Tidyverse Vroom Versions Save

It is now possible (again?) to read from a list of connections (@bairdj, #514).

Internal change for compatibility with cpp11 >= 0.4.6 (@DavisVaughan, #512).

str() now works in a colorized context in the presence of a column of class integer64, i.e. parsed with col_big_integer() (@bart1, #477).

The embedded implementation of the Grisu algorithm for printing floating point numbers now uses snprintf() instead of sprintf() and likewise for vroom's own code (@jeroen, #480).

vroom(col_select=) now handles column selection by numeric position when id column is provided (#455).

vroom(id = "path", col_select = a:c) is treated like vroom(id = "path", col_select = c(path, a:c)). If an id column is provided, it is automatically included in the output (#416).

vroom_write(append = TRUE) does not modify an existing file when appending an empty data frame. In particular, it does not overwrite (delete) the existing contents of that file (https://github.com/tidyverse/readr/issues/1408, #451).

vroom::problems() now defaults to .Last.value for its primary input, similar to how readr::problems() works (#443).

The warning that indicates the existence of parsing problems has been improved, which should make it easier for the user to follow-up (https://github.com/tidyverse/readr/issues/1322).

vroom() reads more reliably from filepaths containing non-ascii characters, in a non-UTF-8 locale (#394, #438).

vroom_format() and vroom_write() only quote values that contain a delimiter, quote, or newline. Specifically values that are equal to the na string (or that start with it) are no longer quoted (#426).

Fixed segfault when reading in multiple files and the first file has only a header row of column names, but subsequent files have at least one row (#430).

Fixed segfault when vroom_format() is given an empty data frame (#425)

Fixed a segfault that could occur when the final field of the final line is missing and the file also does not end in a newline (#429).

Fixed recursive garbage collection error that could occur during vroom_write() when output_column() generates an ALTREP vector (#389).

vroom_progress() uses rlang::is_interactive() instead of base::interactive().

col_factor(levels = NULL) honors the na strings of vroom() and its own include_na argument, as described in the docs, and now reproduces the behaviour of readr's first edition parser (#396).

Jenny Bryan is now the official maintainer.

Fix uninitialized bool detected by CRAN's UBSAN check (https://github.com/r-lib/vroom/pull/386)

Fix buffer overflow when trying to parse an integer field that is over 64 characters long (https://github.com/tidyverse/readr/issues/1326)

Fix subset indexing when indexes span a file boundary multiple times (#383)

vroom(col_select=) now works if col_names = FALSE as intended (#381)

vroom(n_max=) now correctly handles cases when reading from a connection and the file does not end with a newline (https://github.com/tidyverse/readr/issues/1321)

vroom() no longer issues a spurious warning when the parsing needs to be restarted due to the presence of embedded newlines (https://github.com/tidyverse/readr/issues/1313)

Fix performance issue when materializing subsetted vectors (#378)

vroom_format() now uses the same internal multi-threaded code as vroom_write(), improving its performance in most cases (#377)

vroom_fwf() no longer omits the last line if it does not end with a newline (https://github.com/tidyverse/readr/issues/1293)

Empty files or files with only a header line and no data no longer cause a crash if read with multiple files (https://github.com/tidyverse/readr/issues/1297)

Files with a header but no contents, or a empty file if col_names = FALSE no longer cause a hang when progress = TRUE (https://github.com/tidyverse/readr/issues/1297)

Commented lines with comments at the end of lines no longer hang R (https://github.com/tidyverse/readr/issues/1309)

Comment lines containing unpaired quotes are no longer treated as unterminated quotations (https://github.com/tidyverse/readr/issues/1307)

Values with only a Inf or NaN prefix but additional data afterwards, like Inform or no longer inappropriately guessed as doubles (https://github.com/tidyverse/readr/issues/1319)

Time types now support %h format to denote hour durations greater than 24, like readr (https://github.com/tidyverse/readr/issues/1312)

Fix performance issue when materializing subsetted vectors (#378)

vroom() now supports files with only carriage return newlines (\r). (#360, https://github.com/tidyverse/readr/issues/1236)

vroom() now parses single digit datetimes more consistently as readr has done (https://github.com/tidyverse/readr/issues/1276)

vroom() now parses Inf values as doubles (https://github.com/tidyverse/readr/issues/1283)

vroom() now parses NaN values as doubles (https://github.com/tidyverse/readr/issues/1277)

VROOM_CONNECTION_SIZE is now parsed as a double, which supports scientific notation (#364)

vroom() now works around specifying a \n as the delimiter (#365, https://github.com/tidyverse/dplyr/issues/5977)

vroom() no longer crashes if given a col_name and col_type both less than the number of columns (https://github.com/tidyverse/readr/issues/1271)

vroom() no longer hangs if given an empty value for locale(grouping_mark=) (https://github.com/tidyverse/readr/issues/1241)

Fix performance regression when guessing with large numbers of rows (https://github.com/tidyverse/readr/issues/1267)

vroom(col_types=) now accepts column type names like those accepted by utils::read.table. e.g. vroom::vroom(col_types = list(a = "integer", b = "double", c = "skip"))

vroom() now respects the quote parameter properly in the first two lines of the file (https://github.com/tidyverse/readr/issues/1262)

vroom_write() now always correctly writes its output including column names in UTF-8 (https://github.com/tidyverse/readr/issues/1242)

vroom_write() now creates an empty file when given a input without any columns (https://github.com/tidyverse/readr/issues/1234)

vroom(col_types=) now truncates the column types if the user passes too many types. (#355)

vroom() now always includes the last row when guessing (#352)

vroom(trim_ws = TRUE) now trims field content within quotes as well as without (#354). Previously vroom explicitly retained field content inside quotes regardless of the value of trim_ws.