graphql-pydantic-converter 1.2.1

Convert pydantic schema to pydantic datamodel and build request from it

GraphQL to Pydantic Converter & Pydantic to Query Builder

Overview

The GraphQL to Pydantic Converter is a Python package designed to simplify the process of transforming GraphQL schemas in JSON format into Pydantic models. This tool is particularly useful for developers working with GraphQL APIs who want to generate Pydantic models from GraphQL types for efficient data validation and serialization/deserialization.

Features

• Converts GraphQL schemas in JSON format into Pydantic models.
• Build query or mutation from pydantic dataclass

Installation

You can install the GraphQL to Pydantic Transformer package via pip:

pip install graphql-pydantic-converter
# or
poetry add git+https://github.com/FRINXio/frinx-services-python-api.git@main#subdirectory=utils/graphql-pydantic-converter

Usage

Cli tool to transform GraphQL JSON to Pydantic

graphql-pydantic-converter [-h] [-i INPUT_FILE] [-o OUTPUT_FILE] [--url URL] [--headers HEADERS [HEADERS ...] ]

options:
  -h, --help            show this help message and exit
  -i INPUT_FILE, --input-file INPUT_FILE 
  -o OUTPUT_FILE, --output-file OUTPUT_FILE      
  --url URL 
  --headers HEADERS [HEADERS ...] # --headers "HeaderName: HeaderValue" "HeaderName: HeaderValue"

Output from cli tool

import typing

from pydantic import Field
from pydantic import PrivateAttr
from graphql_pydantic_converter.graphql_types import Input
from graphql_pydantic_converter.graphql_types import Mutation
from graphql_pydantic_converter.graphql_types import Payload

Boolean: typing.TypeAlias = bool
DateTime: typing.TypeAlias = typing.Any
Float: typing.TypeAlias = float
ID: typing.TypeAlias = str
Int: typing.TypeAlias = int
JSON: typing.TypeAlias = typing.Any
String: typing.TypeAlias = str

class CreateScheduleInput(Input):
    name: String
    workflow_name: String = Field(default=None, alias='workflowName')
    workflow_version: String = Field(default=None, alias='workflowVersion')
    cron_string: String = Field(default=None, alias='cronString')
    enabled: typing.Optional[Boolean] = Field(default=None)
    parallel_runs: typing.Optional[Boolean] = Field(default=None, alias='parallelRuns')
    workflow_context: typing.Optional[String] = Field(default=None, alias='workflowContext')
    from_date: typing.Optional[DateTime] = Field(default=None, alias='fromDate')
    to_date: typing.Optional[DateTime] = Field(default=None, alias='toDate')

class Schedule(Payload):
    name: typing.Optional[bool] = Field(default=False)
    enabled: typing.Optional[bool] = Field(default=False)
    parallel_runs: typing.Optional[bool] = Field(alias='parallelRuns', default=False)
    workflow_name: typing.Optional[bool] = Field(alias='workflowName', default=False)
    workflow_version: typing.Optional[bool] = Field(alias='workflowVersion', default=False)
    cron_string: typing.Optional[bool] = Field(alias='cronString', default=False)
    workflow_context: typing.Optional[bool] = Field(alias='workflowContext', default=False)
    from_date: typing.Optional[bool] = Field(alias='fromDate', default=False)
    to_date: typing.Optional[bool] = Field(alias='toDate', default=False)
    status: typing.Optional[bool] = Field(default=False)


class CreateScheduleMutation(Mutation):
    _name: str = PrivateAttr('createSchedule')
    input: CreateScheduleInput
    payload: Schedule

CreateScheduleInput.model_rebuild()
CreateScheduleMutation.model_rebuild()
Schedule.model_rebuild()

Query & Mutation builder

from schedule_api import Schedule, CreateScheduleMutation, CreateScheduleInput

SCHEDULE: Schedule = Schedule(
    name=True,
    enabled=True,
    workflow_name=True,
    workflow_version=True,
    cron_string=True
)

mutation = CreateScheduleMutation(
    payload=SCHEDULE,
    input=CreateScheduleInput(
        name='name',
        workflow_name='workflowName',
        workflow_version='workflowVersion',
        cron_string='* * * * *',
        enabled=True,
        parallel_runs=False,
    )
)

Created query with inlined variables as string

mutation.render(form='inline')

mutation {
  createSchedule(
    input: {
      name: "name"
      workflowName: "workflowName"
      workflowVersion: "workflowVersion"
      cronString: "* * * * *"
      enabled: true
      parallelRuns: false
    }
  ) {
    name
    enabled
    workflowName
    workflowVersion
    cronString
  }
}

Created query with extracted variables

mutation, variables = mutation.render(form='extracted)

# mutation as a string
mutation ($input: CreateScheduleInput!) { 
  createSchedule(input: $input) { 
    name 
    enabled 
    workflowName
    workflowVersion 
    cronString 
  } 
}

# variables as a dict[str, Any]
{
  "input": {
    "name": "name",
    "workflowName": "workflowName",
    "workflowVersion": "workflowVersion",
    "cronString": "* * * * *",
    "enabled": true,
    "parallelRuns": false
  }
}

Response parser

Example of generated model.py

import typing
from pydantic import BaseModel, Field
from graphql_pydantic_converter.graphql_types import ENUM

Boolean: typing.TypeAlias = bool
DateTime: typing.TypeAlias = typing.Any
Float: typing.TypeAlias = float
ID: typing.TypeAlias = str
Int: typing.TypeAlias = int
JSON: typing.TypeAlias = typing.Any
String: typing.TypeAlias = str

class Status(ENUM):
    UNKNOWN = 'UNKNOWN'
    COMPLETED = 'COMPLETED'
    FAILED = 'FAILED'
    PAUSED = 'PAUSED'
    RUNNING = 'RUNNING'
    TERMINATED = 'TERMINATED'
    TIMED_OUT = 'TIMED_OUT'

    
class SchedulePayload(BaseModel):
    name: typing.Optional[typing.Optional[String]] = Field(default=None)
    enabled: typing.Optional[typing.Optional[Boolean]] = Field(default=None)
    parallel_runs: typing.Optional[typing.Optional[Boolean]] = Field(default=None, alias='parallelRuns')
    workflow_name: typing.Optional[typing.Optional[String]] = Field(default=None, alias='workflowName')
    workflow_version: typing.Optional[typing.Optional[String]] = Field(default=None, alias='workflowVersion')
    cron_string: typing.Optional[typing.Optional[String]] = Field(default=None, alias='cronString')
    workflow_context: typing.Optional[typing.Optional[String]] = Field(default=None, alias='workflowContext')
    from_date: typing.Optional[typing.Optional[DateTime]] = Field(default=None, alias='fromDate')
    to_date: typing.Optional[typing.Optional[DateTime]] = Field(default=None, alias='toDate')
    status: typing.Optional[typing.Optional[Status]] = Field(default=None)


class CreateScheduleData(BaseModel):
    create_schedule: SchedulePayload = Field(default=None, alias='createSchedule')

    
class CreateScheduleResponse(BaseModel):
    data: typing.Optional[CreateScheduleData] = Field(default=None)
    errors: typing.Optional[typing.Any] = Field(default=None)

Example of response

from model import CreateScheduleResponse

# send previously created request to backend service
payload = {'query': mutation.render()}
resp = requests.post(SCHELLAR_URL, json=payload)
response = resp.json()

# Example of response
# { 
#    'data': {
#         'createSchedule': {
#              'name': 'name', 
#              'enabled': True, 
#              'workflowName': 'workflowName', 
#              'workflowVersion': 'workflowVersion', 
#              'cronString': '* * * * *'
#          }
#     }
# }

schedule = CreateScheduleResponse(**response)

if schedule.errors is None:
    print(schedule.data.create_schedule.workflow_name)
else:
    print(schedule.errors)

1.0.0

• Migration to pydantic v2

1.0.1

• Change Map type to dict (key, value)

1.0.2

• Stringify mutation input strings
• Add __typename to payload
• create test folder as a module

1.1.0

• Support inline and extracted variables for mutation and query
• Stringify fix

1.1.1

• Fix missing inputs for mutation and query render in extracted format
• ENUM keys always uppercase

1.2.0

• Allow to populate by name

1.2.1

• Fix rendering issue when payload is bool