zstdfile: ensure we do not read more than size / IO_BLOCK_SIZE

[joelynch]

In the previous implementation, _ZtsdFileReader.read could produce output of arbirary size. This can cause memory spikes while decompressing a file. Instead, we should use a ZstdDecompressor.stream_reader which decompresses incrementally into a fixed size output buffer.

About this change - What it does

Resolves: #xxxxx

Why this way

[Copilot]

Pull request overview

This PR addresses memory spikes during zstd decompression by refactoring the _ZtsdFileReader class to use ZstdDecompressor.stream_reader() instead of decompressobj(). The new approach enables incremental decompression with a bounded output buffer, ensuring that read() operations respect the requested size parameter and do not produce arbitrarily large outputs.

• Replaced decompressobj() with stream_reader() for controlled, incremental decompression
• Modified the read() method to honor size parameter and limit output to requested bytes
• Added comprehensive test coverage for compression and decompression with size-bounded reads

Reviewed changes

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

File	Description
rohmu/zstdfile.py	Refactored _ZtsdFileReader to use stream_reader() for bounded decompression and updated read() method to respect size parameter
test/test_zstdfile.py	Added new test case validating compression/decompression with size-limited reads to verify the fix

---

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