Remove OMF (#78)

* Remove OMF

* remove submodule from github workflow

.github/workflows/checks.yml

@@ -9,10 +9,6 @@ jobs:
     steps:
       - name: Setup | Checkout
         uses: actions/checkout@v4
-        with:
-          submodules: 'true'
-          # TODO: Remove this once the repo is public
-          token: ${{ secrets.ADIL_GITHUB_TOKEN }}
       - name: Setup | Rust
         run: rustup toolchain install stable --profile minimal
       - name: Run Clippy
@@ -25,10 +21,6 @@ jobs:
     steps:
       - name: Setup | Checkout
         uses: actions/checkout@v4
-        with:
-          submodules: 'true'
-          # TODO: Remove this once the repo is public
-          token: ${{ secrets.ADIL_GITHUB_TOKEN }}
       - name: Setup | Rust
         run: rustup toolchain install stable --profile minimal
       - name: Run Rustfmt
@@ -40,10 +32,6 @@ jobs:
     steps:
       - name: Setup | Checkout
         uses: actions/checkout@v4
-        with:
-          submodules: 'true'
-          # TODO: Remove this once the repo is public
-          token: ${{ secrets.ADIL_GITHUB_TOKEN }}
       - name: Setup | Rust
         run: rustup toolchain install stable --profile minimal
       - name: Setup | Install wasm toolchain

.gitmodules

@@ -1,3 +0,0 @@
-[submodule "open-message-format"]
-	path = open-message-format
-	url = git@github.com:open-llm-initiative/open-message-format-private.git

envoyfilter/Cargo.toml

@@ -14,7 +14,6 @@ serde = { version = "1.0", features = ["derive"] }
 serde_yaml = "0.9.34"
 serde_json = "1.0"
 md5 = "0.7.0"
-open-message-format-embeddings = { path = "../open-message-format/clients/omf-embeddings-rust" }
 public_types = { path = "../public_types" }
 http = "1.1.0"
 governor = { version = "0.6.3", default-features = false, features = ["no_std"]}

envoyfilter/Dockerfile

@@ -5,7 +5,6 @@ WORKDIR /envoyfilter
 COPY envoyfilter/src /envoyfilter/src
 COPY envoyfilter/Cargo.toml /envoyfilter/
 COPY envoyfilter/Cargo.lock /envoyfilter/
-COPY open-message-format /open-message-format
 COPY public_types /public_types
 
 RUN cargo build --release --target wasm32-wasi

envoyfilter/src/filter_context.rs

@@ -3,13 +3,13 @@ use crate::ratelimit;
 use crate::stats::{Counter, Gauge, RecordingMetric};
 use crate::stream_context::StreamContext;
 use log::debug;
-use open_message_format_embeddings::models::{
-    CreateEmbeddingRequest, CreateEmbeddingRequestInput, CreateEmbeddingResponse,
-};
 use proxy_wasm::traits::*;
 use proxy_wasm::types::*;
 use public_types::common_types::EmbeddingType;
 use public_types::configuration::{Configuration, Overrides, PromptGuards, PromptTarget};
+use public_types::embeddings::{
+    CreateEmbeddingRequest, CreateEmbeddingRequestInput, CreateEmbeddingResponse,
+};
 use serde_json::to_string;
 use std::collections::HashMap;
 use std::rc::Rc;

envoyfilter/src/stream_context.rs

@@ -11,9 +11,6 @@ use crate::tokenizer;
 use acap::cos;
 use http::StatusCode;
 use log::{debug, info, warn};
-use open_message_format_embeddings::models::{
-    CreateEmbeddingRequest, CreateEmbeddingRequestInput, CreateEmbeddingResponse,
-};
 use proxy_wasm::traits::*;
 use proxy_wasm::types::*;
 use public_types::common_types::open_ai::{
@@ -26,6 +23,9 @@ use public_types::common_types::{
     ZeroShotClassificationResponse,
 };
 use public_types::configuration::{Overrides, PromptGuards, PromptTarget, PromptType};
+use public_types::embeddings::{
+    CreateEmbeddingRequest, CreateEmbeddingRequestInput, CreateEmbeddingResponse,
+};
 use std::collections::HashMap;
 use std::num::NonZero;
 use std::rc::Rc;

envoyfilter/tests/integration.rs

@@ -1,9 +1,4 @@
 use http::StatusCode;
-use open_message_format_embeddings::models::{
-    create_embedding_response::{self, CreateEmbeddingResponse},
-    create_embedding_response_usage::CreateEmbeddingResponseUsage,
-    embedding, Embedding,
-};
 use proxy_wasm_test_framework::tester::{self, Tester};
 use proxy_wasm_test_framework::types::{
     Action, BufferType, LogLevel, MapType, MetricType, ReturnType,
@@ -12,6 +7,10 @@ use public_types::common_types::{
     open_ai::{ChatCompletionsResponse, Choice, Message, Usage},
     BoltFCToolsCall, IntOrString, ToolCallDetail,
 };
+use public_types::embeddings::embedding::Object;
+use public_types::embeddings::{
+    create_embedding_response, CreateEmbeddingResponse, CreateEmbeddingResponseUsage, Embedding,
+};
 use public_types::{common_types::ZeroShotClassificationResponse, configuration::Configuration};
 use serial_test::serial;
 use std::collections::HashMap;
@@ -93,7 +92,7 @@ fn normal_flow(module: &mut Tester, filter_context: i32, http_context: i32) {
         data: vec![Embedding {
             index: 0,
             embedding: vec![],
-            object: embedding::Object::default(),
+            object: Object::default(),
         }],
         model: String::from("test"),
         object: create_embedding_response::Object::default(),

open-message-format

@@ -1 +0,0 @@
-Subproject commit 1e838f3f4019e802099309687ca404509cb114a2

public_types/Cargo.toml

@@ -5,5 +5,4 @@ edition = "2021"
 
 [dependencies]
 serde = { version = "1.0", features = ["derive"] }
-open-message-format-embeddings = { path = "../open-message-format/clients/omf-embeddings-rust" }
 serde_yaml = "0.9.34"

public_types/src/embeddings/create_embedding_request.rs

@@ -0,0 +1,56 @@
+/*
+ * OMF Embeddings
+ *
+ * No description provided (generated by Openapi Generator https://github.com/openapitools/openapi-generator)
+ *
+ * The version of the OpenAPI document: 1.0.0
+ *
+ * Generated by: https://openapi-generator.tech
+ */
+
+use crate::embeddings;
+use serde::{Deserialize, Serialize};
+
+#[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
+pub struct CreateEmbeddingRequest {
+    #[serde(rename = "input")]
+    pub input: Box<embeddings::CreateEmbeddingRequestInput>,
+    /// ID of the model to use. You can use the [List models](/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](/docs/models/overview) for descriptions of them.
+    #[serde(rename = "model")]
+    pub model: String,
+    /// The format to return the embeddings in. Can be either `float` or [`base64`](https://pypi.org/project/pybase64/).
+    #[serde(rename = "encoding_format", skip_serializing_if = "Option::is_none")]
+    pub encoding_format: Option<EncodingFormat>,
+    /// The number of dimensions the resulting output embeddings should have. Only supported in `text-embedding-3` and later models.
+    #[serde(rename = "dimensions", skip_serializing_if = "Option::is_none")]
+    pub dimensions: Option<i32>,
+    /// A unique identifier representing your end-user, which can help to monitor and detect abuse. [Learn more](/docs/guides/safety-best-practices/end-user-ids).
+    #[serde(rename = "user", skip_serializing_if = "Option::is_none")]
+    pub user: Option<String>,
+}
+
+impl CreateEmbeddingRequest {
+    pub fn new(input: embeddings::CreateEmbeddingRequestInput, model: String) -> CreateEmbeddingRequest {
+        CreateEmbeddingRequest {
+            input: Box::new(input),
+            model,
+            encoding_format: None,
+            dimensions: None,
+            user: None,
+        }
+    }
+}
+/// The format to return the embeddings in. Can be either `float` or [`base64`](https://pypi.org/project/pybase64/).
+#[derive(Clone, Copy, Debug, Eq, PartialEq, Ord, PartialOrd, Hash, Serialize, Deserialize)]
+pub enum EncodingFormat {
+    #[serde(rename = "float")]
+    Float,
+    #[serde(rename = "base64")]
+    Base64,
+}
+
+impl Default for EncodingFormat {
+    fn default() -> EncodingFormat {
+        Self::Float
+    }
+}

public_types/src/embeddings/create_embedding_request_input.rs

@@ -0,0 +1,29 @@
+/*
+ * OMF Embeddings
+ *
+ * No description provided (generated by Openapi Generator https://github.com/openapitools/openapi-generator)
+ *
+ * The version of the OpenAPI document: 1.0.0
+ *
+ * Generated by: https://openapi-generator.tech
+ */
+
+use crate::embeddings;
+use serde::{Deserialize, Serialize};
+
+/// CreateEmbeddingRequestInput : Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single request, pass an array of strings or array of token arrays. The input must not exceed the max input tokens for the model (8192 tokens for `text-embedding-ada-002`), cannot be an empty string, and any array must be 2048 dimensions or less. for counting tokens.
+/// Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single request, pass an array of strings or array of token arrays. The input must not exceed the max input tokens for the model (8192 tokens for `text-embedding-ada-002`), cannot be an empty string, and any array must be 2048 dimensions or less. for counting tokens.
+#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
+#[serde(untagged)]
+pub enum CreateEmbeddingRequestInput {
+    /// The string that will be turned into an embedding.
+    String(String),
+    /// The array of integers that will be turned into an embedding.
+    Array(Vec<i32>),
+}
+
+impl Default for CreateEmbeddingRequestInput {
+    fn default() -> Self {
+        Self::String(Default::default())
+    }
+}

public_types/src/embeddings/create_embedding_response.rs

@@ -0,0 +1,50 @@
+/*
+ * OMF Embeddings
+ *
+ * No description provided (generated by Openapi Generator https://github.com/openapitools/openapi-generator)
+ *
+ * The version of the OpenAPI document: 1.0.0
+ *
+ * Generated by: https://openapi-generator.tech
+ */
+
+use crate::embeddings;
+use serde::{Deserialize, Serialize};
+
+#[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
+pub struct CreateEmbeddingResponse {
+    /// The list of embeddings generated by the model.
+    #[serde(rename = "data")]
+    pub data: Vec<embeddings::Embedding>,
+    /// The name of the model used to generate the embedding.
+    #[serde(rename = "model")]
+    pub model: String,
+    /// The object type, which is always \"list\".
+    #[serde(rename = "object")]
+    pub object: Object,
+    #[serde(rename = "usage")]
+    pub usage: Box<embeddings::CreateEmbeddingResponseUsage>,
+}
+
+impl CreateEmbeddingResponse {
+    pub fn new(data: Vec<embeddings::Embedding>, model: String, object: Object, usage: embeddings::CreateEmbeddingResponseUsage) -> CreateEmbeddingResponse {
+        CreateEmbeddingResponse {
+            data,
+            model,
+            object,
+            usage: Box::new(usage),
+        }
+    }
+}
+/// The object type, which is always \"list\".
+#[derive(Clone, Copy, Debug, Eq, PartialEq, Ord, PartialOrd, Hash, Serialize, Deserialize)]
+pub enum Object {
+    #[serde(rename = "list")]
+    List,
+}
+
+impl Default for Object {
+    fn default() -> Object {
+        Self::List
+    }
+}

public_types/src/embeddings/create_embedding_response_usage.rs

@@ -0,0 +1,33 @@
+/*
+ * OMF Embeddings
+ *
+ * No description provided (generated by Openapi Generator https://github.com/openapitools/openapi-generator)
+ *
+ * The version of the OpenAPI document: 1.0.0
+ *
+ * Generated by: https://openapi-generator.tech
+ */
+
+use crate::embeddings;
+use serde::{Deserialize, Serialize};
+
+/// CreateEmbeddingResponseUsage : The usage information for the request.
+#[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
+pub struct CreateEmbeddingResponseUsage {
+    /// The number of tokens used by the prompt.
+    #[serde(rename = "prompt_tokens")]
+    pub prompt_tokens: i32,
+    /// The total number of tokens used by the request.
+    #[serde(rename = "total_tokens")]
+    pub total_tokens: i32,
+}
+
+impl CreateEmbeddingResponseUsage {
+    /// The usage information for the request.
+    pub fn new(prompt_tokens: i32, total_tokens: i32) -> CreateEmbeddingResponseUsage {
+        CreateEmbeddingResponseUsage {
+            prompt_tokens,
+            total_tokens,
+        }
+    }
+}

public_types/src/embeddings/embedding.rs

@@ -0,0 +1,49 @@
+/*
+ * OMF Embeddings
+ *
+ * No description provided (generated by Openapi Generator https://github.com/openapitools/openapi-generator)
+ *
+ * The version of the OpenAPI document: 1.0.0
+ *
+ * Generated by: https://openapi-generator.tech
+ */
+
+use crate::embeddings;
+use serde::{Deserialize, Serialize};
+
+/// Embedding : Represents an embedding vector returned by embedding endpoint.
+#[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
+pub struct Embedding {
+    /// The index of the embedding in the list of embeddings.
+    #[serde(rename = "index")]
+    pub index: i32,
+    /// The embedding vector, which is a list of floats. The length of vector depends on the model as listed in the [embedding guide](/docs/guides/embeddings).
+    #[serde(rename = "embedding")]
+    pub embedding: Vec<f64>,
+    /// The object type, which is always \"embedding\"
+    #[serde(rename = "object")]
+    pub object: Object,
+}
+
+impl Embedding {
+    /// Represents an embedding vector returned by embedding endpoint.
+    pub fn new(index: i32, embedding: Vec<f64>, object: Object) -> Embedding {
+        Embedding {
+            index,
+            embedding,
+            object,
+        }
+    }
+}
+/// The object type, which is always \"embedding\"
+#[derive(Clone, Copy, Debug, Eq, PartialEq, Ord, PartialOrd, Hash, Serialize, Deserialize)]
+pub enum Object {
+    #[serde(rename = "embedding")]
+    Embedding,
+}
+
+impl Default for Object {
+    fn default() -> Object {
+        Self::Embedding
+    }
+}

public_types/src/embeddings/mod.rs

@@ -0,0 +1,10 @@
+pub mod create_embedding_request;
+pub use self::create_embedding_request::CreateEmbeddingRequest;
+pub mod create_embedding_request_input;
+pub use self::create_embedding_request_input::CreateEmbeddingRequestInput;
+pub mod create_embedding_response;
+pub use self::create_embedding_response::CreateEmbeddingResponse;
+pub mod create_embedding_response_usage;
+pub use self::create_embedding_response_usage::CreateEmbeddingResponseUsage;
+pub mod embedding;
+pub use self::embedding::Embedding;

public_types/src/lib.rs

@@ -1,2 +1,5 @@
+#![allow(unused_imports)]
+
 pub mod common_types;
 pub mod configuration;
+pub mod embeddings;