Skip to content

Commit

Permalink
NEWS for writing min/max
Browse files Browse the repository at this point in the history
[CI skip]
  • Loading branch information
gaborcsardi committed Sep 22, 2024
1 parent 3f96d6c commit e43fc37
Showing 1 changed file with 6 additions and 0 deletions.
6 changes: 6 additions & 0 deletions NEWS.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,8 @@
`read_parquet_schema()` or the new `infer_parquet_schema()` function
instead.

* Other improvements:

- The new `parquet_schema()` function creates a Parquet schema from
scratch. You can use this schema as the new `schema` argument of
`write_parquet()`, to specify how the columns of a data frame should
Expand All @@ -25,6 +27,10 @@
at most 10 million rows into a single row group. You can choose the
row groups manually with the `row_groups` argument.

- `write_parquet()` now writes minimum and maximum values per row group
for most types. See `?parquet_options()` for turning this off. It also
writes out the number of non-missing values.

- Newly supported type conversions in `write_parquet()` via the
schema argument:

Expand Down

0 comments on commit e43fc37

Please sign in to comment.