Merge branch 'utilityai:main' into silas-redo-build-script

Author: SilasMarvin

Cargo.lock

@@ -17,7 +17,7 @@ criterion = "0.5.1"
 pprof = "0.13.0"
 bindgen = "0.69.4"
 cc = "1.0.94"
-anyhow = "1.0.82"
+anyhow = "1.0.86"
 clap = "4.5.4"
 encoding_rs = "0.8.34"
 

Cargo.toml

@@ -1,12 +1,12 @@
 [package]
 name = "embeddings"
-version = "0.1.52"
+version = "0.1.54"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-llama-cpp-2 = { path = "../llama-cpp-2", version = "0.1.52" }
+llama-cpp-2 = { path = "../llama-cpp-2", version = "0.1.54" }
 hf-hub = { workspace = true }
 clap = { workspace = true , features = ["derive"] }
 anyhow = { workspace = true }

embeddings/Cargo.toml

@@ -1,15 +1,15 @@
 [package]
 name = "llama-cpp-2"
 description = "llama.cpp bindings for Rust"
-version = "0.1.52"
+version = "0.1.54"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/utilityai/llama-cpp-rs"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.52" }
+llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.54" }
 thiserror = { workspace = true }
 tracing = { workspace = true }
 

llama-cpp-2/Cargo.toml

@@ -113,7 +113,11 @@ impl LlamaModel {
     /// # Errors
     ///
     /// See [`TokenToStringError`] for more information.
-    pub fn token_to_str(&self, token: LlamaToken, special: Special) -> Result<String, TokenToStringError> {
+    pub fn token_to_str(
+        &self,
+        token: LlamaToken,
+        special: Special,
+    ) -> Result<String, TokenToStringError> {
         self.token_to_str_with_size(token, 32, special)
     }
 
@@ -122,7 +126,11 @@ impl LlamaModel {
     /// # Errors
     ///
     /// See [`TokenToStringError`] for more information.
-    pub fn token_to_bytes(&self, token: LlamaToken, special: Special) -> Result<Vec<u8>, TokenToStringError> {
+    pub fn token_to_bytes(
+        &self,
+        token: LlamaToken,
+        special: Special,
+    ) -> Result<Vec<u8>, TokenToStringError> {
         self.token_to_bytes_with_size(token, 32, special)
     }
 
@@ -131,9 +139,17 @@ impl LlamaModel {
     /// # Errors
     ///
     /// See [`TokenToStringError`] for more information.
-    pub fn tokens_to_str(&self, tokens: &[LlamaToken], special: Special) -> Result<String, TokenToStringError> {
+    pub fn tokens_to_str(
+        &self,
+        tokens: &[LlamaToken],
+        special: Special,
+    ) -> Result<String, TokenToStringError> {
         let mut builder = String::with_capacity(tokens.len() * 4);
-        for str in tokens.iter().copied().map(|t| self.token_to_str(t, special)) {
+        for str in tokens
+            .iter()
+            .copied()
+            .map(|t| self.token_to_str(t, special))
+        {
             builder += &str?;
         }
         Ok(builder)
@@ -451,12 +467,14 @@ impl LlamaModel {
                 content: c.content.as_ptr(),
             })
             .collect();
+
         // Set the tmpl pointer
         let tmpl = tmpl.map(CString::new);
-        let tmpl_ptr = match tmpl {
-            Some(str) => str?.as_ptr(),
+        let tmpl_ptr = match &tmpl {
+            Some(str) => str.as_ref().map_err(|e| e.clone())?.as_ptr(),
             None => std::ptr::null(),
         };
+
         let formatted_chat = unsafe {
             let res = llama_cpp_sys_2::llama_chat_apply_template(
                 self.model.as_ptr(),

llama-cpp-2/src/model.rs

@@ -1,7 +1,7 @@
 [package]
 name = "llama-cpp-sys-2"
 description = "Low Level Bindings to llama.cpp"
-version = "0.1.52"
+version = "0.1.54"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/utilityai/llama-cpp-rs"

llama-cpp-sys-2/Cargo.toml

@@ -1,12 +1,12 @@
 [package]
 name = "simple"
-version = "0.1.52"
+version = "0.1.54"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-llama-cpp-2 = { path = "../llama-cpp-2", version = "0.1.52" }
+llama-cpp-2 = { path = "../llama-cpp-2", version = "0.1.54" }
 hf-hub = { workspace = true }
 clap = { workspace = true , features = ["derive"] }
 anyhow = { workspace = true }