updated long pin, retrieval function

DESCRIPTION

@@ -2,7 +2,7 @@ Package: arktax
 Title: Archive: NUCC Taxonomy Codesets, 2009 - 2024
 Version: 0.0.0.9000
 Authors@R: 
-    person("Andrew", "Bruce", , "andrewallenbruce@gmail.com", role = c("aut", "cre", "cph"))
+    person(c("Andrew", "Allen"), "Bruce", , "andrewallenbruce@gmail.com", role = c("aut", "cre", "cph"))
 Maintainer: Andrew Bruce <andrewallenbruce@gmail.com>
 Description: What the package does (one paragraph).
 License: MIT + file LICENSE
@@ -17,7 +17,7 @@ Imports:
 Suggests:
     qs,
     roxyglobals,
-    testthat (>= 3.0.0),
+    testthat (>= 3.0.0)
 Remotes: 
     andrewallenbruce/fuimus
 Config/roxyglobals/filename: generated-globals.R

R/arktax.R

@@ -2,23 +2,26 @@
 #'
 #' @param year `<int>` year of source file release; options are `2009:2024`
 #'
-#' @param code `<chr>` Taxonomy code
+#' @param version `<int>` version of source file; options are `0` or `1`
 #'
-#' @param which `<chr>` wide or long version of the taxonomy; options are `wide` or `long`
+#' @param code `<chr>`  Health Care Provider Taxonomy code, a unique alphanumeric code, ten characters in length
+#'
+#' @param which `<chr>` `wide` or `long` version of the taxonomy
 #'
 #' @returns `<tibble>` of search results
 #'
 #' @examples
 #' retrieve_ark(year = 2024, code = "101Y00000X", which = "wide")
 #'
-#' retrieve_ark(year = 2024, code = "101Y00000X", which = "long")
+#' retrieve_ark(code = "101Y00000X", which = "long")
 #'
 #' @importFrom fuimus search_in_if
 #'
 #' @autoglobal
 #'
 #' @export
 retrieve_ark <- function(year = NULL,
+                         version = NULL,
                          code = NULL,
                          which = c("wide", "long")) {
 
@@ -28,7 +31,10 @@ retrieve_ark <- function(year = NULL,
     long = get_pin("ark_long")
   )
 
-  ark <- search_in_if(ark, ark[["year"]], year)
+  if (which == "wide") {
+    ark <- search_in_if(ark, ark[["year"]], year)
+    ark <- search_in_if(ark, ark[["version"]], version)
+  }
 
   ark <- search_in_if(ark, ark[["code"]], code)
 

data-raw/nucc_pins.R

@@ -155,6 +155,9 @@ ark_long |>
   dplyr::arrange(year) |>
   print(n = 100)
 
+# ark_long <- get_pin("ark_long") |>
+#   dplyr::distinct(code, level, description)
+
 pin_update(
   ark_long,
   name = "ark_long",
@@ -168,6 +171,3 @@ ark_long |>
   dplyr::arrange(code, level, year) |>
   dplyr::filter(code == "133VN1201X") |>
   print(n = 100)
-
-
-

data-raw/raw/Taxonomy Notes.csv

@@ -0,0 +1,14 @@
+[1] Medicare no longer enrolls this provider type.
+"[2] This is an arrangement whereby immunization services are rendered by health care providers who are under contract to the entity/individual who is enrolled in Medicare as a ""Mass Immunization Roster Biller"". The Healthcare Provider Taxonomy Code(s) would be that (those) of the health care provider(s) who render the immunization services.  "
+"[3] Medicare carriers use Specialty Code 87 for Pharmacies when processing prescription drugs that are paid through the Medicare Physician Fee Schedule.  However, the description for Specialty Code 87 remains All Other Suppliers."
+"[4] Because the Medicare Provider Type Physician/Undefined Physician Type applies to physicians,Unknown Supplier/Provider Specialty applies to other than physicians.  Because Medicare does not enroll every non-physician supplier/provider type in the Healthcare Provider Taxonomy Code Set, we cannot arbitrarily link this Medicare Provider Type to all remaining supplier/provider types in the Healthcare Provider Taxonomy Code Set.  In truth, this Medicare Provider Type would link to every Healthcare Provider Taxonomy Code (except those that describe physicians) that is listed in this document.  We do not list them here because doing so would greatly increase the size of this document."
+"[5] Because the Medicare Provider Type indicates that the provider/supplier is a Physician, we have linked this to the most generic Allopathic & Osteopathic Physicians classification.  It could technically link to every Allopathic & Osteopathic Physician."
+[6] Medicare specialty code designation is for Part B DMEPOS purposes.
+[7] Medicare specialty code designation is for Part B DMEPOS purposes.
+[8] Medicare specialty code designation is for Part B DMEPOS purposes.
+[9] Medicare specialty code designation is for Part B DMEPOS purposes.
+[10] Medicare specialty code designation is for Part B DMEPOS purposes.
+[11] Medicare specialty code designation is for Part B DMEPOS purposes.
+"[12] There is no linkage to specific ""Healthcare Provider Taxonomy Codes"" because an ""Indian Health Service""facility can be various provider types/classifications."
+[13] Medicare specialty code designation is for Part B DMEPOS purposes.
+[14] Does not meet the definition of health care provider at 45 CFR 160.103 is not eligible for an NPI.

data-raw/splitting.R

@@ -0,0 +1,56 @@
+source(here::here("data-raw", "pins_internal.R"))
+
+wide <- arktax::retrieve_ark(year = 2024, which = "wide") |>
+  dplyr::select(-modified, -effective, -deactivated)
+
+base <- wide |>
+  dplyr::select(code, notes) |>
+  dplyr::distinct(code, notes, .keep_all = TRUE) |>
+  dplyr::mutate(
+    .id = dplyr::row_number(),
+    updates = stringr::str_extract_all(notes, "\\[.*\\]"),
+    sources = stringr::str_remove_all(notes, "\\s?\\[.*\\]") |> dplyr::na_if(""),
+    notes = NULL) |>
+  tidyr::unnest(updates, keep_empty = TRUE)
+
+updates <- base |>
+  dplyr::select(-sources) |>
+  dplyr::filter(not_na(updates)) |>
+  tidyr::separate_longer_delim(
+    cols = updates,
+    delim = stringr::fixed("; ")) |>
+  dplyr::mutate(
+    updates = stringr::str_remove_all(updates, "(^\\[)|(\\]$)"),
+    first = stringr::str_extract_all(updates, ", \\d{1,2}")) |>
+  tidyr::unnest(first, keep_empty = TRUE) |>
+  dplyr::mutate(
+    first = str_remove(first, ", "),
+    multi = str_count(updates, ", \\d{1}/\\d{1}/\\d{4}"))
+
+updates <- vctrs::vec_rbind(
+  updates |> dplyr::filter(multi == 0),
+  updates |>
+    dplyr::filter(multi > 0) |>
+    tidyr::separate_longer_delim(
+      cols = updates,
+      delim = stringr::regex(", \\d{1,2}")) |>
+    dplyr::mutate(updates = dplyr::if_else(
+      stringr::str_detect(updates, "^[/]"),
+      stringr::str_glue("{first}{updates}"), updates))
+) |>
+  dplyr::select(-first, -multi) |>
+  dplyr::mutate(
+    dates = stringr::str_extract_all(updates, "\\d{1,2}[/]\\d{1,2}[/]\\d{4}"),
+    updates = stringr::str_remove_all(updates, "\\d{1,2}[/]\\d{1,2}[/]\\d{4}[:, ]\\s?")) |>
+  tidyr::unnest(dates, keep_empty = TRUE) |>
+  dplyr::mutate(dates = clock::date_parse(dates, format = "%m/%d/%Y"))
+
+updates |>
+  dplyr::mutate(
+    date_type = dplyr::case_when(
+      stringr::str_detect(updates, "[Nn]ew") ~ "effective",
+      stringr::str_detect(updates, "inactive") ~ "deactivated",
+      .default = "modified") |> forcats::as_factor()) |>
+  dplyr::arrange(code, dates) |>
+  dplyr::count(updates, sort = TRUE) |>
+  print(n = 50)

inst/extdata/pins/_pins.yaml

@@ -1,4 +1,4 @@
 ark_long:
-- ark_long/20241124T223231Z-c6e54/
+- ark_long/20241129T111715Z-13827/
 ark_taxonomy:
 - ark_taxonomy/20241124T214441Z-05252/

inst/extdata/pins/ark_long/20241124T223231Z-c6e54/data.txt → inst/extdata/pins/ark_long/20241129T111715Z-13827/data.txt

@@ -1,10 +1,10 @@
 file: ark_long.qs
-file_size: 26012
-pin_hash: c6e54d54d4b7fdd4
+file_size: 13738
+pin_hash: 13827f0ad127660e
 type: qs
 title: NUCC Taxonomy Archive 2009-2024 (Long)
 description: Health Care Provider Taxonomy Code Set Archive 2009-2024 (Long)
 tags: ~
 urls: ~
-created: 20241124T223231Z
+created: 20241129T111715Z
 api_version: 1