CRAN submission changes (#70)

* adjusting URLs * updating Rd files * trying to be more specific for interactive session detection. * further improving tests * add error case for auth * fixing tests * adjusting dbfs tests to clean up after themselves better * Further adjustments of tests to use withr * Change to native pipe to drop magrittr dependency * Remove dependency on data.table * simplify multipart body code * Fixing REPL error and removing test that may cause CRAN issues --------- Co-authored-by: Zac Davies <[email protected]>
databrickslabs · Nov 14, 2024 · c032a87 · c032a87
1 parent 4df4b3e
commit c032a87
Show file tree

Hide file tree

Showing 42 changed files with 279 additions and 217 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -20,18 +20,18 @@ Description: Collection of utilities that improve using 'Databricks' from R.
 License: Apache License (>= 2)
 Encoding: UTF-8
 LazyData: true
+Depends:
+  R (>= 4.1.0)
 Imports:
     arrow,
     base64enc,
     cli,
     curl,
-    data.table,
     dplyr,
     glue,
     httr2,
     ini,
     jsonlite,
-    magrittr,
     purrr,
     reticulate,
     R6 (>= 2.4.0),
@@ -42,12 +42,10 @@ Suggests:
     testthat (>= 3.0.0),
     huxtable,
     htmltools,
-    htmlwidgets,
     knitr,
     magick,
     rmarkdown,
     rstudioapi,
-    rvest,
     withr
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.3.2

diff --git a/NAMESPACE b/NAMESPACE
@@ -220,7 +220,6 @@ import(cli)
 import(httr2)
 import(tibble)
 importFrom(glue,glue)
-importFrom(magrittr,`%>%`)
 importFrom(rlang,.data)
 importFrom(stats,setNames)
 importFrom(utils,object.size)
diff --git a/NOTICE b/NOTICE
@@ -22,9 +22,6 @@ Copyright 2021 httr2 authors
 jeroen/jsonlite - https://github.com/jeroen/jsonlite
 Copyright 2020 Jeroen Ooms
 
-tidyverse/magrittr - https://github.com/tidyverse/magrittr
-Copyright 2023 magrittr authors
-
 tidyverse/purrr - https://github.com/tidyverse/purrr
 Copyright 2023 purrr authors
 
@@ -41,13 +38,6 @@ rstudio/rstudionapi - https://github.com/rstudio/rstudioapi
 Copyright 2015 RStudio
 tidyverse/rvest - https://github.com/tidyverse/rvest
 Copyright 2023 rvest authors
-
-__________
-This Software contains code from the following open source projects, licensed under the MPL-2 license:
-
-Rdatatable/data.table - https://github.com/Rdatatable/data.table
-Copyright data.table authors
-
 __________
 This Software contains code from the following open source projects, licensed under the GPL-2/GPL-3 licenses:
 

diff --git a/R/clusters.R b/R/clusters.R
@@ -325,7 +325,7 @@ db_cluster_action <- function(cluster_id,
     token = token
   )
 
-  req <- req %>%
+  req <- req |>
     httr2::req_body_json(body)
 
   if (perform_request) {
@@ -558,12 +558,12 @@ db_cluster_get <- function(cluster_id,
     token = token
   )
 
-  req <- req %>%
+  req <- req |>
     httr2::req_body_json(body)
 
   if (perform_request) {
-    req %>%
-      httr2::req_perform() %>%
+    req |>
+      httr2::req_perform() |>
       httr2::resp_body_json()
   } else {
     req
@@ -872,26 +872,26 @@ get_latest_dbr <- function(lts, ml, gpu, photon,
 
   runtimes <- db_cluster_runtime_versions(host = host, token = token)
 
-  runtimes_adj <- runtimes[[1]] %>%
+  runtimes_adj <- runtimes[[1]] |>
     purrr::map_dfr(function(x) {
       list(key = x[["key"]], name = x[["name"]])
-    }) %>%
+    }) |>
     dplyr::mutate(
       version = as.numeric(gsub("^(\\d+\\.\\d)\\..*", "\\1", .data$key)),
       lts = grepl("LTS", .data$name),
       ml = grepl("ml", .data$key),
       gpu = grepl("gpu", .data$key),
       photon = grepl("photon", .data$key),
-    ) %>%
+    ) |>
     dplyr::arrange(dplyr::desc(version))
 
-  runtime_matches <- runtimes_adj %>%
+  runtime_matches <- runtimes_adj |>
     dplyr::filter(
       .data$lts == {{lts}},
       .data$ml == {{ml}},
       .data$gpu == {{gpu}},
       .data$photon == {{photon}}
-    ) %>%
+    ) |>
     dplyr::slice_head(n = 1)
 
   list(

diff --git a/R/connection-pane.R b/R/connection-pane.R
@@ -140,7 +140,7 @@ get_uc_model_versions <- function(catalog, schema, model, host, token,
   aliases <- purrr::map(
     model_info$aliases, ~{
       setNames(.x$version_num, .x$alias_name)
-    }) %>%
+    }) |>
     unlist()
 
   version_names <- purrr::map_chr(versions, function(x) {
@@ -281,8 +281,8 @@ get_schema_objects <- function(catalog, schema, host, token) {
 
   # how many objects of each type exist
   # only show when objects exist within
-  sizes <- purrr::map_int(objects, nrow) %>%
-    purrr::keep(~.x > 0) %>%
+  sizes <- purrr::map_int(objects, nrow) |>
+    purrr::keep(~.x > 0) |>
     purrr::imap_chr(~ glue::glue("{.y} ({.x})"))
 
   data.frame(

diff --git a/R/data-structures.R b/R/data-structures.R
@@ -315,7 +315,7 @@ is.file_storage_info <- function(x) {
 #' `bucket-owner-full-control`. If `canned_acl` is set, the cluster instance
 #' profile must have `s3:PutObjectAcl` permission on the destination bucket and
 #' prefix. The full list of possible canned ACLs can be found in
-#' [docs](https://docs.aws.amazon.com/AmazonS3/latest/dev/acl-overview.html#canned-acl).
+#' [docs](https://docs.aws.amazon.com/AmazonS3/latest/userguide/acl-overview.html#canned-acl).
 #' By default only the object owner gets full control. If you are using cross
 #' account role for writing data, you may want to set
 #' `bucket-owner-full-control` to make bucket owner able to read the logs.

diff --git a/R/dbfs.R b/R/dbfs.R
@@ -454,14 +454,10 @@ db_dbfs_put <- function(path, file = NULL, contents = NULL, overwrite = FALSE,
   )
 
   if (perform_request) {
-    req %>%
-      httr2::req_body_multipart(
-        path = body$path,
-        contents = body$contents,
-        overwrite = body$overwrite
-      ) %>%
-      httr2::req_error(body = db_req_error_body) %>%
-      httr2::req_perform() %>%
+    req |>
+      httr2::req_body_multipart(!!!body) |>
+      httr2::req_error(body = db_req_error_body) |>
+      httr2::req_perform() |>
       httr2::resp_body_json()
   } else {
     req

diff --git a/R/execution-context.R b/R/execution-context.R
@@ -98,7 +98,7 @@ db_context_status <- function(cluster_id,
     token = token
   )
 
-  req <- req %>%
+  req <- req |>
     httr2::req_url_query(
       clusterId = cluster_id,
       contextId = context_id
@@ -254,7 +254,7 @@ db_context_command_status <- function(cluster_id,
     token = token
   )
 
-  req <- req %>%
+  req <- req |>
     httr2::req_url_query(
       clusterId = cluster_id,
       contextId = context_id,
@@ -292,7 +292,7 @@ db_context_command_cancel <- function(cluster_id,
     token = token
   )
 
-  req <- req %>%
+  req <- req |>
     httr2::req_url_query(
       clusterId = cluster_id,
       contextId = context_id,
@@ -327,13 +327,16 @@ db_context_command_parse <- function(x, language = c("r", "py", "scala", "sql"))
   }
 
   if (x$results$resultType == "table") {
-    schema <- data.table::rbindlist(x$results$schema)
-    tbl <- data.table::rbindlist(x$results$data)
-    names(tbl) <- schema$name
+    schema <- dplyr::bind_rows(x$results$schema)
 
-    output_tbl <- huxtable::hux(tbl) %>%
-      huxtable::set_all_borders(TRUE) %>%
-      huxtable::set_font_size(10) %>%
+    tbl <- purrr::list_transpose(x$results$data) |>
+      as.data.frame()
+
+    names(tbl) <- schema$names
+
+    output_tbl <- huxtable::hux(tbl) |>
+      huxtable::set_all_borders(TRUE) |>
+      huxtable::set_font_size(10) |>
       huxtable::set_position("left")
 
     huxtable::print_screen(output_tbl)

diff --git a/R/experiments.R b/R/experiments.R
@@ -54,7 +54,7 @@ db_experiments_get <- function(name = NULL, id = NULL,
     body = body,
     host = host,
     token = token
-  ) %>%
+  ) |>
     httr2::req_url_path_append(endpoint_suffix)
 
   if (perform_request) {

diff --git a/R/feature-store.R b/R/feature-store.R
@@ -42,7 +42,7 @@ db_feature_tables_get <- function(feature_table,
     version = "2.0",
     host = host,
     token = token
-  ) %>%
+  ) |>
     httr2::req_url_query(name = feature_table)
 
   if (perform_request) {
@@ -63,7 +63,7 @@ db_feature_table_features <- function(feature_table,
     version = "2.0",
     host = host,
     token = token
-  ) %>%
+  ) |>
     httr2::req_url_query(feature_table = feature_table)
 
   if (perform_request) {

diff --git a/R/misc-helpers.R b/R/misc-helpers.R
@@ -41,9 +41,9 @@ db_current_workspace_id <- function(host = db_host(), token = db_token(),
   )
 
   if (perform_request) {
-    resp <- req %>%
-      httr2::req_error(body = db_req_error_body) %>%
-      httr2::req_perform() %>%
+    resp <- req |>
+      httr2::req_error(body = db_req_error_body) |>
+      httr2::req_perform() |>
       httr2::resp_headers()
 
     # workspace id can be extracted from response headers

diff --git a/R/package-auth.R b/R/package-auth.R
@@ -68,7 +68,7 @@ db_host <- function(id = NULL, prefix = NULL, profile = default_config_profile()
 #'
 #' @description
 #' The function will check for a token in the `DATABRICKS_HOST` environment variable.
-#' `.databrickscfg` will be searched if `db_profile` and `use_databrickscfg` are set or 
+#' `.databrickscfg` will be searched if `db_profile` and `use_databrickscfg` are set or
 #' if Posit Workbench managed OAuth credentials are detected.
 #' If none of the above are found then will default to using OAuth U2M flow.
 #'
@@ -98,7 +98,7 @@ db_token <- function(profile = default_config_profile()) {
 #' @description
 #' Workspace ID, optionally specified to make connections pane more powerful.
 #' Specified as an environment variable `DATABRICKS_WSID`.
-#' `.databrickscfg` will be searched if `db_profile` and `use_databrickscfg` are set or 
+#' `.databrickscfg` will be searched if `db_profile` and `use_databrickscfg` are set or
 #' if Posit Workbench managed OAuth credentials are detected.
 #'
 #' Refer to [api authentication docs](https://docs.databricks.com/dev-tools/api/latest/authentication.html)
@@ -223,7 +223,6 @@ read_env_var <- function(key = c("token", "host", "wsid"),
 
   value <- Sys.getenv(key_name)
 
-
   if (value == "") {
     if (error) {
       stop(cli::format_error(c(
@@ -273,7 +272,7 @@ db_oauth_client <- function(host = db_host()) {
 #' Returns the default config profile
 #' @details Returns the config profile first looking at `DATABRICKS_CONFIG_PROFILE`
 #' and then the `db_profile` option.
-#' 
+#'
 #' @return profile name
 #' @keywords internal
 default_config_profile <- function() {
@@ -288,7 +287,7 @@ default_config_profile <- function() {
 #' Returns whether or not to use a `.databrickscfg` file
 #' @details Indicates `.databrickscfg` should be used instead of environment variables when
 #' either the `use_databrickscfg` option is set or Posit Workbench managed OAuth credentials are detected.
-#' 
+#'
 #' @return boolean
 #' @keywords internal
 use_databricks_cfg <- function() {
@@ -297,4 +296,27 @@ use_databricks_cfg <- function() {
     use_databricks_cfg <- TRUE
   }
   return(use_databricks_cfg)
-}
+}
+
+
+# Extended from {odbc}
+#
+# Try to determine whether we can redirect the user's browser to a server on
+# localhost, which isn't possible if we are running on a hosted platform.
+#
+# This is based on the strategy pioneered by the {gargle} package and {httr2}.
+is_hosted_session <- function() {
+
+  if (on_databricks()) {
+    return(TRUE)
+  }
+
+  if (nzchar(Sys.getenv("COLAB_RELEASE_TAG"))) {
+    return(TRUE)
+  }
+
+  # If RStudio Server or Posit Workbench is running locally (which is possible,
+  # though unusual), it's not acting as a hosted environment.
+  Sys.getenv("RSTUDIO_PROGRAM_MODE") == "server" &&
+    !grepl("localhost", Sys.getenv("RSTUDIO_HTTP_REFERER"), fixed = TRUE)
+}