Skip to content

Split CopyContext implementation into modular include files#1759

Merged
oferchen merged 1 commit into
masterfrom
prepare-production-release-and-optimize-code
Oct 31, 2025
Merged

Split CopyContext implementation into modular include files#1759
oferchen merged 1 commit into
masterfrom
prepare-production-release-and-optimize-code

Conversation

@oferchen
Copy link
Copy Markdown
Owner

@oferchen oferchen commented Oct 31, 2025

Summary

  • split the CopyContext inherent impl into include! fragments under context_impl/ to keep each file within hygiene limits
  • update context.rs to reference the new parts while preserving the existing preamble/postamble layout
  • document the include-based layout for the local copy context in internal docs

Testing

  • cargo fmt
  • cargo check

https://chatgpt.com/codex/tasks/task_e_6904b750a80c8323a2833cdc5725f364

@oferchen oferchen merged commit e3618ea into master Oct 31, 2025
@oferchen oferchen deleted the prepare-production-release-and-optimize-code branch October 31, 2025 13:53
oferchen added a commit that referenced this pull request May 14, 2026
…e-buffered (#4033)

Task #1759 asked for double-buffering in the retained checksum pipeline.
The work is already done in `crates/checksums/src/pipelined/`: the reader
pre-allocates two buffers, seeds one into a recycle channel, and uses a
bounded `sync_channel(1)` so the I/O thread reads block N+1 while the
caller hashes block N. Functional parity with the serialized hasher is
covered by unit and proptest cases. This audit documents where the
implementation lives, the two-buffer invariant, the sync fallback, and
the existing test coverage so future contributors do not reintroduce a
parallel primitive.
oferchen added a commit that referenced this pull request May 18, 2026
…e-buffered (#4033)

Task #1759 asked for double-buffering in the retained checksum pipeline.
The work is already done in `crates/checksums/src/pipelined/`: the reader
pre-allocates two buffers, seeds one into a recycle channel, and uses a
bounded `sync_channel(1)` so the I/O thread reads block N+1 while the
caller hashes block N. Functional parity with the serialized hasher is
covered by unit and proptest cases. This audit documents where the
implementation lives, the two-buffer invariant, the sync fallback, and
the existing test coverage so future contributors do not reintroduce a
parallel primitive.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant