That would work right now, yes — that's how you can use the Cache API for anything, not just D1.

MMatt Silverlock This is a holdover from the v4 API response format we use at Cloudflare. We don'...

R

Rex•6/12/23, 4:58 PM

so it will stay that way? its not the world ofc, i can work around it easily, just wanting to know

GG Thanks. Do you know of any plans that will streamline this? Or even somehow usin...

M

Matt SilverlockOP•6/12/23, 4:58 PM

On the roadmap. Not soon. LOTS of work to do first

RRex so it will stay that way? its not the world ofc, i can work around it easily, ju...

M

Matt SilverlockOP•6/12/23, 4:59 PM

TBD.

R

Rex•6/12/23, 4:59 PM

im confused on what that means

sorry

RRex im confused on what that means 😅 sorry

M

Matt SilverlockOP•6/12/23, 4:59 PM

"To be determined" - aka, I can't give you an answer either way right now (vs. just ignoring you)

R

Rex•6/12/23, 4:59 PM

ouh i see, thanks for your help

GG D1 speed: appreciate this is in alpha, but does anyone know of any speed compari...

M

Matt SilverlockOP•6/12/23, 7:35 PM

It really depends on your access patterns. If you are trying to read the same KV pair repeatedly, in the same location, KV is likely going to be faster. The trade-off is that (naturally) you need to fit into a KV-type data model.

D

Dahoom•6/13/23, 7:45 AM

Hello, is there a newer d1 api? Im trying to access my d1 db from my sveltekit cloudflare pages project but the only api i can find seems to be a little dated? https://www.npmjs.com/package/@cloudflare/d1?activeTab=readme

npm

@cloudflare/d1

The D1 Client API supports:. Latest version: 1.4.1, last published: a year ago. Start using @cloudflare/d1 in your project by running

npm i @cloudflare/d1

npm i @cloudflare/d1

npm i @cloudflare/d1

npm i @cloudflare/d1. There are no other projects in the npm registry using @cloudflare/d1.

DDahoom Hello, is there a newer d1 api? Im trying to access my d1 db from my sveltekit c...

H

Hello, I’m Allie!•6/13/23, 7:47 AM

IIRC, that package existed before the D1 shim was added directly to

wrangler

wrangler

, so you shouldn't need to install it for it to work

D

Dahoom•6/13/23, 7:48 AM

How do i import that into my project? afaik wrangler is a cli thing no?

DDahoom How do i import that into my project? afaik wrangler is a cli thing no?

H

Hello, I’m Allie!•6/13/23, 7:50 AM

You don't. If you add a D1 Binding to your Worker, then it will just work

export default {
  async fetch(req, env) {
    // ...
    const result = await env.D1.exec("SELECT * FROM db;");
    // ...
  }
}

export default {
  async fetch(req, env) {
    // ...
    const result = await env.D1.exec("SELECT * FROM db;");
    // ...
  }
}

export default {
  async fetch(req, env) {
    // ...
    const result = await env.D1.exec("SELECT * FROM db;");
    // ...
  }
}

export default {
  async fetch(req, env) {
    // ...
    const result = await env.D1.exec("SELECT * FROM db;");
    // ...
  }
}

D

Dahoom•6/13/23, 7:50 AM

thing that confuses me is that im trying to use a pages function here

DDahoom thing that confuses me is that im trying to use a pages function here

H

Hello, I’m Allie!•6/13/23, 7:52 AM

Pages Functions is the same, it should be present on the

ctx.env

ctx.env

ctx.env

ctx.env object.

export const onRequestGet = ctx => {
  const result = await ctx.env.D1.exec("SELECT * FROM db;");
  // ...
};

export const onRequestGet = ctx => {
  const result = await ctx.env.D1.exec("SELECT * FROM db;");
  // ...
};

export const onRequestGet = ctx => {
  const result = await ctx.env.D1.exec("SELECT * FROM db;");
  // ...
};

export const onRequestGet = ctx => {
  const result = await ctx.env.D1.exec("SELECT * FROM db;");
  // ...
};

D

Dahoom•6/13/23, 8:04 AM

I'm a little stupid, for the .env item what should i add? my wrangler.toml file looks like this but i dont think its working:

[[d1_databases]]
binding = "DataB"
database_name = "thing"
database_id = "dwadwadwad3213dsfasfdaeiu2"

[[d1_databases]]
binding = "DataB"
database_name = "thing"
database_id = "dwadwadwad3213dsfasfdaeiu2"

[[d1_databases]]
binding = "DataB"
database_name = "thing"
database_id = "dwadwadwad3213dsfasfdaeiu2"

[[d1_databases]]
binding = "DataB"
database_name = "thing"
database_id = "dwadwadwad3213dsfasfdaeiu2"

if i have to create a .env file would i format it the same way?

DDahoom I'm a little stupid, for the .env item what should i add? my wrangler.toml file ...

H

Hello, I’m Allie!•6/13/23, 8:05 AM

You don't use

wrangler.toml

wrangler.toml

for Pages, you pass bindings via the CLI

HHello, I’m Allie!You don't use `wrangler.toml` for Pages, you pass bindings via the CLI

D

Dahoom•6/13/23, 8:40 AM

do you mean wrangler pages dev --d1 db -- vite or some other method?

I

Isak•6/13/23, 11:10 AM

Is it possible to cache database queries using the cache API inside workers? Do you simply make a response with the data with some random url and call it a day?

DDahoom do you mean wrangler pages dev --d1 db -- vite or some other method?

M

Matt SilverlockOP•6/13/23, 11:14 AM

https://developers.cloudflare.com/pages/platform/functions/bindings/#d1-databases

Bindings · Cloudflare Pages docs

A binding enables your Pages Functions to interact with resources on the Cloudflare developer platform. Use bindings to integrate your Pages Functions …

IIsak Is it possible to cache database queries using the cache API inside workers? Do ...

M

Matt SilverlockOP•6/13/23, 11:15 AM

d1-database

MMatt Silverlock https://discord.com/channels/595317990191398933/992060581832032316/1117859495134...

I

Isak•6/13/23, 12:15 PM

Thanks was to lazy to search the threads.

S

sokorototo•6/13/23, 12:21 PM

Any idea why

wrangler d1 execute EmailTracker --command='SELECT * FROM SenderInfo WHERE email = "drnewt536@gmail.com";'

wrangler d1 execute EmailTracker --command='SELECT * FROM SenderInfo WHERE email = "drnewt536@gmail.com";'

wrangler d1 execute EmailTracker --command='SELECT * FROM SenderInfo WHERE email = "drnewt536@gmail.com";'

wrangler d1 execute EmailTracker --command='SELECT * FROM SenderInfo WHERE email = "drnewt536@gmail.com";' fails, yet

wrangler d1 execute EmailTracker --command="SELECT * FROM SenderInfo WHERE email = 'drnewt536@gmail.com';"

wrangler d1 execute EmailTracker --command="SELECT * FROM SenderInfo WHERE email = 'drnewt536@gmail.com';"

wrangler d1 execute EmailTracker --command="SELECT * FROM SenderInfo WHERE email = 'drnewt536@gmail.com';"

wrangler d1 execute EmailTracker --command="SELECT * FROM SenderInfo WHERE email = 'drnewt536@gmail.com';" passes

S

sokorototo•6/13/23, 12:21 PM

I just changed it from

"--"

"--"

"--"

"--" to

'--'

'--'

'--'

'--'

Ssokorototo Any idea why `wrangler d1 execute EmailTracker --command='SELECT * FROM SenderIn...

M

Max (@rozenmd)•6/13/23, 1:12 PM

"..."

"..."

"..."

"..." means column, or general database objects - you want

'...'

'...'

'...'

'...' for strings

Ttjenkinson hi with the `batch` api is it possible to get the results of all statements up t...

M

Matt SilverlockOP•6/13/23, 7:53 PM

It is not. The batch runs in an implicit transaction, so that a batch doesn’t “partially” complete and leave your database in an unexpected state.

Ttjenkinson Thanks. Could it not still return the results from earlier stages though, even i...

M

Matt SilverlockOP•6/13/23, 7:58 PM

That’s not how a transaction works. What if some of those earlier queries insert data? Modify the schema? Or return nothing? etc.

If you need to treat them as discrete queries, issue multiple separate queries

Ttjenkinson I think it would still be helpful to see what the output was with the understand...

M

Matt SilverlockOP•6/13/23, 8:22 PM

We plan to expose a transaction API, but you can’t do this right now

A

Adi•6/14/23, 1:33 AM

Some of the discussion actually reminded me of a question I have had since wrangler3/miniflare3 came out.
With those old miniflare packages (https://www.npmjs.com/package/@miniflare/d1) when in a test environment, you could write code that would simulate d1 in the following way

import { D1Database, D1DatabaseAPI } from "@miniflare/d1";
import { drizzle } from "drizzle-orm/d1";

const d1 = new D1Database(new D1DatabaseAPI(sqliteDb));
const db = drizzle(d1); // now I can can write unit tests for drizzle

import { D1Database, D1DatabaseAPI } from "@miniflare/d1";
import { drizzle } from "drizzle-orm/d1";

const d1 = new D1Database(new D1DatabaseAPI(sqliteDb));
const db = drizzle(d1); // now I can can write unit tests for drizzle

import { D1Database, D1DatabaseAPI } from "@miniflare/d1";
import { drizzle } from "drizzle-orm/d1";

const d1 = new D1Database(new D1DatabaseAPI(sqliteDb));
const db = drizzle(d1); // now I can can write unit tests for drizzle

import { D1Database, D1DatabaseAPI } from "@miniflare/d1";
import { drizzle } from "drizzle-orm/d1";

const d1 = new D1Database(new D1DatabaseAPI(sqliteDb));
const db = drizzle(d1); // now I can can write unit tests for drizzle

How can I write this with miniflare 3?

MMatt Silverlock Yes, but not > 100 at this stage. I’d like to get a better understanding of your...

J

Jason Hostetter•6/14/23, 2:53 AM

I ended up refactoring my query into two separate queries, which now seems to keep it under the depth of 20 limit. I initially had the queries combined with nested CTEs so that I could avoid two DB round-trips, but it seems like the speed of the new backend more than makes up for it. It's really fast now.

That would work right now, yes — that's how you can use the Cache API for anything, not just D1.

Similar Threads