filePersistence.ts
utils/filePersistence/filePersistence.ts
288
Lines
7763
Bytes
3
Exports
10
Imports
10
Keywords
What this is
This page documents one file from the repository and includes its full source so you can read it without leaving the docs site.
Beginner explanation
This file is one piece of the larger system. Its name, directory, imports, and exports show where it fits. Start by reading the exports and related files first.
How it is used
Start from the exports list and related files. Those are the easiest clues for where this file fits into the system.
Expert explanation
Architecturally, this file intersects with file-tools. It contains 288 lines, 10 detected imports, and 3 detected exports.
Important relationships
Detected exports
runFilePersistenceexecuteFilePersistenceisFilePersistenceEnabled
Keywords
filesresultfilelengthpersistenceturnstarttimefailedlogdebugmodesignal
Detected imports
bun:bundlepath../../services/analytics/index.js../../services/api/filesApi.js../cwd.js../errors.js../log.js../sessionIngressAuth.js./outputsScanner.js./types.js
Source notes
This page embeds the full file contents. Small or leaf files are still indexed honestly instead of being over-explained.
Full source
/**
* File persistence orchestrator
*
* This module provides the main orchestration logic for persisting files
* at the end of each turn:
* - BYOC mode: Upload files to Files API and collect file IDs
* - 1P/Cloud mode: Query Files API listDirectory for file IDs (rclone handles sync)
*/
import { feature } from 'bun:bundle'
import { join, relative } from 'path'
import {
type AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS,
logEvent,
} from '../../services/analytics/index.js'
import {
type FilesApiConfig,
uploadSessionFiles,
} from '../../services/api/filesApi.js'
import { getCwd } from '../cwd.js'
import { errorMessage } from '../errors.js'
import { logError } from '../log.js'
import { getSessionIngressAuthToken } from '../sessionIngressAuth.js'
import {
findModifiedFiles,
getEnvironmentKind,
logDebug,
} from './outputsScanner.js'
import {
DEFAULT_UPLOAD_CONCURRENCY,
type FailedPersistence,
FILE_COUNT_LIMIT,
type FilesPersistedEventData,
OUTPUTS_SUBDIR,
type PersistedFile,
type TurnStartTime,
} from './types.js'
/**
* Execute file persistence for modified files in the outputs directory.
*
* Assembles all config internally:
* - Checks environment kind (CLAUDE_CODE_ENVIRONMENT_KIND)
* - Retrieves session access token
* - Requires CLAUDE_CODE_REMOTE_SESSION_ID for session ID
*
* @param turnStartTime - The timestamp when the turn started
* @param signal - Optional abort signal for cancellation
* @returns Event data, or null if not enabled or no files to persist
*/
export async function runFilePersistence(
turnStartTime: TurnStartTime,
signal?: AbortSignal,
): Promise<FilesPersistedEventData | null> {
const environmentKind = getEnvironmentKind()
if (environmentKind !== 'byoc') {
return null
}
const sessionAccessToken = getSessionIngressAuthToken()
if (!sessionAccessToken) {
return null
}
const sessionId = process.env.CLAUDE_CODE_REMOTE_SESSION_ID
if (!sessionId) {
logError(
new Error(
'File persistence enabled but CLAUDE_CODE_REMOTE_SESSION_ID is not set',
),
)
return null
}
const config: FilesApiConfig = {
oauthToken: sessionAccessToken,
sessionId,
}
const outputsDir = join(getCwd(), sessionId, OUTPUTS_SUBDIR)
// Check if aborted
if (signal?.aborted) {
logDebug('Persistence aborted before processing')
return null
}
const startTime = Date.now()
logEvent('tengu_file_persistence_started', {
mode: environmentKind as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS,
})
try {
let result: FilesPersistedEventData
if (environmentKind === 'byoc') {
result = await executeBYOCPersistence(
turnStartTime,
config,
outputsDir,
signal,
)
} else {
result = await executeCloudPersistence()
}
// Nothing to report
if (result.files.length === 0 && result.failed.length === 0) {
return null
}
const durationMs = Date.now() - startTime
logEvent('tengu_file_persistence_completed', {
success_count: result.files.length,
failure_count: result.failed.length,
duration_ms: durationMs,
mode: environmentKind as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS,
})
return result
} catch (error) {
logError(error)
logDebug(`File persistence failed: ${error}`)
const durationMs = Date.now() - startTime
logEvent('tengu_file_persistence_completed', {
success_count: 0,
failure_count: 0,
duration_ms: durationMs,
mode: environmentKind as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS,
error:
'exception' as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS,
})
return {
files: [],
failed: [
{
filename: outputsDir,
error: errorMessage(error),
},
],
}
}
}
/**
* Execute BYOC mode persistence: scan local filesystem for modified files,
* then upload to Files API.
*/
async function executeBYOCPersistence(
turnStartTime: TurnStartTime,
config: FilesApiConfig,
outputsDir: string,
signal?: AbortSignal,
): Promise<FilesPersistedEventData> {
// Find modified files via local filesystem scan
// Uses same directory structure as downloads: {cwd}/{sessionId}/outputs
const modifiedFiles = await findModifiedFiles(turnStartTime, outputsDir)
if (modifiedFiles.length === 0) {
logDebug('No modified files to persist')
return { files: [], failed: [] }
}
logDebug(`Found ${modifiedFiles.length} modified files`)
if (signal?.aborted) {
return { files: [], failed: [] }
}
// Enforce file count limit
if (modifiedFiles.length > FILE_COUNT_LIMIT) {
logDebug(
`File count limit exceeded: ${modifiedFiles.length} > ${FILE_COUNT_LIMIT}`,
)
logEvent('tengu_file_persistence_limit_exceeded', {
file_count: modifiedFiles.length,
limit: FILE_COUNT_LIMIT,
})
return {
files: [],
failed: [
{
filename: outputsDir,
error: `Too many files modified (${modifiedFiles.length}). Maximum: ${FILE_COUNT_LIMIT}.`,
},
],
}
}
const filesToProcess = modifiedFiles
.map(filePath => ({
path: filePath,
relativePath: relative(outputsDir, filePath),
}))
.filter(({ relativePath }) => {
// Security: skip files that resolve outside the outputs directory
if (relativePath.startsWith('..')) {
logDebug(`Skipping file outside outputs directory: ${relativePath}`)
return false
}
return true
})
logDebug(`BYOC mode: uploading ${filesToProcess.length} files`)
// Upload files in parallel
const results = await uploadSessionFiles(
filesToProcess,
config,
DEFAULT_UPLOAD_CONCURRENCY,
)
// Separate successful and failed uploads
const persistedFiles: PersistedFile[] = []
const failedFiles: FailedPersistence[] = []
for (const result of results) {
if (result.success) {
persistedFiles.push({
filename: result.path,
file_id: result.fileId,
})
} else {
failedFiles.push({
filename: result.path,
error: result.error,
})
}
}
logDebug(
`BYOC persistence complete: ${persistedFiles.length} uploaded, ${failedFiles.length} failed`,
)
return {
files: persistedFiles,
failed: failedFiles,
}
}
/**
* Execute Cloud (1P) mode persistence.
* TODO: Read file_id from xattr on output files. xattr-based file IDs are
* currently being added for 1P environments.
*/
function executeCloudPersistence(): FilesPersistedEventData {
logDebug('Cloud mode: xattr-based file ID reading not yet implemented')
return { files: [], failed: [] }
}
/**
* Execute file persistence and emit result via callback.
* Handles errors internally.
*/
export async function executeFilePersistence(
turnStartTime: TurnStartTime,
signal: AbortSignal,
onResult: (result: FilesPersistedEventData) => void,
): Promise<void> {
try {
const result = await runFilePersistence(turnStartTime, signal)
if (result) {
onResult(result)
}
} catch (error) {
logError(error)
}
}
/**
* Check if file persistence is enabled.
* Requires: feature flag ON, valid environment kind, session access token,
* and CLAUDE_CODE_REMOTE_SESSION_ID.
* This ensures only public-api/sessions users trigger file persistence,
* not normal Claude Code CLI users.
*/
export function isFilePersistenceEnabled(): boolean {
if (feature('FILE_PERSISTENCE')) {
return (
getEnvironmentKind() === 'byoc' &&
!!getSessionIngressAuthToken() &&
!!process.env.CLAUDE_CODE_REMOTE_SESSION_ID
)
}
return false
}