Merge pull request #4 from smallcloudai/code_scratchpads

merge with code_scratchpad repo

.github/workflows/build.yml

@@ -0,0 +1,170 @@
+name: CI build
+
+on:
+  # Trigger the workflow on pushes to only the 'main' branch (this avoids duplicate checks being run e.g. for dependabot pull requests)
+  push:
+    branches: [main]
+  # Trigger the workflow on any pull request
+  pull_request:
+  workflow_dispatch:
+
+
+env:
+  CARGO_INCREMENTAL: 0
+  CARGO_NET_RETRY: 10
+  # RUSTFLAGS: "-D warnings -W unreachable-pub"
+  RUSTUP_MAX_RETRIES: 10
+  FETCH_DEPTH: 0 # pull in the tags for the version string
+  MACOSX_DEPLOYMENT_TARGET: 10.15
+  CARGO_TARGET_AARCH64_UNKNOWN_LINUX_GNU_LINKER: aarch64-linux-gnu-gcc
+  CARGO_TARGET_ARM_UNKNOWN_LINUX_GNUEABIHF_LINKER: arm-linux-gnueabihf-gcc
+
+jobs:
+  dist:
+    strategy:
+      matrix:
+        include:
+          - os: windows-latest
+            target: x86_64-pc-windows-msvc
+            code-target: win32-x64
+          - os: windows-latest
+            target: i686-pc-windows-msvc
+            code-target: win32-ia32
+          - os: windows-latest
+            target: aarch64-pc-windows-msvc
+            code-target: win32-arm64
+          - os: ubuntu-20.04
+            target: x86_64-unknown-linux-gnu
+            code-target: linux-x64
+            container: ubuntu:18.04
+          - os: ubuntu-20.04
+            target: aarch64-unknown-linux-gnu
+            code-target: linux-arm64
+          - os: ubuntu-20.04
+            target: arm-unknown-linux-gnueabihf
+            code-target: linux-armhf
+          - os: macos-11
+            target: x86_64-apple-darwin
+            code-target: darwin-x64
+          - os: macos-11
+            target: aarch64-apple-darwin
+            code-target: darwin-arm64
+
+    env:
+      LLM_LS_TARGET: ${{ matrix.target }}
+
+    name: dist (${{ matrix.target }})
+    runs-on: ${{ matrix.os }}
+    container: ${{ matrix.container }}
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: ${{ env.FETCH_DEPTH }}
+
+      - name: Install toolchain dependencies
+        if: matrix.container == 'ubuntu:18.04'
+        shell: bash
+        run: |
+          apt-get update && apt-get install -y build-essential curl libssl-dev pkg-config
+          curl --proto '=https' --tlsv1.2 --retry 10 --retry-connrefused -fsSL "https://sh.rustup.rs" | sh -s -- --profile minimal --default-toolchain none -y
+          echo "${CARGO_HOME:-$HOME/.cargo}/bin" >> $GITHUB_PATH
+
+      - name: Install Rust toolchain
+        run: |
+          rustup update --no-self-update stable
+          rustup target add ${{ matrix.target }}
+          rustup component add rust-src
+
+      - name: Update apt repositories
+        if: matrix.target == 'aarch64-unknown-linux-gnu' || matrix.target == 'arm-unknown-linux-gnueabihf'
+        run: sudo apt-get update
+
+      - name: Install AArch64 target toolchain
+        if: matrix.target == 'aarch64-unknown-linux-gnu'
+        run: sudo apt-get install gcc-aarch64-linux-gnu
+
+      - name: Install ARM target toolchain
+        if: matrix.target == 'arm-unknown-linux-gnueabihf'
+        run: sudo apt-get install gcc-multilib-arm-linux-gnueabihf
+
+      - name: Build artifact
+        if: matrix.target == 'x86_64-unknown-linux-gnu'
+        run: OPENSSL_STATIC=1 OPENSSL_LIB_DIR=/usr/lib/x86_64-linux-gnu OPENSSL_INCLUDE_DIR=/usr/include/openssl cargo build --release --target-dir dist/
+
+      - name: Build artifact
+        if: matrix.target != 'x86_64-unknown-linux-gnu'
+        run: cargo build --release --target-dir dist/
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v3
+        with:
+          name: dist-${{ matrix.target }}
+          if-no-files-found: ignore
+          path: |
+            ./dist/release/code-scratchpads
+            ./dist/release/code-scratchpads.exe
+
+  dist-x86_64-unknown-linux-musl:
+    name: dist (x86_64-unknown-linux-musl)
+    runs-on: ubuntu-latest
+    env:
+      LLM_LS_TARGET: x86_64-unknown-linux-musl
+      # For some reason `-crt-static` is not working for clang without lld
+      RUSTFLAGS: "-C link-arg=-fuse-ld=lld -C target-feature=-crt-static"
+    container:
+      image: rust:alpine
+      volumes:
+        - /usr/local/cargo/registry:/usr/local/cargo/registry
+
+    steps:
+      - name: Install dependencies
+        run: apk add --no-cache git clang lld musl-dev nodejs npm openssl-dev pkgconfig g++
+
+      - name: Checkout repository
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: ${{ env.FETCH_DEPTH }}
+
+      - name: Dist
+        run: cargo build --release --target-dir dist/
+
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v1
+        with:
+          name: dist-x86_64-unknown-linux-musl
+          path: ./dist/release/code-scratchpads
+
+  # dist-arm64-unknown-linux-musl:
+  #   name: dist (arm64-unknown-linux-musl)
+  #   runs-on: ubuntu-latest
+
+  #   steps:
+  #     - name: Checkout repository
+  #       uses: actions/checkout@v3
+  #       with:
+  #         fetch-depth: ${{ env.FETCH_DEPTH }}
+
+  #     - uses: uraimo/run-on-arch-action@v2
+  #       name: Build artifact
+  #       id: build
+  #       with:
+  #         arch: aarch64
+  #         distro: rust:alpine
+  #         dockerRunArgs: |
+  #           --volume "${PWD}:/data"
+  #         env: |
+  #           RUSTFLAGS: "-C link-arg=-fuse-ld=lld -C target-feature=-crt-static"
+  #         shell: /bin/sh
+  #         install: |
+  #           apk add --no-cache git clang lld musl-dev nodejs npm openssl-dev pkgconfig g++
+  #         run: |
+  #           cd /data
+  #           cargo build --release --target-dir dist/
+
+  #     - name: Upload artifacts
+  #       uses: actions/upload-artifact@v1
+  #       with:
+  #         name: dist-arm64-unknown-linux-musl
+  #         path: ./dist/release/code-scratchpads

Cargo.toml

@@ -0,0 +1,41 @@
+[package]
+name = "code-scratchpads"
+version = "0.1.0"
+edition = "2021"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+hyper = { version = "0.14", features = ["server", "stream"] }
+reqwest = { version = "0.11", features = ["json", "stream"] }
+tokio = { version = "1", features = ["fs", "io-std", "io-util", "macros", "rt-multi-thread", "signal"] }
+reqwest-eventsource = "0.4.0"
+url = "2.4.1"
+serde = { version = "1", features = ["derive"] }
+serde_json = "1"
+tower = "0.4"
+tower-lsp = "0.20"
+tracing = "0.1"
+tracing-appender = "0.2"
+tracing-subscriber = { version = "0.3", features = ["env-filter", "json"] }
+tokenizers = "0.13"
+ropey = "1.6"
+home = "0.5"
+structopt = "0.3"
+futures = "0.3"
+futures-core = "0.3"
+futures-util = "0.3"
+async-stream = "0.3.5"
+chrono = "0.4.31"
+difference = "2.0.0"
+regex = "1.9.5"
+
+#use = "0.0.0"
+#async-trait = "0.1.73"
+#route-recognizer = "0.3.1"
+#bytes = "0.5"
+#eventsource-client = "0.11.0"
+#reqwest-streams = { version = "0.3.0", features = ["json"] }
+#tokio-stream = "0.1.14"
+#eventsource-stream = "0.2.3"
+

Dockerfile

@@ -0,0 +1,7 @@
+FROM python:3.8
+
+ADD . /opt/app
+RUN pip install -v /opt/app
+
+EXPOSE 8001
+CMD huggingface-cli login --token $HUGGINGFACE_TOKEN && python -m code_scratchpads.http_server

README.md

@@ -0,0 +1,35 @@
+# Code Scratchpads
+
+This code converts high level code completion calls to low level prompts and converts result back. This is useful for many IDE plugins (VS Code, JB) as a common code that handles the low level.
+
+
+## Usage
+
+Simple example:
+
+```
+curl http://127.0.0.1:8001/v1/code-completion -k \
+  -H 'Content-Type: application/json' \
+  -d '{
+  "inputs": {
+    "sources": {"hello.py": "def hello_world():"},
+    "cursor": {
+      "file": "hello.py",
+      "line": 0,
+      "character": 18
+    },
+    "multiline": true
+  },
+  "model": "bigcode/starcoder",
+  "stream": false,
+  "parameters": {
+    "temperature": 0.1,
+    "max_new_tokens": 20
+  }
+}'
+```
+
+Output is `[{"code_completion": "\n    return \"Hello World!\"\n"}]`.
+
+To check out more examples, look at [code_scratchpads/tests/test_api.py](code_scratchpads/tests/test_api.py).
+

setup.py

@@ -1,8 +1,7 @@
-from setuptools import setup
+from setuptools import setup, find_packages
+
 
 setup(
-    name="refact-lsp",
-    py_modules=["refact_lsp"],
     version="0.0.1",
     url="https://github.com/smallcloudai/refact_lsp",
     summary="LSP server for Refact, suitable for Sublime Text, and other editors",

src/cached_tokenizers.rs

@@ -0,0 +1,48 @@
+use reqwest::header::AUTHORIZATION;
+use tracing::info;
+use tokio::io::AsyncWriteExt;
+use std::path::Path;
+
+
+#[derive(Debug, Clone, PartialEq, Eq)]
+pub struct Error {
+    pub message: String,
+    pub data: Option<serde_json::Value>,
+}
+
+pub async fn download_tokenizer_file(
+    http_client: &reqwest::Client,
+    http_path: &str,
+    api_token: String,
+    to: impl AsRef<Path>,
+) -> Result<(), String> {
+    if to.as_ref().exists() {
+        return Ok(());
+    }
+    info!("downloading tokenizer \"{}\" to {}...", http_path, to.as_ref().display());
+    tokio::fs::create_dir_all(
+            to.as_ref().parent().ok_or_else(|| "tokenizer path has no parent")?,
+        )
+        .await
+        .map_err(|e| format!("failed to create parent dir: {}", e))?;
+    let mut req = http_client.get(http_path);
+    if !api_token.is_empty() {
+        req = req.header(AUTHORIZATION, format!("Bearer {api_token}"))
+    }
+    let res = req
+        .send()
+        .await
+        .map_err(|e| format!("failed to get response: {}", e))?
+        .error_for_status()
+        .map_err(|e| format!("failed to get response: {}", e))?;
+    let mut file = tokio::fs::OpenOptions::new()
+        .write(true)
+        .create(true)
+        .open(to)
+        .await
+        .map_err(|e| format!("failed to open file: {}", e))?;
+    file.write_all(&res.bytes().await
+        .map_err(|e| format!("failed to fetch bytes: {}", e))?
+    ).await.map_err(|e| format!("failed to write to file: {}", e))?;
+    Ok(())
+}

src/call_validation.rs

@@ -0,0 +1,58 @@
+use serde::Deserialize;
+use serde::Serialize;
+use std::collections::HashMap;
+
+
+#[derive(Debug, Serialize, Deserialize, Clone, Default)]
+pub struct CursorPosition {
+    pub file: String,
+    pub line: i32,
+    pub character: i32,
+}
+
+#[derive(Debug, Serialize, Deserialize, Clone, Default)]
+pub struct CodeCompletionInputs {
+    pub sources: HashMap<String, String>,
+    pub cursor: CursorPosition,
+    pub multiline: bool,
+}
+
+#[derive(Debug, Serialize, Deserialize, Clone, Default)]
+pub struct SamplingParameters {
+    #[serde(default)]
+    pub max_new_tokens: usize,
+    pub temperature: Option<f32>,
+    pub top_p: Option<f32>,
+    pub stop: Option<Vec<String>>,
+}
+
+#[derive(Debug, Deserialize, Clone)]
+pub struct CodeCompletionPost {
+    pub inputs: CodeCompletionInputs,
+    #[serde(default)]
+    pub parameters: SamplingParameters,
+    #[serde(default)]
+    pub model: String,
+    #[serde(default)]
+    pub scratchpad: String,
+    #[serde(default)]
+    pub stream: bool,
+}
+
+#[derive(Debug, Deserialize, Clone)]
+pub struct ChatMessage {
+    pub role: String,
+    pub content: String,
+}
+
+#[derive(Debug, Deserialize, Clone)]
+pub struct ChatPost {
+    pub messages: Vec<ChatMessage>,
+    #[serde(default)]
+    pub parameters: SamplingParameters,
+    #[serde(default)]
+    pub model: String,
+    #[serde(default)]
+    pub scratchpad: String,
+    pub stream: Option<bool>,
+}