refactor(tvix/castore): use Directory struct separate from proto one r/8484

This uses our own data type to deal with Directories in the castore model.

It makes some undesired states unrepresentable, removing the need for conversions and checking in various places:

 - In the protobuf, blake3 digests could have a wrong length, as proto doesn't know fixed-size fields. We now use `B3Digest`, which makes cloning cheaper, and removes the need to do size-checking everywhere.
 - In the protobuf, we had three different lists for `files`, `symlinks` and `directories`. This was mostly a protobuf size optimization, but made interacting with them a bit awkward. This has now been replaced with a list of enums, and convenience iterators to get various nodes, and add new ones.

Change-Id: I7b92691bb06d77ff3f58a5ccea94a22c16f84f04
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12057
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>

3 files changed, 52 insertions, 64 deletions

diff --git a/tvix/store/src/nar/import.rs b/tvix/store/src/nar/import.rs
index 32c2f4e58061..9f3d0e0d74ea 100644
--- a/tvix/store/src/nar/import.rs
+++ b/tvix/store/src/nar/import.rs
@@ -7,12 +7,11 @@ use tokio::{
 };
 use tvix_castore::{
     blobservice::BlobService,
-    directoryservice::DirectoryService,
+    directoryservice::{DirectoryService, NamedNode, Node},
     import::{
         blobs::{self, ConcurrentBlobUploader},
         ingest_entries, IngestionEntry, IngestionError,
     },
-    proto::{node::Node, NamedNode},
     PathBuf,
 };
 
@@ -99,7 +98,7 @@ where
     let (_, node) = try_join!(produce, consume)?;
 
     // remove the fake "root" name again
-    debug_assert_eq!(&node.get_name(), b"root");
+    debug_assert_eq!(&node.get_name()[..], b"root");
     Ok(node.rename("".into()))
 }
 
@@ -172,12 +171,13 @@ mod test {
     use rstest::*;
     use tokio_stream::StreamExt;
     use tvix_castore::blobservice::BlobService;
-    use tvix_castore::directoryservice::DirectoryService;
+    use tvix_castore::directoryservice::{
+        Directory, DirectoryNode, DirectoryService, FileNode, Node, SymlinkNode,
+    };
     use tvix_castore::fixtures::{
         DIRECTORY_COMPLICATED, DIRECTORY_WITH_KEEP, EMPTY_BLOB_DIGEST, HELLOWORLD_BLOB_CONTENTS,
         HELLOWORLD_BLOB_DIGEST,
     };
-    use tvix_castore::proto as castorepb;
 
     use crate::tests::fixtures::{
         blob_service, directory_service, NAR_CONTENTS_COMPLICATED, NAR_CONTENTS_HELLOWORLD,
@@ -199,10 +199,13 @@ mod test {
         .expect("must parse");
 
         assert_eq!(
-            castorepb::node::Node::Symlink(castorepb::SymlinkNode {
-                name: "".into(), // name must be empty
-                target: "/nix/store/somewhereelse".into(),
-            }),
+            Node::Symlink(
+                SymlinkNode::new(
+                    "".into(), // name must be empty
+                    "/nix/store/somewhereelse".into(),
+                )
+                .unwrap()
+            ),
             root_node
         );
     }
@@ -222,12 +225,15 @@ mod test {
         .expect("must parse");
 
         assert_eq!(
-            castorepb::node::Node::File(castorepb::FileNode {
-                name: "".into(), // name must be empty
-                digest: HELLOWORLD_BLOB_DIGEST.clone().into(),
-                size: HELLOWORLD_BLOB_CONTENTS.len() as u64,
-                executable: false,
-            }),
+            Node::File(
+                FileNode::new(
+                    "".into(), // name must be empty
+                    HELLOWORLD_BLOB_DIGEST.clone(),
+                    HELLOWORLD_BLOB_CONTENTS.len() as u64,
+                    false,
+                )
+                .unwrap()
+            ),
             root_node
         );
 
@@ -250,11 +256,14 @@ mod test {
         .expect("must parse");
 
         assert_eq!(
-            castorepb::node::Node::Directory(castorepb::DirectoryNode {
-                name: "".into(), // name must be empty
-                digest: DIRECTORY_COMPLICATED.digest().into(),
-                size: DIRECTORY_COMPLICATED.size(),
-            }),
+            Node::Directory(
+                DirectoryNode::new(
+                    "".into(), // name must be empty
+                    DIRECTORY_COMPLICATED.digest(),
+                    DIRECTORY_COMPLICATED.size(),
+                )
+                .unwrap()
+            ),
             root_node,
         );
 
@@ -262,7 +271,7 @@ mod test {
         assert!(blob_service.has(&EMPTY_BLOB_DIGEST).await.unwrap());
 
         // directoryservice must contain the directories, at least with get_recursive.
-        let resp: Result<Vec<castorepb::Directory>, _> = directory_service
+        let resp: Result<Vec<Directory>, _> = directory_service
             .get_recursive(&DIRECTORY_COMPLICATED.digest())
             .collect()
             .await;
diff --git a/tvix/store/src/nar/mod.rs b/tvix/store/src/nar/mod.rs
index 8cbb091f1aab..8a19f6bd6c47 100644
--- a/tvix/store/src/nar/mod.rs
+++ b/tvix/store/src/nar/mod.rs
@@ -8,16 +8,14 @@ pub use import::ingest_nar_and_hash;
 pub use renderer::calculate_size_and_sha256;
 pub use renderer::write_nar;
 pub use renderer::SimpleRenderer;
-use tvix_castore::proto as castorepb;
+use tvix_castore::directoryservice::Node;
 
 #[async_trait]
 pub trait NarCalculationService: Send + Sync {
     /// Return the nar size and nar sha256 digest for a given root node.
     /// This can be used to calculate NAR-based output paths.
-    async fn calculate_nar(
-        &self,
-        root_node: &castorepb::node::Node,
-    ) -> Result<(u64, [u8; 32]), tvix_castore::Error>;
+    async fn calculate_nar(&self, root_node: &Node)
+        -> Result<(u64, [u8; 32]), tvix_castore::Error>;
 }
 
 #[async_trait]
@@ -27,7 +25,7 @@ where
 {
     async fn calculate_nar(
         &self,
-        root_node: &castorepb::node::Node,
+        root_node: &Node,
     ) -> Result<(u64, [u8; 32]), tvix_castore::Error> {
         self.as_ref().calculate_nar(root_node).await
     }
diff --git a/tvix/store/src/nar/renderer.rs b/tvix/store/src/nar/renderer.rs
index e3cb54dd229f..bb60f7835810 100644
--- a/tvix/store/src/nar/renderer.rs
+++ b/tvix/store/src/nar/renderer.rs
@@ -10,8 +10,7 @@ use tracing::{instrument, Span};
 use tracing_indicatif::span_ext::IndicatifSpanExt;
 use tvix_castore::{
     blobservice::BlobService,
-    directoryservice::DirectoryService,
-    proto::{self as castorepb, NamedNode},
+    directoryservice::{DirectoryService, NamedNode, Node},
 };
 
 pub struct SimpleRenderer<BS, DS> {
@@ -36,7 +35,7 @@ where
 {
     async fn calculate_nar(
         &self,
-        root_node: &castorepb::node::Node,
+        root_node: &Node,
     ) -> Result<(u64, [u8; 32]), tvix_castore::Error> {
         calculate_size_and_sha256(
             root_node,
@@ -52,7 +51,7 @@ where
 /// NAR output.
 #[instrument(skip_all, fields(indicatif.pb_show=1))]
 pub async fn calculate_size_and_sha256<BS, DS>(
-    root_node: &castorepb::node::Node,
+    root_node: &Node,
     blob_service: BS,
     directory_service: DS,
 ) -> Result<(u64, [u8; 32]), RenderError>
@@ -80,13 +79,13 @@ where
     Ok((cw.count(), h.finalize().into()))
 }
 
-/// Accepts a [castorepb::node::Node] pointing to the root of a (store) path,
+/// Accepts a [Node] pointing to the root of a (store) path,
 /// and uses the passed blob_service and directory_service to perform the
 /// necessary lookups as it traverses the structure.
 /// The contents in NAR serialization are writen to the passed [AsyncWrite].
 pub async fn write_nar<W, BS, DS>(
     mut w: W,
-    proto_root_node: &castorepb::node::Node,
+    proto_root_node: &Node,
     blob_service: BS,
     directory_service: DS,
 ) -> Result<(), RenderError>
@@ -115,7 +114,7 @@ where
 /// This consumes the node.
 async fn walk_node<BS, DS>(
     nar_node: nar_writer::Node<'_, '_>,
-    proto_node: &castorepb::node::Node,
+    proto_node: &Node,
     blob_service: BS,
     directory_service: DS,
 ) -> Result<(BS, DS), RenderError>
@@ -124,23 +123,17 @@ where
     DS: DirectoryService + Send,
 {
     match proto_node {
-        castorepb::node::Node::Symlink(proto_symlink_node) => {
+        Node::Symlink(proto_symlink_node) => {
             nar_node
-                .symlink(&proto_symlink_node.target)
+                .symlink(proto_symlink_node.target())
                 .await
                 .map_err(RenderError::NARWriterError)?;
         }
-        castorepb::node::Node::File(proto_file_node) => {
-            let digest_len = proto_file_node.digest.len();
-            let digest = proto_file_node.digest.clone().try_into().map_err(|_| {
-                RenderError::StoreError(io::Error::new(
-                    io::ErrorKind::Other,
-                    format!("invalid digest len {} in file node", digest_len),
-                ))
-            })?;
+        Node::File(proto_file_node) => {
+            let digest = proto_file_node.digest();
 
             let mut blob_reader = match blob_service
-                .open_read(&digest)
+                .open_read(digest)
                 .await
                 .map_err(RenderError::StoreError)?
             {
@@ -153,36 +146,24 @@ where
 
             nar_node
                 .file(
-                    proto_file_node.executable,
-                    proto_file_node.size,
+                    proto_file_node.executable(),
+                    proto_file_node.size(),
                     &mut blob_reader,
                 )
                 .await
                 .map_err(RenderError::NARWriterError)?;
         }
-        castorepb::node::Node::Directory(proto_directory_node) => {
-            let digest_len = proto_directory_node.digest.len();
-            let digest = proto_directory_node
-                .digest
-                .clone()
-                .try_into()
-                .map_err(|_| {
-                    RenderError::StoreError(io::Error::new(
-                        io::ErrorKind::InvalidData,
-                        format!("invalid digest len {} in directory node", digest_len),
-                    ))
-                })?;
-
+        Node::Directory(proto_directory_node) => {
             // look it up with the directory service
             match directory_service
-                .get(&digest)
+                .get(proto_directory_node.digest())
                 .await
                 .map_err(|e| RenderError::StoreError(e.into()))?
             {
                 // if it's None, that's an error!
                 None => Err(RenderError::DirectoryNotFound(
-                    digest,
-                    proto_directory_node.name.clone(),
+                    proto_directory_node.digest().clone(),
+                    proto_directory_node.get_name().clone(),
                 ))?,
                 Some(proto_directory) => {
                     // start a directory node
@@ -206,7 +187,7 @@ where
 
                         (blob_service, directory_service) = Box::pin(walk_node(
                             child_node,
-                            &proto_node,
+                            proto_node,
                             blob_service,
                             directory_service,
                         ))