groceristar / graphql-server

Second generation of groceristar server

graphql-server

Data

Static data

imported from @groceristar/groceristar-fetch plugin and partially from Food-static-data: https://github.com/GroceriStar/sd More about this plugin here:

Database schema

Details about current DB scheme: https://medium.com/groceristar/database-schema-of-groceristar-current-release-20dc1b4be7b9

This is a strucuture that we have in our models: https://github.com/ChickenKyiv/database-visuals/tree/master/groceristar/models same structure we use at groceristar-fetch, so we can easy grab data, that we need. database(we'll use it later):

"url": "mongodb://heroku_p3w65n77:[email protected]:35065/heroku_p3w65n77",
"name": "groceryDS",

Rest API methods

This is a list of methods, that we have in our Frontend Version. I can be like a checklist for us. In order to fully migrate from REST API server to graphQL we'll need to cover all of this cases: https://medium.com/groceristar/groceristar-website-methods-list-75b57e1414ae

GroceryStar Documentation

This is our main documentation space, related to Groceristar project: https://groceristar.github.io/documentation/

Running the development server

npm install

Graphql server will be running in http://localhost:3000/

You can explore all schema types

GraphQL Resources

I have some collection of tutorials, websites, related to GraphQL topic, storing it here: https://github.com/ChickenKyiv/awesome-graphql-beginner-links

as default webpack config didn't support babel 7 yet, we have both versions at package.json

install?? looks like it's old version // "babel-plugin-syntax-async-functions": "^6.13.0", // "babel-plugin-transform-regenerator": "^6.26.0", // // "plugins": [ // "syntax-async-functions", // "transform-regenerator" // ] // https://itnext.io/building-a-graphql-api-using-graphql-yoga-lambda-and-mongodb-c0c195621f47 https://medium.com/@xoor/coding-a-graphql-api-with-node-js-c02d617f49f4

1. relations. i think we should debate each relation. because I think our schema(picture that we have) have more old-fashion logic. For example - UserGrocery table - it can be just part of User type. I saw a lot of big projects, that just put everything in one place and then parse it.
2. for me its hard to keep in mind that we have Items. I mean I have in my head structure of groceristar without items, and all documentation, methods related to it. maybe it will be more easy for me - when I finish dealing with our old servers. So I propose to have our development in two ways: a) just recreating/replacing our current GS server - so when we create for example a query or method - we just open my old code - see how it works and does the same logic

b) we can work with items in parallel, but we need to be aware that this part of schema doesn't have our documentation coverage, so we'll need to create everything from scratch. it's not a problem for me, but I'll be more supportive to items version when I'll know that we did basic important methods for ingredients, grocery, departments. these 3 types have some complex things and I want to address them first.

btw, if you have questions - ask them more specifically. and I think at another task.

graphql-server's Issues

update test environment similar to fake-api setup

Food-Static-Data/fake-api#16

create relations between types

when i was reading the specification - i tried to create some sort of schema at paper. i just find it, and decide to move it here, for discussion. I'm 100% sure that my schema wrong, because i didn't finished my learning of graphQL yet. but maybe it will give us a topic to discuss and will help you. listing it below.

type UserGrocery {
user: User!
groceries: [Grocery]!
id: ID!
}

type Department {
name: String!
desc:String!
ingredients: [Ingredients]!
id: ID!
}

type Favorites {
user:User!
favorites: [Ingredient]!
id: ID!
}

type Ingredient {
name
description
custom: Boolean!
department: Department!
}

type User {
tokens: [Token]!
identities:[]!
credentials:[]!
groceries:[Grocery]!
favorites:[Favorite]!
id: ID!
}

type Grocery {
name: String!
image: Binary!
description: String!
slug: String!
departments: [Department]! {Show/Hide}
purchased: [Ingredient]!
ingredients:[Ingredient]!
id: ID!
}

type Item {
name: String!
description: String!
quantity: Number!
purchase: Boolean!
id: ID!
}

creating types

we need to create tables(types) for our database schema

• User Grocery
• Favorites
• Ingredient
• Grocery
• Item
• User
• Department
• Grocery

documentation [move sample queries]

#28

https://github.com/marmelab/graphql-schema-from-json

import data to mongodb via graphql

i'm curious to know if we have some way to 'migrate' data(different from graphQL migration) from js arrays into database. i'm pretty sure that gql eco-system have some cool plugin that can save us time

testing

https://github.com/ChickenKyiv/awesome-graphql-beginner-links/blob/master/README.md#testing
Plus https://github.com/ZolaApp/zola-api/blob/develop/src/tests/resetDatabase.js

read

https://codeburst.io/large-offline-datasets-with-apollo-client-c19c987e4a98
https://developer.mozilla.org/en-US/docs/Web/API/IndexedDB_API

cool strucutre

https://github.com/ParabolInc/action/tree/master/src/server/graphql

move code, created from playing-* repository, if it more advanced that we have here

getItems method is missing in the fetch

install our `fetch` plugin

• read an article about plugin
• sync latest changes from main repo
• create/finish examples of static data for each of types that we have. Vadim will use it in order to compare results.
• reply to our comments, advice at our questions
• confirm that method from fetch, related to department works well

---

• get some data, related to Grocery, Department, Ingredient (separately)
• show me, how, by changing query - we'll get a different set of data from our static data as result

table userFav

The table has ingredient_id as PM and FK, I don't understand how it's possible?

migrate to babel 7

plus use babel.json instead of .babelrc - new rules

install sd-wrapper, connect it to resolvers and see if everything will work or not

0. setup project locally Done
1. install sd-wrapper. our module https://github.com/GroceriStar/sd#quick-start Done
2. comment previous versions and use a new version here: https://github.com/GroceriStar/graphql-server/blob/master/src/data/selector.js
   as we make it in smart way previously, all methods located in one file, instead of checking all files
3. run this server, run tests - does it work? nothing crashing? it's strange :)

Create new methods at fetch plugin, that will fit to current server requirements (prev. static data)

#6 (comment)

and we need to have that data for all types, that we specified here: https://github.com/GroceriStar/graphql-server/tree/master/src/typeDefs

---

create new methods

Departments

• Moved data to selector.js here?
• Completed and tested here with approval from Nada

---

Grocery

Schema
Static Data that was tested at graphql server

• Was Grocery Methods published at fetch plugin?
• Did we update package at this repo?
• Moved data to selector.js here?
• Completed and tested here with approval from Nada

---

Ingredients

Schema
don't have file Static Data that was tested at graphql server

• Was Ingredient Methods published at fetch plugin?
• Did we update package at this repo?
• Moved data to selector.js here?
• Completed and tested here with approval from Nada

---

users

Schema
Static Data that was tested at graphql server

• Was users Methods published at fetch plugin?
• Did we update package at this repo?
• Moved data to selector.js here?
• Completed and tested here with approval from Nada

---

next types will explain later

Adding queries, resolvers, mutations

Define all used queries, resolvers, mutations for each type.

strange situation

Vadim made some updates at latest release in our fetch plugin. and we know that this should crash the server. Builds from CI should prevent this situation. But they didn't.... explore more later
https://travis-ci.org/GroceriStar/graphql-server/jobs/482185706
#29

bug

https://travis-ci.org/GroceriStar/graphql-server/jobs/494666176

Grocery-Department relation

It is many-many, right?
is the id_1 in Grocery table refer to department_id?

Dependency Dashboard

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

Warning

These dependencies are deprecated:

Datasource	Name	Replacement PR?
npm	@babel/plugin-proposal-class-properties
npm	merge-graphql-schemas

Open

These updates have all been created already. Click a checkbox below to force a retry/rebase of any.

• Update dependency express to v4.19.2 [SECURITY]
• Update dependency snyk to v1.1064.0 [SECURITY]
• Update dependency merge-graphql-schemas to v1.7.8
• Update dependency nodemon to v2.0.22
• Update babel monorepo (@babel/cli, @babel/core, @babel/node, @babel/plugin-proposal-class-properties, @babel/preset-env)
• Update dependency regenerator-runtime to v0.14.1
• Update jest monorepo (babel-jest, jest)
• Update dependency babel-plugin-module-resolver to v5
• Update dependency graphql-yoga to v5
• Update dependency nodemon to v3
• Update dependency supertest to v7
• Click on this checkbox to rebase all open PRs at once

Ignored or Blocked

These are blocked by an existing closed PR and will not be recreated unless you click a checkbox below.

• Update dependency standard to v14.3.4
• Update dependency babel-plugin-module-resolver to v4.1.0
• Update dependency standard to v17
• Update jest monorepo to v29 (major) (babel-jest, jest)

Detected dependencies

npm

package.json

• @groceristar/groceristar-fetch ^2.2.7
• @groceristar/sd-wrapper ^4.15.0
• express ^4.17.1
• graphql-yoga ^1.18.3
• jest ^25.0.0
• merge-graphql-schemas ^1.7.3
• snyk ^1.258.2
• supertest ^4.0.2
• @babel/cli 7.8.4
• @babel/core 7.9.0
• @babel/node 7.8.7
• @babel/plugin-proposal-class-properties 7.8.3
• @babel/preset-env 7.9.0
• babel-core 7.0.0-bridge.0
• babel-jest 25.2.6
• babel-plugin-module-resolver 4.0.0
• nodemon 2.0.2
• regenerator-runtime 0.13.5
• standard 14.3.3

travis

.travis.yml

• node 10
• node 11

---

• Check this box to trigger a request for Renovate to run again on this repository

discussion

what approach we'll have for our graphql server?
i mean there a lot of servers like yoga or apollo stuff, please advice

add to readme how to setup, run server and run query editor locally

https://developer.ibm.com/blogs/how-to-validate-your-graphql-queries-with-eslint/

test how our server works in real world

• should be deployed and visible from outside
• should be connected to one of our react projects(maybe at first time we should just have a clean CRA only)
• instead of using 5-7 methods from our fetch plugin - we should get data by using some queries...

at resolvers we need to try and replace source of our static data.

right now all of our methods are separated, same to our data.
But in order to make it work with our old structure - we can test how it works with our nearly created file/method with Big object of grocery lists

what should we use for our ID fields and relations based on FK IDs..

based on #18 (comment)
and #18 (comment)

when we'll resolve this question, we'll be able to update/clean up our fetch plugin, by removing methods, that we don't need anymore...

https://moonhighway.com/building-a-graphql-server-in-five-minutes

complex structures replacement from groceristar-fetch to sd-wrapper

based on this comment

links

https://www.npmjs.com/package/graphql-import
https://www.apollographql.com/docs/graphql-tools/generate-schema/
https://graphql.org/learn/schema/
https://www.prisma.io/blog/graphql-server-basics-the-schema-ac5e2950214e
https://www.prisma.io/blog/datamodel-v11-lrzqy1f56c90/
https://github.com/jquense/yup