use dummy address components for better addr parsing of tiger street …

…range names (#35)

* structure tiger street names for addr parsing

* print addr objects in color and with style to visually represent tags (#34)

* example code for working with s2 parent cells and mapping with  rdeck (#33)

* parse tiger ranges into addr with dummy addr tags

* cleanup, update authors

---------

Co-authored-by: Cole Brokamp <[email protected]>

DESCRIPTION

@@ -2,8 +2,8 @@ Package: addr
 Title: Clean, Parse, Harmonize, Match, and Geocode Messy Real-World Addresses
 Version: 0.4.0.9020
 Authors@R: 
-    person("Cole", "Brokamp", , "[email protected]", role = c("aut", "cre"),
-           comment = c(ORCID = "0000-0002-0289-3151"))
+    c(person("Cole", "Brokamp", email = "[email protected]", role = c("aut", "cre")),
+      person("Erika", "Manning", role = c("aut")))
 Description: Addresses that were not validated at the time of collection are often heterogenously formatted, making them difficult to compare or link to other sets of addresses. The addr package is designed to clean character strings of addresses, use the `usaddress` library to tag address components, and paste together select components to create a normalized address. Normalized addresses can be hashed to create hashdresses that can be used to merge with other sets of addresses.
 URL: https://github.com/cole-brokamp/addr, https://cole-brokamp.github.io/addr/
 BugReports: https://github.com/cole-brokamp/addr/issues

R/addr_tiger_match.R

@@ -54,6 +54,9 @@ get_tiger_street_ranges <- function(county, year = "2022") {
 #' a street range tibble with zero rows indicates that although a street was matched,
 #' there was no range containing the street number
 #' @export
+#' @details
+#' To best parse street names and types, this function appends dummy address components just
+#' for the purposes of matching tiger street range names (e.g., `1234 {tiger_street_name} Anytown AB 00000`)
 #' @examples
 #' my_addr <- as_addr(c("224 Woolper Ave", "3333 Burnet Ave", "33333 Burnet Ave", "609 Walnut St"))
 #' 
@@ -78,7 +81,7 @@ addr_match_tiger_street_ranges <- function(x,
 
   street_matches <-
     addr_match_street(ia,
-      suppressWarnings(as_addr(names(d_tiger))),
+      suppressWarnings(as_addr(glue::glue("1234 {names(d_tiger)} Anytown AB 00000"))),
       stringdist_match = "osa_lt_1",
       match_street_type = TRUE
     ) |>

tests/testthat/test-addr_match_geocode.R

@@ -8,7 +8,7 @@ test_that("addr_match_geocode() works", {
   table(my_geocodes$match_method) |>
     expect_identical(
       structure(
-        c(ref_addr = 216L, tiger_range = 11L, tiger_street = 2L, none = 22L),
+        c(ref_addr = 216L, tiger_range = 16L, tiger_street = 2L, none = 17L),
         dim = 4L,
         dimnames = structure(list(c("ref_addr", "tiger_range", "tiger_street", "none")), names = ""),
         class = "table"

tests/testthat/test-addr_tiger_match.R

@@ -11,15 +11,16 @@ test_that("get_tiger_street_ranges() works", {
 })
 
 test_that("addr_match_tiger_street_ranges() works", {
-  addr_match_tiger_street_ranges(as_addr(c("224 Woolper Ave", "3333 Burnet Ave", "33333 Burnet Ave", "609 Walnut St")),
+  addr_match_tiger_street_ranges(as_addr(c("224 Woolper Ave", "3333 Burnet Ave", "33333 Burnet Ave", "609 Walnut St", "609 Weknut Street")),
     street_only_match = "none"
   ) |>
     purrr::map(nrow) |>
     expect_identical(list(
       `224 Woolper Avenue` = 1L,
       `3333 Burnet Avenue` = 2L,
       `33333 Burnet Avenue` = 0L,
-      `609 Walnut Street` = NULL
+      `609 Walnut Street` = 2L,
+      `609 Weknut Street` = NULL
     ))
 
   addr_match_tiger_street_ranges(as_addr(c("224 Woolper Ave", "3333 Burnet Ave", "33333 Burnet Ave", "609 Walnut St")),
@@ -30,7 +31,7 @@ test_that("addr_match_tiger_street_ranges() works", {
       `224 Woolper Avenue` = 1L,
       `3333 Burnet Avenue` = 2L,
       `33333 Burnet Avenue` = 20L,
-      `609 Walnut Street` = NULL
+      `609 Walnut Street` = 2L
     ))
 
   addr_match_tiger_street_ranges(as_addr(c("224 Woolper Ave", "3333 Burnet Ave", "33333 Burnet Ave", "609 Walnut St")),
@@ -41,6 +42,6 @@ test_that("addr_match_tiger_street_ranges() works", {
       `224 Woolper Avenue` = 1L,
       `3333 Burnet Avenue` = 2L,
       `33333 Burnet Avenue` = 1L,
-      `609 Walnut Street` = NULL
+      `609 Walnut Street` = 2L
     ))
 })