[
https://issues.apache.org/jira/browse/PARQUET-1092?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16293027#comment-16293027
]
ASF GitHub Bot commented on PARQUET-1092:
-----------------------------------------
wesm commented on a change in pull request #426: PARQUET-1092: Support writing
chunked arrow::Table columns
URL: https://github.com/apache/parquet-cpp/pull/426#discussion_r157273275
##########
File path: src/parquet/arrow/writer.cc
##########
@@ -106,21 +106,19 @@ class LevelBuilder {
NOT_IMPLEMENTED_VISIT(Struct)
NOT_IMPLEMENTED_VISIT(Union)
NOT_IMPLEMENTED_VISIT(Dictionary)
- NOT_IMPLEMENTED_VISIT(Interval)
Status GenerateLevels(const Array& array, const std::shared_ptr<Field>&
field,
- int64_t* values_offset, ::arrow::Type::type*
values_type,
- int64_t* num_values, int64_t* num_levels,
- std::shared_ptr<Buffer>* def_levels,
- std::shared_ptr<Buffer>* rep_levels,
+ int64_t* values_offset, int64_t* num_values, int64_t*
num_levels,
+ const std::shared_ptr<PoolBuffer>& def_levels_scratch,
Review comment:
In the nested case, the definition levels are generated using a builder.
Otherwise we can reuse the same pre-allocated memory for the levels. This could
all be refactored potentially to reuse a common shared buffer, maybe we can do
that later
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> [C++] Write Arrow tables with chunked columns
> ---------------------------------------------
>
> Key: PARQUET-1092
> URL: https://issues.apache.org/jira/browse/PARQUET-1092
> Project: Parquet
> Issue Type: Improvement
> Components: parquet-cpp
> Reporter: Wes McKinney
> Assignee: Wes McKinney
> Fix For: cpp-1.3.0
>
>
> Requires incoming patch in ARROW-232
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
