Val Town Code SearchReturn to Val Town

88Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.
89
90### OpenAI
91
92```ts
93import { OpenAI } from "https://esm.town/v/std/openai";
94const openai = new OpenAI();
95const completion = await openai.chat.completions.create({
96  messages: [
97    { role: "user", content: "Say hello in a creative way" },

94Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.
95
96### OpenAI
97
98```ts
99import { OpenAI } from "https://esm.town/v/std/openai";
100const openai = new OpenAI();
101const completion = await openai.chat.completions.create({
102  messages: [
103    { role: "user", content: "Say hello in a creative way" },

1import { OpenAI } from "https://esm.town/v/std/openai";
2import { Video, ClipSuggestion, KeyframeInfo, TranscriptSegment } from '../../shared/types.ts';
3
4// Initialize OpenAI client
5const openai = new OpenAI();
6
7/**
25    
26    // In a real implementation, we would analyze the actual video content
27    // For this demo, we'll simulate video analysis using OpenAI
28    
29    // First, check if we have transcript data
70async function simulateTranscriptGeneration(video: Video): Promise<TranscriptSegment[]> {
71  // In a real implementation, we would use a speech-to-text service
72  // For this demo, we'll generate a fake transcript using OpenAI
73  
74  const prompt = `Generate a realistic transcript for a ${video.duration || 300} second video titled "${video.title}".
78Format the response as a JSON array of objects with 'start', 'end', and 'text' properties.`;
79
80  const completion = await openai.chat.completions.create({
81    model: "gpt-4o-mini",
82    messages: [{ role: "user", content: prompt }],
114): Promise<KeyframeInfo[]> {
115  // In a real implementation, we would analyze video frames and audio
116  // For this demo, we'll use OpenAI to identify potential key moments from the transcript
117  
118  const transcriptText = transcript.map(segment => segment.text).join(" ");
129Format the response as a JSON array of objects with 'time', 'confidence', and 'description' properties.`;
130
131  const completion = await openai.chat.completions.create({
132    model: "gpt-4o-mini",
133    messages: [{ role: "user", content: prompt }],
254): Promise<string> {
255  // In a real implementation, we would use more sophisticated NLP
256  // For this demo, we'll use OpenAI to generate a title
257  
258  const transcriptText = relevantTranscript.map(segment => segment.text).join(" ");
267The title should be attention-grabbing and optimized for social media engagement.`;
268
269  const completion = await openai.chat.completions.create({
270    model: "gpt-4o-mini",
271    messages: [{ role: "user", content: prompt }],
286): Promise<string> {
287  // In a real implementation, we would use more sophisticated NLP
288  // For this demo, we'll use OpenAI to generate a description
289  
290  const transcriptText = relevantTranscript.map(segment => segment.text).join(" ");
301Include relevant hashtags if appropriate.`;
302
303  const completion = await openai.chat.completions.create({
304    model: "gpt-4o-mini",
305    messages: [{ role: "user", content: prompt }],

36- **Backend**: Hono (API framework)
37- **Database**: SQLite
38- **AI Services**: OpenAI for content analysis, external video processing APIs
39- **Authentication**: JWT-based auth system
40- **Storage**: Val Town blob storage for metadata, external storage for video files

1import { OpenAI } from "https://esm.town/v/std/openai";
2import { Resume, JobRequirement, ResumeScore } from "../shared/types.ts";
3
4// Initialize OpenAI client
5const openai = new OpenAI();
6
7/**
28  `;
29
30  const response = await openai.chat.completions.create({
31    model: "gpt-4o-mini",
32    messages: [{ role: "user", content: prompt }],
48
49/**
50 * Calculate similarity score between two texts using OpenAI embeddings
51 */
52export async function calculateSimilarity(text1: string, text2: string): Promise<number> {
53  const response = await openai.embeddings.create({
54    model: "text-embedding-3-small",
55    input: [text1, text2],

15- **Frontend**: HTML, CSS (Tailwind), and TypeScript with React
16- **Database**: SQLite for storing resumes and job requirements
17- **NLP/ML**: OpenAI embeddings for semantic matching
18
19## Project Structure

1# Small  subset of OpenAI API
2
3>![WARNING]
9
10```bash
11curl 'https://www.val.town/v/YOUR_USER_NAME/openai_api' \
12  -H 'Content-Type: application/json' \
13  -X POST \
26
27```js
28const response = await fetch('https://www.val.town/v/YOUR_USER_NAME/openai_api', {
29  method: 'POST',
30  headers: {

1import Ajv from "https://esm.sh/ajv@8.13.0";
2import { OpenAI } from "https://esm.town/v/std/openai";
3
4const openai = new OpenAI();
5const sharedHeaders = {
6	"Access-Control-Allow-Origin": "*",
62      </head>
63      <body>
64          <h1>Small subset of OpenAI API</h1>
65        
66        <blockquote>
75        
76
77        <pre><code>curl 'https://www.val.town/v/YOUR_USER_NAME/openai_api' 
78  -H 'Content-Type: application/json' 
79  -X POST 
89        <p>or in JS</p>
90
91        <pre><code>const response = await fetch('https://www.val.town/v/YOUR_USER_NAME/openai_api', {
92  method: 'POST',
93  headers: {
145	}
146
147	const completion = await openai.chat.completions.create({
148		...body,
149		model: "gpt-3.5-turbo",

88Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.
89
90### OpenAI
91
92```ts
93import { OpenAI } from "https://esm.town/v/std/openai";
94const openai = new OpenAI();
95const completion = await openai.chat.completions.create({
96  messages: [
97    { role: "user", content: "Say hello in a creative way" },

94Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.
95
96### OpenAI
97
98```ts
99import { OpenAI } from "https://esm.town/v/std/openai";
100const openai = new OpenAI();
101const completion = await openai.chat.completions.create({
102  messages: [
103    { role: "user", content: "Say hello in a creative way" },