[I] [C++][Parquet] Optimize TypedRecordReader::DelimitRecords for nested type [arrow]

Tue, 23 Apr 2024 20:40:29 -0700 


mapleFU opened a new issue, #41361:
URL: https://github.com/apache/arrow/issues/41361

   ### Describe the enhancement requested
   
   When read nested type( for example, `list<list<int>>` ) , from the 
flamegraph, the bottlenect is `DelimitRecords`. It scan the rep-levels and 
def-levels, and return the number of records/non-null-values to user.
   
   
   
![1eeab6336e834b7d210071e9a8c2525e](https://github.com/apache/arrow/assets/24351052/81588c99-5515-4846-bc8d-9cd8eb631ed1)
   
   The bottleneck of this is branching checking, we can optimize this.
   
   
   
   ### Component(s)
   
   C++


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to