[FEA] Add the Spark 3.1.0 shim layer for GpuWindowInPandasExec #844
Labels
feature request
New feature or request
Comments
firestarman
added
feature request
This was referenced Sep 25, 2020
tgravescs
pushed a commit
to tgravescs/spark-rapids
that referenced
this issue
Nov 30, 2023
* Add `MapUtils.java` * Add clang-format style * Fix comment * Add empty files * Fix compile issue and update clang-format * Add Java test * Concatenate the input json strings * Misc * Misc * Print debug * Update Java test * Add more test * Implement several more computation * Add comments * Implement node-to-token-index map * Compute node range from node indices, not token indices * Extract node ranges for key-value of non-nested types * Add more pairs to node section * Get node ranges for nested nodes * Extract json key-value pairs * Extract parent node ids of keys * Compute offsets for the output lists * Fix offsets computation Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Print debug for the output Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * More efficient substring operation Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Update Java test Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Remove parameter Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Rewrite docs Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Rewrite for easier benchmark Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Extract out functions Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Refactor and cleanup Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Handle empty and nulls input rows Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Update Java test Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Cleanup headers Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Implement UTF-8 support Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Add Java test Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Fix error Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Move header into .cu file Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Update copyright headers Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Update function name Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Add `assert` Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Remove wrong comment Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Extract debug code into a separate header Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Simplify `output_size` computation Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Fix typo Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Cleanup unused variable Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Fix a bug Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Rename variable Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Print debug input when error Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Change the error message Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Optimize error report Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> * Change comment Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com> Signed-off-by: Nghia Truong <nghiatruong.vn@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Is your feature request related to a problem? Please describe.
Its parent class
WindowExecBase, an abstract class in Spark 3.0.x, becomes a trait in Spark 3.1.0.Describe the solution you'd like
So need to add a shim layer to support Spark 3.1.0
The text was updated successfully, but these errors were encountered: