orca/src/main/native-chat/transcript-opencode-sqlite-query.test.ts

import { mkdtempSync, rmSync } from 'node:fs'
import { tmpdir } from 'node:os'
import { join } from 'node:path'
import { pathToFileURL } from 'node:url'
import { afterEach, describe, expect, it, vi } from 'vitest'
import { AGENT_SESSION_HOST_STATUS_COPY } from '../../shared/agent-session-host-status-rows'
import Database from '../sqlite/sync-database'
import {
  readOpenCodeTranscriptPage,
  readOpenCodeTranscriptSignal
} from './transcript-opencode-sqlite-query'

let tempDirs: string[] = []
let openDbs: Database.Database[] = []

afterEach(() => {
  // Close before removal: Windows locks open databases.
  for (const db of openDbs) {
    db.close()
  }
  openDbs = []
  for (const dir of tempDirs) {
    rmSync(dir, { recursive: true, force: true })
  }
  tempDirs = []
})

function createTempDb(): { db: Database.Database; path: string } {
  const dir = mkdtempSync(join(tmpdir(), 'orca-opencode-native-chat-'))
  tempDirs.push(dir)
  const path = join(dir, 'opencode.db')
  const db = new Database(path)
  openDbs.push(db)
  return { db, path }
}

function applySchema(db: Database.Database): void {
  db.exec(`
    CREATE TABLE session (id TEXT PRIMARY KEY);
    CREATE TABLE message (
      id TEXT PRIMARY KEY,
      session_id TEXT NOT NULL,
      time_created INTEGER NOT NULL,
      time_updated INTEGER NOT NULL,
      data TEXT NOT NULL
    );
    CREATE TABLE part (
      id TEXT PRIMARY KEY,
      message_id TEXT NOT NULL,
      session_id TEXT NOT NULL,
      time_updated INTEGER NOT NULL,
      data TEXT NOT NULL
    );
    INSERT INTO session (id) VALUES ('ses-1'), ('ses-2');
  `)
}

function insertMessage(
  db: Database.Database,
  args: { id: string; sessionId?: string; time: number; role?: string }
): void {
  db.prepare(
    'INSERT INTO message (id, session_id, time_created, time_updated, data) VALUES (?, ?, ?, ?, ?)'
  ).run(
    args.id,
    args.sessionId ?? 'ses-1',
    args.time,
    args.time,
    JSON.stringify({ role: args.role ?? 'user' })
  )
}

function insertPart(
  db: Database.Database,
  args: { id: string; messageId: string; sessionId?: string; time: number; data: unknown }
): void {
  db.prepare(
    'INSERT INTO part (id, message_id, session_id, time_updated, data) VALUES (?, ?, ?, ?, ?)'
  ).run(args.id, args.messageId, args.sessionId ?? 'ses-1', args.time, JSON.stringify(args.data))
}

function messageRowid(db: Database.Database, id: string): number {
  // oxlint-disable-next-line typescript/consistent-type-assertions -- SAFETY: node:sqlite rows are unknown; the SELECT projects exactly one `r` integer column.
  return (db.prepare('SELECT rowid AS r FROM message WHERE id = ?').get(id) as { r: number }).r
}

describe('readOpenCodeTranscriptPage', () => {
  it.each(['message', 'part'] as const)(
    'omits an oversized %s without losing identity or pagination',
    (table) => {
      const { db, path } = createTempDb()
      applySchema(db)
      for (let index = 1; index <= 3; index++) {
        insertMessage(db, { id: `msg-${index}`, time: index })
        insertPart(db, {
          id: `prt-${index}`,
          messageId: `msg-${index}`,
          time: index,
          data: { type: 'text', text: `message ${index}` }
        })
      }
      db.prepare(`UPDATE ${table} SET data = ? WHERE id = ?`).run(
        JSON.stringify({
          role: 'user',
          type: 'file',
          mime: 'image/png',
          url: 'x'.repeat(2 * 1024 * 1024)
        }),
        table === 'message' ? 'msg-2' : 'prt-2'
      )
      const args = { dbPath: path, sessionId: 'ses-1', limit: 2 }
      const page = readOpenCodeTranscriptPage(args)
      expect(page?.items.map((item) => item.message.id)).toEqual(['msg-2', 'msg-3'])
      expect(page?.items[0]?.message.blocks).toEqual([
        {
          type: 'text',
          text: AGENT_SESSION_HOST_STATUS_COPY['history-item-too-large'],
          presentation: 'history-item-too-large'
        }
      ])
      expect(page?.items[0]?.rowid).toBe(messageRowid(db, 'msg-2'))
      expect(readOpenCodeTranscriptPage(args)?.items[0]?.fingerprint).toBe(
        page?.items[0]?.fingerprint
      )
      db.prepare(`UPDATE ${table} SET time_updated = 10 WHERE id = ?`).run(
        table === 'message' ? 'msg-2' : 'prt-2'
      )
      expect(readOpenCodeTranscriptPage(args)?.items[0]?.fingerprint).not.toBe(
        page?.items[0]?.fingerprint
      )
      expect(
        readOpenCodeTranscriptPage({
          ...args,
          beforeMessageRowId: page?.beforeMessageRowId ?? undefined
        })?.items.map((item) => item.message.id)
      ).toEqual(['msg-1'])
    }
  )

  it('preserves ordinary text beside an oversized UTF-8 attachment part', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    insertMessage(db, { id: 'msg', time: 1 })
    insertPart(db, {
      id: 'text',
      messageId: 'msg',
      time: 1,
      data: { type: 'text', text: 'see this' }
    })
    insertPart(db, {
      id: 'large',
      messageId: 'msg',
      time: 1,
      data: { type: 'file', mime: 'image/png', url: '😀'.repeat(600_000) }
    })
    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 1 })
    expect(page?.items.map((item) => item.message)).toMatchObject([
      { id: 'msg', role: 'user', blocks: [{ type: 'text', text: 'see this' }] },
      {
        id: 'msg:omission',
        role: 'system',
        blocks: [{ type: 'text', presentation: 'history-item-too-large' }]
      }
    ])
    expect(page?.items[0]?.rowid).toBe(page?.items[1]?.rowid)
    expect(page?.items[0]?.message.transcriptOffset).toBe(page?.items[1]?.message.transcriptOffset)
  })

  it('returns null when the session row does not exist', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    expect(readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'missing', limit: 10 })).toBeNull()
  })

  it('treats a session_v2-only (OpenCode 2) DB as not-found instead of throwing', () => {
    const { db, path } = createTempDb()
    // A v2-only store must fail softly in the v1 reader.
    db.exec('CREATE TABLE session_v2 (id TEXT PRIMARY KEY)')
    expect(readOpenCodeTranscriptSignal(path, 'ses-1')).toBeNull()
    expect(readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 10 })).toBeNull()
  })

  it('returns the NEWEST limit messages when more history exists', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    for (let index = 1; index <= 10; index++) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: { type: 'text', text: `message ${index}` }
      })
    }
    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 3 })
    expect(page).not.toBeNull()
    expect(page!.items.map((item) => item.message.id)).toEqual(['msg-8', 'msg-9', 'msg-10'])
    expect(page!.hasMore).toBe(true)
    expect(page!.beforeMessageRowId).toBe(page!.items[0]!.rowid)
  })

  it('pages strictly older messages from the raw-row cursor', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    for (let index = 1; index <= 10; index++) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: { type: 'text', text: `message ${index}` }
      })
    }
    const first = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 3 })
    const second = readOpenCodeTranscriptPage({
      dbPath: path,
      sessionId: 'ses-1',
      limit: 3,
      beforeMessageRowId: first!.beforeMessageRowId!
    })
    expect(second!.items.map((item) => item.message.id)).toEqual(['msg-5', 'msg-6', 'msg-7'])
  })

  it('counts the limit in renderable messages, batching past non-renderable rows', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    for (let index = 1; index <= 6; index++) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: index % 2 === 0 ? { type: 'text', text: `message ${index}` } : { type: 'step-start' }
      })
    }
    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 2 })
    expect(page!.items.map((item) => item.message.id)).toEqual(['msg-4', 'msg-6'])
    expect(page!.hasMore).toBe(true)
    // Cursor tracks the oldest raw row, including filtered rows, so the probe survives for the next page.
    expect(page!.beforeMessageRowId).toBe(messageRowid(db, 'msg-3'))
    const next = readOpenCodeTranscriptPage({
      dbPath: path,
      sessionId: 'ses-1',
      limit: 2,
      beforeMessageRowId: page!.beforeMessageRowId!
    })
    expect(next!.items.map((item) => item.message.id)).toEqual(['msg-2'])
    expect(next!.hasMore).toBe(false)
  })

  it('walks a fully non-renderable history to its oldest raw row without dead-ending', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    for (let index = 1; index <= 4; index++) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: { type: 'step-start' }
      })
    }
    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 2 })
    expect(page!.items).toEqual([])
    expect(page!.hasMore).toBe(false)
    expect(page!.beforeMessageRowId).toBe(messageRowid(db, 'msg-1'))
  })

  it('re-reads messages trimmed by an overshoot batch instead of skipping them', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    // Sparse first batch then dense second batch must trim without skipping overflow.
    for (const index of [1, 2, 3, 5]) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: { type: 'text', text: `message ${index}` }
      })
    }
    insertMessage(db, { id: 'msg-4', time: 4 })
    insertPart(db, {
      id: 'prt-4',
      messageId: 'msg-4',
      time: 4,
      data: { type: 'step-start' }
    })
    const seen: string[] = []
    let cursor: number | null = null
    for (;;) {
      const page = readOpenCodeTranscriptPage({
        dbPath: path,
        sessionId: 'ses-1',
        limit: 2,
        ...(cursor !== null ? { beforeMessageRowId: cursor } : {})
      })
      seen.push(...page!.items.map((item) => item.message.id))
      if (!page!.hasMore) {
        break
      }
      cursor = page!.beforeMessageRowId
    }
    // Pages arrive newest-first; preserve order within each page and include every message once.
    expect(seen).toEqual(['msg-3', 'msg-5', 'msg-1', 'msg-2'])
  })

  it('keeps hasMore true when a trim happens on the final short batch', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    // A final batch can overshoot despite reaching table end; trimming must retain hasMore.
    for (const index of [1, 2, 4]) {
      insertMessage(db, { id: `msg-${index}`, time: index })
      insertPart(db, {
        id: `prt-${index}`,
        messageId: `msg-${index}`,
        time: index,
        data: { type: 'text', text: `message ${index}` }
      })
    }
    insertMessage(db, { id: 'msg-3', time: 3 })
    insertPart(db, {
      id: 'prt-3',
      messageId: 'msg-3',
      time: 3,
      data: { type: 'step-start' }
    })
    const first = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 2 })
    expect(first!.hasMore).toBe(true)
    const second = readOpenCodeTranscriptPage({
      dbPath: path,
      sessionId: 'ses-1',
      limit: 2,
      beforeMessageRowId: first!.beforeMessageRowId!
    })
    expect(second!.items.map((item) => item.message.id)).toEqual(['msg-1'])
    expect(second!.hasMore).toBe(false)
  })

  it('maps image file parts to image refs and skips non-image files', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    // Only native-platform file URLs decode to local paths; foreign paths stay opaque.
    const localPath = process.platform === 'win32' ? 'C:\\Users\\u\\pic.jpg' : '/home/user/pic.jpg'
    const fileUrl = pathToFileURL(localPath).href
    insertMessage(db, { id: 'msg-1', time: 1, role: 'user' })
    insertPart(db, {
      id: 'prt-data',
      messageId: 'msg-1',
      time: 1,
      data: {
        type: 'file',
        mime: 'image/png',
        url: 'data:image/png;base64,AAA',
        filename: 'paste.png'
      }
    })
    insertPart(db, {
      id: 'prt-file-url',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'file', mime: 'image/jpeg', url: fileUrl }
    })
    insertPart(db, {
      id: 'prt-plain-path',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'file', mime: 'image/webp', url: '/tmp/opencode/paste.webp' }
    })
    insertPart(db, {
      id: 'prt-pdf',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'file', mime: 'application/pdf', url: 'file:///tmp/doc.pdf' }
    })
    insertPart(db, {
      id: 'prt-text',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'text', text: 'prompt body' }
    })

    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 10 })
    expect(page!.items[0]!.message.blocks).toEqual([
      { type: 'image-ref', url: 'data:image/png;base64,AAA', alt: 'paste.png' },
      { type: 'image-ref', path: localPath },
      { type: 'image-ref', path: '/tmp/opencode/paste.webp' },
      { type: 'text', text: 'prompt body' }
    ])
  })

  it('filters synthetic text parts and maps tool parts to call/result blocks', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    insertMessage(db, { id: 'msg-1', time: 1, role: 'user' })
    insertPart(db, {
      id: 'prt-syn',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'text', text: 'injected context', synthetic: true }
    })
    insertPart(db, {
      id: 'prt-text',
      messageId: 'msg-1',
      time: 1,
      data: { type: 'text', text: 'real prompt' }
    })
    insertMessage(db, { id: 'msg-2', time: 2, role: 'assistant' })
    insertPart(db, {
      id: 'prt-tool-done',
      messageId: 'msg-2',
      time: 2,
      data: {
        type: 'tool',
        tool: 'bash',
        callID: 'call-1',
        state: { status: 'completed', input: { command: 'ls' }, output: 'file list' }
      }
    })
    insertPart(db, {
      id: 'prt-tool-pending',
      messageId: 'msg-2',
      time: 2,
      data: { type: 'tool', tool: 'read', callID: 'call-2', state: { status: 'running' } }
    })

    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 10 })
    const user = page!.items.find((item) => item.message.id === 'msg-1')!
    expect(user.message.blocks).toEqual([{ type: 'text', text: 'real prompt' }])
    const assistant = page!.items.find((item) => item.message.id === 'msg-2')!
    expect(assistant.message.blocks).toEqual([
      {
        type: 'tool-call',
        callId: 'call-1',
        name: 'bash',
        state: 'completed',
        input: { command: 'ls' }
      },
      { type: 'tool-result', callId: 'call-1', output: 'file list' },
      { type: 'tool-call', callId: 'call-2', name: 'read', state: 'running', input: undefined }
    ])
  })

  it('surfaces a tool error string as the result output', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    insertMessage(db, { id: 'msg-1', time: 1, role: 'assistant' })
    insertPart(db, {
      id: 'prt-tool-error',
      messageId: 'msg-1',
      time: 1,
      data: {
        type: 'tool',
        tool: 'bash',
        callID: 'call-1',
        state: { status: 'error', input: { command: 'ls' }, error: 'exit 1: boom' }
      }
    })

    const page = readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 10 })
    expect(page!.items[0]!.message.blocks).toEqual([
      {
        type: 'tool-call',
        callId: 'call-1',
        name: 'bash',
        state: 'failed',
        input: { command: 'ls' }
      },
      { type: 'tool-result', callId: 'call-1', output: 'exit 1: boom', isError: true }
    ])
  })
})

describe('readOpenCodeTranscriptSignal', () => {
  it('aggregates counts and maxes per session', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    insertMessage(db, { id: 'msg-1', time: 1 })
    insertMessage(db, { id: 'msg-2', time: 2 })
    insertMessage(db, { id: 'other-1', time: 3, sessionId: 'ses-2' })
    insertPart(db, { id: 'prt-1', messageId: 'msg-1', time: 10, data: { type: 'text', text: 'a' } })
    insertPart(db, { id: 'prt-2', messageId: 'msg-2', time: 20, data: { type: 'text', text: 'b' } })

    db.exec(
      "ALTER TABLE session ADD COLUMN directory TEXT; UPDATE session SET directory = '/folder'"
    )
    const signal = readOpenCodeTranscriptSignal(path, 'ses-1')
    expect(signal).toEqual({
      cwd: '/folder',
      messageCount: 2,
      partCount: 2,
      maxMessageRowId: 2,
      maxPartTimeUpdated: 20
    })
  })

  it('returns null when the session row does not exist', () => {
    const { db, path } = createTempDb()
    applySchema(db)
    expect(readOpenCodeTranscriptSignal(path, 'missing')).toBeNull()
  })
})

it('stops before fetching a third part batch when earlier batches exhaust the page byte budget', () => {
  const { db, path } = createTempDb()
  applySchema(db)
  const text = 'x'.repeat(100_000)
  db.exec('BEGIN')
  {
    for (let i = 0; i < 300; i++) {
      insertMessage(db, { id: `budget-${i}`, time: i })
      insertPart(db, {
        id: `part-${i}`,
        messageId: `budget-${i}`,
        time: i,
        data: { type: 'text', text }
      })
    }
  }
  db.exec('COMMIT')
  const prepared = vi.spyOn(Database.prototype, 'prepare')
  try {
    expect(() =>
      readOpenCodeTranscriptPage({ dbPath: path, sessionId: 'ses-1', limit: 300 })
    ).toThrow('read limit')
    const partQueries = prepared.mock.calls.filter(
      ([sql]) => sql.includes('FROM part') && sql.includes('message_id IN')
    )
    expect(partQueries).toHaveLength(2)
  } finally {
    prepared.mockRestore()
  }
})