Token Package

Just contains various bits used across micro services

python3 setup.py sdist bdist_wheel

How to use the search and filter module

In your graphql schema you can add the inputs variable that contains several graphql inputs suitable to use for different data types

full_schema = [load_schema_from_path("schema.graphql"), *inputs]

Example Input

input UuidFilterInput { 
    eq: Uuid neq: Uuid 
}

This input can be applied to UUID fields, so eq is equals and neq not equals, to match the record from the database

Graphql Schema

In order to use this in the graphql schemas you need to create a filter input that will contain these inputs that are created in the search and filter module

for example:

input OrderBookingFilter {
  id: UuidFilterInput
  reference: StringFilterInput
  trackingNumber: IntegerFilterInput
  status: BookingStatus
  scheduled: DateTimeFilterInput
  created: DateTimeFilterInput
}

And if you want to use this as a query field you can like this:

type Query {
  bookings(filter: OrderBookingFilter, search: String, first: Int, after: String): [OrderBooking]
}

This module comes with filter and search functionality, the search one just requires a String input and Model that you want to do the search on, the results will be coming from get_search_query(model, search).

The filter functionality comes from get_filter_query(model, filter), filter is dictionary, like:

filter = {
    "id": {
        "eq": "something",
    }
}

There's also a ModelResolver that accepts:

• formatter (the way model fields are displayed through graphql)
• model
• kwargs (this can be filter or search, plus the parameters required for the paginator)

After initiating this resolver you get:

• It will create RelayPaginator as a property accessible through instance.paginator.
• It creates base queryset using Model.all(), and if you call the instance.get_data() without specified qs it will return all the paginated records of the Model.
• You can override this base queryset inside the instance.get_data(qs=YourQuerySet) if needed. The output of this method is paginated records of the Model.
• If the instance of the resolver contains filter or search in the extra fields then it will prioritise those ones and make the queryset use one of get_search_query() or get_filter_query().

If you decide to use the ModelResolver's get_data method to get the data then the output of your function will be a Connection type.

You can easily build your connection type using the graphql interfaces in this package.

The package comes with PageInfo type:

type PageInfo {
    hasNextPage: Boolean!
    startCursor: String
    endCursor: String
}

And interfaces:

interface Edge {
    cursor: String
}
interface Connection {
    pageInfo: PageInfo!
}

So you can use these interfaces in your graphql schema to build your types used for the RelayPaginator. For example:

type ModelEdge implements Edge {
    cursor: String
    node: Model
}

type ModelConnection implements Connection {
    pageInfo: PageInfo!
    edges: [ModelEdge]
}

How to fix the workflows once search and filter module implemented.

Changes for Prod, Stage and Schema workflow

replace - uses: actions/checkout@v2 with

- name: Wait for tests
  uses: fountainhead/[email protected]
  id: wait-for-build
  with:
    token: ${{ secrets.GITHUB_TOKEN }}
    checkName: test
    ref: ${{ github.event.pull_request.head.sha || github.sha }}
    intervalSeconds: 20
- uses: actions/checkout@v2
- name: Download the graphql artifact
  if: steps.wait-for-build.outputs.conclusion == 'success'
  uses: dawidd6/action-download-artifact@v2
  with:
    workflow: python-app.yml
    workflow_conclusion: success
    name: output-schema-file
    path: ./lib_schema

replace --localSchemaFile=./schema.graphql with --localSchemaFile=./schema.graphql,./lib_schema/schema.graphql

Changes for Test workflow

ensure the file that contains the test is called python-app.yaml

add the following above steps

env:
  PYTHON_VERSION: 3.8

alter the Set up Python 3.8 step to be the following

- name: Set up Python 3.8
  uses: actions/setup-python@v2
  with:
    python-version: ${{ env.PYTHON_VERSION }}

Finally, add the following step to the end

- name: Upload Libary Schema
  uses: actions/upload-artifact@v1
  with:
    name: output-schema-file
    path: ${{ env.LD_LIBRARY_PATH }}/python${{ env.PYTHON_VERSION }}/site-packages/aj_micro_utils/search_and_filter/schema.graphql