search

tjohnson1106 / saga

V1
0 stars 0 forks source link

Create GraphQL Server #8

Open tjohnson1106 opened 6 years ago

tjohnson1106 commented 6 years ago

Create a graphQL framework; initially fetching static data from Mongo with Express

tjohnson1106 commented 6 years ago

Need to talk to Damian about Amazon back-end

damianesteban commented 6 years ago

What about specifically?

damianesteban commented 6 years ago

I would recommend using PostgreSQL w/ TypeORM as your ORM. Then hook that up to your GraphQL Server.

I can make a basic server based on your current schema if you want me to.

tjohnson1106 commented 6 years ago

Whatever you think is best on the back in the long run; I set this up so I could fetch data from a database and use a Apollo to see the UI evolve. I currently have it running with Node and Mongo but I just need to have a dev environment setup , it's hard to work with just static UI changes.

Let me tweak the GraphQL server, tomorrow morning you can pull it down, I only have two resolvers right now user and "tweet" just so I can add Mock data. But that should be a good start and I will add to that once I get a feed going.

On Mon, Dec 4, 2017 at 7:56 PM, Damian Esteban notifications@github.com wrote:

I would recommend using PostgreSQL w/ TypeORM https://github.com/typeorm/typeorm as your ORM. Then hook that up to your GraphQL Server.

I can make a basic server based on your current schema if you want me to.

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/tjohnson1106/saga/issues/8#issuecomment-349159214, or mute the thread https://github.com/notifications/unsubscribe-auth/AGH8XII58t8qk0wlujO11eEyr17Cj6Tdks5s9JTMgaJpZM4QtymI .

-- Thomas

tjohnson1106 commented 6 years ago

Whatever you think is best on the back in the long run; I set this up so I could fetch data from a database and use a Apollo to see the UI evolve. I currently have it running with Node and Mongo but I just need to have a dev environment setup , it's hard to work with just static UI changes.

Let me tweak the GraphQL server, tomorrow morning you can pull it down, I only have two resolvers right now user and "tweet" just so I can add Mock data. But that should be a good start and I will add to that once I get a feed going. In terms of the back-end you said that had set some services up, I just needed a reminder as to what you had set up.

tjohnson1106 commented 6 years ago

Just to make a record of this, we decided against PostgreSQL and decided to use Mongo, specifically Mongo Atlas with AWS. We looked at the data fields that we'd utilize from Facebook, one thing we didn't talk about who is actually going to handle coding the graphQL resolvers. (of course the schema will need to tweaked) Is this something best done with the implementation of the back-end architecture? Or should it be handled based on the rendering the data as components. I am not sure what would be the best practice.

tjohnson1106 commented 6 years ago

Basic GQL Server is in place. This will obviously be an evolving process but I believe the premise is there and there a greater understanding of what is needed. Saga beta-version 0.4(sagaBetaV4 repository) is effectively fetching and rendering image URLs which was problematic with the previous version and the structure is superior, it should be a good basis for the production server. Moving to review.

tjohnson1106 commented 6 years ago

Moving this out of QA, as stated before this is an evolving process. The development back-end is now running on Phoenix/Elixir with Absinthe which simplifies maintaining a development server and I believe makes the code more flexible when porting to a AWS server-less environment. This is far from a finished product and I should never have moved it to QA. Moving to Refactor due to evolving nature.