Cloudflare Developers•3y ago

Only if you are enterprise. The recommended way would be to run an HTTP proxy worker so you don’t ha

Only if you are enterprise. The recommended way would be to run an HTTP proxy worker so you don’t have a rate limit

Cyb3r-Jak3OP•11/12/23, 2:15 AM

Something like https://github.com/elithrar/http-api-d1-example

GitHub

GitHub - elithrar/http-api-d1-example: An example HTTP API for Clou...

An example HTTP API for Cloudflare's D1 database: https://developers.cloudflare.com/d1/

GitHub - elithrar/http-api-d1-example: An example HTTP API for Cloudflare's D1 database: ht...

Wwadefletch Is there a slick way in D1 (generated columns, maybe?) to support updatedAt colu...

Isaac McFadyen•11/12/23, 2:35 AM

Generated columns are done at SELECT time not INSERT time, so there's not really a way aside from what you've done, no.

Isaac McFadyen•11/12/23, 2:36 AM

D1 is just SQLite so most things that work with SQLite will work with D1, FYI.

IIsaac McFadyen Generated columns are done at SELECT time not INSERT time, so there's not really...

Matt Silverlock•11/12/23, 12:55 PM

There are two types of generated column: https://developers.cloudflare.com/d1/learning/generated-columns/#types-of-generated-columns

Generated columns · Cloudflare D1 docs

D1 allows you to define generated columns based on the values of one or more other columns, SQL functions, or even extracted JSON values.

MMatt Silverlock There are two types of generated column: https://developers.cloudflare.com/d1/le...

Isaac McFadyen•11/12/23, 1:12 PM

Oh - interesting! TIL, thank you.

jiangok•11/12/23, 1:57 PM

How do I use d1 with a different env? could anyone help a remix related question? thanks.

Jjiangok https://discord.com/channels/595317990191398933/1172999291451736164/117310614809...

Matt Silverlock•11/12/23, 2:54 PM

You’re better off responding/creating a new thread, but I’ve linked https://developers.cloudflare.com/d1/examples/d1-and-remix/ for you

Query D1 from Remix · Cloudflare D1 docs

Remix is a full-stack web framework that operates on both client and server. You can query your D1 database(s) easily from Remix with Remix’s data …

FFerry While we are wainting for transactions is it possible to nest statement methods ...

GuillaumeLakano•11/12/23, 4:36 PM

db.prepare is not interactive, it's similar to a parser that format correctly your SQL statement.
Fetch your statements from your table, and push them inside an array to db.batch.

Something like:

const { results: myviewStatements } = await db.prepare('select statetement from myview where tb=?1").bind(t1).all()
await db.batch([ db.prepare("create table.."), ...myviewStatements.map(s => db.prepare(s.statement)) ]

GuillaumeLakano•11/12/23, 4:57 PM

I have a request, and it's may be a bug in the pricing calculation.

The goal of « partial indexes » ( https://www.sqlite.org/partialindex.html ) is to attach a WHERE condition to reduce the required writes / storage. So the additional write for the index is not always done.

I've done a test, and this is always counted as an extra written row, even when the WHERE condition is not true.
It's a bug ?

FFerry But how if one of the statements fails?How to rollback?

GuillaumeLakano•11/12/23, 5:11 PM

the rollback is automatic in the batch, so if one of your statement fail, all the batch is rolled back

jiangok•11/13/23, 4:14 AM

hi, I am using CF pages + d1 + remix + drizzle orm + vitest. is there any example for adding a unit test for remix loader loading d1 data from a local sqlite? Thanks.

rostero•11/13/23, 5:38 PM

Hey. I'm also here from d1 + remix (but updated to the latest libs). I'm getting this when clicking on some of the avengers: Failed to get worker definitions TypeError: fetch failedFailed to get worker definitions TypeError: fetch failed and underneath it: UND_ERR_HEADERS_TIMEOUT.UND_ERR_HEADERS_TIMEOUT.

rostero•11/13/23, 5:45 PM

this is all local w/ wrangler

rostero•11/13/23, 5:52 PM

also, are there any docs on the read/write behaviors of D1, similar to https://docs.turso.tech/reference/data-consistency?

SmallShen•11/14/23, 9:07 AM

does delete rows count how many rows read?

SSmallShen does delete rows count how many rows read?

GuillaumeLakano•11/14/23, 9:44 AM

You can check the rows_read / rows_written in your response, but I guess this will be the same number of reads from the same WHERE condition on a SELECT + the 1x write of delete per row deleted

meow•11/14/23, 10:06 AM

Is it dangerous to commit the database name and id in

wrangler.toml

wrangler.toml

? Should that be a secret?

Mmeow Is it dangerous to commit the database name and id in `wrangler.toml`? Should th...

Hello, I’m Allie!•11/14/23, 10:07 AM

No, they are not a secret

Rrostero Hey. I'm also here from d1 + remix (but updated to the latest libs). I'm getti...

Max (@rozenmd)•11/14/23, 10:48 AM

errrr that wouldn't happen to be https://github.com/rozenmd/d1-drizzle-remix-example would it?

Max (@rozenmd)•11/14/23, 10:49 AM

I've just updated this

to work on Remix v2

MMax (@rozenmd)errrr that wouldn't happen to be https://github.com/rozenmd/d1-drizzle-remix-exa...

rostero•11/14/23, 12:56 PM

Yep! Thanks for updating. This was very helpful. Also, I think { mode: "timestamp" } is broken{ mode: "timestamp" } is broken could be { mode: "timestamp_ms" }{ mode: "timestamp_ms" }, but I'm new to drizzle.

Rrostero Yep! Thanks for updating. This was very helpful. Also, I think `{ mode: "time...

Max (@rozenmd)•11/14/23, 12:59 PM

looks like they've removed mode: "timestamp"mode: "timestamp" entirely from drizzle

Max (@rozenmd)•11/14/23, 12:59 PM

used to be in their docs

rostero•11/14/23, 12:59 PM

ahh. ok.

rostero•11/14/23, 1:04 PM

Another D1 question. Can a JS stored procedure use drizzle/prisma?

Rrostero Another D1 question. Can a JS stored procedure use drizzle/prisma?

Max (@rozenmd)•11/14/23, 1:12 PM

We don't have JS stored procedures yet

bkyerv•11/15/23, 12:12 PM

When working locally with D1 and Workers, I've noticed that a miniflare folder, which holds a local database copy, is created inconsistently. Sometimes it's there, and other times it's not. What determines the creation of this folder in certain situations but not in others?

Bbkyerv When working locally with D1 and Workers, I've noticed that a miniflare folder, ...

Hello, I’m Allie!•11/15/23, 12:15 PM

Hm, that sounds like an old version of

wrangler

wrangler

... Current versions create a folder named .wrangler.wrangler at start

bkyerv•11/15/23, 12:17 PM

i mean inside .wrangler/state/d1.....

Bbkyerv When working locally with D1 and Workers, I've noticed that a miniflare folder, ...

Max (@rozenmd)•11/15/23, 12:22 PM

as of wrangler 3.15.0
wrangler d1wrangler d1 commands default to prod, and need to be passed

--local

--local

to run locally

wrangler dev

wrangler dev

defaults to local, and needs be passed

--remote

--remote

to run against your prod db

could explain why it might seem inconsistent

jiangok•11/15/23, 4:37 PM

I use CF pages + d1. On local development, pnpm run build & npx wrangler pages dev ./public --compatibility-date=2023-11-15 --d1=DBpnpm run build & npx wrangler pages dev ./public --compatibility-date=2023-11-15 --d1=DB creates a d1 database under .wrangler/state/v3/d1/miniflare-D1DatabaseObject.wrangler/state/v3/d1/miniflare-D1DatabaseObject folder. This is different from CF worker which store database under .wrangler/state/v3/{database_id}.wrangler/state/v3/{database_id} which can be operated like wrangler d1 execute $DATABASE_NAME --file=./seed.sql --localwrangler d1 execute $DATABASE_NAME --file=./seed.sql --local which observes

wrangler.toml

wrangler.toml

. How can I use wrangler to operate the local databases generated by CF pages like those generated by CF workers? It will be great if CF pages & workers have consistent local development model for D1. Thanks for any hints.

MMax (@rozenmd)as of wrangler 3.15.0 `wrangler d1` commands default to prod, and need to be pa...

bkyerv•11/15/23, 5:49 PM

what is the difference between the db file in the {database_id}{database_id} folder and the one in the miniflare-D1DatabaseObjectminiflare-D1DatabaseObject folder? which one is local and which one is prod?

Bbkyerv what is the difference between the db file in the `{database_id}` folder and the...

Hello, I’m Allie!•11/15/23, 5:58 PM

Neither of them are prod. AFAIK, the miniflareminiflare-prefixed SQLite files are used for internal storage in

wrangler

wrangler

, like the names of bindings, metrics, etc

Bbkyerv what is the difference between the db file in the `{database_id}` folder and the...

Max (@rozenmd)•11/15/23, 6:17 PM

miniflare-D1DatabaseObject uses miniflare to create the sqlite file, while the other uses better-sqlite3 directly. both are local/neither are prod. I don't know why they aren't in the same location tbh

Cyb3r-Jak3OP•11/15/23, 6:42 PM

I thought the newer way was with the miniflare prefix and stuff was migrated over

_tweak^prone•11/15/23, 9:42 PM

We have a product thats about to release around late december 2023 - I would really like to include D1 for this project. How likely is it that it would be safe to use D1 - and/or that its actually released late december 2023 ?

(I can quickly switch to another db - its not a "major issue")

__tweak^prone We have a product thats about to release around late december 2023 - I would rea...

Matt Silverlock•11/15/23, 10:15 PM

https://blog.cloudflare.com/d1-open-beta-is-here/

We’re setting a clear goal: we want to make D1 “generally available” (GA) for production use-cases by early next year (Q1 2024).

The Cloudflare Blog

D1: open beta is here

D1 is now in open beta, and the theme is “scale”: with higher per-database storage limits and the ability to create more databases, we’re unlocking the ability for developers to build production-scale applications on D1

MMatt Silverlock https://blog.cloudflare.com/d1-open-beta-is-here/ > We’re setting a clear goal: ...

_tweak^prone•11/15/23, 11:34 PM

nice - thank you

MMatt Silverlock wrt D1 in South America - we’re working on it.

tadeumaia•11/16/23, 10:35 AM

Hello. Any news on this or maybe some alpha/beta/waitlist? This is holding me to use D1 for a project going live soon. Going turso right now as they already have 2 POP right now, But would love to avoid the migration.

Ddmio_co Hey, question about how D1 works. We have 1 central place where db is hosted/sto...

Matt Silverlock•11/16/23, 11:28 AM

Per https://blog.cloudflare.com/d1-open-beta-is-here/

read replication is something we’re working on

The Cloudflare Blog

D1: open beta is here

Ttadeumaia Hello. Any news on this or maybe some alpha/beta/waitlist? This is holding me to...

Matt Silverlock•11/16/23, 11:28 AM

No news yet. Next year.

busterroni•11/16/23, 3:27 PM

Hi, I'm working on a flight search engine and am considering D1. I'm concerned that the 2GB database limit is too small to hold information for all upcoming flights -- there will be a couple million rows. We could also have one database per airport, to distribute the number of rows in each database. Do you think this is a good approach? Is D1 a good option for this kind of project?

Only if you are enterprise. The recommended way would be to run an HTTP proxy worker so you don’t ha

Similar Threads