feat: Add shared array and buffer to nanoarrow.h#864
Draft
paleolimbot wants to merge 9 commits intoapache:mainfrom
Draft
feat: Add shared array and buffer to nanoarrow.h#864paleolimbot wants to merge 9 commits intoapache:mainfrom
paleolimbot wants to merge 9 commits intoapache:mainfrom
Conversation
paleolimbot
added a commit
that referenced
this pull request
Apr 18, 2026
This PR uses the previous steps to output arrays (including the ArrowArrayStream reader). This lets us wire it in to all the tests as well. The main follow up is that this PR currently deep copies the dictionary for every batch that arrives, negating much of the point of dictionary encoding. This is a fairly self-contained change that I'll do separately: #864 I also added dictionary index validation while I was here! It is fairly compact (compared to the other code). Closes #845.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
For #861 (actual decoding of dictionary arrays), we really need shared arrays for this to be reasonable (e.g., to not deep copy each dictionary for each batch!).
This PR (1) moves shared buffers to the C library instead of the IPC extension, (2) implements a second kind of shared buffer, which is borrowed from an owned reference-counted array, and (3) implements a "clone" that mostly uses the second concept to explode an array into 100% reference counted buffers that we can clone.
I had hoped we could replace the R and Python versions of these but the logic is fuzzier there because once an array has been referenced we can't mess with it or any of its parents (or it will cause a crash). My mind explodes (and I get a lot of failing R tests cases) whenever I mess with that piece and so I'll try to deal with that a different day.
This still needs tests for the shared array move and clone.