Evenfire is an intelligence layer platform for running LLM agents that take real actions, on your infrastructure
TypeScript
// @vitest-environment jsdom
import { afterEach, describe, expect, it } from 'vitest'
import { cleanup, render, screen } from '@testing-library/react'
import type { SessionTokensLite } from '../../hooks/useChatStore'
import { SessionTokensIndicator } from '../agents/SessionTokensIndicator'
afterEach(cleanup)
describe('SessionTokensIndicator', () => {
it('shows input and output as separate compact stats', () => {
const tokens: SessionTokensLite = { input: 11_900, output: 500 }
const { container } = render(<SessionTokensIndicator tokens={tokens} />)
const stats = container.querySelectorAll('.token-usage-stat')
expect(stats[0]?.textContent).toBe('11.9k')
expect(stats[1]?.textContent).toBe('500')
})
it('omits the cache glyph and cache rows when the model does not report cache', () => {
const tokens: SessionTokensLite = { input: 100, output: 40 }
const { container } = render(<SessionTokensIndicator tokens={tokens} />)
expect(container.querySelector('.token-usage-stat--cache')).toBeNull()
const tooltip = container.querySelector('.token-usage-tooltip')
expect(tooltip?.textContent).toContain('Input')
expect(tooltip?.textContent).toContain('Output')
expect(tooltip?.textContent).not.toContain('Cache')
})
it('shows the cache glyph and tooltip cache rows when the model reports cache (even at 0)', () => {
const tokens: SessionTokensLite = { input: 200, output: 80, cacheRead: 13, cacheWrite: 0 }
const { container } = render(<SessionTokensIndicator tokens={tokens} />)
expect(container.querySelector('.token-usage-stat--cache')).toBeTruthy()
const tooltip = container.querySelector('.token-usage-tooltip')
expect(tooltip?.textContent).toContain('Cache read')
expect(tooltip?.textContent).toContain('13')
expect(tooltip?.textContent).toContain('Cache write')
})
it('renders nothing when the total is zero', () => {
const { container } = render(<SessionTokensIndicator tokens={{ input: 0, output: 0 }} />)
expect(container.firstChild).toBeNull()
})
it('exposes an accessible label with the exact breakdown', () => {
render(<SessionTokensIndicator tokens={{ input: 1500, output: 200 }} />)
// Compute the expected separator via toLocaleString so the assertion is
// locale-independent (CI may not default to en-US).
const expected = `Conversation token usage — Input ${(1500).toLocaleString()} · Output 200`
expect(screen.getByLabelText(expected)).toBeTruthy()
})
})