@zeyue0329/xiaoma-cli 1.0.49 → 6.0.0-alpha.13

package/tools/flattener/projectRoot.js

@@ -1,10 +1,10 @@
-const fs = require("fs-extra");
-const path = require("node:path");
+const fs = require('fs-extra');
+const path = require('node:path');
 
 // Deno/Node compatibility: explicitly import process
-const process = require("node:process");
-const { execFile } = require("node:child_process");
-const { promisify } = require("node:util");
+const process = require('node:process');
+const { execFile } = require('node:child_process');
+const { promisify } = require('node:util');
 const execFileAsync = promisify(execFile);
 
 // Simple memoization across calls (keyed by realpath of startDir)
@@ -18,7 +18,7 @@ async function _tryRun(cmd, args, cwd, timeoutMs = 500) {
       windowsHide: true,
       maxBuffer: 1024 * 1024,
     });
-    const out = String(stdout || "").trim();
+    const out = String(stdout || '').trim();
     return out || null;
   } catch {
     return null;
@@ -27,21 +27,15 @@ async function _tryRun(cmd, args, cwd, timeoutMs = 500) {
 
 async function _detectVcsTopLevel(startDir) {
   // Run common VCS root queries in parallel; ignore failures
-  const gitP = _tryRun("git", ["rev-parse", "--show-toplevel"], startDir);
-  const hgP = _tryRun("hg", ["root"], startDir);
+  const gitP = _tryRun('git', ['rev-parse', '--show-toplevel'], startDir);
+  const hgP = _tryRun('hg', ['root'], startDir);
   const svnP = (async () => {
-    const show = await _tryRun(
-      "svn",
-      ["info", "--show-item", "wc-root"],
-      startDir,
-    );
+    const show = await _tryRun('svn', ['info', '--show-item', 'wc-root'], startDir);
     if (show) return show;
-    const info = await _tryRun("svn", ["info"], startDir);
+    const info = await _tryRun('svn', ['info'], startDir);
     if (info) {
-      const line = info
-        .split(/\r?\n/)
-        .find((l) => l.toLowerCase().startsWith("working copy root path:"));
-      if (line) return line.split(":").slice(1).join(":").trim();
+      const line = info.split(/\r?\n/).find((l) => l.toLowerCase().startsWith('working copy root path:'));
+      if (line) return line.split(':').slice(1).join(':').trim();
     }
     return null;
   })();
@@ -77,81 +71,80 @@ async function findProjectRoot(startDir) {
     const checks = [];
 
     const add = (rel, weight) => {
-      const makePath = (d) =>
-        Array.isArray(rel) ? path.join(d, ...rel) : path.join(d, rel);
+      const makePath = (d) => (Array.isArray(rel) ? path.join(d, ...rel) : path.join(d, rel));
       checks.push({ makePath, weight });
     };
 
     // Highest priority: explicit sentinel markers
-    add(".project-root", 110);
-    add(".workspace-root", 110);
-    add(".repo-root", 110);
+    add('.project-root', 110);
+    add('.workspace-root', 110);
+    add('.repo-root', 110);
 
     // Highest priority: VCS roots
-    add(".git", 100);
-    add(".hg", 95);
-    add(".svn", 95);
+    add('.git', 100);
+    add('.hg', 95);
+    add('.svn', 95);
 
     // Monorepo/workspace indicators
-    add("pnpm-workspace.yaml", 90);
-    add("lerna.json", 90);
-    add("turbo.json", 90);
-    add("nx.json", 90);
-    add("rush.json", 90);
-    add("go.work", 90);
-    add("WORKSPACE", 90);
-    add("WORKSPACE.bazel", 90);
-    add("MODULE.bazel", 90);
-    add("pants.toml", 90);
+    add('pnpm-workspace.yaml', 90);
+    add('lerna.json', 90);
+    add('turbo.json', 90);
+    add('nx.json', 90);
+    add('rush.json', 90);
+    add('go.work', 90);
+    add('WORKSPACE', 90);
+    add('WORKSPACE.bazel', 90);
+    add('MODULE.bazel', 90);
+    add('pants.toml', 90);
 
     // Lockfiles and package-manager/top-level locks
-    add("yarn.lock", 85);
-    add("pnpm-lock.yaml", 85);
-    add("package-lock.json", 85);
-    add("bun.lockb", 85);
-    add("Cargo.lock", 85);
-    add("composer.lock", 85);
-    add("poetry.lock", 85);
-    add("Pipfile.lock", 85);
-    add("Gemfile.lock", 85);
+    add('yarn.lock', 85);
+    add('pnpm-lock.yaml', 85);
+    add('package-lock.json', 85);
+    add('bun.lockb', 85);
+    add('Cargo.lock', 85);
+    add('composer.lock', 85);
+    add('poetry.lock', 85);
+    add('Pipfile.lock', 85);
+    add('Gemfile.lock', 85);
 
     // Build-system root indicators
-    add("settings.gradle", 80);
-    add("settings.gradle.kts", 80);
-    add("gradlew", 80);
-    add("pom.xml", 80);
-    add("build.sbt", 80);
-    add(["project", "build.properties"], 80);
+    add('settings.gradle', 80);
+    add('settings.gradle.kts', 80);
+    add('gradlew', 80);
+    add('pom.xml', 80);
+    add('build.sbt', 80);
+    add(['project', 'build.properties'], 80);
 
     // Language/project config markers
-    add("deno.json", 75);
-    add("deno.jsonc", 75);
-    add("pyproject.toml", 75);
-    add("Pipfile", 75);
-    add("requirements.txt", 75);
-    add("go.mod", 75);
-    add("Cargo.toml", 75);
-    add("composer.json", 75);
-    add("mix.exs", 75);
-    add("Gemfile", 75);
-    add("CMakeLists.txt", 75);
-    add("stack.yaml", 75);
-    add("cabal.project", 75);
-    add("rebar.config", 75);
-    add("pubspec.yaml", 75);
-    add("flake.nix", 75);
-    add("shell.nix", 75);
-    add("default.nix", 75);
-    add(".tool-versions", 75);
-    add("package.json", 74); // generic Node project (lower than lockfiles/workspaces)
+    add('deno.json', 75);
+    add('deno.jsonc', 75);
+    add('pyproject.toml', 75);
+    add('Pipfile', 75);
+    add('requirements.txt', 75);
+    add('go.mod', 75);
+    add('Cargo.toml', 75);
+    add('composer.json', 75);
+    add('mix.exs', 75);
+    add('Gemfile', 75);
+    add('CMakeLists.txt', 75);
+    add('stack.yaml', 75);
+    add('cabal.project', 75);
+    add('rebar.config', 75);
+    add('pubspec.yaml', 75);
+    add('flake.nix', 75);
+    add('shell.nix', 75);
+    add('default.nix', 75);
+    add('.tool-versions', 75);
+    add('package.json', 74); // generic Node project (lower than lockfiles/workspaces)
 
     // Changesets
-    add([".changeset", "config.json"], 70);
-    add(".changeset", 70);
+    add(['.changeset', 'config.json'], 70);
+    add('.changeset', 70);
 
     // Custom markers via env (comma-separated names)
     if (process.env.PROJECT_ROOT_MARKERS) {
-      for (const name of process.env.PROJECT_ROOT_MARKERS.split(",")
+      for (const name of process.env.PROJECT_ROOT_MARKERS.split(',')
         .map((s) => s.trim())
         .filter(Boolean)) {
         add(name, 72);
@@ -160,10 +153,10 @@ async function findProjectRoot(startDir) {
 
     /** Check for package.json with "workspaces" */
     const hasWorkspacePackageJson = async (d) => {
-      const pkgPath = path.join(d, "package.json");
+      const pkgPath = path.join(d, 'package.json');
       if (!(await exists(pkgPath))) return false;
       try {
-        const raw = await fs.readFile(pkgPath, "utf8");
+        const raw = await fs.readFile(pkgPath, 'utf8');
         const pkg = JSON.parse(raw);
         return Boolean(pkg && pkg.workspaces);
       } catch {
@@ -181,13 +174,10 @@ async function findProjectRoot(startDir) {
 
     while (true) {
       // Special check: package.json with "workspaces"
-      if ((await hasWorkspacePackageJson(dir)) && (!best || 90 >= best.weight))
-        best = { dir, weight: 90 };
+      if ((await hasWorkspacePackageJson(dir)) && (!best || 90 >= best.weight)) best = { dir, weight: 90 };
 
       // Evaluate all other checks in parallel
-      const results = await Promise.all(
-        checks.map(async (c) => ({ c, ok: await exists(c.makePath(dir)) })),
-      );
+      const results = await Promise.all(checks.map(async (c) => ({ c, ok: await exists(c.makePath(dir)) })));
 
       for (const { c, ok } of results) {
         if (!ok) continue;

package/tools/flattener/prompts.js

@@ -1,11 +1,11 @@
-const os = require("node:os");
-const path = require("node:path");
-const readline = require("node:readline");
-const process = require("node:process");
+const os = require('node:os');
+const path = require('node:path');
+const readline = require('node:readline');
+const process = require('node:process');
 
 function expandHome(p) {
   if (!p) return p;
-  if (p.startsWith("~")) return path.join(os.homedir(), p.slice(1));
+  if (p.startsWith('~')) return path.join(os.homedir(), p.slice(1));
   return p;
 }
 
@@ -27,18 +27,16 @@ function promptQuestion(question) {
 }
 
 async function promptYesNo(question, defaultYes = true) {
-  const suffix = defaultYes ? " [Y/n] " : " [y/N] ";
-  const ans = (await promptQuestion(`${question}${suffix}`))
-    .trim()
-    .toLowerCase();
+  const suffix = defaultYes ? ' [Y/n] ' : ' [y/N] ';
+  const ans = (await promptQuestion(`${question}${suffix}`)).trim().toLowerCase();
   if (!ans) return defaultYes;
-  if (["y", "yes"].includes(ans)) return true;
-  if (["n", "no"].includes(ans)) return false;
+  if (['y', 'yes'].includes(ans)) return true;
+  if (['n', 'no'].includes(ans)) return false;
   return promptYesNo(question, defaultYes);
 }
 
 async function promptPath(question, defaultValue) {
-  const prompt = `${question}${defaultValue ? ` (default: ${defaultValue})` : ""}: `;
+  const prompt = `${question}${defaultValue ? ` (default: ${defaultValue})` : ''}: `;
   const ans = (await promptQuestion(prompt)).trim();
   return expandHome(ans || defaultValue);
 }

package/tools/flattener/stats.helpers.js

@@ -1,11 +1,11 @@
-"use strict";
+'use strict';
 
-const fs = require("node:fs/promises");
-const path = require("node:path");
-const zlib = require("node:zlib");
-const { Buffer } = require("node:buffer");
-const crypto = require("node:crypto");
-const cp = require("node:child_process");
+const fs = require('node:fs/promises');
+const path = require('node:path');
+const zlib = require('node:zlib');
+const { Buffer } = require('node:buffer');
+const crypto = require('node:crypto');
+const cp = require('node:child_process');
 
 const KB = 1024;
 const MB = 1024 * KB;
@@ -13,17 +13,13 @@ const MB = 1024 * KB;
 const formatSize = (bytes) => {
   if (bytes < 1024) return `${bytes} B`;
   if (bytes < 1024 * 1024) return `${(bytes / 1024).toFixed(1)} KB`;
-  if (bytes < 1024 * 1024 * 1024)
-    return `${(bytes / (1024 * 1024)).toFixed(1)} MB`;
+  if (bytes < 1024 * 1024 * 1024) return `${(bytes / (1024 * 1024)).toFixed(1)} MB`;
   return `${(bytes / (1024 * 1024 * 1024)).toFixed(2)} GB`;
 };
 
 const percentile = (sorted, p) => {
   if (sorted.length === 0) return 0;
-  const idx = Math.min(
-    sorted.length - 1,
-    Math.max(0, Math.ceil((p / 100) * sorted.length) - 1),
-  );
+  const idx = Math.min(sorted.length - 1, Math.max(0, Math.ceil((p / 100) * sorted.length) - 1));
   return sorted[idx];
 };
 
@@ -38,10 +34,10 @@ async function enrichAllFiles(textFiles, binaryFiles) {
   const allFiles = [];
 
   async function enrich(file, isBinary) {
-    const ext = (path.extname(file.path) || "").toLowerCase();
-    const dir = path.dirname(file.path) || ".";
+    const ext = (path.extname(file.path) || '').toLowerCase();
+    const dir = path.dirname(file.path) || '.';
     const depth = file.path.split(path.sep).filter(Boolean).length;
-    const hidden = file.path.split(path.sep).some((seg) => seg.startsWith("."));
+    const hidden = file.path.split(path.sep).some((seg) => seg.startsWith('.'));
     let mtimeMs = 0;
     let isSymlink = false;
     try {
@@ -73,19 +69,15 @@ async function enrichAllFiles(textFiles, binaryFiles) {
 
 function buildHistogram(allFiles) {
   const buckets = [
-    [1 * KB, "0–1KB"],
-    [10 * KB, "1–10KB"],
-    [100 * KB, "10–100KB"],
-    [1 * MB, "100KB–1MB"],
-    [10 * MB, "1–10MB"],
-    [100 * MB, "10–100MB"],
-    [Infinity, ">=100MB"],
+    [1 * KB, '0–1KB'],
+    [10 * KB, '1–10KB'],
+    [100 * KB, '10–100KB'],
+    [1 * MB, '100KB–1MB'],
+    [10 * MB, '1–10MB'],
+    [100 * MB, '10–100MB'],
+    [Infinity, '>=100MB'],
   ];
-  const histogram = buckets.map(([_, label]) => ({
-    label,
-    count: 0,
-    bytes: 0,
-  }));
+  const histogram = buckets.map(([_, label]) => ({ label, count: 0, bytes: 0 }));
   for (const f of allFiles) {
     for (const [i, bucket] of buckets.entries()) {
       if (f.size < bucket[0]) {
@@ -101,7 +93,7 @@ function buildHistogram(allFiles) {
 function aggregateByExtension(allFiles) {
   const byExtension = new Map();
   for (const f of allFiles) {
-    const key = f.ext || "<none>";
+    const key = f.ext || '<none>';
     const v = byExtension.get(key) || { ext: key, count: 0, bytes: 0 };
     v.count++;
     v.bytes += f.size;
@@ -119,13 +111,13 @@ function aggregateByDirectory(allFiles) {
     byDirectory.set(dir, v);
   }
   for (const f of allFiles) {
-    const parts = f.dir === "." ? [] : f.dir.split(path.sep);
-    let acc = "";
+    const parts = f.dir === '.' ? [] : f.dir.split(path.sep);
+    let acc = '';
     for (let i = 0; i < parts.length; i++) {
       acc = i === 0 ? parts[0] : acc + path.sep + parts[i];
       addDirBytes(acc, f.size);
     }
-    if (parts.length === 0) addDirBytes(".", f.size);
+    if (parts.length === 0) addDirBytes('.', f.size);
   }
   return [...byDirectory.values()].sort((a, b) => b.bytes - a.bytes);
 }
@@ -139,9 +131,7 @@ function computeDepthAndLongest(allFiles) {
     .sort((a, b) => b.path.length - a.path.length)
     .slice(0, 25)
     .map((f) => ({ path: f.path, length: f.path.length, size: f.size }));
-  const depthDist = [...depthDistribution.entries()]
-    .sort((a, b) => a[0] - b[0])
-    .map(([depth, count]) => ({ depth, count }));
+  const depthDist = [...depthDistribution.entries()].sort((a, b) => a[0] - b[0]).map(([depth, count]) => ({ depth, count }));
   return { depthDist, longestPaths };
 }
 
@@ -149,18 +139,15 @@ function computeTemporal(allFiles, nowMs) {
   let oldest = null,
     newest = null;
   const ageBuckets = [
-    { label: "> 1 year", minDays: 365, maxDays: Infinity, count: 0, bytes: 0 },
-    { label: "6–12 months", minDays: 180, maxDays: 365, count: 0, bytes: 0 },
-    { label: "1–6 months", minDays: 30, maxDays: 180, count: 0, bytes: 0 },
-    { label: "7–30 days", minDays: 7, maxDays: 30, count: 0, bytes: 0 },
-    { label: "1–7 days", minDays: 1, maxDays: 7, count: 0, bytes: 0 },
-    { label: "< 1 day", minDays: 0, maxDays: 1, count: 0, bytes: 0 },
+    { label: '> 1 year', minDays: 365, maxDays: Infinity, count: 0, bytes: 0 },
+    { label: '6–12 months', minDays: 180, maxDays: 365, count: 0, bytes: 0 },
+    { label: '1–6 months', minDays: 30, maxDays: 180, count: 0, bytes: 0 },
+    { label: '7–30 days', minDays: 7, maxDays: 30, count: 0, bytes: 0 },
+    { label: '1–7 days', minDays: 1, maxDays: 7, count: 0, bytes: 0 },
+    { label: '< 1 day', minDays: 0, maxDays: 1, count: 0, bytes: 0 },
   ];
   for (const f of allFiles) {
-    const ageDays = Math.max(
-      0,
-      (nowMs - (f.mtimeMs || nowMs)) / (24 * 60 * 60 * 1000),
-    );
+    const ageDays = Math.max(0, (nowMs - (f.mtimeMs || nowMs)) / (24 * 60 * 60 * 1000));
     for (const b of ageBuckets) {
       if (ageDays >= b.minDays && ageDays < b.maxDays) {
         b.count++;
@@ -172,37 +159,21 @@ function computeTemporal(allFiles, nowMs) {
     if (!newest || f.mtimeMs > newest.mtimeMs) newest = f;
   }
   return {
-    oldest: oldest
-      ? {
-          path: oldest.path,
-          mtime: oldest.mtimeMs ? new Date(oldest.mtimeMs).toISOString() : null,
-        }
-      : null,
-    newest: newest
-      ? {
-          path: newest.path,
-          mtime: newest.mtimeMs ? new Date(newest.mtimeMs).toISOString() : null,
-        }
-      : null,
+    oldest: oldest ? { path: oldest.path, mtime: oldest.mtimeMs ? new Date(oldest.mtimeMs).toISOString() : null } : null,
+    newest: newest ? { path: newest.path, mtime: newest.mtimeMs ? new Date(newest.mtimeMs).toISOString() : null } : null,
     ageBuckets,
   };
 }
 
 function computeQuality(allFiles, textFiles) {
   const zeroByteFiles = allFiles.filter((f) => f.size === 0).length;
-  const emptyTextFiles = textFiles.filter(
-    (f) => (f.size || 0) === 0 || (f.lines || 0) === 0,
-  ).length;
+  const emptyTextFiles = textFiles.filter((f) => (f.size || 0) === 0 || (f.lines || 0) === 0).length;
   const hiddenFiles = allFiles.filter((f) => f.hidden).length;
   const symlinks = allFiles.filter((f) => f.isSymlink).length;
   const largeThreshold = 50 * MB;
   const suspiciousThreshold = 100 * MB;
-  const largeFilesCount = allFiles.filter(
-    (f) => f.size >= largeThreshold,
-  ).length;
-  const suspiciousLargeFilesCount = allFiles.filter(
-    (f) => f.size >= suspiciousThreshold,
-  ).length;
+  const largeFilesCount = allFiles.filter((f) => f.size >= largeThreshold).length;
+  const suspiciousLargeFilesCount = allFiles.filter((f) => f.size >= suspiciousThreshold).length;
   return {
     zeroByteFiles,
     emptyTextFiles,
@@ -231,8 +202,8 @@ function computeDuplicates(allFiles, textFiles) {
     for (const tf of textGroup) {
       try {
         const src = textFiles.find((x) => x.absolutePath === tf.absolutePath);
-        const content = src ? src.content : "";
-        const h = crypto.createHash("sha1").update(content).digest("hex");
+        const content = src ? src.content : '';
+        const h = crypto.createHash('sha1').update(content).digest('hex');
         const g = contentHashGroups.get(h) || [];
         g.push(tf);
         contentHashGroups.set(h, g);
@@ -243,7 +214,7 @@ function computeDuplicates(allFiles, textFiles) {
     for (const [_h, g] of contentHashGroups.entries()) {
       if (g.length > 1)
         duplicateCandidates.push({
-          reason: "same-size+text-hash",
+          reason: 'same-size+text-hash',
           size: Number(sizeKey),
           count: g.length,
           files: g.map((f) => f.path),
@@ -251,7 +222,7 @@ function computeDuplicates(allFiles, textFiles) {
     }
     if (otherGroup.length > 1) {
       duplicateCandidates.push({
-        reason: "same-size",
+        reason: 'same-size',
         size: Number(sizeKey),
         count: otherGroup.length,
         files: otherGroup.map((f) => f.path),
@@ -269,7 +240,7 @@ function estimateCompressibility(textFiles) {
       const sampleLen = Math.min(256 * 1024, tf.size || 0);
       if (sampleLen <= 0) continue;
       const sample = tf.content.slice(0, sampleLen);
-      const gz = zlib.gzipSync(Buffer.from(sample, "utf8"));
+      const gz = zlib.gzipSync(Buffer.from(sample, 'utf8'));
       compSampleBytes += sampleLen;
       compCompressedBytes += gz.length;
     } catch {
@@ -291,19 +262,19 @@ function computeGitInfo(allFiles, rootDir, largeThreshold) {
   try {
     if (!rootDir) return info;
     const top = cp
-      .execFileSync("git", ["rev-parse", "--show-toplevel"], {
+      .execFileSync('git', ['rev-parse', '--show-toplevel'], {
         cwd: rootDir,
-        stdio: ["ignore", "pipe", "ignore"],
+        stdio: ['ignore', 'pipe', 'ignore'],
       })
       .toString()
       .trim();
     if (!top) return info;
     info.isRepo = true;
-    const out = cp.execFileSync("git", ["ls-files", "-z"], {
+    const out = cp.execFileSync('git', ['ls-files', '-z'], {
       cwd: rootDir,
-      stdio: ["ignore", "pipe", "ignore"],
+      stdio: ['ignore', 'pipe', 'ignore'],
     });
-    const tracked = new Set(out.toString().split("\0").filter(Boolean));
+    const tracked = new Set(out.toString().split('\0').filter(Boolean));
     let trackedBytes = 0,
       trackedCount = 0,
       untrackedBytes = 0,
@@ -314,8 +285,7 @@ function computeGitInfo(allFiles, rootDir, largeThreshold) {
       if (isTracked) {
         trackedCount++;
         trackedBytes += f.size;
-        if (f.size >= largeThreshold)
-          lfsCandidates.push({ path: f.path, size: f.size });
+        if (f.size >= largeThreshold) lfsCandidates.push({ path: f.path, size: f.size });
       } else {
         untrackedCount++;
         untrackedBytes += f.size;
@@ -325,9 +295,7 @@ function computeGitInfo(allFiles, rootDir, largeThreshold) {
     info.trackedBytes = trackedBytes;
     info.untrackedCount = untrackedCount;
     info.untrackedBytes = untrackedBytes;
-    info.lfsCandidates = lfsCandidates
-      .sort((a, b) => b.size - a.size)
-      .slice(0, 50);
+    info.lfsCandidates = lfsCandidates.sort((a, b) => b.size - a.size).slice(0, 50);
   } catch {
     /* git not available or not a repo, ignore */
   }
@@ -344,63 +312,39 @@ function computeLargestFiles(allFiles, totalBytes) {
       size: f.size,
       sizeFormatted: formatSize(f.size),
       percentOfTotal: toPct(f.size, totalBytes),
-      ext: f.ext || "",
+      ext: f.ext || '',
       isBinary: f.isBinary,
       mtime: f.mtimeMs ? new Date(f.mtimeMs).toISOString() : null,
     }));
 }
 
 function mdTable(rows, headers) {
-  const header = `| ${headers.join(" | ")} |`;
-  const sep = `| ${headers.map(() => "---").join(" | ")} |`;
-  const body = rows.map((r) => `| ${r.join(" | ")} |`).join("\n");
+  const header = `| ${headers.join(' | ')} |`;
+  const sep = `| ${headers.map(() => '---').join(' | ')} |`;
+  const body = rows.map((r) => `| ${r.join(' | ')} |`).join('\n');
   return `${header}\n${sep}\n${body}`;
 }
 
-function buildMarkdownReport(
-  largestFiles,
-  byExtensionArr,
-  byDirectoryArr,
-  totalBytes,
-) {
+function buildMarkdownReport(largestFiles, byExtensionArr, byDirectoryArr, totalBytes) {
   const toPct = (num, den) => (den === 0 ? 0 : (num / den) * 100);
   const md = [];
   md.push(
-    "\n### Top Largest Files (Top 50)\n",
+    '\n### Top Largest Files (Top 50)\n',
     mdTable(
-      largestFiles.map((f) => [
-        f.path,
-        f.sizeFormatted,
-        `${f.percentOfTotal.toFixed(2)}%`,
-        f.ext || "",
-        f.isBinary ? "binary" : "text",
-      ]),
-      ["Path", "Size", "% of total", "Ext", "Type"],
+      largestFiles.map((f) => [f.path, f.sizeFormatted, `${f.percentOfTotal.toFixed(2)}%`, f.ext || '', f.isBinary ? 'binary' : 'text']),
+      ['Path', 'Size', '% of total', 'Ext', 'Type'],
     ),
-    "\n\n### Top Extensions by Bytes (Top 20)\n",
+    '\n\n### Top Extensions by Bytes (Top 20)\n',
   );
   const topExtRows = byExtensionArr
     .slice(0, 20)
-    .map((e) => [
-      e.ext,
-      String(e.count),
-      formatSize(e.bytes),
-      `${toPct(e.bytes, totalBytes).toFixed(2)}%`,
-    ]);
-  md.push(
-    mdTable(topExtRows, ["Ext", "Count", "Bytes", "% of total"]),
-    "\n\n### Top Directories by Bytes (Top 20)\n",
-  );
+    .map((e) => [e.ext, String(e.count), formatSize(e.bytes), `${toPct(e.bytes, totalBytes).toFixed(2)}%`]);
+  md.push(mdTable(topExtRows, ['Ext', 'Count', 'Bytes', '% of total']), '\n\n### Top Directories by Bytes (Top 20)\n');
   const topDirRows = byDirectoryArr
     .slice(0, 20)
-    .map((d) => [
-      d.dir,
-      String(d.count),
-      formatSize(d.bytes),
-      `${toPct(d.bytes, totalBytes).toFixed(2)}%`,
-    ]);
-  md.push(mdTable(topDirRows, ["Directory", "Files", "Bytes", "% of total"]));
-  return md.join("\n");
+    .map((d) => [d.dir, String(d.count), formatSize(d.bytes), `${toPct(d.bytes, totalBytes).toFixed(2)}%`]);
+  md.push(mdTable(topDirRows, ['Directory', 'Files', 'Bytes', '% of total']));
+  return md.join('\n');
 }
 
 module.exports = {

package/tools/flattener/stats.js

@@ -1,4 +1,4 @@
-const H = require("./stats.helpers.js");
+const H = require('./stats.helpers.js');
 
 async function calculateStatistics(aggregatedContent, xmlFileSize, rootDir) {
   const { textFiles, binaryFiles, errors } = aggregatedContent;
@@ -26,12 +26,7 @@ async function calculateStatistics(aggregatedContent, xmlFileSize, rootDir) {
   const compressibilityRatio = H.estimateCompressibility(textFiles);
   const git = H.computeGitInfo(allFiles, rootDir, quality.largeThreshold);
   const largestFiles = H.computeLargestFiles(allFiles, totalBytes);
-  const markdownReport = H.buildMarkdownReport(
-    largestFiles,
-    byExtensionArr,
-    byDirectoryArr,
-    totalBytes,
-  );
+  const markdownReport = H.buildMarkdownReport(largestFiles, byExtensionArr, byDirectoryArr, totalBytes);
 
   return {
     // Back-compat summary