cas types changes

.gitignore

@@ -1,8 +1,8 @@
-.idea
+**/.idea
 # Generated by Cargo
 # will have compiled files and executables
 debug/
-target/
+**/target/
 
 # These are backup files generated by rustfmt
 **/*.rs.bk
@@ -15,3 +15,4 @@ target/
 
 # VS Code configs
 .vscode
+venv

Cargo.toml

@@ -21,6 +21,10 @@ members = [
     "xet_error",
 ]
 
+exclude = [
+    "hf_xet",
+]
+
 [profile.release]
 opt-level = 3
 lto = true

cas_types/src/key.rs

@@ -14,4 +14,80 @@ impl Display for Key {
     fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
         write!(f, "{}/{:x}", self.prefix, self.hash)
     }
-}
+}
+
+mod hex {
+    pub mod serde {
+        use merklehash::MerkleHash;
+        use serde::de::{self, Visitor};
+        use serde::{Deserializer, Serializer};
+        use std::fmt;
+
+        pub fn serialize<S>(value: &MerkleHash, serializer: S) -> Result<S::Ok, S::Error>
+        where
+            S: Serializer,
+        {
+            let hex = value.hex();
+            serializer.serialize_str(&hex)
+        }
+
+        pub fn deserialize<'de, D>(deserializer: D) -> Result<MerkleHash, D::Error>
+        where
+            D: Deserializer<'de>,
+        {
+            deserializer.deserialize_str(HexVisitor)
+        }
+
+        // Visitor for deserialization
+        struct HexVisitor;
+
+        impl<'de> Visitor<'de> for HexVisitor {
+            type Value = MerkleHash;
+
+            fn expecting(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+                formatter.write_str("a merklehash")
+            }
+
+            fn visit_str<E>(self, v: &str) -> Result<Self::Value, E>
+            where
+                E: de::Error,
+            {
+                MerkleHash::from_hex(v).map_err(|e| serde::de::Error::custom(e))
+            }
+        }
+    }
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize, Default)]
+pub struct HexMerkleHash(#[serde(with = "hex::serde")] pub MerkleHash);
+
+impl From<MerkleHash> for HexMerkleHash {
+    fn from(value: MerkleHash) -> Self {
+        HexMerkleHash(value)
+    }
+}
+
+impl From<HexMerkleHash> for MerkleHash {
+    fn from(value: HexMerkleHash) -> Self {
+        value.0
+    }
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize, Default)]
+pub struct HexKey {
+    pub prefix: String,
+    #[serde(with = "hex::serde")]
+    pub hash: MerkleHash,
+}
+
+impl From<HexKey> for Key {
+    fn from(HexKey { prefix, hash }: HexKey) -> Self {
+        Key { prefix, hash }
+    }
+}
+
+impl From<Key> for HexKey {
+    fn from(Key { prefix, hash }: Key) -> Self {
+        HexKey { prefix, hash }
+    }
+}

cas_types/src/lib.rs

@@ -20,9 +20,10 @@ pub struct Range {
 
 #[derive(Debug, Serialize, Deserialize, Clone)]
 pub struct CASReconstructionTerm {
-    pub hash: String,
+    pub hash: HexMerkleHash,
     pub unpacked_length: u64,
     pub range: Range,
+    pub url: String,
 }
 
 #[derive(Debug, Serialize, Deserialize, Clone)]

data/src/lib.rs

@@ -6,7 +6,7 @@ mod clean;
 pub mod configurations;
 mod constants;
 mod data_processing;
-mod errors;
+pub mod errors;
 mod metrics;
 mod pointer_file;
 mod remote_shard_interface;

data/src/pointer_file.rs

@@ -12,7 +12,7 @@ const CURRENT_VERSION: &str = "0";
 /// A struct that wraps a Xet pointer file.
 /// Xet pointer file format is a TOML file,
 /// and the first line must be of the form "# xet version <x.y>"
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Debug, Clone, PartialEq, Eq, Default)]
 pub struct PointerFile {
     /// The version string of the pointer file
     version_string: String,