Initialize v2 work
This commit is contained in:
Fatih Kadir Akın
@@ -1,3 +1 @@
|
||||
# These are supported funding model platforms
|
||||
|
||||
github: [f]
|
||||
|
||||
@@ -1,20 +0,0 @@
|
||||
# Add New Prompt
|
||||
|
||||
You'll need to add your prompt into README.md, and to the `prompts.csv` file. If
|
||||
your prompt includes quotes, you will need to double-quote them to escape in CSV
|
||||
file.
|
||||
|
||||
If the prompt is generated by AI, please add `<mark>Generated by AI</mark>` to
|
||||
the end of the contribution line.
|
||||
|
||||
- [ ] I've confirmed the prompt works well
|
||||
- [ ] I've added
|
||||
`Contributed by: [@yourusername](https://github.com/yourusername)`
|
||||
- [ ] I've added to the README.md
|
||||
- [ ] I've added to the `prompts.csv`
|
||||
- [ ] Escaped quotes by double-quoting them
|
||||
- [ ] No spaces after commas after double quotes. e.g. `"Hello","hi"`, not
|
||||
`"Hello", "hi"`
|
||||
- [ ] Removed "Act as" from the title on CSV
|
||||
|
||||
Please make sure you've completed all the checklist.
|
||||
@@ -1,623 +0,0 @@
|
||||
name: AI Bot
|
||||
|
||||
on:
|
||||
issues:
|
||||
types: [opened]
|
||||
issue_comment:
|
||||
types: [created]
|
||||
pull_request_review_comment:
|
||||
types: [created]
|
||||
pull_request:
|
||||
types: [opened, edited, synchronize]
|
||||
|
||||
jobs:
|
||||
respond-to-commands:
|
||||
runs-on: ubuntu-latest
|
||||
if: |
|
||||
(github.actor == 'f') &&
|
||||
((github.event_name == 'issues' && contains(github.event.issue.body, '/ai')) ||
|
||||
(github.event_name == 'issue_comment' && contains(github.event.comment.body, '/ai')) ||
|
||||
(github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '/ai')) ||
|
||||
(github.event_name == 'pull_request' && contains(github.event.pull_request.body, '/ai')))
|
||||
permissions:
|
||||
contents: write
|
||||
pull-requests: write
|
||||
issues: write
|
||||
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
with:
|
||||
fetch-depth: 0
|
||||
token: ${{ secrets.PAT_TOKEN }}
|
||||
|
||||
- name: Setup Node.js
|
||||
uses: actions/setup-node@v3
|
||||
with:
|
||||
node-version: "18"
|
||||
|
||||
- name: Install dependencies
|
||||
run: npm install openai@^4.0.0 @octokit/rest@^19.0.0
|
||||
|
||||
- name: Process command
|
||||
id: process
|
||||
env:
|
||||
GH_TOKEN: ${{ secrets.PAT_TOKEN }}
|
||||
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
|
||||
run: |
|
||||
node << 'EOF'
|
||||
const OpenAI = require('openai');
|
||||
const { Octokit } = require('@octokit/rest');
|
||||
|
||||
async function main() {
|
||||
const openai = new OpenAI({
|
||||
apiKey: process.env.OPENAI_API_KEY
|
||||
});
|
||||
|
||||
const octokit = new Octokit({
|
||||
auth: process.env.GH_TOKEN
|
||||
});
|
||||
|
||||
const eventName = process.env.GITHUB_EVENT_NAME;
|
||||
const eventPath = process.env.GITHUB_EVENT_PATH;
|
||||
const event = require(eventPath);
|
||||
|
||||
// Double check user authorization
|
||||
const actor = event.sender?.login || event.pull_request?.user?.login || event.issue?.user?.login;
|
||||
if (actor !== 'f') {
|
||||
console.log('Unauthorized user attempted to use the bot:', actor);
|
||||
return;
|
||||
}
|
||||
|
||||
// Get command and context
|
||||
let command = '';
|
||||
let issueNumber = null;
|
||||
let isPullRequest = false;
|
||||
|
||||
if (eventName === 'issues') {
|
||||
command = event.issue.body;
|
||||
issueNumber = event.issue.number;
|
||||
} else if (eventName === 'issue_comment') {
|
||||
command = event.comment.body;
|
||||
issueNumber = event.issue.number;
|
||||
isPullRequest = !!event.issue.pull_request;
|
||||
} else if (eventName === 'pull_request_review_comment') {
|
||||
command = event.comment.body;
|
||||
issueNumber = event.pull_request.number;
|
||||
isPullRequest = true;
|
||||
} else if (eventName === 'pull_request') {
|
||||
command = event.pull_request.body;
|
||||
issueNumber = event.pull_request.number;
|
||||
isPullRequest = true;
|
||||
}
|
||||
|
||||
if (!command.startsWith('/ai')) {
|
||||
return;
|
||||
}
|
||||
|
||||
// Extract the actual command after /ai
|
||||
const aiCommand = command.substring(3).trim();
|
||||
|
||||
// Handle resolve conflicts command
|
||||
if (aiCommand === 'resolve' || aiCommand === 'fix conflicts') {
|
||||
if (!isPullRequest) {
|
||||
console.log('Command rejected: Not a pull request');
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: '❌ The resolve command can only be used on pull requests.'
|
||||
});
|
||||
return;
|
||||
}
|
||||
|
||||
try {
|
||||
console.log('Starting resolve command execution...');
|
||||
|
||||
// Get PR details
|
||||
console.log('Fetching PR details...');
|
||||
const { data: pr } = await octokit.pulls.get({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber
|
||||
});
|
||||
console.log(`Original PR found: #${issueNumber} from ${pr.user.login}`);
|
||||
|
||||
// Get the PR diff to extract the new prompt
|
||||
console.log('Fetching PR file changes...');
|
||||
const { data: files } = await octokit.pulls.listFiles({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber
|
||||
});
|
||||
console.log(`Found ${files.length} changed files`);
|
||||
|
||||
// Extract prompt from changes
|
||||
console.log('Analyzing changes to extract prompt information...');
|
||||
const prompts = new Map(); // Use Map to deduplicate prompts by title
|
||||
|
||||
// Helper function to normalize prompt titles
|
||||
const normalizeTitle = (title) => {
|
||||
title = title.trim();
|
||||
// Remove "Act as" or "Act as a" or "Act as an" from start if present
|
||||
title = title.replace(/^Act as (?:a |an )?/i, '');
|
||||
|
||||
// Capitalize each word except common articles and prepositions
|
||||
const lowercaseWords = ['a', 'an', 'the', 'and', 'but', 'or', 'for', 'nor', 'on', 'at', 'to', 'for', 'with', 'in'];
|
||||
const capitalized = title.toLowerCase().split(' ').map((word, index) => {
|
||||
// Always capitalize first and last word
|
||||
if (index === 0 || !lowercaseWords.includes(word)) {
|
||||
return word.charAt(0).toUpperCase() + word.slice(1);
|
||||
}
|
||||
return word;
|
||||
}).join(' ');
|
||||
|
||||
// Add "Act as" prefix
|
||||
return `Act as ${capitalized}`;
|
||||
};
|
||||
|
||||
// First try to find prompts in README
|
||||
let foundInReadme = false;
|
||||
for (const file of files) {
|
||||
console.log(`Processing file: ${file.filename}`);
|
||||
if (file.filename === 'README.md') {
|
||||
const patch = file.patch || '';
|
||||
const addedLines = patch.split('\n')
|
||||
.filter(line => line.startsWith('+'))
|
||||
.map(line => line.substring(1))
|
||||
.join('\n');
|
||||
|
||||
console.log('Attempting to extract prompts from README changes...');
|
||||
const promptMatches = [...addedLines.matchAll(/## (?:Act as (?:a |an )?)?([^\n]+)\n(?:Contributed by:[^\n]*\n)?(?:> )?([^#]+?)(?=\n##|\n\n##|$)/ig)];
|
||||
|
||||
for (const match of promptMatches) {
|
||||
const actName = normalizeTitle(match[1]);
|
||||
const promptText = match[2].trim()
|
||||
.replace(/^(?:Contributed by:?[^\n]*\n\s*)+/i, '')
|
||||
.trim();
|
||||
const contributorLine = addedLines.match(/Contributed by: \[@([^\]]+)\]\(https:\/\/github\.com\/([^\)]+)\)/);
|
||||
const contributorInfo = contributorLine
|
||||
? `Contributed by: [@${contributorLine[1]}](https://github.com/${contributorLine[2]})`
|
||||
: `Contributed by: [@${pr.user.login}](https://github.com/${pr.user.login})`;
|
||||
|
||||
prompts.set(actName.toLowerCase(), { actName, promptText, contributorInfo });
|
||||
console.log(`Found prompt in README: "${actName}"`);
|
||||
foundInReadme = true;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Only look in CSV if we didn't find anything in README
|
||||
if (!foundInReadme) {
|
||||
console.log('No prompts found in README, checking CSV...');
|
||||
for (const file of files) {
|
||||
if (file.filename === 'prompts.csv') {
|
||||
const patch = file.patch || '';
|
||||
const addedLines = patch.split('\n')
|
||||
.filter(line => line.startsWith('+'))
|
||||
.map(line => line.substring(1))
|
||||
.filter(line => line.trim()); // Remove empty lines
|
||||
|
||||
console.log('Attempting to extract prompts from CSV changes...');
|
||||
for (const line of addedLines) {
|
||||
// Parse CSV line considering escaped quotes
|
||||
const matches = [...line.matchAll(/"([^"]*(?:""[^"]*)*)"/g)];
|
||||
if (matches.length >= 2) {
|
||||
const actName = normalizeTitle(matches[0][1].replace(/""/g, '"').trim());
|
||||
const promptText = matches[1][1].replace(/""/g, '"').trim()
|
||||
.replace(/^(?:Contributed by:?[^\n]*\n\s*)+/i, '')
|
||||
.trim();
|
||||
|
||||
const contributorInfo = `Contributed by: [@${pr.user.login}](https://github.com/${pr.user.login})`;
|
||||
prompts.set(actName.toLowerCase(), { actName, promptText, contributorInfo });
|
||||
console.log(`Found prompt in CSV: "${actName}"`);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if (prompts.size === 0) {
|
||||
console.log('Failed to extract prompt information');
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: '❌ Could not extract prompt information from changes'
|
||||
});
|
||||
return;
|
||||
}
|
||||
|
||||
// Get content from main branch
|
||||
console.log('Fetching current content from main branch...');
|
||||
const { data: readmeFile } = await octokit.repos.getContent({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'README.md',
|
||||
ref: 'main'
|
||||
});
|
||||
|
||||
const { data: csvFile } = await octokit.repos.getContent({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'prompts.csv',
|
||||
ref: 'main'
|
||||
});
|
||||
|
||||
// Prepare new content
|
||||
console.log('Preparing content updates...');
|
||||
let readmeContent = Buffer.from(readmeFile.content, 'base64').toString('utf-8');
|
||||
let csvContent = Buffer.from(csvFile.content, 'base64').toString('utf-8');
|
||||
if (!csvContent.endsWith('\n')) csvContent += '\n';
|
||||
|
||||
// Convert Map to array for processing
|
||||
const promptsArray = Array.from(prompts.values());
|
||||
|
||||
// Process each prompt
|
||||
for (const { actName, promptText, contributorInfo } of promptsArray) {
|
||||
// Remove markdown quote character and trim whitespace
|
||||
const cleanPrompt = promptText.replace(/^>\s*/gm, '').trim();
|
||||
|
||||
// For README: Add quote to each line
|
||||
const readmePrompt = cleanPrompt.split('\n')
|
||||
.map(line => `> ${line.trim()}`)
|
||||
.join('\n');
|
||||
const newSection = `## ${actName}\n${contributorInfo}\n\n${readmePrompt}\n\n`;
|
||||
|
||||
// For CSV: Convert to single paragraph
|
||||
const csvPrompt = cleanPrompt.replace(/\n+/g, ' ').trim();
|
||||
|
||||
// Insert the new section before Contributors in README
|
||||
const contributorsIndex = readmeContent.indexOf('## Contributors');
|
||||
if (contributorsIndex === -1) {
|
||||
console.log('Contributors section not found, appending to end');
|
||||
readmeContent += newSection;
|
||||
} else {
|
||||
console.log('Inserting before Contributors section');
|
||||
readmeContent = readmeContent.slice(0, contributorsIndex) + newSection + readmeContent.slice(contributorsIndex);
|
||||
}
|
||||
|
||||
// Add to CSV content
|
||||
csvContent += `"${actName.replace(/^Act as an?/i, '').replace(/"/g, '""')}","${csvPrompt.replace(/"/g, '""')}"\n`;
|
||||
}
|
||||
|
||||
// Create new branch
|
||||
const branchName = `prompt/${promptsArray.map(p => p.actName.toLowerCase().replace(/[^a-z0-9]+/g, '-')).join('-')}`;
|
||||
console.log(`Creating new branch: ${branchName}`);
|
||||
|
||||
// Check if branch exists and delete it
|
||||
try {
|
||||
console.log('Checking if branch already exists...');
|
||||
const { data: existingRef } = await octokit.git.getRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: `heads/${branchName}`
|
||||
});
|
||||
|
||||
if (existingRef) {
|
||||
// Check for existing PRs from this branch
|
||||
console.log('Checking for existing PRs from this branch...');
|
||||
const { data: existingPRs } = await octokit.pulls.list({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
head: `${event.repository.owner.login}:${branchName}`,
|
||||
state: 'open'
|
||||
});
|
||||
|
||||
// Close any existing PRs
|
||||
for (const pr of existingPRs) {
|
||||
console.log(`Closing existing PR #${pr.number}...`);
|
||||
await octokit.pulls.update({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: pr.number,
|
||||
state: 'closed'
|
||||
});
|
||||
}
|
||||
|
||||
console.log('Branch exists, deleting it...');
|
||||
await octokit.git.deleteRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: `heads/${branchName}`
|
||||
});
|
||||
console.log('Existing branch deleted');
|
||||
}
|
||||
} catch (error) {
|
||||
// 404 means branch doesn't exist, which is fine
|
||||
if (error.status !== 404) {
|
||||
throw error;
|
||||
}
|
||||
console.log('Branch does not exist, proceeding with creation');
|
||||
}
|
||||
|
||||
// Get main branch ref
|
||||
const { data: mainRef } = await octokit.git.getRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: 'heads/main'
|
||||
});
|
||||
|
||||
// Create new branch
|
||||
await octokit.git.createRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: `refs/heads/${branchName}`,
|
||||
sha: mainRef.object.sha
|
||||
});
|
||||
|
||||
// Get current files from the new branch
|
||||
console.log('Getting current file SHAs...');
|
||||
const { data: currentReadme } = await octokit.repos.getContent({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'README.md',
|
||||
ref: branchName
|
||||
});
|
||||
|
||||
const { data: currentCsv } = await octokit.repos.getContent({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'prompts.csv',
|
||||
ref: branchName
|
||||
});
|
||||
|
||||
// Update files with correct author
|
||||
console.log('Updating README.md...');
|
||||
await octokit.repos.createOrUpdateFileContents({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'README.md',
|
||||
message: promptsArray.length === 1
|
||||
? `feat: Add "${promptsArray[0].actName}" to README`
|
||||
: `feat: Add multiple prompts to README`,
|
||||
content: Buffer.from(readmeContent).toString('base64'),
|
||||
branch: branchName,
|
||||
sha: currentReadme.sha,
|
||||
committer: {
|
||||
name: pr.user.login,
|
||||
email: `${pr.user.login}@users.noreply.github.com`
|
||||
},
|
||||
author: {
|
||||
name: pr.user.login,
|
||||
email: `${pr.user.login}@users.noreply.github.com`
|
||||
}
|
||||
});
|
||||
|
||||
console.log('Updating prompts.csv...');
|
||||
await octokit.repos.createOrUpdateFileContents({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: 'prompts.csv',
|
||||
message: promptsArray.length === 1
|
||||
? `feat: Add "${promptsArray[0].actName}" to prompts.csv`
|
||||
: `feat: Add multiple prompts to prompts.csv`,
|
||||
content: Buffer.from(csvContent).toString('base64'),
|
||||
branch: branchName,
|
||||
sha: currentCsv.sha,
|
||||
committer: {
|
||||
name: pr.user.login,
|
||||
email: `${pr.user.login}@users.noreply.github.com`
|
||||
},
|
||||
author: {
|
||||
name: pr.user.login,
|
||||
email: `${pr.user.login}@users.noreply.github.com`
|
||||
}
|
||||
});
|
||||
|
||||
// Create new PR
|
||||
const prTitle = promptsArray.length === 1
|
||||
? `feat: Add "${promptsArray[0].actName}"`
|
||||
: `feat: Add multiple prompts (${promptsArray.map(p => `"${p.actName}"`).join(', ')})`;
|
||||
|
||||
const prBody = promptsArray.length === 1
|
||||
? `This PR supersedes #${issueNumber} with proper formatting. Original PR by @${pr.user.login}. Added "${promptsArray[0].actName}" to README.md and prompts.csv, preserving original attribution.`
|
||||
: `This PR supersedes #${issueNumber} with proper formatting. Original PR by @${pr.user.login}.\n\nAdded the following prompts:\n${promptsArray.map(p => `- "${p.actName}"`).join('\n')}\n\nAll prompts have been added to README.md and prompts.csv, preserving original attribution.`;
|
||||
|
||||
const { data: newPr } = await octokit.pulls.create({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
title: prTitle,
|
||||
head: branchName,
|
||||
base: 'main',
|
||||
body: prBody
|
||||
});
|
||||
|
||||
// Comment on original PR
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: `I've created a new PR #${newPr.number} with your contribution properly formatted. This PR will be closed in favor of the new one.`
|
||||
});
|
||||
|
||||
// Close original PR
|
||||
await octokit.pulls.update({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber,
|
||||
state: 'closed'
|
||||
});
|
||||
|
||||
console.log(`Created new PR #${newPr.number} and closed original PR #${issueNumber}`);
|
||||
|
||||
} catch (error) {
|
||||
console.error('Error details:', error);
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: `❌ Error while trying to create new PR:\n\`\`\`\n${error.message}\n\`\`\``
|
||||
});
|
||||
}
|
||||
return;
|
||||
}
|
||||
|
||||
// Handle rename command specifically
|
||||
if (aiCommand.startsWith('rename') || aiCommand === 'suggest title') {
|
||||
if (!isPullRequest) {
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: '❌ The rename command can only be used on pull requests.'
|
||||
});
|
||||
return;
|
||||
}
|
||||
|
||||
// Get PR details for context
|
||||
const { data: pr } = await octokit.pulls.get({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber
|
||||
});
|
||||
|
||||
// Get the list of files changed in the PR
|
||||
const { data: files } = await octokit.pulls.listFiles({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber
|
||||
});
|
||||
|
||||
// Process file changes
|
||||
const fileChanges = await Promise.all(files.map(async file => {
|
||||
if (file.status === 'removed') {
|
||||
return `Deleted: ${file.filename}`;
|
||||
}
|
||||
|
||||
// Get file content for added or modified files
|
||||
if (file.status === 'added' || file.status === 'modified') {
|
||||
const patch = file.patch || '';
|
||||
return `${file.status === 'added' ? 'Added' : 'Modified'}: ${file.filename}\nChanges:\n${patch}`;
|
||||
}
|
||||
|
||||
return `${file.status}: ${file.filename}`;
|
||||
}));
|
||||
|
||||
const completion = await openai.chat.completions.create({
|
||||
model: "gpt-3.5-turbo",
|
||||
messages: [
|
||||
{
|
||||
role: "system",
|
||||
content: "You are a helpful assistant that generates clear and concise pull request titles. Follow these rules:\n1. Use conventional commit style (feat:, fix:, docs:, etc.)\n2. Focus on WHAT changed, not HOW or WHERE\n3. Keep it short and meaningful\n4. Don't mention file names or technical implementation details\n5. Return ONLY the new title, nothing else\n\nGood examples:\n- feat: Add \"Act as a Career Coach\"\n- fix: Correct typo in Linux Terminal prompt\n- docs: Update installation instructions\n- refactor: Improve error handling"
|
||||
},
|
||||
{
|
||||
role: "user",
|
||||
content: `Based on these file changes, generate a concise PR title:\n\n${fileChanges.join('\n\n')}`
|
||||
}
|
||||
],
|
||||
temperature: 0.5,
|
||||
max_tokens: 60
|
||||
});
|
||||
|
||||
const newTitle = completion.choices[0].message.content.trim();
|
||||
|
||||
// Update PR title
|
||||
await octokit.pulls.update({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
pull_number: issueNumber,
|
||||
title: newTitle
|
||||
});
|
||||
|
||||
// Add comment about the rename
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: `✨ Updated PR title to: "${newTitle}"\n\nBased on the following changes:\n\`\`\`diff\n${fileChanges.join('\n')}\n\`\`\``
|
||||
});
|
||||
return;
|
||||
}
|
||||
|
||||
// Handle other commands
|
||||
const completion = await openai.chat.completions.create({
|
||||
model: "gpt-3.5-turbo",
|
||||
messages: [
|
||||
{
|
||||
role: "system",
|
||||
content: "You are a helpful AI assistant that helps with GitHub repositories. You can suggest code changes, fix issues, and improve code quality."
|
||||
},
|
||||
{
|
||||
role: "user",
|
||||
content: aiCommand
|
||||
}
|
||||
],
|
||||
temperature: 0.7,
|
||||
max_tokens: 2000
|
||||
});
|
||||
|
||||
const response = completion.choices[0].message.content;
|
||||
|
||||
// If response contains code changes, create a new branch and PR
|
||||
if (response.includes('```')) {
|
||||
const branchName = `ai-bot/fix-${issueNumber}`;
|
||||
|
||||
// Create new branch
|
||||
const defaultBranch = event.repository.default_branch;
|
||||
const ref = await octokit.git.getRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: `heads/${defaultBranch}`
|
||||
});
|
||||
|
||||
await octokit.git.createRef({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
ref: `refs/heads/${branchName}`,
|
||||
sha: ref.data.object.sha
|
||||
});
|
||||
|
||||
// Extract code changes and file paths from response
|
||||
const codeBlocks = response.match(/```[\s\S]*?```/g);
|
||||
for (const block of codeBlocks) {
|
||||
const [_, filePath, ...codeLines] = block.split('\n');
|
||||
const content = Buffer.from(codeLines.join('\n')).toString('base64');
|
||||
|
||||
await octokit.repos.createOrUpdateFileContents({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
path: filePath,
|
||||
message: `AI Bot: Apply suggested changes for #${issueNumber}`,
|
||||
content,
|
||||
branch: branchName
|
||||
});
|
||||
}
|
||||
|
||||
// Create PR
|
||||
await octokit.pulls.create({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
title: `AI Bot: Fix for #${issueNumber}`,
|
||||
body: `This PR was automatically generated in response to #${issueNumber}\n\nChanges proposed:\n${response}`,
|
||||
head: branchName,
|
||||
base: defaultBranch
|
||||
});
|
||||
}
|
||||
|
||||
// Add comment with response
|
||||
await octokit.issues.createComment({
|
||||
owner: event.repository.owner.login,
|
||||
repo: event.repository.name,
|
||||
issue_number: issueNumber,
|
||||
body: `AI Bot Response:\n\n${response}`
|
||||
});
|
||||
}
|
||||
|
||||
main().catch(error => {
|
||||
console.error('Error:', error);
|
||||
process.exit(1);
|
||||
});
|
||||
EOF
|
||||
|
||||
- name: Handle errors
|
||||
if: failure()
|
||||
uses: actions/github-script@v6
|
||||
with:
|
||||
script: |
|
||||
const issueNumber = context.issue.number || context.payload.pull_request.number;
|
||||
await github.rest.issues.createComment({
|
||||
owner: context.repo.owner,
|
||||
repo: context.repo.repo,
|
||||
issue_number: issueNumber,
|
||||
body: '❌ Sorry, there was an error processing your command. Please try again or contact the repository maintainers.'
|
||||
});
|
||||
@@ -1,43 +0,0 @@
|
||||
name: Auto AI Commands
|
||||
|
||||
on:
|
||||
pull_request:
|
||||
types: [opened, reopened, synchronize]
|
||||
pull_request_target:
|
||||
types: [opened, reopened, synchronize]
|
||||
|
||||
jobs:
|
||||
check-and-comment:
|
||||
runs-on: ubuntu-latest
|
||||
permissions:
|
||||
pull-requests: write
|
||||
issues: write
|
||||
|
||||
steps:
|
||||
- name: Check PR status and comment
|
||||
uses: actions/github-script@v6
|
||||
with:
|
||||
script: |
|
||||
const pr = context.payload.pull_request;
|
||||
|
||||
// Check if PR has conflicts
|
||||
if (pr.mergeable === false) {
|
||||
console.log('PR has conflicts, commenting /ai resolve');
|
||||
await github.rest.issues.createComment({
|
||||
owner: context.repo.owner,
|
||||
repo: context.repo.repo,
|
||||
issue_number: pr.number,
|
||||
body: '/ai resolve'
|
||||
});
|
||||
}
|
||||
|
||||
// Check if PR title starts with "updated"
|
||||
if (pr.title.toLowerCase().startsWith('updated')) {
|
||||
console.log('PR title starts with "updated", commenting /ai suggest title');
|
||||
await github.rest.issues.createComment({
|
||||
owner: context.repo.owner,
|
||||
repo: context.repo.repo,
|
||||
issue_number: pr.number,
|
||||
body: '/ai suggest title'
|
||||
});
|
||||
}
|
||||
@@ -1,82 +0,0 @@
|
||||
name: CSV Linter and Trailing Whitespaces
|
||||
|
||||
on:
|
||||
push:
|
||||
pull_request:
|
||||
|
||||
jobs:
|
||||
lint_and_check_trailing_whitespaces:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
|
||||
- name: Set up Python
|
||||
uses: actions/setup-python@v2
|
||||
with:
|
||||
python-version: "3.8"
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
python -m pip install --upgrade pip
|
||||
python -m pip install csvkit
|
||||
|
||||
- name: Validate CSV structure
|
||||
run: |
|
||||
echo "Checking CSV structure..."
|
||||
if ! csvclean -n prompts.csv 2>&1 | tee /tmp/csv_errors.log; then
|
||||
echo "::error::CSV validation failed"
|
||||
cat /tmp/csv_errors.log
|
||||
exit 1
|
||||
fi
|
||||
|
||||
- name: Check CSV format
|
||||
run: |
|
||||
echo "Checking CSV format..."
|
||||
if ! python -c '
|
||||
import csv
|
||||
with open("prompts.csv", "r", encoding="utf-8") as f:
|
||||
reader = csv.reader(f)
|
||||
headers = next(reader)
|
||||
if headers != ["act", "prompt", "for_devs", "type"]:
|
||||
print("Error: CSV headers must be exactly [act, prompt, for_devs, type]")
|
||||
exit(1)
|
||||
valid_types = ["TEXT", "JSON"]
|
||||
for row_num, row in enumerate(reader, 2):
|
||||
if len(row) != 4:
|
||||
print(f"Error: Row {row_num} has {len(row)} columns, expected 4")
|
||||
exit(1)
|
||||
if not row[0] or not row[1] or not row[2] or not row[3]:
|
||||
print(f"Error: Row {row_num} has empty values")
|
||||
exit(1)
|
||||
if row[3] not in valid_types:
|
||||
print(f"Error: Row {row_num} has invalid type \"{row[3]}\". Must be TEXT or JSON")
|
||||
exit(1)
|
||||
print("CSV format OK")
|
||||
'; then
|
||||
echo "::error::CSV format check failed"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
- name: Check Trailing Whitespaces
|
||||
run: |
|
||||
echo "Checking for trailing whitespaces..."
|
||||
if grep -q "[[:space:]]$" prompts.csv; then
|
||||
echo "::error::Found trailing whitespaces in prompts.csv"
|
||||
grep -n "[[:space:]]$" prompts.csv | while read -r line; do
|
||||
echo "Line with trailing whitespace: $line"
|
||||
done
|
||||
exit 0
|
||||
fi
|
||||
echo "No trailing whitespaces found"
|
||||
|
||||
- name: Check for UTF-8 BOM and line endings
|
||||
run: |
|
||||
echo "Checking for UTF-8 BOM and line endings..."
|
||||
if file prompts.csv | grep -q "with BOM"; then
|
||||
echo "::error::File contains UTF-8 BOM marker"
|
||||
exit 1
|
||||
fi
|
||||
if file prompts.csv | grep -q "CRLF"; then
|
||||
echo "::error::File contains Windows-style (CRLF) line endings"
|
||||
exit 1
|
||||
fi
|
||||
@@ -1,29 +0,0 @@
|
||||
name: Publish to GitHub Pages
|
||||
|
||||
on:
|
||||
push:
|
||||
branches:
|
||||
- main
|
||||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
|
||||
steps:
|
||||
- name: Checkout repository
|
||||
uses: actions/checkout@v3
|
||||
|
||||
- name: Set up Ruby
|
||||
uses: ruby/setup-ruby@v1
|
||||
with:
|
||||
ruby-version: "3.2"
|
||||
bundler-cache: true
|
||||
|
||||
- name: Build site
|
||||
run: bundle exec jekyll build
|
||||
|
||||
- name: Deploy to GitHub Pages
|
||||
uses: peaceiris/actions-gh-pages@v3
|
||||
with:
|
||||
github_token: ${{ secrets.GITHUB_TOKEN }}
|
||||
publish_dir: ./_site
|
||||
Reference in New Issue
Block a user