Forked from @degordian/avro-to-typescript
avro-to-typescript compiles avro schema files (.avsc) to TypeScript classes and interfaces. Making using avro schematics with node.js easy and comfortable.
This fork removes the schema registry and node-rdkafka dependencies for a lighter-weight install.
- Compiles most if not all avro types (record, enum, primitive, map, array)
Most projects will use avro-to-typescript this way
npm install -g @chasdevs/avro-to-typescript
avro-to-typescript --compile [ schema-directory ] [ output-directory ]This will generate namespaced folders and files for your schemas inside output directory so have that in mind.
You also need to install avro-to-typescript in your project.
npm install @chasdevs/avro-to-typescript --save
This way is if your projects needs to generate avro classes while running.
npm install @chasdevs/avro-to-typescript --save
import { Compiler } from "chasdevs/avro-to-typescript";
const compiler = new Compiler(outputDir);
await compiler.compile(avro);
If you want to take advantage of logical types, you can pass these as an argument to the cli
avro-to-typescript --compile [ schema-directory ] [ output-directory ] --logical-types [avro type] [typescript type]You can even pass more than one if you alternate them:
avro-to-typescript --compile [ schema-directory ] [ output-directory ] --logical-types [avro type] [typescript type] [avro type] [typescript type]You can also pass them to the compiler in your code directly:
import { Compiler } from "chasdevs/avro-to-typescript";
const compiler = new Compiler(outputDir, { date: 'string', 'timestamp-millis': 'string'} });
await compiler.compile(avro);
If there are logical types in the avro schema but you don't provide a mapping for them, they will default to the underlying primitive type.
If you are having issues, please let us know on our issue tracker.
- Issue Tracker: github.com/chasdevs/avro-to-typescript/issues
- Source Code: github.com/chasdevs/avro-to-typescript
The project is licensed under the MIT license.