brainy/src/storage/adapters/fileSystemStorage.ts
David Snelling 7f082df6a3 **test(api): add integration tests for API endpoints and core functionality**
- **Integration Tests**:
  - Introduced `api-integration.test.ts` to validate API functionality:
    - Verifies text insertion, vector embedding generation, and search operations.
    - Confirms HNSW index correctness for vector similarity search.
    - Ensures no dimensional mismatches in embeddings.

- **Test Server**:
  - Added test server utilizing Express for endpoint simulation (`/insert` and `/search/text`).

- **Dependencies**:
  - Introduced `express` and `node-fetch` as new dependencies for testing purposes.

- **Vitest Fix**:
  - Updated `vitest.config.ts` to resolve the `process.memoryUsage` error by setting `logHeapUsage: false`.

- **Package Updates**:
  - Modified `package-lock.json` to include newly added dependencies and updates.

**Purpose**: Guarantees the stability of core API endpoints and vector-related functionality, ensuring reliable behavior for end-to-end scenarios.
2025-07-28 10:04:45 -07:00

588 lines
17 KiB
TypeScript

/**
* File System Storage Adapter
* File system storage adapter for Node.js environments
*/
import { GraphVerb, HNSWNoun } from '../../coreTypes.js'
import {
BaseStorage,
NOUNS_DIR,
VERBS_DIR,
METADATA_DIR,
INDEX_DIR
} from '../baseStorage.js'
// Type aliases for better readability
type HNSWNode = HNSWNoun
type Edge = GraphVerb
// Node.js modules - dynamically imported to avoid issues in browser environments
let fs: any
let path: any
// Try to load Node.js modules
try {
// Using dynamic imports to avoid issues in browser environments
const fsPromise = import('fs')
const pathPromise = import('path')
Promise.all([fsPromise, pathPromise])
.then(([fsModule, pathModule]) => {
fs = fsModule
path = pathModule.default
})
.catch((error) => {
console.error('Failed to load Node.js modules:', error)
})
} catch (error) {
console.error(
'FileSystemStorage: Failed to load Node.js modules. This adapter is not supported in this environment.',
error
)
}
/**
* File system storage adapter for Node.js environments
* Uses the file system to store data in the specified directory structure
*/
export class FileSystemStorage extends BaseStorage {
private rootDir: string
private nounsDir!: string
private verbsDir!: string
private metadataDir!: string
private indexDir!: string
/**
* Initialize the storage adapter
* @param rootDirectory The root directory for storage
*/
constructor(rootDirectory: string) {
super()
this.rootDir = rootDirectory
// Defer path operations until init() when path module is guaranteed to be loaded
}
/**
* Initialize the storage adapter
*/
public async init(): Promise<void> {
if (this.isInitialized) {
return
}
// Check if Node.js modules are available
if (!fs || !path) {
throw new Error(
'FileSystemStorage requires a Node.js environment, but `fs` and `path` modules could not be loaded.'
)
}
try {
// Initialize directory paths now that path module is loaded
this.nounsDir = path.join(this.rootDir, NOUNS_DIR)
this.verbsDir = path.join(this.rootDir, VERBS_DIR)
this.metadataDir = path.join(this.rootDir, METADATA_DIR)
this.indexDir = path.join(this.rootDir, INDEX_DIR)
// Create the root directory if it doesn't exist
await this.ensureDirectoryExists(this.rootDir)
// Create the nouns directory if it doesn't exist
await this.ensureDirectoryExists(this.nounsDir)
// Create the verbs directory if it doesn't exist
await this.ensureDirectoryExists(this.verbsDir)
// Create the metadata directory if it doesn't exist
await this.ensureDirectoryExists(this.metadataDir)
// Create the index directory if it doesn't exist
await this.ensureDirectoryExists(this.indexDir)
this.isInitialized = true
} catch (error) {
console.error('Error initializing FileSystemStorage:', error)
throw error
}
}
/**
* Ensure a directory exists, creating it if necessary
*/
private async ensureDirectoryExists(dirPath: string): Promise<void> {
try {
await fs.promises.mkdir(dirPath, { recursive: true })
} catch (error: any) {
// Ignore EEXIST error, which means the directory already exists
if (error.code !== 'EEXIST') {
throw error
}
}
}
/**
* Save a node to storage
*/
protected async saveNode(node: HNSWNode): Promise<void> {
await this.ensureInitialized()
// Convert connections Map to a serializable format
const serializableNode = {
...node,
connections: this.mapToObject(node.connections, (set) =>
Array.from(set as Set<string>)
)
}
const filePath = path.join(this.nounsDir, `${node.id}.json`)
await fs.promises.writeFile(
filePath,
JSON.stringify(serializableNode, null, 2)
)
}
/**
* Get a node from storage
*/
protected async getNode(id: string): Promise<HNSWNode | null> {
await this.ensureInitialized()
const filePath = path.join(this.nounsDir, `${id}.json`)
try {
const data = await fs.promises.readFile(filePath, 'utf-8')
const parsedNode = JSON.parse(data)
// Convert serialized connections back to Map<number, Set<string>>
const connections = new Map<number, Set<string>>()
for (const [level, nodeIds] of Object.entries(parsedNode.connections)) {
connections.set(Number(level), new Set(nodeIds as string[]))
}
return {
id: parsedNode.id,
vector: parsedNode.vector,
connections
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading node ${id}:`, error)
}
return null
}
}
/**
* Get all nodes from storage
*/
protected async getAllNodes(): Promise<HNSWNode[]> {
await this.ensureInitialized()
const allNodes: HNSWNode[] = []
try {
const files = await fs.promises.readdir(this.nounsDir)
for (const file of files) {
if (file.endsWith('.json')) {
const filePath = path.join(this.nounsDir, file)
const data = await fs.promises.readFile(filePath, 'utf-8')
const parsedNode = JSON.parse(data)
// Convert serialized connections back to Map<number, Set<string>>
const connections = new Map<number, Set<string>>()
for (const [level, nodeIds] of Object.entries(
parsedNode.connections
)) {
connections.set(Number(level), new Set(nodeIds as string[]))
}
allNodes.push({
id: parsedNode.id,
vector: parsedNode.vector,
connections
})
}
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading directory ${this.nounsDir}:`, error)
}
}
return allNodes
}
/**
* Get nodes by noun type
* @param nounType The noun type to filter by
* @returns Promise that resolves to an array of nodes of the specified noun type
*/
protected async getNodesByNounType(nounType: string): Promise<HNSWNode[]> {
await this.ensureInitialized()
const nouns: HNSWNode[] = []
try {
const files = await fs.promises.readdir(this.nounsDir)
for (const file of files) {
if (file.endsWith('.json')) {
const filePath = path.join(this.nounsDir, file)
const data = await fs.promises.readFile(filePath, 'utf-8')
const parsedNode = JSON.parse(data)
// Filter by noun type using metadata
const nodeId = parsedNode.id
const metadata = await this.getMetadata(nodeId)
if (metadata && metadata.noun === nounType) {
// Convert serialized connections back to Map<number, Set<string>>
const connections = new Map<number, Set<string>>()
for (const [level, nodeIds] of Object.entries(
parsedNode.connections
)) {
connections.set(Number(level), new Set(nodeIds as string[]))
}
nouns.push({
id: parsedNode.id,
vector: parsedNode.vector,
connections
})
}
}
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading directory ${this.nounsDir}:`, error)
}
}
return nouns
}
/**
* Delete a node from storage
*/
protected async deleteNode(id: string): Promise<void> {
await this.ensureInitialized()
const filePath = path.join(this.nounsDir, `${id}.json`)
try {
await fs.promises.unlink(filePath)
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error deleting node file ${filePath}:`, error)
throw error
}
}
}
/**
* Save an edge to storage
*/
protected async saveEdge(edge: Edge): Promise<void> {
await this.ensureInitialized()
// Convert connections Map to a serializable format
const serializableEdge = {
...edge,
connections: this.mapToObject(edge.connections, (set) =>
Array.from(set as Set<string>)
)
}
const filePath = path.join(this.verbsDir, `${edge.id}.json`)
await fs.promises.writeFile(
filePath,
JSON.stringify(serializableEdge, null, 2)
)
}
/**
* Get an edge from storage
*/
protected async getEdge(id: string): Promise<Edge | null> {
await this.ensureInitialized()
const filePath = path.join(this.verbsDir, `${id}.json`)
try {
const data = await fs.promises.readFile(filePath, 'utf-8')
const parsedEdge = JSON.parse(data)
// Convert serialized connections back to Map<number, Set<string>>
const connections = new Map<number, Set<string>>()
for (const [level, nodeIds] of Object.entries(parsedEdge.connections)) {
connections.set(Number(level), new Set(nodeIds as string[]))
}
return {
id: parsedEdge.id,
vector: parsedEdge.vector,
connections,
sourceId: parsedEdge.sourceId,
targetId: parsedEdge.targetId,
type: parsedEdge.type,
weight: parsedEdge.weight,
metadata: parsedEdge.metadata
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading edge ${id}:`, error)
}
return null
}
}
/**
* Get all edges from storage
*/
protected async getAllEdges(): Promise<Edge[]> {
await this.ensureInitialized()
const allEdges: Edge[] = []
try {
const files = await fs.promises.readdir(this.verbsDir)
for (const file of files) {
if (file.endsWith('.json')) {
const filePath = path.join(this.verbsDir, file)
const data = await fs.promises.readFile(filePath, 'utf-8')
const parsedEdge = JSON.parse(data)
// Convert serialized connections back to Map<number, Set<string>>
const connections = new Map<number, Set<string>>()
for (const [level, nodeIds] of Object.entries(
parsedEdge.connections
)) {
connections.set(Number(level), new Set(nodeIds as string[]))
}
allEdges.push({
id: parsedEdge.id,
vector: parsedEdge.vector,
connections,
sourceId: parsedEdge.sourceId,
targetId: parsedEdge.targetId,
type: parsedEdge.type,
weight: parsedEdge.weight,
metadata: parsedEdge.metadata
})
}
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading directory ${this.verbsDir}:`, error)
}
}
return allEdges
}
/**
* Get edges by source
*/
protected async getEdgesBySource(sourceId: string): Promise<Edge[]> {
const edges = await this.getAllEdges()
return edges.filter((edge) => edge.sourceId === sourceId)
}
/**
* Get edges by target
*/
protected async getEdgesByTarget(targetId: string): Promise<Edge[]> {
const edges = await this.getAllEdges()
return edges.filter((edge) => edge.targetId === targetId)
}
/**
* Get edges by type
*/
protected async getEdgesByType(type: string): Promise<Edge[]> {
const edges = await this.getAllEdges()
return edges.filter((edge) => edge.type === type)
}
/**
* Delete an edge from storage
*/
protected async deleteEdge(id: string): Promise<void> {
await this.ensureInitialized()
const filePath = path.join(this.verbsDir, `${id}.json`)
try {
await fs.promises.unlink(filePath)
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error deleting edge file ${filePath}:`, error)
throw error
}
}
}
/**
* Save metadata to storage
*/
public async saveMetadata(id: string, metadata: any): Promise<void> {
await this.ensureInitialized()
const filePath = path.join(this.metadataDir, `${id}.json`)
await fs.promises.writeFile(filePath, JSON.stringify(metadata, null, 2))
}
/**
* Get metadata from storage
*/
public async getMetadata(id: string): Promise<any | null> {
await this.ensureInitialized()
const filePath = path.join(this.metadataDir, `${id}.json`)
try {
const data = await fs.promises.readFile(filePath, 'utf-8')
return JSON.parse(data)
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error reading metadata ${id}:`, error)
}
return null
}
}
/**
* Clear all data from storage
*/
public async clear(): Promise<void> {
await this.ensureInitialized()
// Helper function to remove all files in a directory
const removeDirectoryContents = async (dirPath: string): Promise<void> => {
try {
const files = await fs.promises.readdir(dirPath)
for (const file of files) {
const filePath = path.join(dirPath, file)
const stats = await fs.promises.stat(filePath)
if (stats.isDirectory()) {
await removeDirectoryContents(filePath)
await fs.promises.rmdir(filePath)
} else {
await fs.promises.unlink(filePath)
}
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(`Error removing directory contents ${dirPath}:`, error)
throw error
}
}
}
// Remove all files in the nouns directory
await removeDirectoryContents(this.nounsDir)
// Remove all files in the verbs directory
await removeDirectoryContents(this.verbsDir)
// Remove all files in the metadata directory
await removeDirectoryContents(this.metadataDir)
// Remove all files in the index directory
await removeDirectoryContents(this.indexDir)
}
/**
* Get information about storage usage and capacity
*/
public async getStorageStatus(): Promise<{
type: string
used: number
quota: number | null
details?: Record<string, any>
}> {
await this.ensureInitialized()
try {
// Calculate the total size of all files in the storage directories
let totalSize = 0
// Helper function to calculate directory size
const calculateSize = async (dirPath: string): Promise<number> => {
let size = 0
try {
const files = await fs.promises.readdir(dirPath)
for (const file of files) {
const filePath = path.join(dirPath, file)
const stats = await fs.promises.stat(filePath)
if (stats.isDirectory()) {
size += await calculateSize(filePath)
} else {
size += stats.size
}
}
} catch (error: any) {
if (error.code !== 'ENOENT') {
console.error(
`Error calculating size for directory ${dirPath}:`,
error
)
}
}
return size
}
// Calculate size for each directory
const nounsDirSize = await calculateSize(this.nounsDir)
const verbsDirSize = await calculateSize(this.verbsDir)
const metadataDirSize = await calculateSize(this.metadataDir)
const indexDirSize = await calculateSize(this.indexDir)
totalSize = nounsDirSize + verbsDirSize + metadataDirSize + indexDirSize
// Count files in each directory
const nounsCount = (await fs.promises.readdir(this.nounsDir)).filter(
(file: string) => file.endsWith('.json')
).length
const verbsCount = (await fs.promises.readdir(this.verbsDir)).filter(
(file: string) => file.endsWith('.json')
).length
const metadataCount = (
await fs.promises.readdir(this.metadataDir)
).filter((file: string) => file.endsWith('.json')).length
// Count nouns by type using metadata
const nounTypeCounts: Record<string, number> = {}
const metadataFiles = await fs.promises.readdir(this.metadataDir)
for (const file of metadataFiles) {
if (file.endsWith('.json')) {
try {
const filePath = path.join(this.metadataDir, file)
const data = await fs.promises.readFile(filePath, 'utf-8')
const metadata = JSON.parse(data)
if (metadata.noun) {
nounTypeCounts[metadata.noun] =
(nounTypeCounts[metadata.noun] || 0) + 1
}
} catch (error) {
console.error(`Error reading metadata file ${file}:`, error)
}
}
}
return {
type: 'filesystem',
used: totalSize,
quota: null, // File system doesn't provide quota information
details: {
rootDirectory: this.rootDir,
nounsCount,
verbsCount,
metadataCount,
nounsDirSize,
verbsDirSize,
metadataDirSize,
indexDirSize,
nounTypes: nounTypeCounts
}
}
} catch (error) {
console.error('Failed to get storage status:', error)
return {
type: 'filesystem',
used: 0,
quota: null,
details: { error: String(error) }
}
}
}
}