gh_decode accepts factor input (hkwi#26)

dshkol · Dec 7, 2019 · 022aa5c · 022aa5c
1 parent e752940
commit 022aa5c
Show file tree

Hide file tree

Showing 4 changed files with 15 additions and 1 deletion.
diff --git a/NEWS.md b/NEWS.md
@@ -2,6 +2,10 @@
 
 ## v0.3.1
 
+### NEW FEATURES
+
+ 1. `gh_decode` accepts and efficiently processes `factor` input by only decoding each level one time, [#17](https://github.com/MichaelChirico/geohashTools/issues/17). If you're likely to have a fair number of duplicate geohashes in your input, consider storing them as a factor or running `gh_decode(as.factor(x))` for efficiency.
+
 ### BUG FIXES
 
  1. `gh_decode` errors early on non-ASCII input to prevent out-of-memory access, [#19](https://github.com/MichaelChirico/geohashTools/issues/19).

diff --git a/R/gh_decode.R b/R/gh_decode.R
@@ -1,4 +1,8 @@
 gh_decode = function(geohashes, include_delta = FALSE, coord_loc = 'c') {
+  if (is.factor(geohashes)) {
+    return(lapply(gh_decode(levels(geohashes), include_delta, coord_loc),
+                  function(z) z[geohashes]))
+  }
   if (length(coord_loc) > 1L)
     stop("Please provide only one value for 'coord_loc'")
   coord_loc = switch(

diff --git a/man/gh_decode.Rd b/man/gh_decode.Rd
@@ -11,12 +11,14 @@
 gh_decode(geohashes, include_delta = FALSE, coord_loc = 'c')
 }
 \arguments{
-  \item{geohashes}{ \code{character} vector of input geohashes. There's no need for all inputs to be of the same precision. }
+  \item{geohashes}{ \code{character} or \code{factor} vector or of input geohashes. There's no need for all inputs to be of the same precision. }
   \item{include_delta}{ \code{logical}; should the cell half-width delta be included in the output? }
   \item{coord_loc}{ \code{character} specifying where in the cell points should be mapped to; cell centroid is mapped by default; case-insensitive. See Details. }
 }
 \details{
   \code{coord_loc} can be the cell's center (\code{'c'} or \code{'centroid'}), or it can be any of the 8 corners (e.g. \code{'s'}/\code{'south'} for the midpoint of the southern boundary of the cell, or \code{'ne'}/\code{'northeast'} for the upper-right corner.
+
+  For \code{factor} input, decoding will be done on the levels for efficiency.
 }
 \value{
   \code{list} with the following entries:

diff --git a/tests/testthat/test-decode.R b/tests/testthat/test-decode.R
@@ -18,6 +18,10 @@ test_that('geohash decoder works', {
                list(latitude = c(-7.60528564453125, 42.91259765625),
                     longitude = c(110.198364257812, 17.60009765625)))
 
+  # input is factor, #17
+  x = gl(4L, 20L, labels = c(borobudur, akarenga, kalakuta, neum))
+  expect_equal(gh_decode(x), gh_decode(as.character(x)))
+
   # option: include_delta
   expect_equal(gh_decode(borobudur, include_delta = TRUE),
                list(latitude = -7.60528564453125,