Version 0.8.0 (#31)

* improve test

* Increment minor version

* Update variable lists

* Increment version number

* shorten line length per `{lintr}`

* Remove unnecessary brackets.

* shorten line lengths per `{lintr}`

* Now need to get more rows

* Adding variable 'packs' (#28)

* Add a vector with demographic variables.

* Add a vector with LTC variables

* Add a vector with bedday variables

* Add a vector with cost variables.

* Add a vignette for variable packs

* Increment version number

.gitignore

@@ -3,3 +3,4 @@
 .RData
 .Ruserdata
 docs
+inst/doc

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: slfhelper
 Title: Useful functions for working with the Source Linkage Files
-Version: 0.7.1
+Version: 0.8.0
 Authors@R: 
     person("James", "McMahon", , "james.mcmahon@phs.scot", role = c("cre", "aut"),
            comment = c(ORCID = "0000-0002-5380-2029"))
@@ -30,9 +30,13 @@ Imports:
     tibble
 Suggests:
     covr,
+    knitr,
     phsmethods,
+    rmarkdown,
     spelling,
     testthat (>= 3.0.0)
+VignetteBuilder: 
+    knitr
 Remotes: 
     Public-Health-Scotland/phsmethods
 Config/testthat/edition: 3

NEWS.md

@@ -1,3 +1,8 @@
+# slfhelper 0.8.0
+
+* Add a number of 'variable packs' for easily selecting sets of variables: `demog_vars`, `ltc_vars`,  `ep_file_bedday_vars` and `ep_file_cost_vars`.
+* Add a new vignette introducing variable packs.
+
 # slfhelper 0.7.1
 
 * Add check for server and hscdiip access on package load.

R/data.R

@@ -18,6 +18,46 @@
 #' @keywords data
 "indiv_file_vars"
 
+#' Demographic variables
+#'
+#' A vector containing the names of the demographic variables.
+#'
+#' @name demog_vars
+#' @docType data
+#' @keywords data
+"demog_vars"
+
+#' LTC variables
+#'
+#' A vector containing the names of the
+#' Long Term Condition (LTC) variables.
+#'
+#' @name ltc_vars
+#' @docType data
+#' @keywords data
+"ltc_vars"
+
+#' Episode file bedday variables
+#'
+#' A vector containing the names of the
+#' bedday related variables in the
+#' episode file.
+#'
+#' @name ep_file_bedday_vars
+#' @docType data
+#' @keywords data
+"ep_file_bedday_vars"
+
+#' Episode file cost variables
+#'
+#' A vector containing the names of the
+#' cost related variables in the
+#' episode file.
+#'
+#' @name ep_file_cost_vars
+#' @docType data
+#' @keywords data
+"ep_file_cost_vars"
 
 #' HSCP name / code lookup
 #'
@@ -51,8 +91,9 @@
 #' @keywords data
 "recids"
 
-#' A [tibble][tibble::tibble-package] containing an example cohort of CHI numbers under different
-#' variable names. It is used for testing and to illustrate examples.
+#' A [tibble][tibble::tibble-package] containing an example
+#' cohort of CHI numbers under different variable names.
+#' It is used for testing and to illustrate examples.
 #' It is likely that many of the 'chi numbers' are not valid.
 #'
 #' @name chi_cohort

R/read_slf.R

@@ -27,13 +27,13 @@ read_slf <-
     # but the column wasn't selected we need to add it (and remove later)
     remove_partnership_var <- FALSE
     remove_recid_var <- FALSE
-    if (!(is.null(optional_params$columns))) {
-      if (!(is.null(partnerships)) &
+    if (!is.null(optional_params$columns)) {
+      if (!is.null(partnerships) &
         !("hscp2018" %in% optional_params$columns)) {
         optional_params$columns <- c(optional_params$columns, "hscp2018")
         remove_partnership_var <- TRUE
       }
-      if (!(is.null(recids)) & file_version == "episode" &
+      if (!is.null(recids) & file_version == "episode" &
         !("recid" %in% optional_params$columns)) {
         optional_params$columns <- c(optional_params$columns, "recid")
         remove_recid_var <- TRUE
@@ -66,15 +66,15 @@ read_slf <-
     # If a partnership is specified filter first;
     # With testing it seems to usually be faster if we do partnership
     # filtering before recid filtering
-    if (!(is.null(partnerships))) {
+    if (!is.null(partnerships)) {
       slfs_list <- purrr::map(
         slfs_list,
         ~ dplyr::filter(.x, .x$hscp2018 %in% partnerships)
       )
     }
 
     # If a recid is specified filter now
-    if (!(is.null(recids))) {
+    if (!is.null(recids)) {
       slfs_list <- purrr::map(
         slfs_list,
         ~ dplyr::filter(.x, .x$recid %in% recids)
@@ -134,9 +134,9 @@ read_slf_episode <-
       read_slf(
         year = year,
         file_version = "episode",
-        partnerships = partnerships,
-        recids = recids,
-        columns = columns,
+        partnerships = unique(partnerships),
+        recids = unique(recids),
+        columns = unique(columns),
         ...
       )
     )
@@ -174,8 +174,8 @@ read_slf_individual <-
       read_slf(
         year = year,
         file_version = "individual",
-        partnerships = partnerships,
-        columns = columns,
+        partnerships = unique(partnerships),
+        columns = unique(columns),
         ...
       )
     )

inst/WORDLIST

@@ -1,14 +1,16 @@
-NSS
+Bedday
+HSCP
+LTC
 ORCID
-RStudio
+Recid
 Rmd
 SLFs
+bedday
+beddays
 dplyr
 fst
+hscdiip
 hscp
 recid
 recids
 tibble
-hscdiip
-PHS
-ORCID

tests/testthat/test-gen_file_path.R

@@ -1,7 +1,11 @@
 test_that("Produces single episode file path", {
   path <- gen_file_path("1718", "episode")
 
-  expect_identical(path, fs::path("/conf/hscdiip/01-Source-linkage-files/source-episode-file-201718.fst"))
+  expect_identical(path, fs::path(
+    "/conf/hscdiip",
+    "01-Source-linkage-files",
+    "source-episode-file-201718.fst"
+  ))
 
   expect_identical(fs::path_ext(path), "fst")
 
@@ -12,7 +16,11 @@ test_that("Produces single episode file path", {
 test_that("Produces single individual file path", {
   path <- gen_file_path("1718", "individual")
 
-  expect_identical(path, fs::path("/conf/hscdiip/01-Source-linkage-files/source-individual-file-201718.fst"))
+  expect_identical(path, fs::path(
+    "/conf/hscdiip",
+    "01-Source-linkage-files",
+    "source-individual-file-201718.fst"
+  ))
 
   expect_identical(fs::path_ext(path), "fst")
 

vignettes/.gitignore

@@ -0,0 +1,2 @@
+*.html
+*.R