Preview runs against a real D1 DB

J

jacob•8/14/24, 6:04 PM

Are D1 errors just presented as a regular "Error"? How should I handle a scenario where I want to check specifically if the error in my catch block is due to a unique constraint failing

J

jacob•8/14/24, 6:06 PM

I'm effectively trying to do something like this ..

try {
  await this.LandmarkRepository.addLandmark(label, locations[0]);
} catch (e) {
  if (e instanceof SomeUniqueConstraintError) {
            
  }
}

try {
  await this.LandmarkRepository.addLandmark(label, locations[0]);
} catch (e) {
  if (e instanceof SomeUniqueConstraintError) {
            
  }
}

try {
  await this.LandmarkRepository.addLandmark(label, locations[0]);
} catch (e) {
  if (e instanceof SomeUniqueConstraintError) {
            
  }
}

try {
  await this.LandmarkRepository.addLandmark(label, locations[0]);
} catch (e) {
  if (e instanceof SomeUniqueConstraintError) {
            
  }
}

A

arch•8/14/24, 8:32 PM

How do people prefer to manage their Workers + D1 schemas? So far I've been relying on just SQL files I manually load in. Prefer not using an ORM in most cases :p

J

jdruwe•8/14/24, 9:58 PM

Guys, I have a question regarding pushing a lot of data into D1. Basically I want to insert around 300K records (based on data from a CSV file I am passed). I build a UI using Hono inside a worker but I am running into the following error: "Exceeded CPU Time Limits", so to me it seems like a worker is not the best solution. What other option can you suggest to me? Could I just insert the data using the REST api? I just noticed there is also an option for: https://developers.cloudflare.com/workers/testing/local-development/#develop-using-remote-resources-and-bindings. It also results the same error, isn't that weird as I am executing the logic locally? For now I think something like: $ npx wrangler d1 execute example-db --remote --file=users_export.sql is better

A

Ashish Rawat•8/15/24, 7:15 AM

D1 is not usable without sharding. Just for mini projects it's good. I want to know is there any lib to shard d1 with workers.

AAshish Rawat D1 is not usable without sharding. Just for mini projects it's good. I want to k...

I

iska•8/15/24, 9:24 AM

im using drizzle orm to avoid vendor lock in and will move to aws rds whenever i hit 10 gb limit, dealing with sharding is a hassle , will wait for an auto scale solution from cloudflare

I

iska•8/15/24, 9:24 AM

i m building a product , dont want to deal with infra too much

R

rossanmol•8/15/24, 5:26 PM

Getting this error not authorized: SQLITE_AUTH in Workers when using D1? Anyone knows what it could be?

Rrossanmol Getting this error not authorized: SQLITE_AUTH in Workers when using D1? Anyone ...

I

Isaac McFadyen•8/15/24, 5:27 PM

Are you trying to access internal SQLite tables or use SQLite pragmas?

I

Isaac McFadyen•8/15/24, 5:27 PM

Some are blocked on D1 for security reasons.

I

Isaac McFadyen•8/15/24, 5:28 PM

If you are using a database ORM or explorer then it's possible it's making those queries on your behalf to try and inspect the schema of the database (and getting blocked).

J

jdruwe•8/15/24, 9:49 PM

When importing data into a D1 db using the wrangler command it mentions: "This process may take some time, during which your D1 database will be unavailable to serve queries.". I noticed this also in a test as drizzle showed "7500: Currently processing a long-running import." on a select query. Is there any way to get around this? Lets assume that the DB is already being used in production, I would like for the app to continue working. Any suggestions?

O

oof2win2•8/16/24, 5:38 AM

is there any eta on read replicas with d1?

Y

YuryC•8/16/24, 6:48 PM

Hello all! I need to rename a few columns in the existing table. For some reason, the RENAME statement doesn't work for me. Does d1 support the RENAME statement? Thanks in advance.

YYuryC Hello all! I need to rename a few columns in the existing table. For some reason...

I

Isaac McFadyen•8/16/24, 6:50 PM

D1 is SQLite, so see https://stackoverflow.com/questions/805363/how-do-i-rename-a-column-in-an-sqlite-database-table

Stack Overflow

How do I rename a column in an SQLite database table?

I need to rename table columns in an SQLite database. This similar question doesn't mention SQLite. From the documentation of ALTER TABLE I gather that it's not a single statement.
What is a generi...

S

Steven•8/16/24, 7:07 PM

Hi all,
I have an issue I don't understand: when i save values to a boolean column they seem to be saved as strings (true, false, or in the case of accidental error in inserting some random other strings).

This is on a local d1 database in development.

I was under the impression that values would be converted to 0 or 1 automatically (or error thrown if not truthy or falsy). Is this not correct?

SSteven Hi all, I have an issue I don't understand: when i save values to a boolean colu...

I

Isaac McFadyen•8/16/24, 7:12 PM

So yes,

true

true

and

false

false

will be cooerced into integers, but SQLite (which D1 uses) also has the concept of "flexible datatypes" that other DB engines don't have: https://sqlite.org/datatype3.html

I

Isaac McFadyen•8/16/24, 7:12 PM

It means you can insert a string into an integer column, or an integer into a string column, and the DB engine will just accept it

I

Isaac McFadyen•8/16/24, 7:13 PM

You can use STRICT tables if you want datatypes enforced: https://sqlite.org/stricttables.html

S

Steven•8/16/24, 7:14 PM

mm, okay. Thanks. Good to know.

But I do have 'true' and 'false' values in my boolean column. Those should be converted to 1 & 0 then right? Even without STRICT

I

Isaac McFadyen•8/16/24, 7:15 PM

I'm not sure if it's case-sensitive or not?

S

Steven•8/16/24, 7:18 PM

mm, i'll do a typecheck. maybe it's getting turned into a string somewhere

Z

ZeddBread🐣•8/17/24, 10:34 AM

Heyyyy

I'm getting started with D1 databases for my app.
I have 3 separate databases and want to put it into my local wrangler.toml

I don't know how to input multiple databases into it, the docs only show how to do one database like so:

[[d1_databases]]
binding = "DB"
database_name = "test-db"
database_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

[[d1_databases]]
binding = "DB"
database_name = "test-db"
database_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

[[d1_databases]]
binding = "DB"
database_name = "test-db"
database_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

[[d1_databases]]
binding = "DB"
database_name = "test-db"
database_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

If someone could lead me the right way, you would be madd af ngl

V

veeque•8/17/24, 10:58 AM

I made a SvelteKit project but what I don't understand is how to work with D1 in development. I would ideally like to start the SvelteKit app normally (

pnpm dev

pnpm dev

pnpm dev

pnpm dev) to quickly iterate on the UI, but have D1 accessible like I would by having a development docker image of a remote database. Currently I can only have a functioning version of my app when building the ui and then starting

wrangler pages dev .svelte-kit/cloudflare

wrangler pages dev .svelte-kit/cloudflare

wrangler pages dev .svelte-kit/cloudflare

wrangler pages dev .svelte-kit/cloudflare (but since it relies on a build, I would have to re-run the command after every change to have a functioning app in development). What am I missing?

Vveeque I made a SvelteKit project but what I don't understand is how to work with D1 in...

H

Hello, I’m Allie!OP•8/17/24, 11:04 AM

If you are using the Cloudflare Pages Adapter for SvelteKit, then it should be generating bindings for you in

pnpm dev

pnpm dev

pnpm dev

pnpm dev if that is what you mean

HHello, I’m Allie!If you are using the Cloudflare Pages Adapter for SvelteKit, then it should be g...

V

veeque•8/17/24, 11:11 AM

In

src/routes/api/test/+server.ts

src/routes/api/test/+server.ts

src/routes/api/test/+server.ts

src/routes/api/test/+server.ts I have:

import type { RequestHandler } from '@sveltejs/kit';

export const GET: RequestHandler = async ({ request, platform }) => {
  const result = await platform?.env.db.prepare('SELECT * FROM users LIMIT 5').run();

  return new Response(JSON.stringify(result));
};

import type { RequestHandler } from '@sveltejs/kit';

export const GET: RequestHandler = async ({ request, platform }) => {
  const result = await platform?.env.db.prepare('SELECT * FROM users LIMIT 5').run();

  return new Response(JSON.stringify(result));
};

import type { RequestHandler } from '@sveltejs/kit';

export const GET: RequestHandler = async ({ request, platform }) => {
  const result = await platform?.env.db.prepare('SELECT * FROM users LIMIT 5').run();

  return new Response(JSON.stringify(result));
};

import type { RequestHandler } from '@sveltejs/kit';

export const GET: RequestHandler = async ({ request, platform }) => {
  const result = await platform?.env.db.prepare('SELECT * FROM users LIMIT 5').run();

  return new Response(JSON.stringify(result));
};

And in my

src/app.d.ts

src/app.d.ts

src/app.d.ts

src/app.d.ts:

declare global {
  namespace App {
    interface Platform {
      env: {
        db: D1Database;
      };
    }
  }
}

export {};

declare global {
  namespace App {
    interface Platform {
      env: {
        db: D1Database;
      };
    }
  }
}

export {};

declare global {
  namespace App {
    interface Platform {
      env: {
        db: D1Database;
      };
    }
  }
}

export {};

declare global {
  namespace App {
    interface Platform {
      env: {
        db: D1Database;
      };
    }
  }
}

export {};

(where does the

D1Database

D1Database

D1Database

D1Database type come from?)

but when running with

pnpm dev

pnpm dev

pnpm dev

pnpm dev and I navigate to

localhost:5173/api/test

localhost:5173/api/test

localhost:5173/api/test

localhost:5173/api/test I get the following in the console:

TypeError: Cannot read properties of undefined (reading 'prepare')

TypeError: Cannot read properties of undefined (reading 'prepare')

TypeError: Cannot read properties of undefined (reading 'prepare')

TypeError: Cannot read properties of undefined (reading 'prepare'). If I build, then run

wrangler pages dev .svelte-kit/cloudflare --d1 db=[uuid-here]

wrangler pages dev .svelte-kit/cloudflare --d1 db=[uuid-here]

wrangler pages dev .svelte-kit/cloudflare --d1 db=[uuid-here]

wrangler pages dev .svelte-kit/cloudflare --d1 db=[uuid-here] the error is just because I haven't made the table, but would otherwise work.

I don't know about the binding it should be generating, because I don't seem to have any(?), but what also strikes me is that i can't just run

wrangler pages dev

wrangler pages dev

wrangler pages dev

wrangler pages dev, but I have to specify both the folder and the

--d1

--d1

--d1

--d1 binding myself, despite this being my

wrangler.toml

wrangler.toml

file:

name = "[name-here]"
compatibility_date = "2024-07-29"
pages_build_output_dir = ".svelte-kit/cloudflare"

[[d1_databases]]
binding = "DB"
database_name = "prod-main"
database_id = "[uuid-here]"

name = "[name-here]"
compatibility_date = "2024-07-29"
pages_build_output_dir = ".svelte-kit/cloudflare"

[[d1_databases]]
binding = "DB"
database_name = "prod-main"
database_id = "[uuid-here]"

name = "[name-here]"
compatibility_date = "2024-07-29"
pages_build_output_dir = ".svelte-kit/cloudflare"

[[d1_databases]]
binding = "DB"
database_name = "prod-main"
database_id = "[uuid-here]"

name = "[name-here]"
compatibility_date = "2024-07-29"
pages_build_output_dir = ".svelte-kit/cloudflare"

[[d1_databases]]
binding = "DB"
database_name = "prod-main"
database_id = "[uuid-here]"

V

veeque•8/17/24, 11:13 AM

nevermind about the

--d1

--d1

--d1

--d1 thing, it works regardless of it

V

veeque•8/17/24, 11:14 AM

oh, maybe it solved itself after deleting node_modules and retrying

V

veeque•8/17/24, 11:29 AM

ok, the only real issue now is data not persisting between each

pnpm dev

pnpm dev

pnpm dev

pnpm dev run. The documentation only tells me how to deal with this with

wrangler dev

wrangler dev

Vveeque ok, the only real issue now is data not persisting between each `pnpm dev` run. ...

H

Hello, I’m Allie!OP•8/17/24, 11:33 AM

Persistence should be enabled automatically. I don’t actually think you can disable it…

V

veeque•8/17/24, 11:36 AM

that made me check svelte's documentation. to show you all possible options for the adapter they give you a config with every value being used, and the

platformProxy.persist

platformProxy.persist

platformProxy.persist

platformProxy.persist was set to 'false'... and i just copy-pasted it all in a haste

H

Hello, I’m Allie!OP•8/17/24, 11:39 AM

Oh, that’ll do it

V

veeque•8/17/24, 11:39 AM

sorry for bothering you due to my stupidity

H

Hello, I’m Allie!OP•8/17/24, 11:39 AM

Though forgot that was exposed, tbh

Vveeque sorry for bothering you due to my stupidity

H

Hello, I’m Allie!OP•8/17/24, 11:39 AM

All good, you definitely aren’t alone on that front

V

veeque•8/17/24, 11:40 AM

just a question though, is it safe to put my wrangler.toml's config containing the database_id in the git history? is it kind of like an api token?

Vveeque just a question though, is it safe to put my wrangler.toml's config containing t...

H

Hello, I’m Allie!OP•8/17/24, 11:44 AM

Nope. Unless I have access to your account, then the ID is useless. In fact, everything in your

wrangler.toml

wrangler.toml

is safe to commit, assuming you are using secrets where appropriate

V

veeque•8/17/24, 11:44 AM

oh ok, that's nice

V

veeque•8/17/24, 11:44 AM

ty

V

veeque•8/17/24, 12:15 PM

@HardlyWorkin' how would i go about using the database outside of platform.env? (eg use drizzle) i tried looking everywhere but i just can't wrap my head around it

Vveeque @HardlyWorkin' how would i go about using the database outside of platform.env? ...

H

Hello, I’m Allie!OP•8/17/24, 12:20 PM

You construct the

drizzle

drizzle

drizzle

drizzle instance with

platform.env.DB

platform.env.DB

platform.env.DB

platform.env.DB

V

veeque•8/17/24, 12:20 PM

but i have to do that for each api route?

V

veeque•8/17/24, 12:21 PM

don't i need to make one instance for the whole thing?

Vveeque but i have to do that for each api route?

H

Hello, I’m Allie!OP•8/17/24, 12:23 PM

You can construct one inside a Server Hook, but they need to be created fresh on each request

J

jacoblukewood•8/17/24, 12:57 PM

Hi, I'm using D1 in SvelteKit.

I want to prerender all of my pages because the content only needs to be updated occasionally (every few weeks), on build I want to fill in the content of my pages with the content in the DB.

I can't seem to get it to work with adapter-static or adapter-cloudflare because the platform.env.DB object isn't bound at build time. This happens both if I try building locally and on Cloudflare Pages.

Any advice?

Jjacoblukewood Hi, I'm using D1 in SvelteKit. I want to prerender all of my pages because the ...

H

Hello, I’m Allie!OP•8/17/24, 12:57 PM

Bindings like D1 are unfortunately not available at build time

J

jacoblukewood•8/17/24, 12:59 PM

Do you have any guidance on where I could go from here to be able to do a static build? The only thing I can think of is moving my svelte endpoints that access D1 to workers functions and calling those

Similar Threads

Similar Threads

Similar Threads