demographics file

R/create_episode_file.R

@@ -71,6 +71,7 @@ create_episode_file <- function(
         "anon_chi",
         "person_id",
         "social_care_id",
+        "linking_id",
         "gender",
         "dob",
         "gpprac",
@@ -567,61 +568,40 @@ join_sc_client <- function(data,
   }
 
   if (file_type == "episode") {
-    # Match on client variables by chi
-    # Step 1. Link/join ep with sc_client by anon_chi,
-    #         excluding episodes are not joined. We get `data_file_chi_join`
-    # Step 2. For the episodes are not joined in Step 1,
-    #         join them with sc_client again by person_id,
-    #         excluding those are not joined. We get `data_file_pi_join`
-    # Step 3. Episodes that are non-joined, is `data_file_unjoined`
-    # Step 4. bind rows together
-
+    # Match on client variables by linking_id
     # Step 1
     data_sc <- data %>%
       dplyr::filter(.data$recid %in% c("AT", "HC", "CH", "SDS"))
     data_non_sc <- data %>%
       dplyr::filter(!(.data$recid %in% c("AT", "HC", "CH", "SDS")))
 
-    data_file_chi_join <- data_sc %>%
+    data_file_join <- data_sc %>%
       dplyr::inner_join(
         sc_client,
-        by = "anon_chi",
+        by = "linking_id",
         relationship = "many-to-one",
         suffix = c("", "_sc"),
+        # should never have NA in linking_id
         na_matches = "never"
       ) %>%
       dplyr::select(
         -dplyr::ends_with("_sc")
       )
 
     # Step 2
-    data_file_pi_join <- data_sc %>%
+    data_file_unjoined <- data_sc %>%
       dplyr::filter(!(
-        .data$ep_file_row_id %in% dplyr::pull(data_file_chi_join, .data$ep_file_row_id)
-      )) %>%
-      dplyr::inner_join(
-        sc_client,
-        by = "person_id",
-        relationship = "many-to-one",
-        suffix = c("", "_sc"),
-        na_matches = "never"
-      ) %>%
-      dplyr::select(-dplyr::ends_with("_sc"))
+        .data$ep_file_row_id %in%
+          dplyr::pull(data_file_join, .data$ep_file_row_id)
+      ))
 
     # Step 3
-    data_file_unjoined <- data_sc %>%
-      dplyr::filter(!(.data$ep_file_row_id %in% c(
-        dplyr::pull(data_file_chi_join, .data$ep_file_row_id),
-        dplyr::pull(data_file_pi_join, .data$ep_file_row_id)
-      )))
-
-    # Step 4
     data_file <- dplyr::bind_rows(
-      data_file_chi_join,
-      data_file_pi_join,
+      data_file_join,
       data_file_unjoined,
       data_non_sc
-    )
+    ) %>%
+      select(-"linking_id")
   } else {
     data_file <- data %>%
       dplyr::left_join(

R/fy_helpers.R

@@ -0,0 +1,43 @@
+#' Add financial year from Social Care period
+#' Financial year format, eg 2024
+#' @param data social care data frame
+#'
+#' @returns social care data frame with financial year and financial quarter
+add_fy_qtr_from_period <- function(data) {
+  data %>%
+    # create financial_year and financial_quarter variables for sorting
+    dplyr::mutate(
+      financial_year = as.numeric(stringr::str_sub(.data$period, 1, 4)),
+      financial_quarter = stringr::str_sub(.data$period, 6, 6)
+    ) %>%
+    # set financial quarter to 5 when there is only an annual submission -
+    # for ordering periods with annual submission last
+    dplyr::mutate(
+      financial_quarter = dplyr::if_else(
+        is.na(.data$financial_quarter) |
+          .data$financial_quarter == "",
+        "5",
+        .data$financial_quarter
+      )
+    )
+}
+
+
+#' which_fy
+#' Extract financial year from a date
+#'
+#' @param date a date variable
+#' @param format calenddar year format 2024, or financial year format, 2425
+#' @returns financial year
+which_fy <- function(date, format = c("year", "fyear")) {
+  year <- as.numeric(format(date, "%Y"))
+  month <- as.numeric(format(date, "%m"))
+
+  start_year <- ifelse(month < 4, year - 1, year)
+  end_year <- start_year + 1
+  if (format == "year") {
+    return(start_year)
+  } else {
+    return(paste0(substr(start_year, 3, 4), substr(end_year, 3, 4)))
+  }
+}