Improve use of oa_lad21_lookup custom lookup

Author: francisbarton

DESCRIPTION

@@ -1,5 +1,5 @@
 Package: jogger
-Version: 0.2.15
+Version: 0.2.16
 Title: Retrieve Area Boundaries and Data from the ONS Open Geography Portal
 Description: The main function of this package is to download area boundaries
     (in GeoJSON format) using the ONS Open Geography API, for all sub-areas - at a
@@ -12,7 +12,6 @@ Authors@R: person("Francis", "Barton",
   role = c("aut", "cre"))
 License: MIT + file LICENSE
 Encoding: UTF-8
-LazyData: true
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.1.1
 Imports: 

NEWS.md

@@ -1,3 +1,7 @@
+# 0.2.16
+
+* Simplifies `treat_results` routine and use of `oa_lad21_lookup` custom lookup
+
 # 0.2.15
 
 * Fixes internal data issue where datasets weren't being found

R/create_custom_lookup.R

@@ -159,9 +159,9 @@ create_custom_lookup <- function(bounds_level,
     if (within_level == "cty") {
       within_level <- "utla"
     }
-    if (within_level == "ltla") {
-      within_level <- "lad"
-    }
+    # if (within_level == "ltla") {
+    #   within_level <- "lad"
+    # }
   }
 
   if (bounds_level == "lad" && within_level %in% c("utla", "rgn")) {
@@ -246,24 +246,22 @@ create_custom_lookup <- function(bounds_level,
     }
 
     if (return_style == "minimal") {
-      df %>%
+      df1 <- df %>%
         # return just first two columns
-        dplyr::select(!!rlang::sym(fields[1]):!!rlang::sym(fields[2])) %>%
-        dplyr::distinct() %>%
-        janitor::remove_empty("cols")
+        dplyr::select(!!rlang::sym(fields[1]):!!rlang::sym(fields[2]))
     } else if (return_style == "simple") {
-      df %>%
+      df1 <- df %>%
         # return just the columns in "fields"
-        dplyr::select(any_of(fields)) %>%
-        dplyr::distinct() %>%
-        janitor::remove_empty("cols")
+        dplyr::select(any_of(fields))
     } else {
-      df %>%
+      df1 <- df %>%
         # return all columns between first and last specified fields
-        dplyr::select(!!rlang::sym(fields[1]):!!rlang::sym(fields[end_col])) %>%
-        dplyr::distinct() %>%
-        janitor::remove_empty("cols")
+        dplyr::select(!!rlang::sym(fields[1]):!!rlang::sym(fields[end_col]))
     }
+
+    df1 %>%
+      dplyr::distinct() %>%
+      janitor::remove_empty("cols")
   }
 
 
@@ -272,38 +270,29 @@ create_custom_lookup <- function(bounds_level,
   if (within_cd) nth_field <- -2
   if (bounds_cd) nth_field <- 1
 
+  where_level <- dplyr::nth(fields, nth_field)
 
-  # no available ONS API lookup for UTLA:RGN, so use our built-in table
-  # (also prefer it for LAD:RGN lookups!):
-  if (bounds_level %in% c("upper", "utla") && within_level %in% c("region", "rgn")) {
-    df_out <- upper_tier_region_ctry_lookup %>%
-      dplyr::select(!(c(ltla21cd, ltla21nm, ctry21cd, ctry21nm))) %>%
-      dplyr::filter(rgn21nm %in% within) %>%
-      dplyr::distinct()
-  } else if (bounds_level %in% c("lad", "ltla") && within_level %in% c("region", "rgn")) {
-    df_out <- upper_tier_region_ctry_lookup %>%
-      dplyr::select(!(c(ctry21cd, ctry21nm))) %>%
-      dplyr::filter(rgn21nm %in% within)
-  } else if (bounds_level %in% c("upper", "utla") && within_level %in% c("country", "ctry")) {
-    df_out <- upper_tier_region_ctry_lookup %>%
-      dplyr::filter(ctry21nm %in% within) %>%
+  if (all(fields %in% names(oa_lad21_lookup))) {
+    df1 <- oa_lad21_lookup %>%
+      dplyr::filter(.data[[where_level]] %in% within) %>%
       dplyr::distinct()
   } else {
     # and for other things use the API as usual:
-
-    df_out <- within %>%
+    df1 <- within %>%
       batch_it_simple(batch_size = 25) %>% # from my myrmidon pkg
       purrr::map_df( ~ build_api_query(
       ref = table_code_ref,
-      where_level = dplyr::nth(fields, nth_field),
+      where_level = where_level,
       where = .,
       fields = return_fields
       ) %>%
-      extract_lookup() %>%
-      treat_results(return_style = return_style)
+      extract_lookup()
       )
   }
 
+  df_out <- df1 %>%
+    treat_results(return_style = return_style)
+
 
   # if not specified by the user, make an educated decision about
   # including Welsh language MSOA and LAD names (LAD19NMW / MSOA11NMW /

R/oa_lad21_lookup.R

@@ -0,0 +1,21 @@
+#' A lookup of OAs in England and Wales to higher geographies
+#'
+#' @format A data frame with 181408 rows and 15 variables:
+#' \describe{
+#'   \item{oa11cd}{OA code}
+#'   \item{lsoa21cd}{LSOA code}
+#'   \item{lsoa21nm}{LSOA name}
+#'   \item{msoa11cd}{MSOA code}
+#'   \item{msoa11nm}{MSOA name}
+#'   \item{lad20cd}{LAD code}
+#'   \item{lad20nm}{LAD name}
+#'   \item{ltla21cd}{LTLA code}
+#'   \item{ltla21nm}{LTLA name}
+#'   \item{utla21cd}{UTLA code}
+#'   \item{utla21nm}{UTLA name}
+#'   \item{rgn21cd}{Region code}
+#'   \item{rgn21nm}{Region name}
+#'   \item{ctry21cd}{Country code}
+#'   \item{ctry21nm}{Country name}
+#' }
+"oa_lad21_lookup"

R/sysdata.rda

@@ -64,6 +64,28 @@ lad21nmw_lookup <- jsonlite::fromJSON(
 
 
 
+nnhts <- c("Corby", "East Northamptonshire", "Kettering", "Wellingborough")
+wnhts <- c("Daventry", "Northampton", "South Northamptonshire")
+
+
+oa_lad21_lookup <- jsonlite::fromJSON(
+  # https://geoportal.statistics.gov.uk/datasets/ons::output-area-to-lower-layer-super-output-area-to-middle-layer-super-output-area-to-local-authority-district-december-2020-lookup-in-england-and-wales/
+  "https://opendata.arcgis.com/datasets/65664b00231444edb3f6f83c9d40591f_0.geojson"
+) %>%
+  purrr::pluck("features", "properties") %>%
+  janitor::clean_names() %>%
+  dplyr::select(!c(fid, rgn20cd, rgn20nm)) %>%
+  dplyr::mutate(ltla21nm = dplyr::case_when(
+    lad20nm %in% nnhts ~ "North Northamptonshire",
+    lad20nm %in% wnhts ~ "West Northamptonshire",
+    TRUE ~ lad20nm
+  )) %>%
+  dplyr::left_join(upper_tier_region_ctry_lookup) %>%
+  dplyr::relocate(ltla21cd, .before = ltla21nm)
+
+
+
+
 
 hocl_msoa_names <- paste0(
   "https://houseofcommonslibrary.github.io/",
@@ -72,4 +94,4 @@ hocl_msoa_names <- paste0(
   dplyr::select(-Laname)
 
 
-usethis::use_data(lsoa11cdnm, upper_tier_region_ctry_lookup, lad21nmw_lookup, hocl_msoa_names, overwrite = TRUE, internal = TRUE, compress = "bzip2")
+usethis::use_data(lsoa11cdnm, upper_tier_region_ctry_lookup, lad21nmw_lookup, hocl_msoa_names, oa_lad21_lookup, overwrite = TRUE, internal = TRUE, compress = "bzip2")