IPFS Pinning Service API OpenAPI Code Generation Project

IPFS pinning services API codgen project aims to provide a way for pinning services to generate their own SDKs in 30+ languages & frameowrks easily using openapi-generator-cli tools.

Generate OpenApi client SDKs from IPFS Pinning Service OpenAPI Specification.

• Adding new language targets
• Adding custom templates
• Running locally

Latest generated SDKs

• C# (Dotnet Core)
• PHP
• Python
• Ruby
• Typescript

Adding new language targets

We can add support for new language SDKs by following these steps:

• Find the generator name for the language here
• Add a configuration file under config/{generatorName}.yml (most or all generators require at least packageName)
• Find available configuration options by viewing the page for the generator here
• Add the desired configuration values
• Create a separate git repository for the new SDK
• Add a new entry to the matrix in .github/workflows/generate.yml: { repo: "{repoName}", openapi_generator: "{generatorName}" }. The next time the workflow is ran a PR will be created in the new repo with the generated SDK
• Optionally add language templates by creating a directory templates/{generatorName} and placing the .moustache files in here (see Adding custom templates)
• Optionally add a new npm command to package.json (see existing scripts) for generation and update the gen:all npm script

Adding custom templates

OpenApi geneartor uses mustache as its templating engine.

As an example lets use a custom python template:

• Create a directory with the name of the generator under the templates directory (i.e ./templates/python)
• Find the default template that we are updating in the openapi-generator codebase, lets use README.mustache for our example
• Create ./templates/python/README.mustache
• Copy+paste the default template from the openapi-generator codebase in the ./templates/python/README.mustache file
• Make any changes to our custom template, we can add new configuration values using {{configValue}} and then setting the value in the language configuration file ./config/python.yml like so configValue: Hello
• Run generation npm run gen:py, our template changes will be observable in the produced file: out/python/README.md

Running locally

Install dependencies:

• docker, make sure your docker daemon is running before generating
• nodejs & npm

Now perform the following:

• the location of the openapi spec is locked to ipfs pinning service OpenAPI spec repo at: https://raw.githubusercontent.com/ipfs/pinning-services-api-spec/master/ipfs-pinning-service.yaml (so it will always pull the latest version and generate the latest SDKs when you run)
• npm run build
• npm run gen:all to generate all language SDKs OR
• npm run gen:py to generate the python SDK for example

Generated SDKs & Documentation

All generated SDKs can be viewed and tested in /out folder under their respective languages. Test files templates are generated so are the documentation of the SDKs.