Buckets:
| export * as ReadToolFileSystem from "./read-filesystem" | |
| import path from "path" | |
| import { pathToFileURL } from "url" | |
| import { Context, Effect, Layer, Option, Schema } from "effect" | |
| import { FileSystem } from "../filesystem" | |
| import { FSUtil } from "../fs-util" | |
| import { AbsolutePath, PositiveInt, RelativePath } from "../schema" | |
| export const MAX_READ_LINES = 2_000 | |
| export const MAX_READ_BYTES = 50 * 1024 | |
| export const MAX_MEDIA_INGEST_BYTES = 20 * 1024 * 1024 | |
| const MAX_LINE_LENGTH = 2_000 | |
| const MAX_LINE_SUFFIX = `... (line truncated to ${MAX_LINE_LENGTH} chars)` | |
| export class BinaryFileError extends Error { | |
| constructor(readonly resource: string) { | |
| super(`Cannot read binary file: ${resource}`) | |
| this.name = "BinaryFileError" | |
| } | |
| } | |
| export class MediaIngestLimitError extends Error { | |
| constructor( | |
| readonly resource: string, | |
| readonly maximumBytes: number, | |
| ) { | |
| super(`Media exceeds ${maximumBytes} byte ingestion limit: ${resource}`) | |
| this.name = "MediaIngestLimitError" | |
| } | |
| } | |
| export const PageInput = Schema.Struct({ | |
| offset: PositiveInt.pipe(Schema.optional), | |
| limit: PositiveInt.check(Schema.isLessThanOrEqualTo(MAX_READ_LINES)).pipe(Schema.optional), | |
| }) | |
| export type PageInput = typeof PageInput.Type | |
| export class TextPage extends Schema.Class<TextPage>("ReadTool.TextPage")({ | |
| type: Schema.Literal("text-page"), | |
| content: Schema.String, | |
| mime: Schema.String, | |
| offset: PositiveInt, | |
| truncated: Schema.Boolean, | |
| next: PositiveInt.pipe(Schema.optional), | |
| }) {} | |
| export class ListPage extends Schema.Class<ListPage>("ReadTool.ListPage")({ | |
| entries: Schema.Array(FileSystem.Entry), | |
| truncated: Schema.Boolean, | |
| next: PositiveInt.pipe(Schema.optional), | |
| }) {} | |
| export interface Interface { | |
| readonly inspect: (path: AbsolutePath) => Effect.Effect<"file" | "directory"> | |
| readonly read: ( | |
| path: AbsolutePath, | |
| resource: string, | |
| page?: PageInput, | |
| ) => Effect.Effect<FileSystem.Content | TextPage> | |
| readonly list: (path: AbsolutePath, page?: PageInput) => Effect.Effect<ListPage> | |
| } | |
| export class Service extends Context.Service<Service, Interface>()("@opencode/ReadToolFileSystem") {} | |
| const extensions = new Set([ | |
| ".zip", | |
| ".tar", | |
| ".gz", | |
| ".exe", | |
| ".dll", | |
| ".so", | |
| ".class", | |
| ".jar", | |
| ".war", | |
| ".7z", | |
| ".doc", | |
| ".docx", | |
| ".xls", | |
| ".xlsx", | |
| ".ppt", | |
| ".pptx", | |
| ".odt", | |
| ".ods", | |
| ".odp", | |
| ".bin", | |
| ".dat", | |
| ".obj", | |
| ".o", | |
| ".a", | |
| ".lib", | |
| ".wasm", | |
| ".pyc", | |
| ".pyo", | |
| ]) | |
| const startsWith = (bytes: Uint8Array, prefix: number[]) => prefix.every((value, index) => bytes[index] === value) | |
| const imageMime = (bytes: Uint8Array) => { | |
| if (startsWith(bytes, [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a])) return "image/png" | |
| if (startsWith(bytes, [0xff, 0xd8, 0xff])) return "image/jpeg" | |
| if (startsWith(bytes, [0x47, 0x49, 0x46, 0x38])) return "image/gif" | |
| if (startsWith(bytes, [0x52, 0x49, 0x46, 0x46]) && startsWith(bytes.subarray(8), [0x57, 0x45, 0x42, 0x50])) | |
| return "image/webp" | |
| } | |
| const binary = (resource: string, bytes: Uint8Array) => { | |
| if (extensions.has(path.extname(resource).toLowerCase())) return true | |
| if (bytes.length === 0) return false | |
| let nonPrintable = 0 | |
| for (const byte of bytes) { | |
| if (byte === 0) return true | |
| if (byte < 9 || (byte > 13 && byte < 32)) nonPrintable++ | |
| } | |
| return nonPrintable / bytes.length > 0.3 | |
| } | |
| export const inspect = Effect.fn("ReadTool.inspect")(function* (fs: FSUtil.Interface, input: string) { | |
| const info = yield* fs.stat(input).pipe(Effect.orDie) | |
| const type = info.type === "File" ? "file" : info.type === "Directory" ? "directory" : undefined | |
| if (!type) return yield* Effect.die(new Error("Path is not a file or directory")) | |
| return type | |
| }) | |
| export const read = Effect.fn("ReadTool.read")(function* ( | |
| fs: FSUtil.Interface, | |
| input: string, | |
| resource: string, | |
| page: PageInput = {}, | |
| ) { | |
| const real = yield* fs.realPath(input).pipe(Effect.orDie) | |
| return yield* Effect.scoped( | |
| Effect.gen(function* () { | |
| const file = yield* fs.open(real, { flag: "r" }).pipe(Effect.orDie) | |
| const info = yield* file.stat.pipe(Effect.orDie) | |
| if (info.type !== "File") return yield* Effect.die(new Error("Path is not a file")) | |
| const first = Option.getOrElse( | |
| yield* file.readAlloc(Math.min(64 * 1024, Number(info.size) || 4 * 1024)).pipe(Effect.orDie), | |
| () => new Uint8Array(), | |
| ) | |
| const mime = imageMime(first) | |
| if (mime) { | |
| if (info.size > MAX_MEDIA_INGEST_BYTES) | |
| return yield* Effect.die(new MediaIngestLimitError(resource, MAX_MEDIA_INGEST_BYTES)) | |
| const chunks = [first] | |
| let total = first.length | |
| while (total <= MAX_MEDIA_INGEST_BYTES) { | |
| const chunk = yield* file | |
| .readAlloc(Math.min(64 * 1024, MAX_MEDIA_INGEST_BYTES + 1 - total)) | |
| .pipe(Effect.orDie) | |
| if (Option.isNone(chunk)) break | |
| chunks.push(chunk.value) | |
| total += chunk.value.length | |
| } | |
| if (total > MAX_MEDIA_INGEST_BYTES) | |
| return yield* Effect.die(new MediaIngestLimitError(resource, MAX_MEDIA_INGEST_BYTES)) | |
| return { | |
| uri: pathToFileURL(real).href, | |
| name: path.basename(real), | |
| content: Buffer.concat( | |
| chunks.map((chunk) => Buffer.from(chunk)), | |
| total, | |
| ).toString("base64"), | |
| encoding: "base64" as const, | |
| mime, | |
| } | |
| } | |
| if (startsWith(first, [0x25, 0x50, 0x44, 0x46]) || binary(resource, first)) | |
| return yield* Effect.die(new BinaryFileError(resource)) | |
| const paged = info.size > MAX_READ_BYTES || page.offset !== undefined || page.limit !== undefined | |
| if (!paged) { | |
| const decoder = new TextDecoder("utf-8", { fatal: true }) | |
| const text = [yield* Effect.sync(() => decoder.decode(first, { stream: true }))] | |
| while (true) { | |
| const chunk = yield* file.readAlloc(64 * 1024).pipe(Effect.orDie) | |
| if (Option.isNone(chunk)) break | |
| if (chunk.value.includes(0)) return yield* Effect.die(new BinaryFileError(resource)) | |
| text.push(yield* Effect.sync(() => decoder.decode(chunk.value, { stream: true }))) | |
| } | |
| text.push(yield* Effect.sync(() => decoder.decode())) | |
| return { | |
| uri: pathToFileURL(real).href, | |
| name: path.basename(real), | |
| content: text.join(""), | |
| encoding: "utf8" as const, | |
| mime: FSUtil.mimeType(real), | |
| } | |
| } | |
| const offset = page.offset ?? 1 | |
| const limit = Math.min(page.limit ?? MAX_READ_LINES, MAX_READ_LINES) | |
| const lines: string[] = [] | |
| const decoder = new TextDecoder("utf-8", { fatal: true }) | |
| let pending = "" | |
| let discard = false | |
| let line = 1 | |
| let bytes = 0 | |
| let found = false | |
| let truncated = false | |
| let next: number | undefined | |
| const append = (input: string) => { | |
| if (line < offset) { | |
| line++ | |
| return | |
| } | |
| if (lines.length >= limit || bytes >= MAX_READ_BYTES) { | |
| truncated = true | |
| next ??= line++ | |
| return | |
| } | |
| found = true | |
| const text = input.length > MAX_LINE_LENGTH ? input.slice(0, MAX_LINE_LENGTH) + MAX_LINE_SUFFIX : input | |
| const size = Buffer.byteLength(text, "utf-8") + (lines.length > 0 ? 1 : 0) | |
| if (bytes + size > MAX_READ_BYTES) { | |
| truncated = true | |
| next ??= line++ | |
| return | |
| } | |
| lines.push(text) | |
| bytes += size | |
| line++ | |
| } | |
| const consume = (chunk: Uint8Array) => { | |
| if (chunk.includes(0)) throw new BinaryFileError(resource) | |
| let text = decoder.decode(chunk, { stream: true }) | |
| while (true) { | |
| const index = text.indexOf("\n") | |
| if (index === -1) { | |
| if (!discard) { | |
| pending += text | |
| if (pending.length > MAX_LINE_LENGTH) { | |
| pending = pending.slice(0, MAX_LINE_LENGTH + 1) | |
| discard = true | |
| } | |
| } | |
| break | |
| } | |
| const current = pending + (discard ? "" : text.slice(0, index)) | |
| pending = "" | |
| discard = false | |
| text = text.slice(index + 1) | |
| append(current.endsWith("\r") ? current.slice(0, -1) : current) | |
| } | |
| } | |
| yield* Effect.sync(() => consume(first)) | |
| while (true) { | |
| const chunk = yield* file.readAlloc(64 * 1024).pipe(Effect.orDie) | |
| if (Option.isNone(chunk)) break | |
| yield* Effect.sync(() => consume(chunk.value)) | |
| } | |
| const tail = yield* Effect.sync(() => decoder.decode()) | |
| if (!discard) pending += tail | |
| if (pending) append(pending.endsWith("\r") ? pending.slice(0, -1) : pending) | |
| if (!found && offset !== 1) return yield* Effect.die(new Error(`Offset ${offset} is out of range`)) | |
| return new TextPage({ | |
| type: "text-page", | |
| content: lines.join("\n"), | |
| mime: FSUtil.mimeType(real), | |
| offset, | |
| truncated, | |
| ...(next === undefined ? {} : { next }), | |
| }) | |
| }), | |
| ) | |
| }) | |
| export const list = Effect.fn("ReadTool.list")(function* (fs: FSUtil.Interface, input: string, page: PageInput = {}) { | |
| const real = yield* fs.realPath(input).pipe(Effect.orDie) | |
| const items = yield* fs.readDirectoryEntries(real).pipe(Effect.orDie) | |
| const offset = page.offset ?? 1 | |
| const limit = Math.min(page.limit ?? MAX_READ_LINES, MAX_READ_LINES) | |
| const entries = yield* Effect.forEach( | |
| items, | |
| (item) => | |
| Effect.gen(function* () { | |
| const absolute = path.join(real, item.name) | |
| const target = yield* fs.realPath(absolute).pipe(Effect.catch(() => Effect.void)) | |
| if (!target || !FSUtil.contains(real, target)) return | |
| const info = yield* fs.stat(target).pipe(Effect.catch(() => Effect.void)) | |
| const type = info?.type === "Directory" ? "directory" : info?.type === "File" ? "file" : undefined | |
| if (!type) return | |
| return new FileSystem.Entry({ | |
| path: RelativePath.make(item.name + (type === "directory" ? path.sep : "")), | |
| type, | |
| mime: type === "directory" ? "application/x-directory" : FSUtil.mimeType(target), | |
| }) | |
| }), | |
| { concurrency: 16 }, | |
| ) | |
| const visible = entries | |
| .filter((item): item is FileSystem.Entry => item !== undefined) | |
| .sort((a, b) => (a.type === b.type ? a.path.localeCompare(b.path) : a.type === "directory" ? -1 : 1)) | |
| const selected = visible.slice(offset - 1, offset - 1 + limit) | |
| const truncated = offset - 1 + selected.length < visible.length | |
| return new ListPage({ entries: selected, truncated, ...(truncated ? { next: offset + selected.length } : {}) }) | |
| }) | |
| export const layer = Layer.effect( | |
| Service, | |
| Effect.gen(function* () { | |
| const fs = yield* FSUtil.Service | |
| return Service.of({ | |
| inspect: (path) => inspect(fs, path), | |
| read: (path, resource, page) => read(fs, path, resource, page), | |
| list: (path, page) => list(fs, path, page), | |
| }) | |
| }), | |
| ) | |
Xet Storage Details
- Size:
- 11 kB
- Xet hash:
- 6ce513b9e7bddfea0640b7c3d9de17202d88c35d1be83d50e80bca2c6cdda8ca
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.