docs: fix encoding spec correctness (follow-up to #898) #901
Conversation
…vectors\n- Document RawTreeMove encoding + usage\n- Remove incorrect serde_columnar row-count claims\n\nCo-authored-by: lody <[email protected]>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 05e2d17917
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| - [x] ContainerWrapper encoding/decoding | ||
| - [x] Change Block full parsing | ||
| - [x] ContainerArena encoding/decoding (serde_columnar) | ||
| - [x] ContainerArena encoding/decoding (postcard Vec) |
There was a problem hiding this comment.
The checklist now says ContainerArena encoding/decoding (postcard Vec), but the implementation uses serde_columnar::to_vec / serde_columnar::from_bytes in crates/loro-internal/src/encoding/arena.rs (lines 94–101). Anyone implementing a decoder from this spec will build the wrong format and fail to parse ContainerArena bytes. The checklist should continue to reference serde_columnar rather than postcard.
Useful? React with 👍 / 👎.
Follow-up fixes to the encoding docs discovered while reviewing #898.
Changes:
docs/encoding-xxhash32.md.RawTreeMovewire format + clarify ValueKind tag 7/16 indocs/encoding.md.docs/encoding-container-states.md.Note:
decode_block_range()incrates/loro-internal/src/oplog/change_store/block_encode.rsappears inconsistent with the actual block encoding, but this PR keeps scope to docs only.