rto-graph 0.0.4

//! A thin `gix` wrapper exposing exactly the git facts the sync engine needs:
//! the HEAD tree id, the blobs in that tree, and blob contents. Kept small so
//! all `gix` coupling lives in one place.

use std::path::Path;

/// A blob in a tree: its repository-relative path and hex object id.
#[derive(Debug, Clone, PartialEq, Eq)]
pub struct BlobRef {
    /// Repository-relative path (forward-slash separated).
    pub path: String,
    /// Hex-encoded git blob object id.
    pub oid: String,
}

/// Errors raised while reading from a git repository.
#[derive(Debug, thiserror::Error)]
pub enum GitError {
    /// A `gix` operation failed (message preserved).
    #[error("git error: {0}")]
    Git(String),
    /// A tree entry path was not valid UTF-8.
    #[error("non-utf8 path in tree: {0:?}")]
    NonUtf8Path(Vec<u8>),
}

fn ge<E: std::fmt::Display>(e: E) -> GitError {
    GitError::Git(e.to_string())
}

/// A discovered git repository.
pub struct Repo {
    inner: gix::Repository,
}

impl Repo {
    /// Discover the repository containing `path` (walking upwards to the `.git`).
    ///
    /// # Errors
    /// Returns [`GitError::Git`] if no repository is found or it cannot be opened.
    pub fn discover(path: &Path) -> Result<Self, GitError> {
        Ok(Self {
            inner: gix::discover(path).map_err(ge)?,
        })
    }

    /// The repository's *common* git directory. The cache lives under here so it
    /// is shared across linked worktrees (which each have their own git dir).
    #[must_use]
    pub fn common_dir(&self) -> &Path {
        self.inner.common_dir()
    }

    /// This worktree's git directory (per-worktree; the graph DB lives here).
    #[must_use]
    pub fn git_dir(&self) -> &Path {
        self.inner.git_dir()
    }

    /// Hex object id of the tree at `HEAD`.
    ///
    /// # Errors
    /// Returns [`GitError::Git`] if `HEAD` cannot be resolved to a tree.
    pub fn head_tree_id(&self) -> Result<String, GitError> {
        let tree = self.inner.head_tree().map_err(ge)?;
        Ok(tree.id().to_hex().to_string())
    }

    /// Every blob reachable from the `HEAD` tree, with full paths.
    ///
    /// # Errors
    /// Returns [`GitError`] if the tree cannot be traversed or a path is not
    /// valid UTF-8.
    pub fn walk_blobs(&self) -> Result<Vec<BlobRef>, GitError> {
        let tree = self.inner.head_tree().map_err(ge)?;
        let mut recorder = gix::traverse::tree::Recorder::default();
        tree.traverse().breadthfirst(&mut recorder).map_err(ge)?;

        let mut out = Vec::new();
        for entry in recorder.records {
            if !entry.mode.is_blob() {
                continue;
            }
            let path = String::from_utf8(entry.filepath.into())
                .map_err(|e| GitError::NonUtf8Path(e.into_bytes()))?;
            out.push(BlobRef {
                path,
                oid: entry.oid.to_hex().to_string(),
            });
        }
        Ok(out)
    }

    /// Read the bytes of the blob with hex object id `oid`.
    ///
    /// # Errors
    /// Returns [`GitError::Git`] if the id is malformed or the object is absent.
    pub fn read_blob(&self, oid: &str) -> Result<Vec<u8>, GitError> {
        let id = gix::ObjectId::from_hex(oid.as_bytes()).map_err(ge)?;
        // `detach()` moves the owned data out without cloning; `Object` itself
        // implements `Drop`, so the bare field cannot be moved out directly.
        Ok(self.inner.find_object(id).map_err(ge)?.detach().data)
    }
}