(improvement) (Cython only) deserializers: use direct PyUnicode_DecodeUTF8/ASCII fr…

[mykaul]

…om C buffer pointer

Replace the two-step to_bytes(buf).decode('utf8') pattern in DesUTF8Type and DesAsciiType with direct CPython C API calls (PyUnicode_DecodeUTF8 and PyUnicode_DecodeASCII). This eliminates an intermediate bytes object allocation per text cell — the old code created a Python bytes object from the C buffer pointer via to_bytes(buf), then immediately decoded it to str and discarded the bytes.

Text (UTF8Type/VarcharType) is the most common CQL column type, so this optimization applies to the majority of cells in typical workloads.

Benchmark results (Cython row parsing pipeline, median times):

Scenario	Before (original)	After (direct decode)	Speedup
UTF8 1row x 1col short (11B)	565 ns	454 ns	1.24x
UTF8 1row x 10col short	1,594 ns	1,023 ns	1.56x
UTF8 100rows x 5col medium	61,396 ns	28,766 ns	2.13x
UTF8 1000rows x 5col medium	547,145 ns	290,361 ns	1.88x
UTF8 100rows x 5col long(200B)	57,940 ns	35,680 ns	1.62x
UTF8 100rows x 5col multibyte	125,149 ns	103,370 ns	1.21x
ASCII 100rows x 5col medium	41,608 ns	35,817 ns	1.16x
ASCII 1000rows x 5col medium	416,350 ns	374,341 ns	1.11x
Mixed 100rows 3text+2int	44,646 ns	31,189 ns	1.43x

All existing unit tests pass (62 type tests, 116 total across key suites).

Pre-review checklist

• I have split my patch into logically separate commits.
• All commit messages clearly explain what they change and why.
• I added relevant tests for new features and bug fixes.
• All commits compile, pass static checks and pass test.
• PR description sums up the changes and reasons why they should be introduced.
• I have provided docstrings for the public items that I want to introduce.
• I have adjusted the documentation in ./docs/source/.
• I added appropriate Fixes: annotations to PR description.

[Copilot]

Pull request overview

Optimizes text deserialization performance in the Cython row parser by decoding UTF-8/ASCII directly from the underlying buffer, and adds a benchmark module intended to measure the improvement.

Changes:

• Replace to_bytes(buf).decode(...) with PyUnicode_DecodeASCII/UTF8(buf.ptr, buf.size, NULL) in Cython deserializers.
• Add a new pytest-based benchmark suite for UTF-8/ASCII decode and pipeline parsing.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
cassandra/deserializers.pyx	Uses CPython unicode decode APIs to avoid intermediate bytes allocations during ASCII/UTF-8 deserialization.
benchmarks/test_utf8_decode_benchmark.py	Adds end-to-end and microbenchmarks plus correctness checks for the new decode approach.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

This PR optimizes Cython-based UTF-8 and ASCII text deserialization by decoding directly from the underlying C buffer (avoiding an intermediate bytes allocation), and adds correctness tests plus a benchmark to validate/measure the change.

Changes:

• Switch DesUTF8Type/DesAsciiType to use PyUnicode_DecodeUTF8 / PyUnicode_DecodeASCII on Buffer.ptr/Buffer.size.
• Add unit tests covering common and edge-case text decode scenarios (empty, ASCII, multibyte, long strings, NULLs, multiple rows).
• Add a pytest-benchmark benchmark file to compare end-to-end Cython row parsing before/after the optimization.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
cassandra/deserializers.pyx	Replace to_bytes(...).decode(...) with direct CPython Unicode decoding calls for UTF-8/ASCII deserializers.
tests/unit/cython/test_deserializers.py	New correctness-focused unit tests for the optimized Cython decoding path.
benchmarks/utf8_decode_benchmark.py	New benchmark suite to measure the performance impact of the UTF-8/ASCII decode optimization.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.