Style code (GHA)

Author: DougManuel

R/medications.R

@@ -57,8 +57,7 @@
 #'
 #' @export
 create_cat_var <- function(var_raw, cycle, variable_details, variables = NULL,
-                            length, df_mock, prop_NA = NULL, seed = 100) {
-
+                           length, df_mock, prop_NA = NULL, seed = 100) {
   # Level 1: Get variable details for this raw variable + cycle
   var_details <- get_variable_details_for_raw(var_raw, cycle, variable_details, variables)
 
@@ -106,7 +105,6 @@ create_cat_var <- function(var_raw, cycle, variable_details, variables = NULL,
       new = sample(labels, length, replace = TRUE),
       stringsAsFactors = FALSE
     )
-
   } else {
     # Case with NA values using NA codes
     set.seed(seed)
@@ -192,9 +190,8 @@ create_cat_var <- function(var_raw, cycle, variable_details, variables = NULL,
 #'
 #' @export
 create_con_var <- function(var_raw, cycle, variable_details, variables = NULL,
-                            length, df_mock, prop_NA = NULL, seed = 100,
-                            distribution = "uniform") {
-
+                           length, df_mock, prop_NA = NULL, seed = 100,
+                           distribution = "uniform") {
   # Level 1: Get variable details for this raw variable + cycle
   var_details <- get_variable_details_for_raw(var_raw, cycle, variable_details, variables)
 
@@ -261,7 +258,7 @@ create_con_var <- function(var_raw, cycle, variable_details, variables = NULL,
 
   # Handle infinity
   if (is.infinite(overall_min)) overall_min <- 0
-  if (is.infinite(overall_max)) overall_max <- overall_min + 100  # Arbitrary upper bound
+  if (is.infinite(overall_max)) overall_max <- overall_min + 100 # Arbitrary upper bound
 
   # Level 2: Extract NA codes (if prop_NA specified)
   na_labels <- NULL
@@ -285,13 +282,12 @@ create_con_var <- function(var_raw, cycle, variable_details, variables = NULL,
   if (distribution == "normal") {
     # Normal distribution centered at midpoint
     midpoint <- (overall_min + overall_max) / 2
-    spread <- (overall_max - overall_min) / 4  # Use 1/4 of range as SD
+    spread <- (overall_max - overall_min) / 4 # Use 1/4 of range as SD
 
     values <- rnorm(n_regular, mean = midpoint, sd = spread)
 
     # Clip to range
     values <- pmax(overall_min, pmin(overall_max, values))
-
   } else {
     # Uniform distribution (default)
     values <- runif(n_regular, min = overall_min, max = overall_max)

R/mockdata-generators.R

@@ -54,13 +54,14 @@
 #'
 #' # Exclude derived variables
 #' cycle1_original <- get_cycle_variables("cycle1", variables, variable_details,
-#'                                         include_derived = FALSE)
+#'   include_derived = FALSE
+#' )
 #'
 #' @seealso \code{\link{parse_variable_start}}
 #'
 #' @export
 get_cycle_variables <- function(cycle, variables, variable_details,
-                                 include_derived = TRUE) {
+                                include_derived = TRUE) {
   # Basic validation
   if (is.null(cycle) || cycle == "") {
     return(data.frame(
@@ -99,7 +100,9 @@ get_cycle_variables <- function(cycle, variables, variable_details,
   # Extract raw variable names using parse_variable_start
   cycle_vars$variable_raw <- sapply(cycle_vars$variableStart, function(vs) {
     raw_name <- parse_variable_start(vs, cycle)
-    if (is.null(raw_name)) return(NA_character_)
+    if (is.null(raw_name)) {
+      return(NA_character_)
+    }
     return(raw_name)
   })
 
@@ -181,10 +184,11 @@ get_cycle_variables <- function(cycle, variables, variable_details,
 #'
 #' @export
 get_raw_variables <- function(cycle, variables, variable_details,
-                               include_derived = FALSE) {
+                              include_derived = FALSE) {
   # Get all cycle variables (harmonized)
   cycle_vars <- get_cycle_variables(cycle, variables, variable_details,
-                                     include_derived = include_derived)
+    include_derived = include_derived
+  )
 
   # Remove rows with NA raw variable names (e.g., DerivedVar that couldn't be parsed)
   cycle_vars <- cycle_vars[!is.na(cycle_vars$variable_raw), ]
@@ -301,7 +305,8 @@ get_variable_details_for_raw <- function(var_raw, cycle, variable_details, varia
     # Only match if variableStart is EXACTLY the var_raw (no :: or [] or Func:: or DerivedVar::)
     plain_matches <- variable_details[
       variable_details$variableStart == var_raw &
-      grepl(cycle, variable_details$databaseStart, fixed = TRUE), ]
+        grepl(cycle, variable_details$databaseStart, fixed = TRUE),
+    ]
 
     matches <- plain_matches
   }
@@ -385,24 +390,19 @@ get_variable_categories <- function(var_details, include_na = FALSE) {
         # If values not expanded, just use min-max representation
         all_values <- c(all_values, as.character(value))
       }
-
     } else if (parsed$type == "single_value") {
       # Single numeric value
       all_values <- c(all_values, as.character(parsed$value))
-
     } else if (parsed$type == "continuous") {
       # For continuous ranges, keep as-is (don't expand)
       # These will be used for continuous variable generation
       all_values <- c(all_values, as.character(value))
-
     } else if (parsed$type == "special") {
       # Special codes: copy, else, NA::a, NA::b
       all_values <- c(all_values, parsed$value)
-
     } else if (parsed$type == "function") {
       # Function calls: Func::function_name
       all_values <- c(all_values, parsed$value)
-
     } else {
       # Unknown type, use raw value
       all_values <- c(all_values, as.character(value))

R/mockdata-helpers.R

@@ -59,8 +59,12 @@
 #' @export
 parse_variable_start <- function(variable_start, cycle) {
   # Basic validation
-  if (is.null(variable_start) || is.null(cycle)) return(NULL)
-  if (variable_start == "" || cycle == "") return(NULL)
+  if (is.null(variable_start) || is.null(cycle)) {
+    return(NULL)
+  }
+  if (variable_start == "" || cycle == "") {
+    return(NULL)
+  }
 
   # Trim whitespace
   variable_start <- trimws(variable_start)
@@ -99,8 +103,8 @@ parse_variable_start <- function(variable_start, cycle) {
   # Strategy 3: Plain format "varname"
   # Check if it's NOT a DerivedVar, Func, or database-prefixed format
   if (!grepl("^DerivedVar::", variable_start) &&
-      !grepl("^Func::", variable_start) &&
-      !grepl("::", variable_start, fixed = TRUE)) {
+    !grepl("^Func::", variable_start) &&
+    !grepl("::", variable_start, fixed = TRUE)) {
     # Simple variable name, use as-is
     return(trimws(variable_start))
   }
@@ -165,9 +169,9 @@ parse_variable_start <- function(variable_start, cycle) {
 #' # Returns: list(min=30, max=Inf, min_inclusive=TRUE, max_inclusive=FALSE, type="continuous")
 #'
 #' # Special cases
-#' parse_range_notation("NA::a")   # Returns: list(type="special", value="NA::a")
-#' parse_range_notation("copy")    # Returns: list(type="special", value="copy")
-#' parse_range_notation("else")    # Returns: list(type="special", value="else")
+#' parse_range_notation("NA::a") # Returns: list(type="special", value="NA::a")
+#' parse_range_notation("copy") # Returns: list(type="special", value="copy")
+#' parse_range_notation("else") # Returns: list(type="special", value="else")
 #'
 #' @note Adapted from cchsflow v4.0.0 (2025-07-27) - universal across recodeflow projects
 #' @export
@@ -293,7 +297,6 @@ parse_range_notation <- function(range_string, range_type = "auto", expand_integ
       min_inclusive = min_inclusive,
       max_inclusive = max_inclusive
     ))
-
   } else if (range_type == "continuous") {
     return(list(
       type = "continuous",

R/mockdata-parsers.R

@@ -246,13 +246,13 @@ source("R/mockdata-parsers.R")
 var_start <- "cycle1::amsdmva1, [ammdmva1]"
 
 # What does cycle1 parse to?
-parse_variable_start(var_start, "cycle1")  # Should return "amsdmva1"
+parse_variable_start(var_start, "cycle1") # Should return "amsdmva1"
 
 # What does cycle2 parse to?
-parse_variable_start(var_start, "cycle2")  # Should return "ammdmva1"
+parse_variable_start(var_start, "cycle2") # Should return "ammdmva1"
 
 # What does cycle6 parse to?
-parse_variable_start(var_start, "cycle6")  # Should return "ammdmva1"
+parse_variable_start(var_start, "cycle6") # Should return "ammdmva1"
 ```
 
 ### Example 3: Understand range notation parsing
@@ -261,9 +261,9 @@ parse_variable_start(var_start, "cycle6")  # Should return "ammdmva1"
 source("R/mockdata-parsers.R")
 
 # Test different range notations
-parse_range_notation("[7,9]")        # Returns: "7,8,9"
-parse_range_notation("[18.5,25)")    # Returns: "18.5,25" with note "(continuous)"
-parse_range_notation("else")         # Returns: "else"
+parse_range_notation("[7,9]") # Returns: "7,8,9"
+parse_range_notation("[18.5,25)") # Returns: "18.5,25" with note "(continuous)"
+parse_range_notation("else") # Returns: "else"
 
 # Complex example from variable_details
 recodes <- "[18.5,25) :: 1; [7,9] :: .a(Not applicable); else :: .b(Missing)"
@@ -281,7 +281,7 @@ variables <- read.csv("inst/extdata/variables.csv", stringsAsFactors = FALSE)
 
 # Find variables with mixed format (cycle::var, [var])
 mixed_format <- variables[grepl("::", variables$variableStart) &
-                          grepl("\\[", variables$variableStart), ]
+  grepl("\\[", variables$variableStart), ]
 
 cat("Found", nrow(mixed_format), "variables using mixed format\n\n")
 
@@ -304,7 +304,7 @@ variable_details <- read.csv("inst/extdata/variable-details.csv", stringsAsFacto
 # Get all categorical variables for cycle1
 cycle1_vars <- get_cycle_variables("cycle1", variables, variable_details)
 categorical <- cycle1_vars[cycle1_vars$variableType == "Categorical" &
-                           !grepl("^DerivedVar::", cycle1_vars$variableStart), ]
+  !grepl("^DerivedVar::", cycle1_vars$variableStart), ]
 
 cat("Generating", nrow(categorical), "categorical variables for cycle1\n")
 
@@ -384,19 +384,19 @@ if (length(duplicates) > 0) {
 
 4. **Test generation**:
    ```{r eval=FALSE}
-   source("R/mockdata-parsers.R")
-   source("R/mockdata-helpers.R")
-   source("R/mockdata-generators.R")
+source("R/mockdata-parsers.R")
+source("R/mockdata-helpers.R")
+source("R/mockdata-generators.R")
 
-   variables <- read.csv("inst/extdata/variables.csv", stringsAsFactors = FALSE)
-   variable_details <- read.csv("inst/extdata/variable-details.csv", stringsAsFactors = FALSE)
+variables <- read.csv("inst/extdata/variables.csv", stringsAsFactors = FALSE)
+variable_details <- read.csv("inst/extdata/variable-details.csv", stringsAsFactors = FALSE)
 
-   df <- data.frame(id = 1:100)
-   result <- create_cat_var("my_new_var", "cycle1", df, variables, variable_details)
+df <- data.frame(id = 1:100)
+result <- create_cat_var("my_new_var", "cycle1", df, variables, variable_details)
 
-   if (!is.null(result)) {
-     table(result$my_new_var)
-   }
+if (!is.null(result)) {
+  table(result$my_new_var)
+}
    ```
 
 ### Task: Fix a variable that fails to parse
@@ -408,11 +408,11 @@ if (length(duplicates) > 0) {
 
 2. **Test parsing manually**:
    ```{r eval=FALSE}
-   source("R/mockdata-parsers.R")
+source("R/mockdata-parsers.R")
 
-   # Try parsing the variableStart
-   parse_variable_start("cycle1::old_name, [new_name]", "cycle1")
-   parse_variable_start("cycle1::old_name, [new_name]", "cycle2")
+# Try parsing the variableStart
+parse_variable_start("cycle1::old_name, [new_name]", "cycle1")
+parse_variable_start("cycle1::old_name, [new_name]", "cycle2")
    ```
 
 3. **Common fixes**:

mockdata-tools/README.qmd

@@ -85,7 +85,8 @@ test_that("get_cycle_variables filters by exact cycle match", {
     cycles <- strsplit(db_start, ",")[[1]]
     cycles <- trimws(cycles)
     expect_true("cycle1" %in% cycles,
-                info = paste("Variable", cycle1_vars$variable[i], "should have cycle1 in databaseStart"))
+      info = paste("Variable", cycle1_vars$variable[i], "should have cycle1 in databaseStart")
+    )
   }
 })
 
@@ -106,8 +107,10 @@ test_that("get_cycle_variables uses exact match (not substring)", {
 
     # If this variable is in cycle1_meds but NOT in cycle1, that's an error
     if ("cycle1_meds" %in% cycles && !"cycle1" %in% cycles) {
-      fail(paste("Found cycle1_meds-only variable in cycle1 results:",
-                 cycle1_vars$variable[i]))
+      fail(paste(
+        "Found cycle1_meds-only variable in cycle1 results:",
+        cycle1_vars$variable[i]
+      ))
     }
   }
 
@@ -144,7 +147,8 @@ test_that("get_raw_variables returns unique raw variable names", {
 
   # Check that all variable_raw are unique
   expect_equal(nrow(raw_vars), length(unique(raw_vars$variable_raw)),
-               info = "All raw variable names should be unique")
+    info = "All raw variable names should be unique"
+  )
 })
 
 test_that("get_raw_variables groups harmonized variables correctly", {
@@ -158,7 +162,8 @@ test_that("get_raw_variables groups harmonized variables correctly", {
   for (i in 1:nrow(raw_vars)) {
     harmonized_list <- strsplit(raw_vars$harmonized_vars[i], ", ")[[1]]
     expect_equal(raw_vars$n_harmonized[i], length(harmonized_list),
-                 info = paste("Count should match list length for", raw_vars$variable_raw[i]))
+      info = paste("Count should match list length for", raw_vars$variable_raw[i])
+    )
   }
 })
 
@@ -172,7 +177,8 @@ test_that("get_raw_variables excludes derived variables by default", {
 
   # Should not have NA in variable_raw (DerivedVar returns NA)
   expect_true(all(!is.na(raw_vars$variable_raw)),
-              info = "No NA raw variable names when derived excluded")
+    info = "No NA raw variable names when derived excluded"
+  )
 })
 
 # ==============================================================================