A parsing library for CircleCI configuration files, powered by the CircleCI Config SDK
Used by the CircleCI Visual Config Editor to generate an interactive and editable visual representation of your CircleCI config.
Using npm:
$ npm i @circleci/circleci-config-parser
Using yarn:
$ yarn add @circleci/circleci-config-parser
In Node.js:
import ConfigParser from '@circleci/circleci-config-parser';
In Browser:
const ConfigParser = require('@circleci/circleci-config-parser');
Loading a Config instance from a config file
import fs from 'fs';
const configSrc = fs.readFileSync('./config.yml', 'utf8');
const config = ConfigParser.parseConfig(configSrc);
Parsing a job config equivalent object, into a CircleCI Config SDK Job
instance.
const jobIn = {
docker: [{ image: 'cimg/base:2022.08' }],
resource_class: 'medium',
steps: [
{
run: {
command: 'echo << parameters.greeting >>',
},
},
],
parameters: {
greeting: {
type: 'string',
},
},
};
// Parsing function
ConfigParser.parseJob('Job Name', jobIn);
The equivalent config-sdk instantiation for that object:
new CircleCI.reusable.ParameterizedJob(
'my_job',
new CircleCI.executors.DockerExecutor('cimg/node:lts'),
new CircleCI.parameters.CustomParametersList([
new CircleCI.parameters.CustomParameter('greeting', 'string'),
]),
[
new CircleCI.commands.Run({
command: 'echo << parameters.greeting >>',
}),
],
);
Parsing Orb references requires an OrbManifest, which is a representation of Orbs outward facing properties.
import fs from 'fs';
const customOrbProps = {
// component type
jobs: {
// name of component
say_hello: {
// component parameters
greeting: {
type: 'string',
},
},
},
commands: {
say_it: {
what: {
type: 'string',
},
},
},
executors: {
python: {
version: {
type: 'string',
default: '1.0.0',
},
},
},
};
const customOrbManifest = ConfigParser.parseOrbManifest(customOrbProps);
const configSrc = fs.readFileSync('./config.yml', 'utf8');
const config = ConfigParser.parseConfig(configSrc, {
'custom-orb': customOrbManifest,
});