schema.graphql has incorrect duplicate relationships.

[qwikag]

Before opening, please confirm:

• I have searched for duplicate or closed issues.
• I have read the guide for submitting bug reports.
• I have done my best to include a minimal, self-contained set of instructions for consistently reproducing the issue.
• I have removed any sensitive information from my code snippets and submission.

App Id

whyputappidinpublicforum

Region

ap-southeast-2

Environment name

develop

Figma File Version (if applicable)

N/A

Amplify CLI Version

No response

If applicable, what version of Node.js are you using?

No response

What operating system are you using?

win10

Browser type?

chrome

Describe the bug

Cannot publish a new relationship with Pascal Case:
e.g.
type Person @model @auth(rules: [{allow: owner, operations: [read, update, create]}, {allow: private, operations: [create, read, update], provider: iam}]) {
id: ID!
Email: AWSEmail!
Username: AWSEmail
Firstname: String
Lastname: String
Mobile: AWSPhone
Birthdate: AWSDate
Address: String
Phone: AWSPhone
SystemLogs: [SystemLog] @hasmany(indexName: "byPerson", fields: ["id"])
Logins: [SystemLog] @hasmany(indexName: "byPerson", fields: ["id"])
Bookings: [Booking] @hasmany(indexName: "byPerson", fields: ["id"])
}

type Booking @model @auth(rules: [{allow: private, operations: [read]}, {allow: owner, operations: [create, read, update]}]) {
id: ID!
Name: String!
Start: AWSDateTime
Stop: AWSDateTime
OrgID: ID! @Index(name: "byOrganisation")
Status: String
OrganiserId: [Booking] @Index(name: "byPerson")
}

I get this error

Expected behavior

It should work just fine.

Reproduction steps

create parent child relationship with aand change the field name from
e.g. "personID" --> "PersonId"

Project Identifier

1d91fff5a13778a3f2da7571d70fa642

Additional information

NB: the duplicated SystemLog relationship on the Logins, is also very incorrect. It should be "Login".
I have multiple instances of this happening.
This too is a separate problem I am logging.

Amplify& AWS is hugely broken, and no longer worth getting involved in until these issues (and many more) get fixed.
I now have to take my business elsewhere.

AWS need to take a look at Salesforce and how successful their easy data modelling has been over the last 20years, seldom a failure.
And this is due to quality implementation of front end delivering on backend capability.

[qwikag]

This is the schema that the docs say is the norm:

type Order @model {
  id: ID!
  customerID: ID! @index(name: "byCustomerByStatusByDate", sortKeyFields: ["status", "date"]) @index(name: "byCustomerByDate", sortKeyFields: ["date"])
  accountRepresentativeID: ID! @index(name: "byRepresentativebyDate", sortKeyFields: ["date"])
  productID: ID! @index(name: "byProduct", sortKeyFields: ["id"])
  status: String!
  amount: Int!
  date: String!
}

type Customer @model {
  id: ID!
  name: String!
  phoneNumber: String
  accountRepresentativeID: ID! @index(name: "byRepresentative", sortKeyFields: ["id"])
  ordersByDate: [Order] @hasMany(indexName: "byCustomerByDate", fields: ["id"])
  ordersByStatusDate: [Order] @hasMany(indexName: "byCustomerByStatusByDate", fields: ["id"])
}

type Employee @model {
  id: ID!
  name: String! @index(name: "byName", queryField: "employeeByName", sortKeyFields: ["id"])
  startDate: String!
  phoneNumber: String!
  warehouseID: ID! @index(name: "byWarehouse", sortKeyFields: ["id"])
  jobTitle: String! @index(name: "byTitle", queryField: "employeesByJobTitle", sortKeyFields: ["id"])
  newHire: String! @index(name: "newHire", queryField: "employeesNewHire", sortKeyFields: ["id"]) @index(name: "newHireByStartDate", queryField: "employeesNewHireByStartDate", sortKeyFields: ["startDate"])
}

type Warehouse @model {
  id: ID!
  employees: [Employee] @hasMany(indexName: "byWarehouse", fields: ["id"])
}

type AccountRepresentative @model {
  id: ID!
  customers: [Customer] @hasMany(indexName: "byRepresentative", fields: ["id"])
  orders: [Order] @hasMany(indexName: "byRepresentativebyDate", fields: ["id"])
  orderTotal: Int
  salesPeriod: String @index(name: "bySalesPeriodByOrderTotal", queryField: "repsByPeriodAndTotal", sortKeyFields: ["orderTotal"])
}

type Inventory @model {
  productID: ID! @primaryKey(sortKeyFields: ["warehouseID"])
  warehouseID: ID! @index(name: "byWarehouseID", queryField: "itemsByWarehouseID")
  inventoryAmount: Int!
}

type Product @model {
  id: ID!
  name: String!
  orders: [Order] @hasMany(indexName: "byProduct", fields: ["id"])
  inventories: [Inventory] @hasMany(fields: ["id"])
}

the difference with this from mine:

1. lowerCamelCase
2. byPerson should be "byPersonLogins" and "byPersonLogs"

It is highly likely the cli I picking up the double "byPerson" and causing a rewrite the the schema.

also the console is not helping by not enabling a write to these relationships and creating a duplicate "byPerson"

[ykethan]

Hey @qwikag, thank you for reaching out. The error appears to be occurring due to OrganiserId: [Booking] @Index(name: "byPerson"). I was able to reproduce this issue using Amplify CLI. How is the relationship being created? could you provide us a screenshot of the models and its relationships in the data manager.

[qwikag]

Hi @ykethan,

Created in Amplify Studio (Setup --> Data)

Maybe update by cli???

Another example:

The problem here is the documentation about how to put together a good data-model, is weak, and needs a lot of work to communicate the idiosyncrasies.

I think console data modeling is broken, because that is where I started, and things have gone nasty.
but I will move to writing it in the schema.graphql, however the documentation is so Aweful.

[ykethan]

Hey @qwikag, I was able to reproduce the issue. Similar to #788 the visual editor creates a duplicate relationship.

Reproduction steps:

1. Create models A, B, C
2. add a oneToMany relationship from A to B
3. Then add a oneToMany relationship from A to C
4. Then change the oneToMany relationship from A to C -> A to B
5. Then change the relationship back to A to C

The B models seems to have a duplicate relationship

Marking as bug.

I do agree the documentation could be improved to provide better information on creating a data modelling schema. Let pass this feedback to the team in improving the documentation. Thank you for taking the time in reaching out and providing this feedback.