TP-68136 | code critic workflow added
This commit is contained in:
Aman Chaturvedi
114
.github/workflows/pr-review.yml
vendored
Normal file
114
.github/workflows/pr-review.yml
vendored
Normal file
@@ -0,0 +1,114 @@
|
||||
name: Code Critic PR Review
|
||||
|
||||
on:
|
||||
pull_request:
|
||||
types: [opened, reopened, ready_for_review]
|
||||
branches: [master]
|
||||
issue_comment:
|
||||
types: [created]
|
||||
|
||||
jobs:
|
||||
pr_review:
|
||||
runs-on: [ docker ]
|
||||
if: github.event_name == 'pull_request' || (github.event_name == 'issue_comment' && contains(github.event.comment.body, '/review'))
|
||||
|
||||
steps:
|
||||
- name: Set up Python
|
||||
uses: actions/setup-python@v4
|
||||
with:
|
||||
python-version: '3.x'
|
||||
|
||||
- name: Print Python version
|
||||
run: python --version
|
||||
|
||||
- name: Create GitHub App token
|
||||
uses: actions/create-github-app-token@v1
|
||||
id: app-token
|
||||
with:
|
||||
# required
|
||||
app-id: ${{ vars.CODE_CRITIC_APP_ID }}
|
||||
private-key: ${{ secrets.CODE_CRITIC_APP_PRIVATE_KEY }}
|
||||
owner: ${{ github.repository_owner }}
|
||||
repositories: "code-critic"
|
||||
|
||||
- name: Get PR head ref for issue comments
|
||||
if: github.event_name == 'issue_comment'
|
||||
id: get-pr-ref
|
||||
run: |
|
||||
PR_API_URL="${{ github.event.issue.pull_request.url }}"
|
||||
PR_JSON=$(curl -s -H "Authorization: token ${{ secrets.GITHUB_TOKEN }}" $PR_API_URL)
|
||||
PR_HEAD_REF=$(echo "$PR_JSON" | jq -r .head.ref)
|
||||
echo "::set-output name=pr_head_ref::$PR_HEAD_REF"
|
||||
env:
|
||||
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
|
||||
- name: Checkout PR branch
|
||||
uses: actions/checkout@v3
|
||||
with:
|
||||
ref: ${{ steps.get-pr-ref.outputs.pr_head_ref || github.event.pull_request.head.ref }}
|
||||
fetch-depth: 0 # Necessary to fetch all history for diff
|
||||
|
||||
- name: Checkout code-critic repository
|
||||
uses: actions/checkout@v3
|
||||
with:
|
||||
repository: navi-medici/code-critic
|
||||
token: ${{ steps.app-token.outputs.token }}
|
||||
ref: main
|
||||
path: code-critic
|
||||
|
||||
- name: Install requirements
|
||||
shell: bash
|
||||
run: |
|
||||
pip install -r code-critic/requirements.txt
|
||||
- name: React with eyes to /review command
|
||||
if: github.event_name == 'issue_comment' && contains(github.event.comment.body, '/review')
|
||||
run: |
|
||||
echo "Reacting with emoji to /review command..."
|
||||
repo_owner=$(echo "${{ github.repository }}" | cut -d '/' -f 1)
|
||||
repo_name=$(echo "${{ github.repository }}" | cut -d '/' -f 2)
|
||||
comment_id=${{ github.event.comment.id }}
|
||||
echo "Repo Owner: $repo_owner, Repo Name: $repo_name, Comment ID: $comment_id"
|
||||
python code-critic/utils/react_on_comment.py $repo_owner $repo_name $comment_id eyes
|
||||
env:
|
||||
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
|
||||
- name: Set up environment
|
||||
run: |
|
||||
echo "GITHUB_REPOSITORY_OWNER=${{ github.repository_owner }}" >> $GITHUB_ENV
|
||||
echo "REPO_NAME=$(echo ${{ github.repository }} | cut -d'/' -f2)" >> $GITHUB_ENV
|
||||
if [ "${{ github.event_name }}" == "pull_request" ]; then
|
||||
echo "PR_NUMBER=${{ github.event.pull_request.number }}" >> $GITHUB_ENV
|
||||
elif [ "${{ github.event_name }}" == "issue_comment" ]; then
|
||||
PR_URL=${{ github.event.issue.pull_request.html_url }}
|
||||
PR_NUMBER=$(basename $PR_URL)
|
||||
echo "PR_NUMBER=$PR_NUMBER" >> $GITHUB_ENV
|
||||
fi
|
||||
- name: List directory contents
|
||||
run: ls -R $GITHUB_WORKSPACE
|
||||
|
||||
- name: Run code review script
|
||||
shell: bash
|
||||
run: python $GITHUB_WORKSPACE/scripts/pr-review.py
|
||||
env:
|
||||
GPT_MODEL_NAME: ${{ secrets.GPT_MODEL_NAME }}
|
||||
AZURE_API_KEY: ${{ secrets.AZURE_API_KEY }}
|
||||
AZURE_API_BASE: ${{ secrets.AZURE_API_BASE }}
|
||||
AZURE_API_VERSION: ${{ secrets.AZURE_API_VERSION }}
|
||||
|
||||
- name: Read file and post comments
|
||||
run: python code-critic/utils/post_comments.py $PR_NUMBER code_review_output.txt
|
||||
env:
|
||||
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
PR_NUMBER: ${{ env.PR_NUMBER }}
|
||||
|
||||
- name: React with rocket marking completion
|
||||
if: github.event_name == 'issue_comment' && contains(github.event.comment.body, '/review')
|
||||
run: |
|
||||
echo "Reacting with rocket"
|
||||
repo_owner=$(echo "${{ github.repository }}" | cut -d '/' -f 1)
|
||||
repo_name=$(echo "${{ github.repository }}" | cut -d '/' -f 2)
|
||||
comment_id=${{ github.event.comment.id }}
|
||||
echo "Repo Owner: $repo_owner, Repo Name: $repo_name, Comment ID: $comment_id"
|
||||
python code-critic/utils/react_on_comment.py $repo_owner $repo_name $comment_id rocket
|
||||
env:
|
||||
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
9
App.tsx
9
App.tsx
@@ -94,6 +94,15 @@ function App() {
|
||||
}
|
||||
};
|
||||
|
||||
const testFunction = () => {
|
||||
const newArray: number[] = [];
|
||||
[1, 2, 3, 4, 5].map(item => {
|
||||
newArray.push(item * 2);
|
||||
});
|
||||
return newArray;
|
||||
};
|
||||
|
||||
|
||||
const getActiveRouteName = (state) => {
|
||||
if (!state || typeof state.index !== 'number') {
|
||||
return 'Unknown';
|
||||
|
||||
242
scripts/pr-review.py
Normal file
242
scripts/pr-review.py
Normal file
@@ -0,0 +1,242 @@
|
||||
#!/usr/bin/env python
|
||||
# coding: utf-8
|
||||
|
||||
# In[12]:
|
||||
|
||||
|
||||
import os
|
||||
import re
|
||||
import base64
|
||||
import sys
|
||||
import subprocess
|
||||
import concurrent.futures
|
||||
from typing import List, Tuple
|
||||
from litellm import completion
|
||||
from tenacity import retry, wait_exponential, stop_after_attempt
|
||||
|
||||
|
||||
# In[13]:
|
||||
OUTPUT_FILE = 'code_review_output.txt'
|
||||
DEFAULT_MAX_WORKERS = 3
|
||||
|
||||
# Function to check if the current directory is a git repository root
|
||||
def is_git_repository_root(directory: str) -> bool:
|
||||
return os.path.isdir(os.path.join(directory, '.git'))
|
||||
|
||||
def install_packages_from_file(filename: str):
|
||||
try:
|
||||
with open(filename, 'r') as f:
|
||||
for line in f:
|
||||
package_name = line.strip()
|
||||
if package_name:
|
||||
subprocess.check_call([sys.executable, '-m', 'pip', 'install', package_name, '--quiet'], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
|
||||
except Exception as e:
|
||||
print("")
|
||||
|
||||
|
||||
def extract_code(directory: str) -> List[Tuple[str, str]]:
|
||||
code_snippets = []
|
||||
include_terms = {'service', 'controller', 'listener', 'scheduler', 'utils', 'client', 'repository', 'dao'}
|
||||
for root, dirs, files in os.walk(directory):
|
||||
# Skip test directories
|
||||
if 'test' in root.lower():
|
||||
continue
|
||||
|
||||
for file in files:
|
||||
if file.endswith(('.java', '.kt')):
|
||||
file_path = os.path.join(root, file)
|
||||
# Only include directories with include terms
|
||||
if not any(term in file_path.lower() for term in include_terms):
|
||||
continue
|
||||
with open(file_path, 'r', encoding='utf-8') as f:
|
||||
code = f.read()
|
||||
# Remove import statements
|
||||
code = re.sub(r'^import .*$', '', code, flags=re.MULTILINE)
|
||||
code_snippets.append((file_path, code))
|
||||
return code_snippets
|
||||
|
||||
# Function to classify code snippets
|
||||
def classify_code(code_snippets: List[Tuple[str, str]]) -> List[Tuple[str, str, str]]:
|
||||
classified_code = []
|
||||
for file_path, code in code_snippets:
|
||||
class_type = 'Unknown'
|
||||
if 'Service' in file_path:
|
||||
class_type = 'Service'
|
||||
elif 'Client' in file_path:
|
||||
class_type = 'Client'
|
||||
elif 'Controller' in file_path:
|
||||
class_type = 'Controller'
|
||||
elif 'Listener' in file_path:
|
||||
class_type = 'Listener'
|
||||
elif 'Util' in file_path:
|
||||
class_type = 'Util'
|
||||
classified_code.append((file_path, code, class_type))
|
||||
return classified_code
|
||||
|
||||
|
||||
# In[15]:
|
||||
|
||||
def create_system_prompt() -> str:
|
||||
encoded_prompt = os.getenv('ENCODED_CODE_REVIEW_SYSTEM_PROMPT', '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')
|
||||
decoded_bytes = base64.b64decode(encoded_prompt)
|
||||
system_prompt = decoded_bytes.decode('utf-8')
|
||||
return system_prompt
|
||||
|
||||
|
||||
def create_gpt_prompt(file_path: str, code: str, class_type: str) -> str:
|
||||
encoded_prompt = os.getenv('ENCODED_CODE_REVIEW_USER_PROMPT', 'Q29kZToKe2NvZGV9')
|
||||
decoded_bytes = base64.b64decode(encoded_prompt)
|
||||
user_prompt = decoded_bytes.decode('utf-8')
|
||||
filled_prompt = user_prompt.format(file_path=file_path, class_type=class_type, code=code)
|
||||
return filled_prompt
|
||||
|
||||
|
||||
# LLM Utils
|
||||
# In[17]:
|
||||
@retry(wait=wait_exponential(multiplier=2, min=5, max=30), stop=stop_after_attempt(3))
|
||||
def call_gpt_model(prompt: str) -> str:
|
||||
try:
|
||||
response = completion(
|
||||
model=os.getenv('GPT_MODEL_NAME'),
|
||||
messages=[
|
||||
{
|
||||
"role": "system",
|
||||
"content": create_system_prompt(),
|
||||
},
|
||||
{
|
||||
"role": "user",
|
||||
"content": prompt,
|
||||
}
|
||||
],
|
||||
)
|
||||
return response['choices'][0]['message']['content'].strip()
|
||||
except Exception as e:
|
||||
print(f"An error occurred: {e}")
|
||||
raise
|
||||
|
||||
# In[16]:
|
||||
def get_pr_diff(directory: str, base_branch: str) -> List[str]:
|
||||
# Change current working directory to the provided directory
|
||||
try:
|
||||
original_dir = os.getcwd()
|
||||
os.chdir(directory)
|
||||
|
||||
# Ensure base branch is up-to-date
|
||||
subprocess.check_call(['git', 'fetch', 'origin', base_branch])
|
||||
|
||||
# Get the diff output between origin/base_branch and HEAD
|
||||
diff_output = subprocess.check_output(['git', 'diff', f'origin/{base_branch}...HEAD'], text=True)
|
||||
|
||||
return diff_output.splitlines()
|
||||
|
||||
except subprocess.CalledProcessError as e:
|
||||
print(f"Error: {e}")
|
||||
return []
|
||||
|
||||
finally:
|
||||
# Restore original working directory
|
||||
os.chdir(original_dir)
|
||||
|
||||
# Function to extract code snippets from PR diff
|
||||
def extract_code_from_diff(diff_lines: List[str]) -> List[Tuple[str, str]]:
|
||||
code_snippets = []
|
||||
file_path = ""
|
||||
code = ""
|
||||
for line in diff_lines:
|
||||
if line.startswith('+++ b/'):
|
||||
if file_path and code:
|
||||
code_snippets.append((file_path, code))
|
||||
code = ""
|
||||
file_path = line[6:]
|
||||
elif line.startswith('+') and not line.startswith('++'):
|
||||
code += line[1:] + '\n'
|
||||
if file_path and code:
|
||||
code_snippets.append((file_path, code))
|
||||
return code_snippets
|
||||
|
||||
|
||||
def generate_review_for_file(file_path: str, code: str, class_type: str) -> str:
|
||||
try:
|
||||
print("Generating review for file path ", file_path)
|
||||
prompt = create_gpt_prompt(file_path, code, class_type)
|
||||
gpt_response = call_gpt_model(prompt)
|
||||
if gpt_response.strip() and "no major issues found" not in gpt_response.lower():
|
||||
return gpt_response
|
||||
except Exception as e:
|
||||
print(f"Failed to generate review for file {file_path}: {e}")
|
||||
return ""
|
||||
|
||||
|
||||
def generate_reviews(classified_code: List[Tuple[str, str, str]], output_file: str, max_workers) -> List[str]:
|
||||
reviews = []
|
||||
with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
|
||||
future_to_file = {
|
||||
executor.submit(generate_review_for_file, file_path, code, class_type): (file_path, code, class_type)
|
||||
for file_path, code, class_type in classified_code
|
||||
}
|
||||
|
||||
for future in concurrent.futures.as_completed(future_to_file):
|
||||
file_path, code, class_type = future_to_file[future]
|
||||
try:
|
||||
gpt_response = future.result()
|
||||
if gpt_response:
|
||||
reviews.append(gpt_response)
|
||||
with open(output_file, 'a') as f:
|
||||
f.write(gpt_response)
|
||||
f.write("\n---*---\n\n") # Separator
|
||||
except Exception as e:
|
||||
print(f"Failed to generate review for file {file_path}: {e}")
|
||||
|
||||
return reviews
|
||||
|
||||
# In[20]:
|
||||
# Modify main function to support PR review and full codebase review
|
||||
def run_analysis(directory, output_file, max_workers):
|
||||
review_mode = os.getenv('REVIEW_MODE', 'pr') # Default to full codebase review
|
||||
if not is_git_repository_root(directory):
|
||||
response = "Failure :: This script must be run at the root of a Git repository."
|
||||
with open(output_file, 'a') as f:
|
||||
f.write(response)
|
||||
f.write("\n---*---\n") # Separator
|
||||
return
|
||||
|
||||
print("Running in mode : ", review_mode)
|
||||
if review_mode == 'pr':
|
||||
base_branch = os.getenv('BASE_BRANCH', 'master')
|
||||
diff_lines = get_pr_diff(directory, base_branch)
|
||||
print("diff ", diff_lines)
|
||||
code_snippets = extract_code_from_diff(diff_lines)
|
||||
else:
|
||||
code_snippets = extract_code(directory)
|
||||
|
||||
print("Identified code snippets list of size ", len(code_snippets))
|
||||
classified_code = classify_code(code_snippets)
|
||||
print("Classified code snippets list of size ", len(classified_code))
|
||||
report = generate_reviews(classified_code, output_file, max_workers)
|
||||
print(report)
|
||||
|
||||
|
||||
# In[21]:
|
||||
|
||||
if __name__ == "__main__":
|
||||
# Environment Variables
|
||||
# Set the following Keys for the Script to Run
|
||||
# REVIEW_MODE - full/pr
|
||||
# GPT_MODEL_NAME
|
||||
# Based on GPT Model - Set the credentials
|
||||
# OPENAI_API_KEY
|
||||
# AZURE_API_KEY
|
||||
# AZURE_API_BASE
|
||||
# AZURE_API_VERSION
|
||||
|
||||
if os.path.exists(OUTPUT_FILE):
|
||||
os.remove(OUTPUT_FILE)
|
||||
if len(sys.argv) != 2:
|
||||
response = "Usage: python generic_code_review directory"
|
||||
with open(OUTPUT_FILE, 'a') as f:
|
||||
f.write(response)
|
||||
f.write("\n---*---\n") # Separator
|
||||
else:
|
||||
directory = sys.argv[1]
|
||||
install_packages_from_file('requirements.txt')
|
||||
run_analysis(directory, OUTPUT_FILE, os.getenv('CODE_CRITIC_MAX_WORKERS', DEFAULT_MAX_WORKERS))
|
||||
Reference in New Issue
Block a user