Add document-format crate (#4234)

* Add document-format crate

* Adapt document-format to merged graph-storage and consolidate errors

* Fix resource write ordering, codec error masking, and export docs

* Remove document-format RFC; tracked in a dedicated PR

* Restructure document-format into gdd/persist/export/resource modules

Split the 963-line lib.rs by responsibility: persist path into persist.rs, resource I/O into resource.rs, export engine into export.rs. Rename the GddV1 layout struct to GddV1Layout and add a GddV1 = Gdd<GddV1Layout> alias.

* Feature-split document-format and graph-storage for minimal builds

* Address PR review: O(1) history-delta lookup and no-default-features test build

* Address PR review: propagate apply_hot_op persist error and assert resource hash

* Make decompression from archive streaming

* Share archive export body between export and export_to_bytes

* Review cleanup

* Export documents with un-retired hot ops losslessly

* Persist last_broadcast_rev in session.json, drop unused last_retired_at

* Move peer_id from manifest to session.json

* Rename gesture to interaction in document-format storage layer
This commit is contained in:
Dennis Kobert
2026-06-21 12:35:47 +00:00
committed by GitHub
parent 194c5b2222
commit de11d29d8e
23 changed files with 2569 additions and 58 deletions
+24 -7
View File
@@ -3,9 +3,11 @@
//! Each codec streams entries in both directions: writers wrap an `io::Write` sink, and
//! `deserialize` reads from any `io::Read + Seek` source and streams entries into any [`Container`].
#[cfg(any(feature = "xz", feature = "zip"))]
use crate::AsyncContainer;
#[cfg(any(feature = "zip", feature = "xz"))]
use crate::ContainerError;
use crate::{Container, Result};
use crate::Result;
use std::io::{Read, Seek, Write};
/// Hard cap on the total decompressed size a codec will materialize from one archive.
@@ -55,12 +57,23 @@ pub trait Archive {
/// Read entries from `source` and write each into `dest`, streaming so neither the full
/// archive nor the full container ever sits in memory at once.
fn open<R: Read + Seek, C: Container>(source: R, dest: &mut C) -> Result<()>;
fn open<R: Read + Seek, C: AsyncContainer>(source: R, dest: &mut C) -> Result<()>;
}
pub trait ArchiveWriter {
pub trait ArchiveWriter: Sized {
type Sink;
fn write_entry(&mut self, path: &str, bytes: &[u8]) -> Result<()>;
fn finish(self) -> Result<()>;
/// Finish the archive and return the underlying sink, for in-memory archives where the caller
/// wants the written bytes (e.g. `Cursor<Vec<u8>>`) back.
fn finish_into(self) -> Result<Self::Sink>;
/// Finish the archive, discarding the sink. For file-backed archives the bytes are already on disk.
fn finish(self) -> Result<()> {
self.finish_into()?;
Ok(())
}
}
/// Archive container formats distinguishable by their leading magic bytes.
@@ -86,12 +99,16 @@ impl ArchiveFormat {
/// Deserialize an archive into `dest`, auto-detecting the format from `bytes`' magic header.
/// Errors if the bytes are neither a recognized xz nor zip archive.
#[cfg(all(feature = "xz", feature = "zip"))]
pub fn open_auto<C: Container>(bytes: &[u8], dest: &mut C) -> Result<()> {
#[cfg(any(feature = "xz", feature = "zip"))]
pub fn open_auto<C: AsyncContainer>(bytes: &[u8], dest: &mut C) -> Result<()> {
let source = std::io::Cursor::new(bytes);
match ArchiveFormat::detect(bytes) {
#[cfg(feature = "xz")]
Some(ArchiveFormat::Xz) => Xz::open(source, dest),
#[cfg(feature = "zip")]
Some(ArchiveFormat::Zip) => Zip::open(source, dest),
None => Err(ContainerError::Codec("unrecognized archive format (not xz or zip)".into())),
#[allow(unreachable_patterns)]
Some(format) => Err(ContainerError::Codec(format!("tried to open {format:?}, but the binary was compiled without the feature enabled "))),
None => Err(ContainerError::Codec("unrecognized archive format (not xz or zip) ".into())),
}
}