apacheGH-18487: [R] Read Text (CSV/JSON) from character vector (apach…

…e#33968)

### Rationale for this change

Allows literal strings to be read directly through the `I()` function in the same way as the `readr::read_csv()` function.
This is useful for checking behavior without the need to create temporary files.

```r
> read_csv_arrow(I("x,y\n1,2\n3,4"))
# A tibble: 2 × 2
      x     y
  <int> <int>
1     1     2
2     3     4
```

```r
> read_csv_arrow(I(c(
        "x,y
        1,2
        3,4"
    )))
# A tibble: 2 × 2
      x     y
  <int> <int>
1     1     2
2     3     4
```

```r
> read_csv_arrow(I(c("x,y", "1,2", "3,4")))
# A tibble: 2 × 2
      x     y
  <int> <int>
1     1     2
2     3     4
```

### What changes are included in this PR?

In `read_csv_arrow` and `read_json_arrow`, if the first argument `file` inherits `AsIs` class, `file` is now interpreted as literal data.
This is consistent with the behavior of `readr::read_csv()`, which is widely used to read text files as data frames.

This is a breaking change; the behavior of wrapping a path as a string with `I()` is changed.

For example

#### readr::read_csv

```r
> readr::read_csv(I(readr::readr_example("mtcars.csv")))
Rows: 0 Columns: 1
── Column specification ────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr (1): /usr/local/lib/R/site-library/readr/extdata/mtcars.csv

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
# A tibble: 0 × 1
# … with 1 variable:
#   /usr/local/lib/R/site-library/readr/extdata/mtcars.csv <chr>
# ℹ Use `colnames()` to see all variable names
```
#### arrow 10.01's arrow::read_csv_arrow

```r
> arrow::read_csv_arrow(I(readr::readr_example("mtcars.csv")))
# A tibble: 32 × 11
     mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
   <dbl> <int> <dbl> <int> <dbl> <dbl> <dbl> <int> <int> <int> <int>
 1  21       6  160    110  3.9   2.62  16.5     0     1     4     4
 2  21       6  160    110  3.9   2.88  17.0     0     1     4     4
 3  22.8     4  108     93  3.85  2.32  18.6     1     1     4     1
 4  21.4     6  258    110  3.08  3.22  19.4     1     0     3     1
 5  18.7     8  360    175  3.15  3.44  17.0     0     0     3     2
 6  18.1     6  225    105  2.76  3.46  20.2     1     0     3     1
 7  14.3     8  360    245  3.21  3.57  15.8     0     0     3     4
 8  24.4     4  147.    62  3.69  3.19  20       1     0     4     2
 9  22.8     4  141.    95  3.92  3.15  22.9     1     0     4     2
10  19.2     6  168.   123  3.92  3.44  18.3     1     0     4     4
# … with 22 more rows
# ℹ Use `print(n = ...)` to see more rows
```

#### This PR's arrow::read_csv_arrow

```r
> arrow::read_csv_arrow(I(readr::readr_example("mtcars.csv")))
Error:
! Invalid: CSV parse error: Empty CSV file or block: cannot infer number of columns
Run `rlang::last_error()` to see where the error occurred.
```
* Closes: apache#18487

Lead-authored-by: SHIMA Tatsuya <[email protected]>
Co-authored-by: eitsupi <[email protected]>
Co-authored-by: Nic Crane <[email protected]>
Signed-off-by: Nic Crane <[email protected]>

r/R/csv.R

@@ -77,12 +77,15 @@
 #' `col_names`, and the CSV file has a header row that would otherwise be used
 #' to idenfity column names, you'll need to add `skip = 1` to skip that row.
 #'
-#' @param file A character file name or URI, `raw` vector, an Arrow input stream,
-#' or a `FileSystem` with path (`SubTreeFileSystem`).
+#' @param file A character file name or URI, literal data (either a single string or a [raw] vector),
+#' an Arrow input stream, or a `FileSystem` with path (`SubTreeFileSystem`).
+#'
 #' If a file name, a memory-mapped Arrow [InputStream] will be opened and
 #' closed when finished; compression will be detected from the file extension
 #' and handled automatically. If an input stream is provided, it will be left
 #' open.
+#'
+#' To be recognised as literal data, the input must be wrapped with `I()`.
 #' @param delim Single character used to separate fields within a record.
 #' @param quote Single character used to quote strings.
 #' @param escape_double Does the file escape quotes by doubling them?
@@ -154,6 +157,10 @@
 #'   tf,
 #'   col_types = schema(x = timestamp(unit = "us", timezone = "UTC"))
 #' )
+#'
+#' # Read directly from strings with `I()`
+#' read_csv_arrow(I("x,y\n1,2\n3,4"))
+#' read_delim_arrow(I(c("x y", "1 2", "3 4")), delim = " ")
 read_delim_arrow <- function(file,
                              delim = ",",
                              quote = '"',
@@ -198,6 +205,15 @@ read_delim_arrow <- function(file,
     )
   }
 
+  if (inherits(file, "AsIs")) {
+    if (is.raw(file)) {
+      # If a raw vector is wrapped by `I()`, we need to unclass the `AsIs` class to read the raw vector.
+      file <- unclass(file)
+    } else {
+      file <- charToRaw(paste(file, collapse = "\n"))
+    }
+  }
+
   if (!inherits(file, "InputStream")) {
     compression <- detect_compression(file)
     file <- make_readable_file(file)

r/R/json.R

@@ -48,12 +48,24 @@
 #'     { "hello": 3.25, "world": null }
 #'     { "hello": 0.0, "world": true, "yo": null }
 #'   ', tf, useBytes = TRUE)
+#'
 #' read_json_arrow(tf)
+#'
+#' # Read directly from strings with `I()`
+#' read_json_arrow(I(c('{"x": 1, "y": 2}', '{"x": 3, "y": 4}')))
 read_json_arrow <- function(file,
                             col_select = NULL,
                             as_data_frame = TRUE,
                             schema = NULL,
                             ...) {
+  if (inherits(file, "AsIs")) {
+    if (is.raw(file)) {
+      file <- unclass(file)
+    } else {
+      file <- charToRaw(paste(file, collapse = "\n"))
+    }
+  }
+
   if (!inherits(file, "InputStream")) {
     compression <- detect_compression(file)
     file <- make_readable_file(file)

r/tests/testthat/test-csv.R

@@ -687,3 +687,14 @@ test_that("CSV reading/parsing/convert options can be passed in as lists", {
 
   expect_equal(tab1, tab2)
 })
+
+test_that("Read literal data directly", {
+  expected <- tibble::tibble(x = c(1L, 3L), y = c(2L, 4L))
+
+  expect_identical(read_csv_arrow(I("x,y\n1,2\n3,4")), expected)
+  expect_identical(read_csv_arrow(I("x,y\r1,2\r3,4")), expected)
+  expect_identical(read_csv_arrow(I("x,y\n\r1,2\n\r3,4")), expected)
+  expect_identical(read_csv_arrow(charToRaw("x,y\n1,2\n3,4")), expected)
+  expect_identical(read_csv_arrow(I(charToRaw("x,y\n1,2\n3,4"))), expected)
+  expect_identical(read_csv_arrow(I(c("x,y", "1,2", "3,4"))), expected)
+})

r/tests/testthat/test-json.R

@@ -253,3 +253,12 @@ test_that("Can read json file with list<struct<T...>> nested columns (ARROW-7740
   expected <- tibble::tibble(a = c(list(one), list(one)))
   expect_equal(read_json_arrow(tf), expected, ignore_attr = TRUE)
 })
+
+test_that("Read literal data directly", {
+  expected <- tibble::tibble(x = c(1L, 3L), y = c(2L, 4L))
+
+  expect_identical(read_json_arrow(I('{"x": 1, "y": 2}\n{"x": 3, "y": 4}')), expected)
+  expect_identical(read_json_arrow(charToRaw('{"x": 1, "y": 2}\n{"x": 3, "y": 4}')), expected)
+  expect_identical(read_json_arrow(I(charToRaw('{"x": 1, "y": 2}\n{"x": 3, "y": 4}'))), expected)
+  expect_identical(read_json_arrow(I(c('{"x": 1, "y": 2}', '{"x": 3, "y": 4}'))), expected)
+})