jorisvandenbossche commented on a change in pull request #11031:
URL: https://github.com/apache/arrow/pull/11031#discussion_r699280468
##########
File path: cpp/src/parquet/type_fwd.h
##########
@@ -19,8 +19,50 @@
namespace parquet {
+/// \brief Feature selection when writing Parquet files
+///
+/// `ParquetVersion::type` governs which data types are allowed and how they
+/// are represented. For example, uint32_t data will be written differently
+/// depending on this value (as INT64 for PARQUET_1_0, as UINT32 for other
+/// versions).
+///
+/// However, some features - such as compression algorithms, encryption,
+/// or the improved "v2" data page format - must be enabled separately in
+/// ArrowWriterProperties.
struct ParquetVersion {
- enum type { PARQUET_1_0, PARQUET_2_0 };
+ enum type : int {
+ /// Enable only pre-2.0 Parquet format features when writing
+ ///
+ /// This setting is useful for maximum compatibility with legacy readers.
+ PARQUET_1_0,
Review comment:
TIMESTAMP_MILLIS is actually only from 2.2, so "pre 2.2" doesn't help
that much ... (but, yeah, this is a mess :))
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
