Evenfire is an intelligence layer platform for running LLM agents that take real actions, on your infrastructure
TypeScript
import { describe, expect, it } from 'vitest'
import {
TELEGRAM_WORKFLOW_RESULT_CALLBACK_DATA,
parseTelegramCallbackData,
telegramWorkflowResultCallbackData,
} from '../telegramCallbackData'
describe('Telegram workflow result callback data', () => {
it('round trips deterministic workflow result callback data', () => {
const workflowRunId = '11111111-2222-3333-4444-555555555555'
const callbackData = telegramWorkflowResultCallbackData(workflowRunId)
expect(callbackData).toBe(`wf:r:${workflowRunId}`)
expect(parseTelegramCallbackData(callbackData!)).toEqual({
kind: 'workflowResult',
workflowRunId,
})
})
it('rejects malformed workflow run IDs', () => {
expect(telegramWorkflowResultCallbackData('Due Diligence')).toBeNull()
expect(telegramWorkflowResultCallbackData('11111111-2222-3333-4444')).toBeNull()
})
it('keeps legacy result callbacks parseable without inventing a workflow run', () => {
expect(parseTelegramCallbackData(TELEGRAM_WORKFLOW_RESULT_CALLBACK_DATA)).toEqual({
kind: 'workflowResult',
})
})
it('keeps previously delivered recipe-name callbacks working', () => {
expect(parseTelegramCallbackData('wf:r:due-diligence-package')).toEqual({
kind: 'workflowResult',
workflowName: 'due-diligence-package',
})
})
})