions, I wrapped the examples in `\dontshow` conditional on
-  `interactive()`. The affected files include:
-  - `cloud_drive_attach.Rd`
-  - `cloud_drive_browse.Rd`
-  - `cloud_drive_ls.Rd`
-  - `cloud_drive_upload.Rd`
-  - `cloud_drive_download.Rd`
-  - `cloud_drive_write.Rd`
-  - `cloud_drive_read.Rd`
-  - `cloud_drive_upload_bulk.Rd`
-  - `cloud_drive_download_bulk.Rd`
-  - `cloud_drive_write_bulk.Rd`
-  - `cloud_drive_read_bulk.Rd`
-  - `cloud_drive_spreadsheet_autofit.Rd`
-  - `cloud_s3_attach.Rd`
-  - `cloud_s3_browse.Rd`
-  - `cloud_s3_upload.Rd`
-  - `cloud_s3_download.Rd`
-  - `cloud_s3_read.Rd`
-  - `cloud_s3_write.Rd`
-  - `cloud_s3_upload_bulk.Rd`
-  - `cloud_s3_download_bulk.Rd`
-  - `cloud_s3_write_bulk.Rd`
-  - `cloud_s3_read_bulk.Rd`  
+This is a resubmission. Below are my responses to the feedback received in the
+previous review.
+
+### User options
+
+```
+Please always make sure to reset to user's options(), working directory or par() after you changed it in examples and vignettes and demos. --> inst/doc/cloudfs.R
+e.g.:
+old <- options(digits = 3)
+...
+options(old)
+```
+
+I've addressed this by removing the `options(width = 150)` command from the
+beginning of the cloudfs.Rmd vignette.
+
+### User Space Integrity
+
+```
+Please ensure that your functions do not write by default or in your
+examples/vignettes/tests in the user's home filespace (including the package
+directory and getwd()). This is not allowed by CRAN policies.
+Please omit any default path in writing functions. In your
+examples/vignettes/tests you can write to tempdir().
+e.g.: man/cloud_drive_spreadsheet_autofit.Rd ; man/cloud_drive_upload.Rd ; ...
+```
+
+I understand the importance of adhering to CRAN policies and ensuring that there
+are no unintended consequences for the users of the package. However, I have
+chosen not to make the suggested modifications, and I'd like to explain the
+rationale behind this decision and why in my opinion the package does not
+violate the policies.
+
+One of the key features of the package is that it enables the use of concise
+relative paths for both the current working directory and associated cloud
+project folders. For instance, consider the task of uploading a local file,
+"models/glm.rds", to a project's S3 folder. Using `aws.s3`, the code would be:
+
+```R
+aws.s3::put_object(
+  file = "models/glm.rds",
+  bucket = "project-data",
+  object = "project-1/models/glm.rds"
+)
+```
+
+With `cloudfs`, it can be achieved with a significantly simpler syntax:
+
+```R
+cloud_s3_upload("models/glm.rds")
+```
+
+Applying `cloud_s3_upload()` to a file located in a temporary folder goes
+against its design intent. Its main objective is to upload files while mirroring
+the folder structure between the current directory and the project's S3.
+Demonstrating this with a temp folder file would misrepresent the function's
+typical application.
+
+That being said I've taken comprehensive measures to ensure no accidental or
+default file writing occurs in the current working directory:
+
+- **Initial Setup**: Most functions, including all `*read*`, `*write*`,
+`*upload*`, `*download*` require users to link their project directory with
+cloud storage during the package's inaugural use. This entails obtaining
+explicit user consent.
+
+- **dontshow**: Consequently, examples where this linkage would activate are 
+wrapped in `\dontshow` conditional on `interactive()`.
+
+- **Read Functions**: These initially pull files from the cloud to a temp folder
+for reading, leaving the working directory untouched. In examples, the working
+directory is also untouched.
+
+- **Write Functions**: Files are first created in a temp folder, then sent to 
+the cloud. The working directory remains untouched.
+
+- **Download Functions**: These do pull files into the working directory, but
+this is their primary purpose and they cannot write anywhere outside of it.
+Also, in examples (shielded with `\dontshow`), I've added code to remove the
+donloaded files.
+
+- **Upload Functions**: In examples, files are generated files in the working
+directory for uploading purposes. Still, cleanup code ensures their removal
+afterward.
+
+- **Vignettes**: Chunks using `cloudfs` functions aren't executed; they're all 
+tagged with `eval=FALSE`.
+
+- **Tests**: These only operate when Google Drive or S3 tokens are available,
+excluding execution on CRAN. When testing, I use temporary folders for project
+creation and employ `withr::with_dir` to execute `cloudfs` code — a strategy
+suitable for testing but not for example clarity.
 
 ## R CMD check results