How to adapt the custom generation to our standards?
We generate our pojos from an API contract which works pretty well. However, there are a few things need to address on the generated pojos that end up requiring some editing on the generated files - either manually or through a script. For instance, we would like to remove the @Generated annotation completely (and, yes, <hideGenerationTimestamp>true</hideGenerationTimestamp> works, but it does not remove the @Generated annotation altogether). A couple of other things we would like to have is to extract enums, override the toString() method, remove the toIndentendString(Object o) method, remove the JsonNullable import, stop using the @Nullable annotation and, finally, have better control over the generated contructors. Unfortunately, we are not using Lombok at this time, but we will adopt that soon. So, eventually, it would be nice to be able to annotate the classes with @Data, @Builder, @NoArgsConstructor and @AllArgsConstructor. I have seen many examples using Mustache, but have not really been able to enforce what we need. The pom we use is attached below and we execute it with mvn clean generate-sources. There are a few mustache templates in the specified directory, but they either doesn't seem to affect the result or they are not being used at all.
How do I validate the key of a object type schema | v3.0.4
I have my api.yaml like this, I want to validate that the lenght of the key of this map shouldn't be empty. How do I do it with swagger itself. I tried the below, the code generated validates the value and not the key. components: schemas: MyObject: properties: labels: title: labels type: object additionalProperties: type: string minLength: 1 pattern: "^.+$" description: JSON dictionary of labels example: someKey: SomeValue someKey2: SomeValue2 x-omitempty: false Is there a way to validate the object (HashMap) key also ? More info: I am using GoLang and in go a map can have a empty string as a key, that what I want to avoid, by using swagger and generated server code itself.
Use anyOf/oneOf with codegen?
Suppose I have a schema: params: is one of Message|Image|Audio _type: the type to expect in params I (currently) have the schema... ... "components": { "schemas: { "Interaction": { "type": "object", "required": [ "_type", "params" ], "properties": { "_type": {"type": "string"}, "params": { "anyOf": [ // or oneOf ?? { "$ref": "#/components/schemas/Message" }, { "$ref": "#/components/schemas/Image" }, { "$ref": "#/components/schemas/Audio" }, ], "discriminator": "_type" } } } What I would hope would be that the codegen code produced would represent a class such as # dart class Interaction{ _type: str; params: dynamic; // Message, Image or Audio ... } Instead I get from codegen class InteractionData { String method; InteractionDataParams params; } class InteractionDataParams { // The union of the attributes of Message, Image or Audio }; which, among other things, seems likely will run into collisions with differently typed attributes. Is there a sensible solution to this when I could expect codegen to directly produce Message/Image/ Audio objects in the params field?
Getting a `file` response in OpenAPI v3 / python
I'm trying to define a `file` response: responses: '200': content: image/png: schema: format: binary type: string In the generated python code, I see that for a file type, the function __deserialize_file is called, and it's exactly what I want: it save the HTTP response body as a temporary file on the filesystem and returns its path. The problem is that the response_type file is never defined, it's always a bytearray. How can I tell the python wrapper generator to write a file response to a temporary file (calling __deserialize_path) ? Thanks !
Handling unknown subtypes with OpenAPI discriminator property
In our API, we return different subclasses of a base class (SearchEntity) using the discriminator property (searchType) to differentiate between various types. This approach helps maintain data consistency and reduces errors during JSON deserialization. When clients generate client code using the /v3/api-docs/disco JSON, it works well with the known discriminator mappings. However, When we introduce a new subtype on the server side, Clients using older versions of the /v3/api-docs/disco JSON encounter JSON deserialization errors, as their generated client code does not recognize the newly added subtype. Could you provide guidance or suggestions on how to implement this fallback mechanism within the OpenAPI specification? Specifically: Is there a recommended approach to define a default or fallback logic for the discriminator property in OpenAPI? Are there alternative solutions to handle unknown subtypes more effectively without requiring frequent client code regeneration? Any best practices or tools to manage compatibility issues in such scenarios? Note: we were using spring boot from the server side. And attached sample open api json for your referenceNeed Help Escalating PR
Hello! I am from PayPal, and am hoping someone can help provide guidance how I can best escalate a PR to get merged based on business priority. We have had an open PR for several months now, and if it is not merged soon - it will impact one of our enterprise initiatives Please let me know if there is a better channel to escalate this in to get better visibility. Thank you in advance - appreciate the partnership! The code generation tool is crucial for developers to accurately validate specification and implement REST APIs at PayPal. This PR fixes a critical bug that impact virtually all OAS3 REST API specifications using examples, parameter, discriminator and callback features, making the codegen tool incompatible in its current state. Without this fix, developers are forced with cumbersome workarounds to choose between restructure their API specifications that hinder reusability or generate java model classes on the resolved specification resulting significant disruption to existing API implementation. This issue not only impacts individual teams but also prevents the API Platform team from effectively testing and ensuring the platform meets customer needs. Ultimately, Failing to merge this PR to address this critical issue could delay API deployments and hinder the overall adoption of the new platform. https://github.com/swagger-api/swagger-parser/pull/2087?new_signup=true?new_signup=trueopenapi-generator package name and version
I'm using the openapi-generator-maven-plugin to generate a python-pydantic-v1 client against my spec file. The generation is working well, but the generated setup.py script has default package name and version values, no matter what settings I configure in the plugin in my pom file. I can see that the mustache file for the setup.py script is using certain template parameters for these values, but changing them in the plugin configuration seems to have no effect on the values of these template parameters. Any thoughts would be much appreciated. Thanks in advance, John.Regarding swagger is not working in nodejs as per the swagger document.
Hi there, I tried to setup swagger on nodejs as per the document par it is not working and present wrong data of swagger UI see the below attached screenshot. So just wanted to know how to setup swagger in nodejs so that all my apis display in one place. I tried lot of solutions on different different websites but same issue appeared. So kindly guide how to do that so that swagger is running properly. Reference site that I tried below: https://swagger.io/docs/open-source-tools/swagger-codegen/#:~:text=The%20Swagger%20Codegen%20is%20an,can%20be%20found%20in%20GitHub. https://levelup.gitconnected.com/how-to-add-swagger-ui-to-existing-node-js-and-express-js-project-2c8bad9364ce https://medium.com/bb-tutorials-and-thoughts/how-to-add-swagger-to-nodejs-rest-api-7a542cfdc5e1 https://plainenglish.io/blog/how-to-implement-and-use-swagger-in-nodejs-d0b95e765245 https://dev.to/kabartolo/how-to-document-an-express-api-with-swagger-ui-and-jsdoc-50do https://itnext.io/setting-up-swagger-in-a-node-js-application-d3c4d7aa56d4 https://github.com/Surnet/swagger-jsdoc/blob/master/README.md Looking forward to hearing from you. Thanks! Swagger.json: { "swagger": "2.0", "info": { "version": "1.0.0", "title": "My User Project CRUD", "description": "My User Project Application API", "license": { "name": "MIT", "url": "https://opensource.org/licenses/MIT" } }, "host": "localhost:3000", "basePath": "/", "tags": [ { "name": "Users", "description": "API for users in the system" } ], "schemes": ["http"], "consumes": ["application/json"], "produces": ["application/json"] } Server.js: import swaggerUi from 'swagger-ui-express'; import swaggerDocument from './swagger.json'; app.use( '/api-docs', swaggerUi.serve, swaggerUi.setup(swaggerDocument) ); app.listen(port, () => { console.log(`Server running at ${process.env.SERVER_URL}`); });Multiple urls for tokenUrl for multiple servers
I have a list of servers like this: servers: - url: https://api.prod.mydomain.com description: Production URL - url: https://api.{environment}.dev.mydomain.com description: Environment URL And would like my securitySchemes for OAuth2 to mention that the tokenUrl will be different depending on the server you use. Either allow tokenUrl to be a list of urls securitySchemes: oAuth2: type: oauth2 description: 'This API is using OAuth 2.0 with password grant type for authentication.' flows: password: tokenUrl: - https://api.environment.dev.mydomain.com/oauth/token - https://api.prod.mydomain.com/oauth/token scopes: read_pet: read your pets write_pet: modify pets in your account Or to use `server` as a variable like this for the tokenUrl string? securitySchemes: oAuth2: type: oauth2 description: 'This API is using OAuth 2.0 with password grant type for authentication.' flows: password: tokenUrl: {server}/oauth/token scopes: read_pet: read your pets write_pet: modify pets in your account Does anyone know of a way to do this?
