I'm a bit of a sucker for trying to keep best practices also when i'm working on a SaaS solution, is there any way to have version controll on the changes? To elaborate more: My ulimate goal is to have changes in git. (in practice on github). I have not seen anyone doing it, and I think it is an enterprise feature request, but if you could connect to my org account on github, create a repo and update on all the changes done from the GUI that would be amazing. I'm migrating away from the realtime database in firebase now, I just feel that they are trying to get my to get a vendor lockin on their system, and I felt the way you have to write your database structure was limiting and not right for my use case.

@Sindre Svendby There is a local project development workflow currently in development, I'm not sure the status, but it's at least in some form of beta that you could test out. Reach out to @nilan for more information

I would guess that version control is on the horizon, considering the generated graphcool file from the CLI includes version number

@Sindre Svendby We do CLI-only, so everything is version controlled. I think you have to be invited to the beta, but the stuff you asking for exists

The CLI is now in open beta, so you don't need an invite. Update (@joar): according to https://www.graph.cool/forum/t/feedback-new-cli-beta/949, participating in the open beta still requires registration first.

Great!

Is the service down?

My requests are timing out

nvm its on my end

@agartha Are projects created with the beta CLI production-ready?

Question: Any easy way to remove duplicate rows from a table?

Lets say there is a field that exists in every record that is uniquely-identifying, and I be sure I can safely delete one of the two records

[Question] How to migrate the database to my own database? Why I asking this? Because what will happen, if sometime this service is shutting down (I Hope not).

@Radiegtya You can export your data in JSON format, but I'm sure that they will not simply shut the servers down in the blink of an eye

thanks @eruby. I can export successfully my data, but my further question, if my data already very large, ex: 2GB, how can export function handle this?

It exports your data separated by tables / types in a zip

Just give it a test run. It won't shut down your project

Should be fairly straightforward

So in conclusion, it can download/export very large data. THank for the answer 👍

What's the best way to implement Auth0 login/sign up flow for a react native app?

https://www.graph.cool/docs/tutorials/react-apollo-auth0-pheiph4ooj/ this uses

localstorage

for the token and `fetchPolicy: 'network-only'. Therefore, on the ssr side, it will always load the

content as string from server --> refetch query --> loading state --> finished state

. Wondering if anyone has an idea on how I can change the way to check for user auth on both client and server side so that when I serve the content via server it will just be the 'finished state'

halo, salam kenal

is there an option for the auto-generated id's to be shorter in length?

Good morning 🙂 I have 2 questions: 1. If I configure Auth0 as Auth Provider, should I expect it to synch all previously created users within the connected Auth0 account into Graphcool? Or do I need to implement this migration?

and 2. What does actually the type

Node

mean? It’s not explained in the documentation

Hi friends, quick question: Does the limit for Subscriptions not firing when new Files are created also apply to request pipeline functions?

is there any way to restrict visibility of a specific field within a type? or do I have to create a separate type and add permissions there?

does anyone have some issues to get the log of a function with the CLI

graphcool logs -f xxx
> Call to <https://wt-graphcool.it.auth0.com/api/run/xxxx/xxx> failed with: Request timeout to <http://wt-graphcool.it.auth0.com/34|wt-graphcool.it.auth0.com/34>
.253.94.166:443 after 5000 ms

is is possible to filter but nodes that have

null

over a connection?

Morning friends - First of all, thank you for the awesome work on Chromeless. I'm trying to work through the examples to run/access Chrome locally in a lambda function. i.e., have a function that responds to a get request spin-up a chrome instance, do some work, then shut down. Is using Chromeless overkill for that -- should I just use the underlying serverless-chrome library? I can't seem to figure out the right init pattern to get it to work...