Replace tibbles with data frames to improve performance

DESCRIPTION

@@ -34,6 +34,7 @@ Imports:
     rprojroot (>= 1.1),
     tibble (>= 1.4.2),
     tools,
+    vctrs (>= 0.4.1),
     withr (>= 1.0.0),
 Suggests: 
     data.tree (>= 0.1.6),

NAMESPACE

@@ -17,7 +17,6 @@ export(style_text)
 export(tidyverse_math_token_spacing)
 export(tidyverse_reindention)
 export(tidyverse_style)
-import(tibble)
 importFrom(magrittr,"%>%")
 importFrom(magrittr,or)
 importFrom(magrittr,set_names)
@@ -43,6 +42,7 @@ importFrom(rlang,is_installed)
 importFrom(rlang,seq2)
 importFrom(rlang,warn)
 importFrom(rlang,with_handlers)
+importFrom(tibble,tribble)
 importFrom(utils,capture.output)
 importFrom(utils,tail)
 importFrom(utils,write.table)

R/compat-dplyr.R

@@ -26,7 +26,7 @@ arrange_pos_id <- function(data) {
 
 bind_rows <- function(x, y = NULL, ...) {
   if (is.null(x) && is.null(y)) {
-    return(new_tibble(list()))
+    return(new_styler_df(list()))
   }
   if (is.null(x)) {
     if (inherits(y, "data.frame")) {
@@ -62,7 +62,7 @@ left_join <- function(x, y, by) {
 
   res <- merge(x, y, by.x = by_x, by.y = by_y, all.x = TRUE, sort = FALSE) %>%
     arrange_pos_id()
-  res <- new_tibble(res)
+  res <- new_styler_df(res)
   # dplyr::left_join set unknown list columns to NULL, merge sets them
   # to NA
   if (exists("child", res) && anyNA(res$child)) {

R/compat-tidyr.R

@@ -6,5 +6,5 @@ nest_ <- function(data, key_col, nest_cols = character()) {
   res <- list()
   res[[key_column]] <- key_levels
   res[[key_col]] <- split(data[, nest_cols], key_factor)
-  new_tibble(res)
+  new_styler_df(res)
 }

R/nest.R

@@ -249,7 +249,7 @@ add_terminal_token_after <- function(pd_flat) {
     filter(terminal) %>%
     arrange_pos_id()
 
-  rhs <- new_tibble(
+  rhs <- new_styler_df(
     list(
       pos_id = terminals$pos_id,
       token_after = lead(terminals$token, default = "")
@@ -266,7 +266,7 @@ add_terminal_token_before <- function(pd_flat) {
     filter(terminal) %>%
     arrange_pos_id()
 
-  rhs <- new_tibble(
+  rhs <- new_styler_df(
     list(
       id = terminals$id,
       token_before = lag(terminals$token, default = "")

R/nested-to-tree.R

@@ -16,6 +16,7 @@ create_tree_from_pd_with_default_style_attributes <- function(pd,
                                                               structure_only = FALSE) {
   pd %>%
     create_node_from_nested_root(structure_only) %>%
+    # don't use `styler_df()` here; `vctrs::data_frame()` only accepts a vector, not a <Node/R6> object
     as.data.frame()
 }
 

R/parse.R

@@ -94,7 +94,7 @@ get_parse_data <- function(text, include_text = TRUE, ...) {
   # avoid https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=16041
   parse_safely(text, keep.source = TRUE)
   parsed <- parse_safely(text, keep.source = TRUE)
-  pd <- as_tibble(
+  pd <- styler_df(
     utils::getParseData(parsed, includeText = include_text),
     .name_repair = "minimal"
   )
@@ -163,7 +163,7 @@ ensure_correct_txt <- function(pd, text) {
     by.y = "id",
     suffixes = c("", "parent")
   ) %>%
-    as_tibble(.name_repair = "minimal")
+    styler_df(.name_repair = "minimal")
 
   if (!lines_and_cols_match(new_text)) {
     abort(paste(

R/style-guides.R

@@ -315,7 +315,7 @@ tidyverse_style <- function(scope = "tokens",
 #' }
 #' set_line_break_before_curly_opening_style <- function() {
 #'   create_style_guide(
-#'     line_break = tibble::lst(set_line_break_before_curly_opening),
+#'     line_break = list(set_line_break_before_curly_opening),
 #'     style_guide_name = "some-style-guide",
 #'     style_guide_version = "some-version"
 #'   )

R/stylerignore.R

@@ -123,8 +123,8 @@ apply_stylerignore <- function(flattened_pd) {
     env_current$stylerignore[, colnames_required_apply_stylerignore],
     by.x = "pos_id", by.y = "first_pos_id_in_segment", all.x = TRUE,
     sort = FALSE
-  ) %>%
-    as_tibble()
+  )
+
   flattened_pd %>%
     stylerignore_consolidate_col("lag_newlines") %>%
     stylerignore_consolidate_col("lag_spaces") %>%

R/token-create.R

@@ -39,7 +39,7 @@ create_tokens <- function(tokens,
                           block = NA,
                           is_cached = FALSE) {
   len_text <- length(texts)
-  new_tibble(
+  new_styler_df(
     list(
       token = tokens,
       text = texts,

R/token-define.R

@@ -1,4 +1,4 @@
-token <- tribble(
+token <- tibble::tribble(
   ~text,           ~class,         ~token,
   "&",          "logical",          "AND",
   "&&",         "logical",         "AND2",

R/transform-files.R

@@ -30,7 +30,7 @@ transform_files <- function(files,
   )
   communicate_summary(changed, max_char)
   communicate_warning(changed, transformers)
-  new_tibble(list(file = files, changed = changed))
+  new_styler_df(list(file = files, changed = changed))
 }
 
 #' Transform a file and output a customized message

R/ui-caching.R

@@ -75,15 +75,16 @@ cache_info <- function(cache_name = NULL, format = "both") {
   rlang::arg_match(format, c("tabular", "lucid", "both"))
   path_cache <- cache_find_path(cache_name)
   files <- list.files(path_cache, full.names = TRUE)
-  file_info <- file.info(files) %>%
-    as_tibble()
-  tbl <- tibble(
+  file_info <- file.info(files)
+
+  tbl <- styler_df(
     n = nrow(file_info),
     size = sum(file_info$size),
     last_modified = suppressWarnings(max(file_info$mtime)),
     created = file.info(path_cache)$ctime,
     location = path_cache,
-    activated = cache_is_activated(cache_name)
+    activated = cache_is_activated(cache_name),
+    stringsAsFactors = FALSE
   )
   if (any(c("lucid", "both") == format)) {
     cat(

R/ui-styling.R

@@ -1,5 +1,5 @@
 #' @keywords api
-#' @import tibble
+#' @importFrom tibble tribble
 #' @importFrom magrittr %>%
 NULL
 

R/utils.R

@@ -4,6 +4,19 @@ line_col_names <- function() {
   c("line1", "line2", "col1", "col2")
 }
 
+#' Wrapper functions to encapsulate data frame creation
+#' @keywords internal
+#' @noRd
+styler_df <- function(..., .size = NULL, .name_repair = "minimal") {
+  vctrs::data_frame(..., .size = .size, .name_repair = .name_repair)
+}
+
+#' @keywords internal
+#' @noRd
+new_styler_df <- function(x) {
+  vctrs::new_data_frame(x)
+}
+
 #' Ensure there is one (and only one) blank line at the end of a vector
 #' @examples
 #' styler:::ensure_last_n_empty("")
@@ -85,7 +98,7 @@ option_read <- function(x, default = NULL, error_if_not_found = TRUE) {
   }
 }
 
-
+#' @keywords internal
 unwhich <- function(x, length) {
   x_ <- rep(FALSE, length)
   x_[x] <- TRUE

inst/WORDLIST

@@ -271,6 +271,7 @@ upsetjs
 usethis
 utf
 Uwe
+vctrs
 VignetteBuilder
 Visit'em
 walthert

tests/testthat/_snaps/cache-with-r-cache.md

@@ -3,28 +3,22 @@
     Code
       cache_info[, c("n", "size", "last_modified", "activated")]
     Output
-      # A tibble: 1 x 4
-            n  size last_modified activated
-        <int> <dbl> <dttm>        <lgl>    
-      1     0     0 -Inf -Inf     FALSE    
+        n size last_modified activated
+      1 0    0          -Inf     FALSE
 
 ---
 
     Code
       cache_info[, c("n", "size", "activated")]
     Output
-      # A tibble: 1 x 3
-            n  size activated
-        <int> <dbl> <lgl>    
-      1     1     0 TRUE     
+        n size activated
+      1 1    0      TRUE
 
 ---
 
     Code
       cache_info[, c("n", "size", "activated")]
     Output
-      # A tibble: 1 x 3
-            n  size activated
-        <int> <dbl> <lgl>    
-      1     2     0 TRUE     
+        n size activated
+      1 2    0      TRUE
 

tests/testthat/roxygen-examples-complete/12-dontshow-dontrun-donttest-in.R

@@ -16,7 +16,7 @@
 #' }
 #' }
 #' set_line_break_before_curly_opening_style <- function() {
-#'   create_style_guide(line_break = tibble::lst(set_line_break_before_curly_opening))
+#'   create_style_guide(line_break = list(set_line_break_before_curly_opening))
 #' }
 #' @examples
 #' \dontrun{
@@ -32,9 +32,9 @@ create_style_guide <- function(initialize = default_style_guide_attributes,
                                indention = NULL,
                                use_raw_indention = FALSE,
                                reindention = tidyverse_reindention()) {
-  lst(
+  list(
     # transformer functions
-    initialize = lst(initialize),
+    initialize = list(initialize),
     line_break,
     space,
     token,
@@ -63,7 +63,7 @@ create_style_guide <- function(initialize = default_style_guide_attributes,
 #' }
 #' }
 #' set_line_break_before_curly_opening_style <- function() {
-#' create_style_guide(line_break= tibble::lst(set_line_break_before_curly_opening))
+#' create_style_guide(line_break= list(set_line_break_before_curly_opening))
 #' }
 #' @examples
 #' \donttest{style_text("a <- function(x) { x }
@@ -78,9 +78,9 @@ create_style_guide <- function(initialize = default_style_guide_attributes,
                                indention = NULL,
                                use_raw_indention = FALSE,
                                reindention = tidyverse_reindention()) {
-  lst(
+  list(
     #transformer functions
-    initialize = lst(initialize),
+    initialize = list(initialize),
     line_break,
     space,
     token,