From 68b3ca61f4638602b846bd336a5be5a10bd0d3ae Mon Sep 17 00:00:00 2001
From: James Lamb <jaylamb20@gmail.com>
Date: Fri, 31 Jan 2025 22:11:17 -0600
Subject: [PATCH] drop {uuid} dependency (fixes #238) (#240)

---
 .ci/setup.sh                 |  2 +-
 r-pkg/DESCRIPTION            |  3 +--
 r-pkg/NAMESPACE              |  1 -
 r-pkg/R/es_search.R          | 23 ++++++++++++++++++-----
 r-pkg/R/helperfuns.R         | 14 ++++++++++++++
 r-pkg/R/unpack_nested_data.R |  7 ++++++-
 6 files changed, 40 insertions(+), 10 deletions(-)

diff --git a/.ci/setup.sh b/.ci/setup.sh
index 11eb42b..79c8305 100755
--- a/.ci/setup.sh
+++ b/.ci/setup.sh
@@ -19,5 +19,5 @@ sudo apt-get install \
     tidy \
     qpdf
 
-Rscript -e "install.packages(c('assertthat', 'covr', 'data.table', 'futile.logger', 'httr', 'jsonlite', 'knitr', 'lintr', 'markdown', 'purrr', 'stringr', 'testthat', 'uuid'), repos = 'https://cran.r-project.org', Ncpus = parallel::detectCores())"
+Rscript -e "install.packages(c('assertthat', 'covr', 'data.table', 'futile.logger', 'httr', 'jsonlite', 'knitr', 'lintr', 'markdown', 'purrr', 'stringr', 'testthat'), repos = 'https://cran.r-project.org', Ncpus = parallel::detectCores())"
 cp test-data/* r-pkg/inst/testdata/
diff --git a/r-pkg/DESCRIPTION b/r-pkg/DESCRIPTION
index a959520..2d6e00c 100644
--- a/r-pkg/DESCRIPTION
+++ b/r-pkg/DESCRIPTION
@@ -29,8 +29,7 @@ Imports:
     jsonlite,
     purrr,
     stringr,
-    utils,
-    uuid
+    utils
 Suggests:
     knitr,
     markdown,
diff --git a/r-pkg/NAMESPACE b/r-pkg/NAMESPACE
index ab2db40..55403a9 100644
--- a/r-pkg/NAMESPACE
+++ b/r-pkg/NAMESPACE
@@ -45,4 +45,3 @@ importFrom(stringr,str_replace_all)
 importFrom(stringr,str_split)
 importFrom(stringr,str_split_fixed)
 importFrom(utils,read.table)
-importFrom(uuid,UUIDgenerate)
diff --git a/r-pkg/R/es_search.R b/r-pkg/R/es_search.R
index 6113962..c441fdc 100644
--- a/r-pkg/R/es_search.R
+++ b/r-pkg/R/es_search.R
@@ -224,7 +224,6 @@ es_search <- function(es_host
 #' @importFrom data.table rbindlist setkeyv
 #' @importFrom jsonlite fromJSON
 #' @importFrom parallel clusterMap makeForkCluster makePSOCKcluster stopCluster
-#' @importFrom uuid UUIDgenerate
 # nolint end
 .fetch_all <- function(es_host
                      , es_index
@@ -277,7 +276,8 @@ es_search <- function(es_host
 
     # Find a safe path to write to and create it
     repeat {
-        out_path <- file.path(intermediates_dir, uuid::UUIDgenerate())
+        random_dirname <- sprintf("tmp-%s", .random_string(36L))
+        out_path <- file.path(intermediates_dir, random_dirname)
         if (!dir.exists(out_path)) {
             break
         }
@@ -329,7 +329,14 @@ es_search <- function(es_host
     scroll_id <- enc2utf8(firstResult[["_scroll_id"]])
 
     # Write to disk
-    write(x = firstResultJSON, file = file.path(out_path, paste0(uuid::UUIDgenerate(), ".json")))
+    write(
+        x = firstResultJSON
+        , file = tempfile(
+            pattern = "es-"
+            , tmpdir = out_path
+            , fileext = ".json"
+        )
+    )
 
     # Clean up memory
     rm("firstResult", "firstResultJSON")
@@ -464,7 +471,6 @@ es_search <- function(es_host
 #                       max_hits.
 #' @importFrom httr add_headers
 #' @importFrom jsonlite fromJSON
-#' @importFrom uuid UUIDgenerate
 .keep_on_pullin <- function(scroll_id
                             , out_path
                             , max_hits
@@ -511,7 +517,14 @@ es_search <- function(es_host
         scroll_id <- resultList[["_scroll_id"]]
 
         # Write out JSON to a temporary file
-        write(x = resultJSON, file = file.path(out_path, paste0(uuid::UUIDgenerate(), ".json")))
+        write(
+            x = resultJSON
+            , file = tempfile(
+                pattern = "es-"
+                , tmpdir = out_path
+                , fileext = ".json"
+            )
+        )
 
         # Increment the count
         hits_pulled <- hits_pulled + hitsInThisPage
diff --git a/r-pkg/R/helperfuns.R b/r-pkg/R/helperfuns.R
index c0af015..1a04317 100644
--- a/r-pkg/R/helperfuns.R
+++ b/r-pkg/R/helperfuns.R
@@ -7,6 +7,20 @@
     return(httr::content(response, as = as))
 }
 
+# [title] Get a random length-n string
+# [name] .random_string
+# [description] Get a random length-n string of lowercase letters.
+#               Note that this uses sample() and so might produce deterministic
+#               results in programs where set.seed() is used to control randomness.
+.random_string <- function(num_characters) {
+    return(
+        paste0(
+            sample(letters, replace = TRUE, size = num_characters)
+            , collapse = ""
+        )
+    )
+}
+
 # [title] Execute an HTTP request and return the result
 # [name] .request
 # [description] Mainly here to making mocking easier in testing, but this
diff --git a/r-pkg/R/unpack_nested_data.R b/r-pkg/R/unpack_nested_data.R
index 5abc109..52d02a0 100644
--- a/r-pkg/R/unpack_nested_data.R
+++ b/r-pkg/R/unpack_nested_data.R
@@ -56,7 +56,12 @@ unpack_nested_data <- function(chomped_df, col_to_unpack)  {
     inDT <- data.table::copy(chomped_df)
 
     # Define a column name to store original row ID
-    joinCol <- uuid::UUIDgenerate()
+    repeat {
+        joinCol <- .random_string(36L)
+        if (!(joinCol %in% names(inDT))) {
+            break
+        }
+    }
     inDT[, (joinCol) := .I]
 
     # Take out the packed column