feat: ✨ helper get_*() for project IDs and directories (#251)

# Description

Closes #243, closes #104

Needs a quick review.

## Checklist

- [x] Ran `just run-all`

---------

Co-authored-by: Signe Kirk Brødbæk <40836345+signekb@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: 3 people

R/get.R

@@ -0,0 +1,84 @@
+#' Get the project ID from the current working directory path
+#'
+#' Gets a numeric project ID from the current working directory path by looking
+#' for a folder name with only digits. Errors if a project ID with an unexpected
+#' length was found.
+#'
+#' @returns A 6-digit character string, or `NA` if no project ID is found in the
+#'   path.
+#' @noRd
+get_project_id <- function() {
+  id <- fs::path_wd() |>
+    stringr::str_extract("/[0-9]+/") |>
+    stringr::str_remove_all("/")
+
+  if (is.na(id) || id == "") {
+    cli::cli_warn(
+      c(
+        "No project ID could be found in the path of the current working directory, so outputting `NA`.",
+        "i" = "Your path is {fs::path_wd()}. Maybe change to a working directory within a project?"
+      )
+    )
+  }
+
+  if (stringr::str_length(id) != 6 && !is.na(id)) {
+    cli::cli_abort(
+      "Found an ID, but it was too long or too short to be a project ID.",
+      c(
+        "i" = "The ID found was {id}. Project IDs are expected to be 6 digits long."
+      )
+    )
+  }
+  id
+}
+
+#' Get the path to the rawdata or workdata directory for the current project
+#'
+#' Looks in the [options()] for `fastreg.project_rawdata_dir` and
+#' `fastreg.project_workdata_dir` first, and if not found, constructs a path
+#' based on the project ID using `get_project_id()`. The constructed path is
+#' `E:/<project_id>/rawdata/` for raw data and `E:/<project_id>/workdata/` for n
+#' work data.
+#'
+#' @returns A path object.
+#' @noRd
+get_project_rawdata_dir <- function() {
+  rawdata_path <- getOption("fastreg.project_rawdata_dir")
+  if (!is.null(rawdata_path)) {
+    return(fs::path(rawdata_path))
+  }
+
+  id <- get_project_id()
+  if (is.na(id) || id == "") {
+    cli::cli_abort(
+      c(
+        "Can't set the {.path rawdata/} path without a project ID.",
+        "i" = "Use {.code options(fastreg.project_rawdata_dir = '<path>')} or change into a directory within a project."
+      )
+    )
+  }
+
+  glue::glue("E:/{id}/rawdata/") |>
+    fs::path()
+}
+
+#' @describeIn get_project_rawdata_dir Gets the project workdata directory.
+#' @noRd
+get_project_workdata_dir <- function() {
+  workdata_path <- getOption("fastreg.project_workdata_dir")
+  if (!is.null(workdata_path)) {
+    return(fs::path(workdata_path))
+  }
+
+  id <- get_project_id()
+  if (is.na(id) || id == "") {
+    cli::cli_abort(
+      c(
+        "Can't set the {.path workdata/} path without a project ID.",
+        "i" = "Use {.code options(fastreg.project_workdata_dir = '<path>')} or change into a working directory within a project."
+      )
+    )
+  }
+  glue::glue("E:/{id}/workdata/") |>
+    fs::path()
+}

tests/testthat/test-get.R

@@ -0,0 +1,77 @@
+test_that("`get_project_id()` extracts correct project ID", {
+  temp_dir <- fs::path_temp("701010/test/project/")
+  fs::dir_create(temp_dir, recurse = TRUE)
+  project_id <- withr::with_dir(
+    temp_dir,
+    {
+      get_project_id()
+    }
+  )
+  expect_identical(project_id, "701010")
+})
+
+test_that("`get_project_id()` errors for IDs not of length 6", {
+  temp_dir <- fs::path_temp("70101/test/project/")
+  fs::dir_create(temp_dir, recurse = TRUE)
+  project_id <- withr::with_dir(
+    temp_dir,
+    {
+      expect_error(get_project_id(), regexp = "project ID")
+    }
+  )
+
+  temp_dir <- fs::path_temp("7010101/test/project/")
+  fs::dir_create(temp_dir, recurse = TRUE)
+  project_id <- withr::with_dir(
+    temp_dir,
+    {
+      expect_error(get_project_id(), regexp = "project ID")
+    }
+  )
+})
+
+test_that("`get_project_id()` warns for not finding a project ID", {
+  temp_dir <- fs::path_temp("non-number/test/project/")
+  fs::dir_create(temp_dir, recurse = TRUE)
+  withr::with_dir(
+    temp_dir,
+    {
+      expect_warning(get_project_id(), regexp = "`NA`")
+      expect_identical(
+        suppressWarnings(get_project_id()),
+        NA_character_
+      )
+    }
+  )
+})
+
+test_that("should fail if no `E:` drive exists", {
+  temp_path <- fs::path_temp("701010/")
+  fs::dir_create(temp_path)
+  withr::with_dir(
+    temp_path,
+    {
+      expect_error(suppressWarnings(get_project_rawdata_dir()))
+      expect_error(suppressWarnings(get_project_workdata_dir()))
+    }
+  )
+})
+
+test_that("takes project directory from options", {
+  withr::with_options(
+    list(
+      fastreg.project_rawdata_dir = fs::path("E:/rawdata/701010/"),
+      fastreg.project_workdata_dir = fs::path("E:/workdata/701010/")
+    ),
+    {
+      expect_identical(
+        get_project_rawdata_dir(),
+        fs::path("E:/rawdata/701010/")
+      )
+      expect_identical(
+        get_project_workdata_dir(),
+        fs::path("E:/workdata/701010/")
+      )
+    }
+  )
+})

tests/testthat/test-use-targets.R

@@ -47,21 +47,21 @@ test_that("targets pipeline template converts SAS files to Parquet", {
 
   # Create temp directory structure.
   test_dir <- fs::path_temp("pipeline-test")
-  input_dir <- fs::path(test_dir, "input")
-  output_dir <- fs::path(test_dir, "output")
-  fs::dir_create(input_dir)
-  fs::dir_create(output_dir)
+  test_input_dir <- fs::path(test_dir, "input")
+  test_output_dir <- fs::path(test_dir, "output")
+  fs::dir_create(test_input_dir)
+  fs::dir_create(test_output_dir)
 
   # Create SAS files.
   bef_list <- simulate_register("bef", c("1999", "2020"))
   lmdb_list <- simulate_register("lmdb", c("2020", "2021"))
-  save_as_sas(bef_list, input_dir)
-  save_as_sas(lmdb_list, input_dir)
+  save_as_sas(bef_list, test_input_dir)
+  save_as_sas(lmdb_list, test_input_dir)
 
   # Read template and replace placeholder paths.
   modified_content <- template_content |>
-    stringr::str_replace("/path/to/sas/directory", input_dir) |>
-    stringr::str_replace("/path/to/output/directory", output_dir)
+    stringr::str_replace("/path/to/sas/directory", test_input_dir) |>
+    stringr::str_replace("/path/to/output/directory", test_output_dir)
 
   # Write and run pipeline.
   withr::with_dir(test_dir, {
@@ -70,7 +70,11 @@ test_that("targets pipeline template converts SAS files to Parquet", {
   })
 
   # Check number of created Parquet files.
-  parquet_files <- fs::dir_ls(output_dir, recurse = TRUE, glob = "*.parquet")
+  parquet_files <- fs::dir_ls(
+    test_output_dir,
+    recurse = TRUE,
+    glob = "*.parquet"
+  )
   expect_equal(
     length(parquet_files),
     sum(length(bef_list), length(lmdb_list))
@@ -81,13 +85,13 @@ test_that("targets pipeline template converts SAS files to Parquet", {
   n_expected_lmdb <- sum(purrr::map_int(lmdb_list, nrow))
 
   n_actual_bef <- arrow::open_dataset(fs::path(
-    output_dir,
+    test_output_dir,
     "bef"
   )) |>
     dplyr::collect() |>
     nrow()
   n_actual_lmdb <- arrow::open_dataset(fs::path(
-    output_dir,
+    test_output_dir,
     "lmdb"
   )) |>
     dplyr::collect() |>

vignettes/design.qmd

@@ -38,6 +38,8 @@ The actions are:
 - `list`: List files in a directory, e.g., SAS or Parquet files.
 - `read`: Read a Parquet register into R as a DuckDB table.
 - `use`: Use a template in the current project.
+- `get`: Get or guess some information, e.g., the project ID, workdata
+  directory, or rawdata directory from the current working directory.
 
 While the objects are:
 
@@ -114,11 +116,10 @@ across chunks of the same partition (e.g. `part-*.parquet`).
 %%| fig-cap: "Expected workflow for reading a Parquet register as a DuckDB table using `read_register()`."
 %%| fig-alt: "A flowchart showing the expected flow of reading a Parquet register created with the fastreg package."
 flowchart TD
-    path[/"path<br>[Character scalar]"/]
+    name[/"name<br>[Character scalar]"/]
     read_register("read_register()")
     output[/"Output<br>[DuckDB table]"/]
 
     %% Edges
-    path --> read_register --> output
-
+    name --> read_register --> output
 ```

vignettes/fastreg.qmd

@@ -74,6 +74,49 @@ save_as_sas(
 show_tree(sas_dir)
 ```
 
+## Settings to correct paths
+
+Many of fastreg's functions depend on the locations of the original SAS
+files and the eventual Parquet files including the conversion, writing,
+and reading functions. Through `options()` you can set these paths in
+two settings: `fastreg.project_rawdata_dir` and
+`fastreg.project_workdata_dir`. You can set these `options()` at the top
+of your R script or Quarto document, in your R Project's `.Rprofile`, or
+in your user-level `.Rprofile`. To add to the file, at the top of an R
+script, write (using a temporary directory here for these examples):
+
+```{r options}
+options(
+  # With a fake project ID.
+  fastreg.project_rawdata_dir = fs::path_temp("rawdata/701010/"),
+  fastreg.project_workdata_dir = fs::path_temp("workdata/701010/")
+)
+```
+
+If you want to set those exact same options in the R Project's
+`.Rprofile`, run the following line in your Console to open up the
+`.Rprofile` file for the project:
+
+```{r options-project-profile}
+#| filename: "Console"
+#| eval: false
+usethis::edit_r_profile("project")
+```
+
+You can then add the same `options()` as shown in the R script example
+above to that file and save it. The next time you open the project,
+those options will be set.
+
+If you want to set these options for all of your R projects and
+sessions, you can add them globally in your user-level `.Rprofile`. To
+open the `.Rprofile`, run:
+
+```{r options-user-profile}
+#| filename: "Console"
+#| eval: false
+usethis::edit_r_profile("user")
+```
+
 ## Converting a single file
 
 Converting one file from SAS to Parquet in fastreg isn't a simple change