findweb 0.1.0 → 0.1.1

package/README.md

@@ -0,0 +1,147 @@
+# findweb
+
+Google search CLI powered by system Chrome, with programmatic ad blocking.
+
+## Install
+
+```bash
+bun install -g findweb
+```
+
+Or run directly:
+
+```bash
+bunx findweb "yc"
+```
+
+## Usage
+
+```bash
+# Single search
+findweb "Y Combinator"
+
+# Batch search
+findweb "yc" "apple" "tesla" --parallel 3
+
+# JSON output
+findweb --json "react useEffect"
+
+# Custom region and language
+findweb --gl kr --lang ko "startup"
+
+# More results
+findweb -n 10 "rust async"
+
+# Prepare a signed-in Chrome profile (reduces rate limiting)
+findweb login
+```
+
+## First Run Behavior
+
+`findweb` requires an initialized Google profile before it will run the first search for a given `--userDataDir`.
+
+- If the profile is already prepared, search runs immediately.
+- If the profile has not been prepared yet, `findweb` automatically opens the login flow first.
+- After you sign in and close the browser window, `findweb` writes a local prepared-profile marker so future searches can start immediately.
+
+In practice, the first search on a fresh profile behaves like this:
+
+```bash
+findweb "yc"
+```
+
+1. detect missing prepared-profile marker
+2. open headed Chrome login flow
+3. wait for you to sign in and close the browser
+4. continue the original search
+
+## Options
+
+| Option             | Default        | Description                          |
+| ------------------ | -------------- | ------------------------------------ |
+| `--gl <country>`   | `us`           | Google region hint                   |
+| `-l, --lang`       | `en`           | Google UI language                   |
+| `-n, --num`        | `3`            | Results per query                    |
+| `--parallel`       | `4`            | Batch tab concurrency                |
+| `--userDataDir`    | auto-detected  | Chrome profile directory             |
+| `--headed`         | `false`        | Show the Chrome window               |
+| `--json`           | `false`        | Print output as JSON                 |
+
+## How It Works
+
+1. Launches system Chrome (`/Applications/Google Chrome.app`) with a free debugging port
+2. Connects via CDP using puppeteer-core
+3. Loads the [Ghostery adblocker](https://github.com/ghostery/adblocker) engine programmatically on each page
+4. Navigates to Google, submits the query through DOM manipulation, and extracts results from the rendered page
+5. Returns results as plain text or JSON, then closes Chrome
+
+No Chromium download. No browser extension. No user confirmation.
+
+## Batch Mode
+
+Pass multiple quoted queries as positional arguments. Each query opens a separate tab in the same browser instance and profile.
+
+```bash
+findweb "yc" "apple" "tesla"
+```
+
+Results are returned in input order. Concurrency is controlled by `--parallel` (default: 4).
+
+## Login
+
+Google rate-limits unauthenticated or fresh-profile searches. `findweb` now enforces an interactive login before the first search on a new profile.
+
+You can trigger that ahead of time with:
+
+```bash
+findweb login
+```
+
+This opens a visible Chrome window with the Google sign-in page. After signing in, close the browser. The session is saved to the profile directory, and `findweb` records that the profile is ready for future searches.
+
+## Output
+
+### Plain text (default)
+
+```
+1. Y Combinator
+https://www.ycombinator.com/
+Y Combinator created a new model for funding early stage startups.
+
+2. Y Combinator - Wikipedia
+https://en.wikipedia.org/wiki/Y_Combinator
+```
+
+### JSON (`--json`)
+
+```json
+[
+  {
+    "title": "Y Combinator",
+    "url": "https://www.ycombinator.com/",
+    "snippet": "Y Combinator created a new model for funding early stage startups."
+  }
+]
+```
+
+## Requirements
+
+- macOS
+- System Chrome (`/Applications/Google Chrome.app`)
+- [Bun](https://bun.sh) >= 1.3.11
+
+## Development
+
+```bash
+git clone https://github.com/ysm-dev/findweb.git
+cd findweb
+bun install
+bun run check    # tsgo typecheck
+bun run test     # unit tests
+bun run dev      # run from source
+bun run build    # bundle to dist/
+```
+
+## License
+
+MIT

package/dist/index.js

@@ -53157,7 +53157,7 @@ var require_ffi_WASM_RELEASE_SYNC = __commonJS((exports) => {
 
 // node_modules/@tootallnate/quickjs-emscripten/dist/generated/emscripten-module.WASM_RELEASE_SYNC.js
 var require_emscripten_module_WASM_RELEASE_SYNC = __commonJS((exports, module) => {
-  var __dirname = "/Users/chris/git/ysm/findweb/node_modules/@tootallnate/quickjs-emscripten/dist/generated", __filename = "/Users/chris/git/ysm/findweb/node_modules/@tootallnate/quickjs-emscripten/dist/generated/emscripten-module.WASM_RELEASE_SYNC.js";
+  var __dirname = "/home/runner/work/findweb/findweb/node_modules/@tootallnate/quickjs-emscripten/dist/generated", __filename = "/home/runner/work/findweb/findweb/node_modules/@tootallnate/quickjs-emscripten/dist/generated/emscripten-module.WASM_RELEASE_SYNC.js";
   var QuickJSRaw = (() => {
     var _scriptDir = typeof document !== "undefined" && document.currentScript ? document.currentScript.src : undefined;
     if (typeof __filename !== "undefined")
@@ -66931,9 +66931,6 @@ async function runMain(cmd, opts = {}) {
   }
 }
 
-// src/cli/commands/login.ts
-import fs7 from "fs/promises";
-
 // src/search/browser.ts
 import { existsSync as existsSync3 } from "fs";
 import { spawn as spawn3 } from "child_process";
@@ -76557,8 +76554,68 @@ async function runLoginSession(options) {
   });
 }
 
-// src/cli/schema.ts
+// src/cli/profile.ts
+import fs7 from "fs/promises";
 import path12 from "path";
+var PROFILE_READY_MARKER = ".findweb-profile-ready";
+async function ensureProfileDir(dirPath) {
+  await fs7.mkdir(dirPath, { recursive: true });
+}
+function readyMarkerPath(userDataDir) {
+  return path12.join(userDataDir, PROFILE_READY_MARKER);
+}
+async function hasPreparedProfile(userDataDir) {
+  try {
+    await fs7.access(readyMarkerPath(userDataDir));
+    return true;
+  } catch {
+    return false;
+  }
+}
+async function markProfilePrepared(userDataDir) {
+  await ensureProfileDir(userDataDir);
+  await fs7.writeFile(readyMarkerPath(userDataDir), `${new Date().toISOString()}
+`, "utf8");
+}
+
+// src/cli/flows/login.ts
+function printLoginInstructions(userDataDir) {
+  console.log(`Login browser launched with profile: ${userDataDir}`);
+  console.log("Sign in to Google to prepare this profile for future searches.");
+  console.log("Close the browser window when you are done.");
+}
+async function runInteractiveLoginFlow(options) {
+  await ensureProfileDir(options.userDataDir);
+  const activeBrowser = await launchSearchBrowser({
+    headed: true,
+    lang: options.lang,
+    userDataDir: options.userDataDir
+  });
+  printLoginInstructions(options.userDataDir);
+  try {
+    await runLoginSession({
+      browser: activeBrowser.browser,
+      gl: options.gl,
+      lang: options.lang
+    });
+  } finally {
+    await closeSearchBrowser(activeBrowser);
+  }
+  await markProfilePrepared(options.userDataDir);
+}
+async function ensureInteractiveLogin(options) {
+  await ensureProfileDir(options.userDataDir);
+  if (await hasPreparedProfile(options.userDataDir)) {
+    return false;
+  }
+  console.log("No prepared Google profile was found for this user-data-dir.");
+  console.log("A login step is required before the first search.");
+  await runInteractiveLoginFlow(options);
+  return true;
+}
+
+// src/cli/schema.ts
+import path13 from "path";
 
 // node_modules/zod/v4/classic/external.js
 var exports_external = {};
@@ -77325,10 +77382,10 @@ function mergeDefs(...defs) {
 function cloneDef(schema) {
   return mergeDefs(schema._zod.def);
 }
-function getElementAtPath(obj, path12) {
-  if (!path12)
+function getElementAtPath(obj, path13) {
+  if (!path13)
     return obj;
-  return path12.reduce((acc, key) => acc?.[key], obj);
+  return path13.reduce((acc, key) => acc?.[key], obj);
 }
 function promiseAllObject(promisesObj) {
   const keys = Object.keys(promisesObj);
@@ -77709,11 +77766,11 @@ function aborted(x, startIndex = 0) {
   }
   return false;
 }
-function prefixIssues(path12, issues) {
+function prefixIssues(path13, issues) {
   return issues.map((iss) => {
     var _a4;
     (_a4 = iss).path ?? (_a4.path = []);
-    iss.path.unshift(path12);
+    iss.path.unshift(path13);
     return iss;
   });
 }
@@ -77896,7 +77953,7 @@ function formatError(error, mapper = (issue2) => issue2.message) {
 }
 function treeifyError(error, mapper = (issue2) => issue2.message) {
   const result = { errors: [] };
-  const processError = (error2, path12 = []) => {
+  const processError = (error2, path13 = []) => {
     var _a4, _b2;
     for (const issue2 of error2.issues) {
       if (issue2.code === "invalid_union" && issue2.errors.length) {
@@ -77906,7 +77963,7 @@ function treeifyError(error, mapper = (issue2) => issue2.message) {
       } else if (issue2.code === "invalid_element") {
         processError({ issues: issue2.issues }, issue2.path);
       } else {
-        const fullpath = [...path12, ...issue2.path];
+        const fullpath = [...path13, ...issue2.path];
         if (fullpath.length === 0) {
           result.errors.push(mapper(issue2));
           continue;
@@ -77938,8 +77995,8 @@ function treeifyError(error, mapper = (issue2) => issue2.message) {
 }
 function toDotPath(_path) {
   const segs = [];
-  const path12 = _path.map((seg) => typeof seg === "object" ? seg.key : seg);
-  for (const seg of path12) {
+  const path13 = _path.map((seg) => typeof seg === "object" ? seg.key : seg);
+  for (const seg of path13) {
     if (typeof seg === "number")
       segs.push(`[${seg}]`);
     else if (typeof seg === "symbol")
@@ -89686,13 +89743,13 @@ function resolveRef(ref, ctx) {
   if (!ref.startsWith("#")) {
     throw new Error("External $ref is not supported, only local refs (#/...) are allowed");
   }
-  const path12 = ref.slice(1).split("/").filter(Boolean);
-  if (path12.length === 0) {
+  const path13 = ref.slice(1).split("/").filter(Boolean);
+  if (path13.length === 0) {
     return ctx.rootSchema;
   }
   const defsKey = ctx.version === "draft-2020-12" ? "$defs" : "definitions";
-  if (path12[0] === defsKey) {
-    const key = path12[1];
+  if (path13[0] === defsKey) {
+    const key = path13[1];
     if (!key || !ctx.defs[key]) {
       throw new Error(`Reference not found: ${ref}`);
     }
@@ -90101,7 +90158,7 @@ var sharedSchema = exports_external.object({
   headed: exports_external.boolean().default(false),
   json: exports_external.boolean().default(false),
   lang: defaultedString("lang", "en"),
-  userDataDir: defaultedString("userDataDir", defaultUserDataDir()).transform((value) => path12.resolve(value))
+  userDataDir: defaultedString("userDataDir", defaultUserDataDir()).transform((value) => path13.resolve(value))
 });
 var searchSchema = sharedSchema.extend({
   num: positiveInteger("num").default(3),
@@ -90145,25 +90202,6 @@ function normalizeLoginOptions(input2) {
 }
 
 // src/cli/commands/login.ts
-async function ensureProfileDir(dirPath) {
-  await fs7.mkdir(dirPath, { recursive: true });
-}
-async function runLoginFlow(gl, lang, userDataDir) {
-  await ensureProfileDir(userDataDir);
-  const activeBrowser = await launchSearchBrowser({
-    headed: true,
-    lang,
-    userDataDir
-  });
-  console.log(`Login browser launched with profile: ${userDataDir}`);
-  console.log("Sign in to Google if you want to reuse a logged-in search profile.");
-  console.log("Close the browser window when you are done.");
-  try {
-    await runLoginSession({ browser: activeBrowser.browser, gl, lang });
-  } finally {
-    await closeSearchBrowser(activeBrowser);
-  }
-}
 function createLoginArgs() {
   return {
     gl: {
@@ -90190,7 +90228,7 @@ async function runLogin(args) {
     lang: typeof args.lang === "string" ? args.lang : undefined,
     userDataDir: typeof args.userDataDir === "string" ? args.userDataDir : undefined
   });
-  await runLoginFlow(options.gl, options.lang, options.userDataDir);
+  await runInteractiveLoginFlow(options);
 }
 function createLoginCommand(commandName = "findweb login") {
   return defineCommand({
@@ -90207,13 +90245,12 @@ function createLoginCommand(commandName = "findweb login") {
 }
 
 // src/cli/commands/search.ts
-import fs9 from "fs/promises";
 import process4 from "process";
 
 // src/search/blocker.ts
 import fs8 from "fs/promises";
 import os9 from "os";
-import path13 from "path";
+import path14 from "path";
 
 // node_modules/tldts-core/dist/es6/src/domain.js
 function shareSameDomainSuffix(hostname3, vhost) {
@@ -99477,8 +99514,8 @@ class FilterEngine extends EventEmitter5 {
     if (caching === undefined) {
       return init();
     }
-    const { path: path13, read, write } = caching;
-    return read(path13).then((buffer) => this.deserialize(buffer)).catch(() => init().then((engine) => write(path13, engine.serialize()).then(() => engine)));
+    const { path: path14, read, write } = caching;
+    return read(path14).then((buffer) => this.deserialize(buffer)).catch(() => init().then((engine) => write(path14, engine.serialize()).then(() => engine)));
   }
   static empty(config3 = {}) {
     return new this({ config: config3 });
@@ -101270,13 +101307,13 @@ class PuppeteerBlocker extends FilterEngine {
 // src/search/blocker.ts
 var blockerPromise;
 function defaultCacheDir() {
-  return process.env.GOOGLE_SEARCH_CACHE_DIR ?? path13.join(os9.homedir(), ".cache", "google-search");
+  return process.env.GOOGLE_SEARCH_CACHE_DIR ?? path14.join(os9.homedir(), ".cache", "google-search");
 }
 async function readCache(filePath) {
   return fs8.readFile(filePath);
 }
 async function writeCache(filePath, buffer) {
-  await fs8.mkdir(path13.dirname(filePath), { recursive: true });
+  await fs8.mkdir(path14.dirname(filePath), { recursive: true });
   await fs8.writeFile(filePath, buffer);
 }
 async function loadBlocker() {
@@ -101285,7 +101322,7 @@ async function loadBlocker() {
       const cacheDir = defaultCacheDir();
       await fs8.mkdir(cacheDir, { recursive: true });
       return PuppeteerBlocker.fromPrebuiltAdsAndTracking(globalThis.fetch.bind(globalThis), {
-        path: path13.join(cacheDir, "ghostery-engine.bin"),
+        path: path14.join(cacheDir, "ghostery-engine.bin"),
         read: readCache,
         write: writeCache
       });
@@ -101339,9 +101376,6 @@ function exitCodeForOutcomes(items) {
 }
 
 // src/cli/commands/search.ts
-async function ensureProfileDir2(dirPath) {
-  await fs9.mkdir(dirPath, { recursive: true });
-}
 function printResults(json2, outcomes) {
   if (json2) {
     printJsonResults(outcomes);
@@ -101402,7 +101436,16 @@ async function runSearch(args) {
     parallel: typeof args.parallel === "string" || typeof args.parallel === "number" ? args.parallel : undefined,
     userDataDir: typeof args.userDataDir === "string" ? args.userDataDir : undefined
   });
-  await ensureProfileDir2(options.userDataDir);
+  await ensureProfileDir(options.userDataDir);
+  const loginWasRequired = await ensureInteractiveLogin({
+    gl: options.gl,
+    lang: options.lang,
+    userDataDir: options.userDataDir
+  });
+  if (loginWasRequired) {
+    console.log(`Login completed. Continuing with search...
+`);
+  }
   const blocker = await loadBlocker();
   const activeBrowser = await launchSearchBrowser({
     headed: options.headed,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "findweb",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "Google search CLI powered by system Chrome",
   "type": "module",
   "packageManager": "bun@1.3.11",