fix(rust): Refactor decompression checks and add support for decompressing JSON #18536
Conversation
This file contains 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
ohanf
force-pushed
the
json-csv-decompression
branch
from
3a52e66
to
e3df118
Compare
CodSpeed Performance ReportMerging #18536 will not alter performanceComparing Summary
|
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #18536 /- ##
==========================================
- Coverage 79.92% 79.92% -0.01%
==========================================
Files 1506 1506
Lines 203031 203030 -1
Branches 2891 2891
==========================================
- Hits 162280 162266 -14
- Misses 40201 40214 13
Partials 550 550 ☔ View full report in Codecov by Sentry. |
|
Thank you @ohanf. This looks like an improvement to me. Can you add a test on the python side, where we compress some json on the fly and decompress it. Please don't add any (compressed) data files. |
ohanf
force-pushed
the
json-csv-decompression
branch
from
e3df118
to
5dbb175
Compare
ohanf
requested review from
alexander-beedie,
MarcoGorelli and
reswqa
as code owners
|
I added some basic tests roughly based on the CSV ones, please let me know what else might be needed, thanks |
ritchie46
approved these changes
Sep 11, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
It looks like the bulk of the work for supporting on the fly decompression (#8323) was already implemented. However there appeared to be a couple of inconsistencies, some of which this PR attempts to address.
The first commit is refactoring the compression detection logic to only be performed in a single function, instead of checking prefix bytes in multiple places. It also moves
maybe_decompress_bytesto live with other compression related code.The second commit is a small change to add automatic decompression for JSON files. This essentially mirrors what is done for the Lazy JSON reader the CSV reader, adapted for use with
simd_jsonThere does seem to be some extra unification work that could be done around CSV and Lazy reader decompression, but I wanted to submit the feature without making too many wide-sweeping changes. Another point of divergence is that decompression is only done for JSON on the rust side, as when reading ND-JSON from Python, the bindings call into the Lazy API which already supported decompression.
As a final note, I wasn't sure if the docs should be updated, especially since they were not when this was originally implemented. The docs probably should mention that any compressed file, read via eager or lazy methods, will read the entire file into memory. The underlying reason being that the polars parsing functions want their readers to be
Seekwhichflate2does not support (see rust-lang/flate2-rs#310)