Gregory Szorc

We might be able to add some checks here. But doing this comprehensively is difficult to impossible. e.g. sometimes an application may want to abort an in-progress (de)compression operation while...

Out of curiosity, is pinning to an older version causing any hardship for you? I'm just trying to understand the end-user impact of pinning versions in `pyproject.toml`.

Bleh. Feel free to submit a PR that removes the version pinning from `pyproject.toml`. It's probably more trouble than it is worth since it is becoming apparent that the modern...

Commit f90e2bbcb800047cc971864162998a99d4c2673f changed the `pyproject.toml` versions to be constraints rather than specific versions. Does this address this issue?

I'll remove `wheel` as an explicit dependency since it is transitive. As for `setuptools`, I'm really not sure what to do. I want to use `pip install --config-settings=` since that...

Need to verify with supplied sample data this but knee jerk is it has something to do with zstd frames. Various APIs in this library stop reading at frame boundaries...

For the `ZstdDecompressionObj.decompress()` API, I consider the lack of reading across frames a bug. These `decompress()` APIs are supposed to read as much as possible in a single operation. Its...

Actually, reading the docs for `ZstdDecompressionObj`, we explicitly call out the single frame reading behavior: > Each instance is single use: once an input frame is decoded, `decompress()` can no...

> Overall, I don't understand the use of any API that only supports decoding a single frame, requires the caller to tell it how much to read, and doesn't tell...

I added support for `read_across_frames=True` to `ZstdDecompressionObj` in the `main` branch. If you have time to test and validate with urllib3, any feedback would be appreciated.