Add county_name to count_fcds() (#86)

Add county_name to count_fcds()

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: fcds
 Title: Process Data from the Florida Cancer Data System
-Version: 0.1.3
+Version: 0.1.4
 Authors@R: 
     c(person(given = "Garrick",
              family = "Aden-Buie",

NEWS.md

@@ -1,3 +1,10 @@
+## fcds 0.1.4
+
+* Added `county_name` argument to `count_fcds()` that adds the county into the
+  count groups. This value can be a vector of counties, or `TRUE` to include all
+  counties in the source data. Alternatively, if the value is `"moffitt"` then
+  the counts are filtered to counties in Moffitt's catchment area.
+
 ## fcds 0.1.2
 
 * Add ICD-O-3 codes from [IACR](http://www.iacr.com.fr/index.php?option=com_content&view=category&layout=blog&id=100&Itemid=577)

R/fcds.R

@@ -62,7 +62,16 @@ join_population_by_year <- function(
 #'
 #' fcds_example %>%
 #'   dplyr::filter(county_name == "Pinellas") %>%
-#'   count_fcds(cancer_site_group, sex = "Male")
+#'   count_fcds(cancer_site_group, sex = "Male", county_name = TRUE) %>%
+#'   head()
+#'
+#' fcds_example %>%
+#'   filter_age_groups(age_gt = 20, age_lt = 25) %>%
+#'   count_fcds(sex = TRUE, county_name = c("Pinellas", "Hillsborough"))
+#'
+#' fcds_example %>%
+#'   count_fcds(county_name = "moffitt") %>%
+#'   head()
 #'
 #' @return A grouped data frame with counts. The output groups includes the
 #'   union of the groups of the original input `data`, the groups specified by
@@ -80,8 +89,13 @@ join_population_by_year <- function(
 #'   `TRUE` to include all values of `race` present in input data
 #' @param origin Character vector of values of `origin` to be included in count,
 #'   or `TRUE` to include all values of `origin` present in input data
-#' @param moffitt_catchment Limit counties to those in the catchment area of the
-#'   [Moffitt Cancer Center](https://moffitt.org).
+#' @param county_name Character vector of values of `county_name` to be included
+#'   in count, or `TRUE` to include all values of `county_name` present in the
+#'   input data, or `"moffitt"` to limit to the counties in the
+#'   catchment area of the [Moffitt Cancer Center](https://moffitt.org).
+#' @param moffitt_catchment **Deprecated.** Please use `county_name =
+#'   "moffitt"` instead to limit counties to those in the catchment
+#'   area of the [Moffitt Cancer Center](https://moffitt.org).
 #' @param default_groups Variables that should be included in the grouping,
 #'   prior to counting cancer cases. Set to `NULL` to use only the groups
 #'   already present in the input data.
@@ -96,14 +110,31 @@ count_fcds <- function(
   sex = NULL,
   race = NULL,
   origin = NULL,
-  moffitt_catchment = FALSE,
+  county_name = NULL,
   default_groups = c("year_group", "year", "age_group"),
-  discard_unseen_levels = TRUE
+  discard_unseen_levels = TRUE,
+  moffitt_catchment = FALSE
 ) {
+  if (!missing(moffitt_catchment)) {
+    .Deprecated(msg = paste(
+      "The moffitt_catchment argument is deprecated, please use",
+      "county_name = \"moffitt\""
+    ))
+    if (!missing(county_name)) {
+      warning(
+        "Both `county` and `moffitt_catchment` were specified, ",
+        "only `county` will be used."
+      )
+    } else {
+      county_name <- if (moffitt_catchment) "moffitt"
+    }
+  }
+
   filters <- list(
     sex = sex,
     race = race,
-    origin = origin
+    origin = origin,
+    county_name = county_name
   )
 
   for (var in names(filters)) {
@@ -116,7 +147,11 @@ count_fcds <- function(
     ))
   }
 
-  if (moffitt_catchment) filters$county_name <- fcds_const("moffitt_catchment")
+  if (!is.null(filters$county_name) && length(filters$county_name) == 1) {
+    if (grepl("moffitt", tolower(filters$county_name))) {
+      filters$county_name <- fcds_const("moffitt_catchment")
+    }
+  }
 
   filters <- purrr::compact(filters)
 

tests/testthat/test-fcds.R

@@ -98,8 +98,40 @@ describe("count_fcds()", {
 
   it("subsets to Moffitt counties", {
     r_count_fcds_moffitt <- fcds::fcds_example %>%
-      count_fcds(moffitt_catchment = TRUE)
-    expect_known_hash(r_count_fcds_moffitt %>% dplyr::ungroup(), "a4ff52c455")
+      count_fcds(county_name = "moffitt_catchment") %>%
+      dplyr::ungroup()
+
+    r_count_fcds_moffitt2 <- fcds::fcds_example %>%
+      count_fcds(county_name = "Moffitt Cancer Center") %>%
+      dplyr::ungroup()
+
+    expect_known_hash(r_count_fcds_moffitt, "a4ff52c455")
+    expect_known_hash(r_count_fcds_moffitt2, "a4ff52c455")
+  })
+
+  it("moffitt_catchment is deprecated", {
+    expect_warning(
+      fcds::fcds_example %>% count_fcds(moffitt_catchment = TRUE),
+      "deprecated"
+    )
+
+    expect_identical(
+      suppressWarnings(fcds::fcds_example %>% count_fcds(moffitt_catchment = TRUE)),
+      fcds::fcds_example %>% count_fcds(county_name = "moffitt_catchment")
+    )
+  })
+
+  it("county_name = TRUE includes county_name in group vars", {
+    fcds_county <- fcds::fcds_example %>% count_fcds(county_name = TRUE)
+    expect_identical(
+      dplyr::group_vars(fcds_county),
+      c("county_name", "year_group", "year", "age_group")
+    )
+
+    expect_identical(
+      fcds_county %>% .$county_name %>% paste() %>% unique() %>% sort(),
+      fcds::fcds_example %>% .$county_name %>% paste() %>% unique() %>% sort()
+    )
   })
 
   it("errors when invalid FCDS constants are provided", {
@@ -138,7 +170,7 @@ describe("count_fcds()", {
   it("removes un-observed factor levels in output groups", {
     r_cfl <- fcds::fcds_example %>%
       filter(year > 2000) %>%
-      count_fcds(moffitt_catchment = TRUE, sex = "Male")
+      count_fcds(county_name = "Moffitt", sex = "Male")
 
     expect_true(
       length(setdiff(levels(r_cfl$county_name), fcds_const("moffitt_catchment"))) == 0