vroom breaking changes

• v1.4.0 \u2192 v1.5.0breaking

  New `vroom(show_col_types=)` argument to more simply control when column types are shown.

• v1.4.0 \u2192 v1.5.0breaking

  `vroom()`, `vroom_fwf()` and `vroom_lines()` now support multi-byte encodings such as UTF-16 and UTF-32 by converting these files to UTF-8 under the hood (#138)

• v1.4.0 \u2192 v1.5.0breaking

  `vroom()` now supports skipping comments and blank lines within data, not just at the start of the file (#294, #302)

• v1.4.0 \u2192 v1.5.0breaking

  `vroom()` now uses the tzdb package when parsing date-times (@DavisVaughan, #273)

• v1.4.0 \u2192 v1.5.0breaking

  `vroom()` now emits a warning of class `vroom_parse_issue` if there are non-fatal parsing issues.

• v1.3.2 \u2192 v1.4.0breaking

  vroom now tracks indexing and parsing errors like readr. The first time an issue is encountered a warning will be signaled. A tibble of all found problems can be retrieved with `vroom::problems()`. (#247)

• v1.3.2 \u2192 v1.4.0breaking

  Data with newlines within quoted fields will now automatically revert to using a single thread and be properly read (#282)

• v1.3.2 \u2192 v1.4.0breaking

  NUL values in character data are now permitted, with a warning.

• v1.3.2 \u2192 v1.4.0breaking

  New `vroom_write_lines()` function to write a character vector to a file (#291)

• v1.3.2 \u2192 v1.4.0breaking

  `vroom_write()` gains a `eol=` parameter to specify the end of line character(s) to use. Use `vroom_write(eol = "\r\n")` to write a file with Windows style newlines (#263).

• v1.0.2 \u2192 v1.2.0breaking

  `vroom_altrep_opts()` and the argument `vroom(altrep_opts =)` have been renamed to `vroom_altrep()` and `altrep` respectively. The prior names have been deprecated.

Get this data programmatically \u2014 free, no authentication.