fix: recognize modern binary media formats

CHANGELOG.md

@@ -3,6 +3,13 @@
 All notable changes to this project are documented here, following
 [Keep a Changelog](https://keepachangelog.com/) and semantic versioning.
 
+## [0.1.17] - 2026-09-23
+
+### Fixed
+
+- Recognize AVIF, HEIC, HEIF, APNG, WebM, Ogg, FLAC and M4A as binary assets, avoiding text reads and inflated token estimates.
+- Repair unresolved merge markers in the package lockfile while updating release metadata.
+
 ## [0.1.16] - 2026-09-17
 
 ### Fixed

README.md

@@ -65,6 +65,8 @@ ctxtrim  ·  my-repo  ·  412 files
   ✓ .cursorignore (created, 34 patterns), .aiexclude (created, 34 patterns)
 ```
 
+Known binary media, including AVIF, HEIC/HEIF, APNG, WebM, Ogg, FLAC and M4A, are not read as text and contribute no text tokens. They remain explicit ignore candidates.
+
 Estimates use the widely-cited ~4-chars-per-token rule (great for ranking and relative savings; pass `--price` to match your model).
 
 Vendored dependency and build output directories (`node_modules`, `dist`, `build`, ...) are reported as a single entry keyed by directory name — sized from a cheap stat walk without reading the files inside. Directory-level categories are decided by the directory name, not by reading its contents.

package.json

@@ -1,6 +1,6 @@
 {
   "name": "ctxtrim",
-  "version": "0.1.16",
+  "version": "0.1.17",
   "description": "Trim what bloats your AI coding context. Scan a repo, find the high-cost/low-value files ballooning your Claude Code / Cursor / Codex context, and write ignore files to cut token cost. Zero dependencies.",
   "type": "module",
   "bin": {

src/classify.js

@@ -27,6 +27,7 @@ const DATA_EXT = new Set([
 
 const BINARY_EXT = new Set([
   ".png", ".jpg", ".jpeg", ".gif", ".webp", ".ico", ".bmp", ".tiff",
+  ".avif", ".heic", ".heif", ".apng", ".webm", ".ogg", ".flac", ".m4a",
   ".pdf", ".zip", ".gz", ".tar", ".tgz", ".7z", ".rar", ".jar", ".war",
   ".woff", ".woff2", ".ttf", ".otf", ".eot", ".mp3", ".mp4", ".mov", ".avi",
   ".wasm", ".so", ".dylib", ".dll", ".exe", ".bin", ".class", ".pyc",

test/modern-binary-formats.test.js

@@ -0,0 +1,57 @@
+import fs from "node:fs";
+import { syncBuiltinESMExports } from "node:module";
+import { tmpdir } from "node:os";
+import { join } from "node:path";
+import { test } from "node:test";
+import assert from "node:assert/strict";
+import { classify, classifyPath } from "../src/classify.js";
+import { scanRepo } from "../src/scan.js";
+
+const extensions = ["avif", "heic", "heif", "apng", "webm", "ogg", "flac", "m4a"];
+
+test("modern image, video and audio formats are binary regardless of case", () => {
+  for (const extension of extensions) {
+    for (const ext of [extension, extension.toUpperCase()]) {
+      const name = `media/asset.${ext}`;
+      const result = classify(name, { tokens: 100 });
+      assert.equal(result.category, "binary", name);
+      assert.equal(result.trim, true, name);
+      assert.equal(result.binary, true, name);
+      assert.deepEqual(classifyPath(name), result);
+    }
+  }
+  assert.equal(classify("src/audio.js", { tokens: 100 }).trim, false);
+});
+
+test("scan never reads modern binary media or counts it as text tokens", (t) => {
+  const root = fs.mkdtempSync(join(tmpdir(), "ctxtrim-modern-media-"));
+  t.after(() => fs.rmSync(root, { recursive: true, force: true }));
+  const names = extensions.map((ext) => `asset.${ext}`);
+  for (const name of names) fs.writeFileSync(join(root, name), Buffer.from([0xff, 0x00, 0xfe]));
+  // Exercise both the full-read and partial-read paths if classification regresses.
+  fs.truncateSync(join(root, "asset.webm"), 5_000_001);
+  fs.writeFileSync(join(root, "source.js"), "export const value = 1;\n");
+  const reads = [];
+  for (const method of ["readFileSync", "openSync"]) {
+    const original = fs[method];
+    t.mock.method(fs, method, (...args) => {
+      reads.push(String(args[0]));
+      return original(...args);
+    });
+  }
+  syncBuiltinESMExports();
+  let result;
+  try { result = scanRepo(root); }
+  finally { t.mock.restoreAll(); syncBuiltinESMExports(); }
+
+  for (const name of names) {
+    assert.equal(reads.includes(join(root, name)), false, name);
+    const file = result.files.find((entry) => entry.rel === name);
+    assert.equal(file.category, "binary", name);
+    assert.equal(file.tokens, 0, name);
+    assert.equal(file.trim, true, name);
+    assert.ok(result.patterns.includes(name));
+  }
+  assert.equal(result.totals.textFiles, 1);
+  assert.equal(result.totals.totalTokens, result.files.find((file) => file.rel === "source.js").tokens);
+});