I can think of a way to do that with polling or a subscription but It feels a bit overkill

Looks like console is down for me,

Cannot query field ‘enumValues’ on type ‘Field’. Did you mean ‘defaultValue’?

The databrowser specifically seems to not function

Hi guys, does anyone have a advice how should i upload photos and videos to Graphcool but resize images and convert video to mp4. The ideia is to not have huge files uploaded... I use react-dropzone and i think i can resize images with it, but dont know about videos... THX

@joao.santos While it's possible to resize image on the browser, video is going to be dificult

I played with https://bgrins.github.io/videoconverter.js/demo/ but the codecs alone were around a 7Mb payload to download in the browser

I think you will have more luck uploading video files without conversion and eventually have a graph.cool function process them post-upload

Thx @florian 😄

@joao.santos You can use the file proxy example I made (https://github.com/graphcool-examples/functions/tree/master/file-proxy) to add something like video conversion to the file upload.

It uses streaming, so if you can find a streaming video converter, you can do it without any memory load

Something like this one: https://www.npmjs.com/package/stream-transcoder

@agartha you mean that proxy is able to process the stream while it's being uploaded ?

Yes

Look at the way the upload is piped to the encryption in the example

You could do the same, but with transcoding

well that's pretty cool !

It is 😄

Do you happen to know if the env that runs those functions includes imageMagick ?

I can't seem to be able to add fields through schema designer atm

hi there, i followed this tutorial https://www.graph.cool/blog/2016-11-26-user-authentication-with-auth0-and-digits-thoh9chaek/ and i cant log in after signing up on the sample app. the users are created properly in auth0 but not in graph.cool and i get this error in the browser console: “Error: GraphQL error: The provided idToken is invalid. Please see https://auth0.com/docs/tokens/id_token for how to obtain a valid idToken.” anyone has an idea how to fix it?

@chris_r ran into the same issue, Make sure encryption is set to HS256 in your auth0 Dashbord > Client Settings > Advanced Settings > Oauth.

now it works. thank you very much @florian

I'm looking into a proper way to structure my data with i18n in mind. Let's say I have a document entity with a title and content.

I was thinking having a translation entity with a field per language

and have a

Document n : 1 Translation

relationship

that would allow me to query individual locale when making queries

Another way would make the title and content fields as JSON structures with locales as properties but doesn't seem very scalable

Question about optimizing data storage: we have a

Person

type and want to store information like phone number, email, etc. relative to that person. The obvious option would be to create a new Type for each piece of data and then to connect that type to

Person

, like

person.phones

, etc. However this introduces a lot of additional queries + logic on the client side (we're using Apollo) because we have to have a

createX, updateX, deleteX

mutation for each piece of profile info (could use nested creates but would still need update and delete). Another option would be to have a "superclass" called

Attribute

or something where all profile info is stored. If we did this we'd differentiate by a "type=PHONE" (etc.) field on the

Attribute

type. However this would likely introduce a lot of columns with null data. Would appreciate any experience/insights on structuring those large models. Thanks everyone!

Is it necessary to wait between mutate calls when iterating through an array? I'm using Lokka to mutate into GraphCool from a forEach(...) on an array returned from another api. It seemed to work when only adding 1-2 fields of each item, but with more fields I start to see some items not get added.