apacheGH-38916: [R] Simplify dataset and table print output (apache#3…

…8917) ### Rationale for this change When printing objects with data with lots of rows, the output is long and unwieldy. ### What changes are included in this PR? * Truncates long schema print output and adds the number of columns to dataset print output. * Add number of columns to output so it's clear how many there are in total ### Are these changes tested? Yes ### Are there any user-facing changes? Yes Before: ``` r library(arrow) x <- tibble::tibble(!!!letters, .rows = 5) InMemoryDataset$create(x) #> InMemoryDataset #> "a": string #> "b": string #> "c": string #> "d": string #> "e": string #> "f": string #> "g": string #> "h": string #> "i": string #> "j": string #> "k": string #> "l": string #> "m": string #> "n": string #> "o": string #> "p": string #> "q": string #> "r": string #> "s": string #> "t": string #> "u": string #> "v": string #> "w": string #> "x": string #> "y": string #> "z": string arrow_table(x) #> Table #> 5 rows x 26 columns #> $"a" <string> #> $"b" <string> #> $"c" <string> #> $"d" <string> #> $"e" <string> #> $"f" <string> #> $"g" <string> #> $"h" <string> #> $"i" <string> #> $"j" <string> #> $"k" <string> #> $"l" <string> #> $"m" <string> #> $"n" <string> #> $"o" <string> #> $"p" <string> #> $"q" <string> #> $"r" <string> #> $"s" <string> #> $"t" <string> #> $"u" <string> #> $"v" <string> #> $"w" <string> #> $"x" <string> #> $"y" <string> #> $"z" <string> record_batch(x) #> RecordBatch #> 5 rows x 26 columns #> $"a" <string> #> $"b" <string> #> $"c" <string> #> $"d" <string> #> $"e" <string> #> $"f" <string> #> $"g" <string> #> $"h" <string> #> $"i" <string> #> $"j" <string> #> $"k" <string> #> $"l" <string> #> $"m" <string> #> $"n" <string> #> $"o" <string> #> $"p" <string> #> $"q" <string> #> $"r" <string> #> $"s" <string> #> $"t" <string> #> $"u" <string> #> $"v" <string> #> $"w" <string> #> $"x" <string> #> $"y" <string> #> $"z" <string> ``` After: ``` r library(arrow) x <- tibble::tibble(!!!letters, .rows = 5) InMemoryDataset$create(x) #> InMemoryDataset #> 26 columns #> "a": string #> "b": string #> "c": string #> "d": string #> "e": string #> "f": string #> "g": string #> "h": string #> "i": string #> "j": string #> "k": string #> "l": string #> "m": string #> "n": string #> "o": string #> "p": string #> "q": string #> "r": string #> "s": string #> "t": string #> ... #> Use `schema()` to see entire schema arrow_table(x) #> Table #> 5 rows x 26 columns #> $"a" <string> #> $"b" <string> #> $"c" <string> #> $"d" <string> #> $"e" <string> #> $"f" <string> #> $"g" <string> #> $"h" <string> #> $"i" <string> #> $"j" <string> #> $"k" <string> #> $"l" <string> #> $"m" <string> #> $"n" <string> #> $"o" <string> #> $"p" <string> #> $"q" <string> #> $"r" <string> #> $"s" <string> #> $"t" <string> #> ... #> Use `schema()` to see entire schema record_batch(x) #> RecordBatch #> 5 rows x 26 columns #> $"a" <string> #> $"b" <string> #> $"c" <string> #> $"d" <string> #> $"e" <string> #> $"f" <string> #> $"g" <string> #> $"h" <string> #> $"i" <string> #> $"j" <string> #> $"k" <string> #> $"l" <string> #> $"m" <string> #> $"n" <string> #> $"o" <string> #> $"p" <string> #> $"q" <string> #> $"r" <string> #> $"s" <string> #> $"t" <string> #> ... #> Use `schema()` to see entire schema ``` * Closes: apache#38916 Lead-authored-by: Nic Crane <[email protected]> Co-authored-by: Bryce Mecum <[email protected]> Signed-off-by: Nic Crane <[email protected]>
vibhatha · Mar 13, 2024 · ac1708c · ac1708c
1 parent bd3fab4
commit ac1708c
Show file tree

Hide file tree

Showing 8 changed files with 47 additions and 8 deletions.
diff --git a/r/R/arrow-tabular.R b/r/R/arrow-tabular.R
@@ -23,7 +23,7 @@ ArrowTabular <- R6Class("ArrowTabular",
   inherit = ArrowObject,
   public = list(
     ToString = function() {
-      sch <- unlist(strsplit(self$schema$ToString(), "\n"))
+      sch <- unlist(strsplit(self$schema$ToString(truncate = TRUE), "\n"))
       sch <- sub("(.*): (.*)", "$\\1 <\\2>", sch)
       dims <- sprintf("%s rows x %s columns", self$num_rows, self$num_columns)
       paste(c(dims, sch), collapse = "\n")

diff --git a/r/R/dataset.R b/r/R/dataset.R
@@ -426,7 +426,7 @@ Dataset <- R6Class("Dataset",
     # Start a new scan of the data
     # @return A [ScannerBuilder]
     NewScan = function() dataset___Dataset__NewScan(self),
-    ToString = function() self$schema$ToString(),
+    ToString = function() format_schema(self),
     WithSchema = function(schema) {
       assert_is(schema, "Schema")
       dataset___Dataset__ReplaceSchema(self, schema)

diff --git a/r/R/record-batch-reader.R b/r/R/record-batch-reader.R
@@ -100,7 +100,7 @@ RecordBatchReader <- R6Class("RecordBatchReader",
     read_table = function() Table__from_RecordBatchReader(self),
     Close = function() RecordBatchReader__Close(self),
     export_to_c = function(stream_ptr) ExportRecordBatchReader(self, stream_ptr),
-    ToString = function() self$schema$ToString(),
+    ToString = function() format_schema(self),
     .unsafe_delete = function() {
       RecordBatchReader__UnsafeDelete(self)
       super$.unsafe_delete()

diff --git a/r/R/schema.R b/r/R/schema.R
@@ -81,8 +81,8 @@
 Schema <- R6Class("Schema",
   inherit = ArrowObject,
   public = list(
-    ToString = function() {
-      fields <- print_schema_fields(self)
+    ToString = function(truncate = FALSE) {
+      fields <- print_schema_fields(self, truncate)
       if (self$HasMetadata) {
         fields <- paste0(fields, "\n\nSee $metadata for additional Schema metadata")
       }
@@ -224,9 +224,19 @@ prepare_key_value_metadata <- function(metadata) {
   map_chr(metadata, as.character)
 }
 
-print_schema_fields <- function(s) {
-  # Alternative to Schema__ToString that doesn't print metadata
-  paste(map_chr(s$fields, ~ .$ToString()), collapse = "\n")
+# Alternative to Schema__ToString that doesn't print metadata
+print_schema_fields <- function(s, truncate = FALSE, max_fields = 20L) {
+  assert_that(max_fields > 0)
+  num_fields <- length(s$fields)
+  if (truncate && num_fields > max_fields) {
+    fields_out <- paste(map_chr(s$fields[seq_len(max_fields)], ~ .$ToString()), collapse = "\n")
+    fields_out <- paste0(fields_out, "\n...\n")
+    fields_out <- paste0(fields_out, num_fields - max_fields, " more columns\n")
+    fields_out <- paste0(fields_out, "Use `schema()` to see entire schema")
+  } else {
+    fields_out <- paste(map_chr(s$fields, ~ .$ToString()), collapse = "\n")
+  }
+  fields_out
 }
 
 #' Create a schema or extract one from an object.
@@ -460,3 +470,14 @@ as.data.frame.Schema <- function(x, row.names = NULL, optional = FALSE, ...) {
 
 #' @export
 `names<-.Schema` <- function(x, value) x$WithNames(value)
+
+#' Get a string representing a Dataset or RecordBatchReader object's schema
+#' @param obj a Dataset or RecordBatchReader
+#' @return A string containing a formatted representation of the schema of `obj`
+#' @keywords internal
+format_schema <- function(obj) {
+  assert_is(obj, c("Dataset", "RecordBatchReader"))
+  n_fields_out <- paste0(length(obj$schema$fields), " columns", "\n")
+  schema <- obj$schema$ToString(truncate = TRUE)
+  paste0(n_fields_out, schema)
+}
diff --git a/r/tests/testthat/_snaps/dplyr-glimpse.md b/r/tests/testthat/_snaps/dplyr-glimpse.md
@@ -91,6 +91,7 @@
       Cannot glimpse() data from a RecordBatchReader because it can only be read one time; call `as_arrow_table()` to consume it first.
     Output
       RecordBatchReader
+      7 columns
       int: int32
       dbl: double
       dbl2: double

diff --git a/r/tests/testthat/test-dataset.R b/r/tests/testthat/test-dataset.R
@@ -910,6 +910,7 @@ test_that("Dataset and query print methods", {
     print(ds),
     paste(
       "FileSystemDataset with 2 Parquet files",
+      "8 columns",
       "int: int32",
       "dbl: double",
       "lgl: bool",

diff --git a/r/tests/testthat/test-record-batch-reader.R b/r/tests/testthat/test-record-batch-reader.R
@@ -179,6 +179,7 @@ test_that("RBR methods", {
   expect_output(
     print(reader),
     "RecordBatchStreamReader
+2 columns
 x: int32
 y: string"
   )

diff --git a/r/tests/testthat/test-schema.R b/r/tests/testthat/test-schema.R
@@ -315,5 +315,20 @@ test_that("schema extraction", {
 
   adq <- as_adq(example_data)
   expect_equal(schema(adq), adq$.data$schema)
+})
+
+test_that("schema print truncation", {
+  tbl <- arrow_table(example_data)
+  out <- print_schema_fields(schema(tbl), truncate = TRUE, max_fields = 1)
+  expect_output(
+    cat(out),
+    "int: int32\n...\n6 more columns\nUse `schema()` to see entire schema",
+    fixed = TRUE
+  )
+
+  expect_error(
+    print_schema_fields(schema(tbl), truncate = TRUE, max_fields = 0),
+    regexp = "max_fields not greater than 0"
+  )
 
 })