added browser rendering for getting clean page content

This commit is contained in:
Dhravya 2024-04-13 11:29:02 -07:00
commit 74206a9cf8
9 changed files with 92 additions and 13 deletions

View file

@ -6,6 +6,7 @@ interface Env {
GOOGLE_AI_API_KEY: string;
MY_QUEUE: Queue<TweetData[]>;
KV: KVNamespace;
MYBROWSER: BrowserWorker;
}
interface TweetData {

View file

@ -4,6 +4,7 @@ import * as apiQuery from './routes/query';
import * as apiAsk from './routes/ask';
import * as apiChat from './routes/chat';
import * as apiBatchUploadTweets from './routes/batchUploadTweets';
import * as apiGetPageContent from './routes/getPageContent';
import { OpenAIEmbeddings } from './OpenAIEmbedder';
import { GenerativeModel } from '@google/generative-ai';
import { Request } from '@cloudflare/workers-types';
@ -28,6 +29,7 @@ routeMap.set('/ask', apiAsk);
routeMap.set('/chat', apiChat);
routeMap.set('/batchUploadTweets', apiBatchUploadTweets);
routeMap.set('/getPageContent', apiGetPageContent);
// Add more route mappings as needed
// routeMap.set('/api/otherRoute', { ... });

View file

@ -0,0 +1,31 @@
import { GenerativeModel } from '@google/generative-ai';
import { OpenAIEmbeddings } from '../OpenAIEmbedder';
import { CloudflareVectorizeStore } from '@langchain/cloudflare';
import { Request } from '@cloudflare/workers-types';
import puppeteer from '@cloudflare/puppeteer';
export async function GET(request: Request, _: CloudflareVectorizeStore, embeddings: OpenAIEmbeddings, model: GenerativeModel, env?: Env) {
const { searchParams } = new URL(request.url);
let url = searchParams.get('url');
let img: Buffer;
if (url) {
url = new URL(url).toString(); // normalize
const browser = await puppeteer.launch(env?.MYBROWSER);
const page = await browser.newPage();
await page.goto(url);
// Innertext of content
const contentElement = await page.$('body');
const content = await page.evaluate((element) => element.innerText, contentElement);
await browser.close();
return new Response(content, {
headers: {
'content-type': 'text/html',
},
});
} else {
return new Response('Please add an ?url=https://example.com/ parameter');
}
}

View file

@ -1,6 +1,7 @@
name = "cf-ai-backend"
main = "src/index.ts"
compatibility_date = "2024-02-23"
compatibility_flags = ['nodejs_compat']
[[vectorize]]
binding = "VECTORIZE_INDEX"
@ -19,6 +20,13 @@ binding = "AI"
[[kv_namespaces]]
binding = "KV"
id = "37a90353da63401e84e20e71165531d0"
preview_id = "c58b6202814f4224acea97627d0c18aa"
[browser]
binding = "MYBROWSER"
[placement]
mode = "smart"
# Variable bindings. These are arbitrary, plaintext strings (similar to environment variables)
# Note: Use secrets to store sensitive data.

View file

@ -16,10 +16,12 @@ import { Loader, Plus, X } from "lucide-react";
import { StoredContent } from "@/server/db/schema";
import { cleanUrl } from "@/lib/utils";
import { motion } from "framer-motion";
import { getMetaData } from "@/server/helpers";
export function AddMemoryPage() {
export function AddMemoryPage({ closeDialog }: { closeDialog: () => void }) {
const { addMemory } = useMemory();
const [loading, setLoading] = useState(false);
const [url, setUrl] = useState("");
const [selectedSpacesId, setSelectedSpacesId] = useState<number[]>([]);
@ -37,38 +39,59 @@ export function AddMemoryPage() {
placeholder="Enter the URL of the page"
type="url"
data-modal-autofocus
className="bg-rgray-4 mt-2 w-full"
className="bg-rgray-4 mt-2 w-full disabled:cursor-not-allowed disabled:opacity-70"
value={url}
onChange={(e) => setUrl(e.target.value)}
disabled={loading}
/>
<DialogFooter>
<FilterSpaces
selectedSpaces={selectedSpacesId}
setSelectedSpaces={setSelectedSpacesId}
className="hover:bg-rgray-5 mr-auto bg-white/5"
className="hover:bg-rgray-5 mr-auto bg-white/5 disabled:cursor-not-allowed disabled:opacity-70"
name={"Spaces"}
disabled={loading}
/>
<button
type={"submit"}
disabled={loading}
onClick={async () => {
// @Dhravya this is adding a memory with insufficient information fix pls
setLoading(true);
const metadata = await getMetaData(url);
await addMemory(
{
title: url,
title: metadata.title,
description: metadata.description,
content: "",
type: "page",
url: url,
image: "/icons/logo_without_bg.png",
image: metadata.image,
savedAt: new Date(),
},
selectedSpacesId,
);
closeDialog();
}}
className="bg-rgray-4 hover:bg-rgray-5 focus-visible:bg-rgray-5 focus-visible:ring-rgray-7 rounded-md px-4 py-2 ring-transparent transition focus-visible:outline-none focus-visible:ring-2"
className="bg-rgray-4 hover:bg-rgray-5 focus-visible:bg-rgray-5 focus-visible:ring-rgray-7 relative rounded-md px-4 py-2 ring-transparent transition focus-visible:outline-none focus-visible:ring-2 disabled:cursor-not-allowed disabled:opacity-70"
>
Add
<motion.div
initial={{ x: "-50%", y: "-100%" }}
animate={loading && { y: "-50%", x: "-50%", opacity: 1 }}
className="absolute left-1/2 top-1/2 -translate-x-1/2 translate-y-[-100%] opacity-0"
>
<Loader className="text-rgray-11 h-5 w-5 animate-spin" />
</motion.div>
<motion.div
initial={{ y: "0%" }}
animate={loading && { opacity: 0, y: "30%" }}
>
Add
</motion.div>
</button>
<DialogClose className="hover:bg-rgray-4 focus-visible:bg-rgray-4 focus-visible:ring-rgray-7 rounded-md px-3 py-2 ring-transparent transition focus-visible:outline-none focus-visible:ring-2">
<DialogClose
disabled={loading}
className="hover:bg-rgray-4 focus-visible:bg-rgray-4 focus-visible:ring-rgray-7 rounded-md px-3 py-2 ring-transparent transition focus-visible:outline-none focus-visible:ring-2 disabled:cursor-not-allowed disabled:opacity-70"
>
Cancel
</DialogClose>
</DialogFooter>

View file

@ -210,7 +210,15 @@ export function MemoryItem({ id, title, image, type }: StoredContent) {
<div className="flex h-24 w-24 items-center justify-center">
{type === "page" ? (
<img className="h-16 w-16" id={id.toString()} src={image!} />
<img
className="h-16 w-16"
id={id.toString()}
src={image!}
onError={(e) => {
(e.target as HTMLImageElement).src =
"/icons/white_without_bg.png";
}}
/>
) : type === "note" ? (
<div className="bg-rgray-4 flex items-center justify-center rounded-md p-2 shadow-md">
<Text className="h-10 w-10" />
@ -406,7 +414,7 @@ export function SpaceMoreButton({
className="focus:bg-red-100 focus:text-red-400 dark:focus:bg-red-100/10"
>
<Trash2 className="mr-2 h-4 w-4" strokeWidth={1.5} />
Move to Trash
Delete
</DropdownMenuItem>
</DialogTrigger>
</DropdownMenuContent>
@ -471,7 +479,7 @@ export function AddMemoryModal({
className="w-max max-w-[auto]"
>
{type === "page" ? (
<AddMemoryPage />
<AddMemoryPage closeDialog={() => setIsDialogOpen(false)} />
) : type === "note" ? (
<NoteAddPage closeDialog={() => setIsDialogOpen(false)} />
) : type === "space" ? (

View file

@ -1,3 +1,4 @@
"use server";
import * as cheerio from "cheerio";
export async function getMetaData(url: string) {

View file

@ -17,4 +17,7 @@ type = "ratelimit"
namespace_id = "1001"
# 25 requests per 10 seconds
simple = { limit = 25, period = 10 }
simple = { limit = 25, period = 10 }
[placement]
mode = "smart"

View file

@ -41,6 +41,7 @@
"@auth/drizzle-adapter": "^0.7.0",
"@cloudflare/ai": "^1.0.52",
"@cloudflare/next-on-pages-next-dev": "^0.0.1",
"@cloudflare/puppeteer": "^0.0.6",
"@crxjs/vite-plugin": "^1.0.14",
"@google/generative-ai": "^0.3.1",
"@heroicons/react": "^2.1.1",
@ -53,6 +54,7 @@
"ai": "^3.0.0",
"better-sqlite3": "^9.4.3",
"class-variance-authority": "^0.7.0",
"cloudflare": "^3.0.0",
"dotenv-cli": "^7.3.0",
"drizzle-orm": "^0.29.4",
"eslint-plugin-next-on-pages": "^1.9.0",