nikcli 0.0.6

package/src/file/ignore.ts

@@ -0,0 +1,83 @@
+import { sep } from "node:path"
+
+export namespace FileIgnore {
+  const FOLDERS = new Set([
+    "node_modules",
+    "bower_components",
+    ".pnpm-store",
+    "vendor",
+    ".npm",
+    "dist",
+    "build",
+    "out",
+    ".next",
+    "target",
+    "bin",
+    "obj",
+    ".git",
+    ".svn",
+    ".hg",
+    ".vscode",
+    ".idea",
+    ".turbo",
+    ".output",
+    "desktop",
+    ".sst",
+    ".cache",
+    ".webkit-cache",
+    "__pycache__",
+    ".pytest_cache",
+    "mypy_cache",
+    ".history",
+    ".gradle",
+  ])
+
+  const FILES = [
+    "**/*.swp",
+    "**/*.swo",
+
+    "**/*.pyc",
+
+    // OS
+    "**/.DS_Store",
+    "**/Thumbs.db",
+
+    // Logs & temp
+    "**/logs/**",
+    "**/tmp/**",
+    "**/temp/**",
+    "**/*.log",
+
+    // Coverage/test outputs
+    "**/coverage/**",
+    "**/.nyc_output/**",
+  ]
+
+  const FILE_GLOBS = FILES.map((p) => new Bun.Glob(p))
+
+  export const PATTERNS = [...FILES, ...FOLDERS]
+
+  export function match(
+    filepath: string,
+    opts?: {
+      extra?: Bun.Glob[]
+      whitelist?: Bun.Glob[]
+    },
+  ) {
+    for (const glob of opts?.whitelist || []) {
+      if (glob.match(filepath)) return false
+    }
+
+    const parts = filepath.split(sep)
+    for (let i = 0; i < parts.length; i++) {
+      if (FOLDERS.has(parts[i])) return true
+    }
+
+    const extra = opts?.extra || []
+    for (const glob of [...FILE_GLOBS, ...extra]) {
+      if (glob.match(filepath)) return true
+    }
+
+    return false
+  }
+}

package/src/file/index.ts

@@ -0,0 +1,411 @@
+import { BusEvent } from "@/bus/bus-event"
+import z from "zod"
+import { $ } from "bun"
+import type { BunFile } from "bun"
+import { formatPatch, structuredPatch } from "diff"
+import path from "path"
+import fs from "fs"
+import ignore from "ignore"
+import { Log } from "../util/log"
+import { Filesystem } from "../util/filesystem"
+import { Instance } from "../project/instance"
+import { Ripgrep } from "./ripgrep"
+import fuzzysort from "fuzzysort"
+import { Global } from "../global"
+
+export namespace File {
+  const log = Log.create({ service: "file" })
+
+  export const Info = z
+    .object({
+      path: z.string(),
+      added: z.number().int(),
+      removed: z.number().int(),
+      status: z.enum(["added", "deleted", "modified"]),
+    })
+    .meta({
+      ref: "File",
+    })
+
+  export type Info = z.infer<typeof Info>
+
+  export const Node = z
+    .object({
+      name: z.string(),
+      path: z.string(),
+      absolute: z.string(),
+      type: z.enum(["file", "directory"]),
+      ignored: z.boolean(),
+    })
+    .meta({
+      ref: "FileNode",
+    })
+  export type Node = z.infer<typeof Node>
+
+  export const Content = z
+    .object({
+      type: z.literal("text"),
+      content: z.string(),
+      diff: z.string().optional(),
+      patch: z
+        .object({
+          oldFileName: z.string(),
+          newFileName: z.string(),
+          oldHeader: z.string().optional(),
+          newHeader: z.string().optional(),
+          hunks: z.array(
+            z.object({
+              oldStart: z.number(),
+              oldLines: z.number(),
+              newStart: z.number(),
+              newLines: z.number(),
+              lines: z.array(z.string()),
+            }),
+          ),
+          index: z.string().optional(),
+        })
+        .optional(),
+      encoding: z.literal("base64").optional(),
+      mimeType: z.string().optional(),
+    })
+    .meta({
+      ref: "FileContent",
+    })
+  export type Content = z.infer<typeof Content>
+
+  async function shouldEncode(file: BunFile): Promise<boolean> {
+    const type = file.type?.toLowerCase()
+    log.info("shouldEncode", { type })
+    if (!type) return false
+
+    if (type.startsWith("text/")) return false
+    if (type.includes("charset=")) return false
+
+    const parts = type.split("/", 2)
+    const top = parts[0]
+    const rest = parts[1] ?? ""
+    const sub = rest.split(";", 1)[0]
+
+    const tops = ["image", "audio", "video", "font", "model", "multipart"]
+    if (tops.includes(top)) return true
+
+    const bins = [
+      "zip",
+      "gzip",
+      "bzip",
+      "compressed",
+      "binary",
+      "pdf",
+      "msword",
+      "powerpoint",
+      "excel",
+      "ogg",
+      "exe",
+      "dmg",
+      "iso",
+      "rar",
+    ]
+    if (bins.some((mark) => sub.includes(mark))) return true
+
+    return false
+  }
+
+  export const Event = {
+    Edited: BusEvent.define(
+      "file.edited",
+      z.object({
+        file: z.string(),
+      }),
+    ),
+  }
+
+  const state = Instance.state(async () => {
+    type Entry = { files: string[]; dirs: string[] }
+    let cache: Entry = { files: [], dirs: [] }
+    let fetching = false
+
+    const isGlobalHome = Instance.directory === Global.Path.home && Instance.project.id === "global"
+
+    const fn = async (result: Entry) => {
+      // Disable scanning if in root of file system
+      if (Instance.directory === path.parse(Instance.directory).root) return
+      fetching = true
+
+      if (isGlobalHome) {
+        const dirs = new Set<string>()
+        const ignore = new Set<string>()
+
+        if (process.platform === "darwin") ignore.add("Library")
+        if (process.platform === "win32") ignore.add("AppData")
+
+        const ignoreNested = new Set(["node_modules", "dist", "build", "target", "vendor"])
+        const shouldIgnore = (name: string) => name.startsWith(".") || ignore.has(name)
+        const shouldIgnoreNested = (name: string) => name.startsWith(".") || ignoreNested.has(name)
+
+        const top = await fs.promises
+          .readdir(Instance.directory, { withFileTypes: true })
+          .catch(() => [] as fs.Dirent[])
+
+        for (const entry of top) {
+          if (!entry.isDirectory()) continue
+          if (shouldIgnore(entry.name)) continue
+          dirs.add(entry.name + "/")
+
+          const base = path.join(Instance.directory, entry.name)
+          const children = await fs.promises.readdir(base, { withFileTypes: true }).catch(() => [] as fs.Dirent[])
+          for (const child of children) {
+            if (!child.isDirectory()) continue
+            if (shouldIgnoreNested(child.name)) continue
+            dirs.add(entry.name + "/" + child.name + "/")
+          }
+        }
+
+        result.dirs = Array.from(dirs).toSorted()
+        cache = result
+        fetching = false
+        return
+      }
+
+      const set = new Set<string>()
+      for await (const file of Ripgrep.files({ cwd: Instance.directory })) {
+        result.files.push(file)
+        let current = file
+        while (true) {
+          const dir = path.dirname(current)
+          if (dir === ".") break
+          if (dir === current) break
+          current = dir
+          if (set.has(dir)) continue
+          set.add(dir)
+          result.dirs.push(dir + "/")
+        }
+      }
+      cache = result
+      fetching = false
+    }
+    fn(cache)
+
+    return {
+      async files() {
+        if (!fetching) {
+          fn({
+            files: [],
+            dirs: [],
+          })
+        }
+        return cache
+      },
+    }
+  })
+
+  export function init() {
+    state()
+  }
+
+  export async function status() {
+    const project = Instance.project
+    if (project.vcs !== "git") return []
+
+    const diffOutput = await $`git diff --numstat HEAD`.cwd(Instance.directory).quiet().nothrow().text()
+
+    const changedFiles: Info[] = []
+
+    if (diffOutput.trim()) {
+      const lines = diffOutput.trim().split("\n")
+      for (const line of lines) {
+        const [added, removed, filepath] = line.split("\t")
+        changedFiles.push({
+          path: filepath,
+          added: added === "-" ? 0 : parseInt(added, 10),
+          removed: removed === "-" ? 0 : parseInt(removed, 10),
+          status: "modified",
+        })
+      }
+    }
+
+    const untrackedOutput = await $`git ls-files --others --exclude-standard`
+      .cwd(Instance.directory)
+      .quiet()
+      .nothrow()
+      .text()
+
+    if (untrackedOutput.trim()) {
+      const untrackedFiles = untrackedOutput.trim().split("\n")
+      for (const filepath of untrackedFiles) {
+        try {
+          const content = await Bun.file(path.join(Instance.directory, filepath)).text()
+          const lines = content.split("\n").length
+          changedFiles.push({
+            path: filepath,
+            added: lines,
+            removed: 0,
+            status: "added",
+          })
+        } catch {
+          continue
+        }
+      }
+    }
+
+    // Get deleted files
+    const deletedOutput = await $`git diff --name-only --diff-filter=D HEAD`
+      .cwd(Instance.directory)
+      .quiet()
+      .nothrow()
+      .text()
+
+    if (deletedOutput.trim()) {
+      const deletedFiles = deletedOutput.trim().split("\n")
+      for (const filepath of deletedFiles) {
+        changedFiles.push({
+          path: filepath,
+          added: 0,
+          removed: 0, // Could get original line count but would require another git command
+          status: "deleted",
+        })
+      }
+    }
+
+    return changedFiles.map((x) => ({
+      ...x,
+      path: path.relative(Instance.directory, x.path),
+    }))
+  }
+
+  export async function read(file: string): Promise<Content> {
+    using _ = log.time("read", { file })
+    const project = Instance.project
+    const full = path.join(Instance.directory, file)
+
+    // TODO: Filesystem.contains is lexical only - symlinks inside the project can escape.
+    // TODO: On Windows, cross-drive paths bypass this check. Consider realpath canonicalization.
+    if (!Instance.containsPath(full)) {
+      throw new Error(`Access denied: path escapes project directory`)
+    }
+
+    const bunFile = Bun.file(full)
+
+    if (!(await bunFile.exists())) {
+      return { type: "text", content: "" }
+    }
+
+    const encode = await shouldEncode(bunFile)
+
+    if (encode) {
+      const buffer = await bunFile.arrayBuffer().catch(() => new ArrayBuffer(0))
+      const content = Buffer.from(buffer).toString("base64")
+      const mimeType = bunFile.type || "application/octet-stream"
+      return { type: "text", content, mimeType, encoding: "base64" }
+    }
+
+    const content = await bunFile
+      .text()
+      .catch(() => "")
+      .then((x) => x.trim())
+
+    if (project.vcs === "git") {
+      let diff = await $`git diff ${file}`.cwd(Instance.directory).quiet().nothrow().text()
+      if (!diff.trim()) diff = await $`git diff --staged ${file}`.cwd(Instance.directory).quiet().nothrow().text()
+      if (diff.trim()) {
+        const original = await $`git show HEAD:${file}`.cwd(Instance.directory).quiet().nothrow().text()
+        const patch = structuredPatch(file, file, original, content, "old", "new", {
+          context: Infinity,
+          ignoreWhitespace: true,
+        })
+        const diff = formatPatch(patch)
+        return { type: "text", content, patch, diff }
+      }
+    }
+    return { type: "text", content }
+  }
+
+  export async function list(dir?: string) {
+    const exclude = [".git", ".DS_Store"]
+    const project = Instance.project
+    let ignored = (_: string) => false
+    if (project.vcs === "git") {
+      const ig = ignore()
+      const gitignore = Bun.file(path.join(Instance.worktree, ".gitignore"))
+      if (await gitignore.exists()) {
+        ig.add(await gitignore.text())
+      }
+      const ignoreFile = Bun.file(path.join(Instance.worktree, ".ignore"))
+      if (await ignoreFile.exists()) {
+        ig.add(await ignoreFile.text())
+      }
+      ignored = ig.ignores.bind(ig)
+    }
+    const resolved = dir ? path.join(Instance.directory, dir) : Instance.directory
+
+    // TODO: Filesystem.contains is lexical only - symlinks inside the project can escape.
+    // TODO: On Windows, cross-drive paths bypass this check. Consider realpath canonicalization.
+    if (!Instance.containsPath(resolved)) {
+      throw new Error(`Access denied: path escapes project directory`)
+    }
+
+    const nodes: Node[] = []
+    for (const entry of await fs.promises
+      .readdir(resolved, {
+        withFileTypes: true,
+      })
+      .catch(() => [])) {
+      if (exclude.includes(entry.name)) continue
+      const fullPath = path.join(resolved, entry.name)
+      const relativePath = path.relative(Instance.directory, fullPath)
+      const type = entry.isDirectory() ? "directory" : "file"
+      nodes.push({
+        name: entry.name,
+        path: relativePath,
+        absolute: fullPath,
+        type,
+        ignored: ignored(type === "directory" ? relativePath + "/" : relativePath),
+      })
+    }
+    return nodes.sort((a, b) => {
+      if (a.type !== b.type) {
+        return a.type === "directory" ? -1 : 1
+      }
+      return a.name.localeCompare(b.name)
+    })
+  }
+
+  export async function search(input: { query: string; limit?: number; dirs?: boolean; type?: "file" | "directory" }) {
+    const query = input.query.trim()
+    const limit = input.limit ?? 100
+    const kind = input.type ?? (input.dirs === false ? "file" : "all")
+    log.info("search", { query, kind })
+
+    const result = await state().then((x) => x.files())
+
+    const hidden = (item: string) => {
+      const normalized = item.replaceAll("\\", "/").replace(/\/+$/, "")
+      return normalized.split("/").some((p) => p.startsWith(".") && p.length > 1)
+    }
+    const preferHidden = query.startsWith(".") || query.includes("/.")
+    const sortHiddenLast = (items: string[]) => {
+      if (preferHidden) return items
+      const visible: string[] = []
+      const hiddenItems: string[] = []
+      for (const item of items) {
+        const isHidden = hidden(item)
+        if (isHidden) hiddenItems.push(item)
+        if (!isHidden) visible.push(item)
+      }
+      return [...visible, ...hiddenItems]
+    }
+    if (!query) {
+      if (kind === "file") return result.files.slice(0, limit)
+      return sortHiddenLast(result.dirs.toSorted()).slice(0, limit)
+    }
+
+    const items =
+      kind === "file" ? result.files : kind === "directory" ? result.dirs : [...result.files, ...result.dirs]
+
+    const searchLimit = kind === "directory" && !preferHidden ? limit * 20 : limit
+    const sorted = fuzzysort.go(query, items, { limit: searchLimit }).map((r) => r.target)
+    const output = kind === "directory" ? sortHiddenLast(sorted).slice(0, limit) : sorted
+
+    log.info("search", { query, kind, results: output.length })
+    return output
+  }
+}