Python OpenAPI 2.0 (Swagger) object model
There are any number of good libraries for using OpenAPI/Swagger in specific ways or with specific frameworks. This isn't one of them.
Rather, this library provides a simple object model for producing and consuming Swagger specifications.
Model types extend Python dict
, list
, or string
to support reading from JSON,
writing to JSON, and validation against (the relevant portion of) the Swagger schema.
As a result, models are interchangeable with their primitive counterparts and can be used as much (or as little) as desired to enhance readability or perform validation.
Models convert contained primitive types to models when accessed using keys or attributes
(for dict
-based types) and when using indexing (for list
-based access). Access to
model internals via iteration bypasses conversion.
-
Either load a schema from json:
from openapi import load with open("/path/to/swagger.json") as fileobj: swagger = load(fileobj)
Or construct one explicitly from the model:
from openapi.model import Swagger, Info, Operation, PathItem, Paths, Response, Responses swagger = Swagger( swagger="2.0", info=Info( title="Example", version="1.0.0", ), basePath="/api", paths=Paths({ "/hello": PathItem( get=Operation( responses=Responses({ "200": Response( description="Returns hello", ) }) ), ), }), )
-
Access model internals using attributes:
print swagger.info print swagger.basePath
Pythonic names are automically converted when using attribute access:
print swagger.base_path
Naturally, attribute names that are illegal or that shadow existing attributes must be accessed using key syntax:
print swagger.paths["/hello"]["get"].responses["200"].description
-
Validate the model:
swagger.validate()
Internal models can be validated independently:
swagger.info.validate()
-
Dump the result to JSON:
print swagger.dumps()