@apmantza/greedysearch-pi 1.1.5 → 1.1.7

package/extractors/bing-copilot.mjs

@@ -14,6 +14,7 @@ import { tmpdir } from 'os';
 import { join, dirname } from 'path';
 import { fileURLToPath } from 'url';
 import { dismissConsent, handleVerification } from './consent.mjs';
+import { SELECTORS } from './selectors.mjs';
 
 const __dir = dirname(fileURLToPath(import.meta.url));
 const CDP = join(__dir, '..', 'cdp.mjs');
@@ -22,6 +23,8 @@ const PAGES_CACHE = `${tmpdir().replace(/\\/g, '/')}/cdp-pages.json`;
 const COPY_POLL_INTERVAL = 700;
 const COPY_TIMEOUT = 60000;
 
+const S = SELECTORS.bing;
+
 // ---------------------------------------------------------------------------
 
 function cdp(args, timeoutMs = 30000) {
@@ -84,7 +87,7 @@ async function waitForCopyButton(tab) {
   while (Date.now() < deadline) {
     await new Promise(r => setTimeout(r, COPY_POLL_INTERVAL));
     const found = await cdp(['eval', tab,
-      `!!document.querySelector('button[data-testid="copy-ai-message-button"]')`
+      `!!document.querySelector('${S.copyButton}')`
     ]).catch(() => 'false');
     if (found === 'true') return;
   }
@@ -92,7 +95,7 @@ async function waitForCopyButton(tab) {
 }
 
 async function extractAnswer(tab) {
-  await cdp(['eval', tab, `document.querySelector('button[data-testid="copy-ai-message-button"]')?.click()`]);
+  await cdp(['eval', tab, `document.querySelector('${S.copyButton}')?.click()`]);
   await new Promise(r => setTimeout(r, 400));
 
   const answer = await cdp(['eval', tab, `window.__bingClipboard || ''`]);
@@ -100,9 +103,9 @@ async function extractAnswer(tab) {
 
   const raw = await cdp(['eval', tab, `
     (function() {
-      var sources = Array.from(document.querySelectorAll('a[href^="http"][target="_blank"]'))
+      var sources = Array.from(document.querySelectorAll('${S.sourceLink}'))
         .map(a => ({ url: a.href, title: a.innerText?.trim().split('\\n')[0] || a.title || '' }))
-        .filter(s => s.url && !s.url.includes('copilot.microsoft.com'))
+        .filter(s => s.url && !s.url.includes('${S.sourceExclude}'))
         .filter((v, i, arr) => arr.findIndex(x => x.url === v.url) === i)
         .slice(0, 10);
       return JSON.stringify(sources);
@@ -158,30 +161,30 @@ async function main() {
       }
     }
 
-    // Wait for React app to mount #userInput (up to 15s, longer after verification)
+    // Wait for React app to mount input (up to 15s, longer after verification)
     const inputDeadline = Date.now() + 15000;
     while (Date.now() < inputDeadline) {
-      const found = await cdp(['eval', tab, `!!document.querySelector('#userInput')`]).catch(() => 'false');
+      const found = await cdp(['eval', tab, `!!document.querySelector('${S.input}')`]).catch(() => 'false');
       if (found === 'true') break;
       await new Promise(r => setTimeout(r, 500));
     }
     await new Promise(r => setTimeout(r, 300));
 
     // Verify input is actually there before proceeding
-    const inputReady = await cdp(['eval', tab, `!!document.querySelector('#userInput')`]).catch(() => 'false');
+    const inputReady = await cdp(['eval', tab, `!!document.querySelector('${S.input}')`]).catch(() => 'false');
     if (inputReady !== 'true') {
       throw new Error('Copilot input not found — verification may have failed or page is in unexpected state');
     }
 
     await injectClipboardInterceptor(tab);
-    await cdp(['click', tab, '#userInput']);
+    await cdp(['click', tab, S.input]);
     await new Promise(r => setTimeout(r, 400));
     await cdp(['type', tab, query]);
     await new Promise(r => setTimeout(r, 400));
 
     // Submit with Enter (most reliable across locales and Chrome instances)
     await cdp(['eval', tab,
-      `document.querySelector('#userInput')?.dispatchEvent(new KeyboardEvent('keydown',{key:'Enter',bubbles:true,keyCode:13})), 'ok'`
+      `document.querySelector('${S.input}')?.dispatchEvent(new KeyboardEvent('keydown',{key:'Enter',bubbles:true,keyCode:13})), 'ok'`
     ]);
 
     await waitForCopyButton(tab);

package/extractors/gemini.mjs

@@ -14,13 +14,16 @@ import { tmpdir } from 'os';
 import { join, dirname } from 'path';
 import { fileURLToPath } from 'url';
 import { dismissConsent, handleVerification } from './consent.mjs';
+import { SELECTORS } from './selectors.mjs';
 
 const __dir = dirname(fileURLToPath(import.meta.url));
 const CDP = join(__dir, '..', 'cdp.mjs');
 const PAGES_CACHE = `${tmpdir().replace(/\\/g, '/')}/cdp-pages.json`;
 
 const COPY_POLL_INTERVAL = 600;
-const COPY_TIMEOUT = 120000;    // wait up to 2 min for copy button to appear
+const COPY_TIMEOUT = 120000;
+
+const S = SELECTORS.gemini;
 
 // ---------------------------------------------------------------------------
 
@@ -53,7 +56,7 @@ async function getOrOpenTab(tabPrefix) {
 async function typeIntoGemini(tab, text) {
   await cdp(['eval', tab, `
     (function(t) {
-      var el = document.querySelector('rich-textarea .ql-editor');
+      var el = document.querySelector('${S.input}');
       if (!el) return false;
       el.focus();
       document.execCommand('insertText', false, t);
@@ -93,7 +96,7 @@ async function waitForCopyButton(tab) {
   while (Date.now() < deadline) {
     await new Promise(r => setTimeout(r, COPY_POLL_INTERVAL));
     const found = await cdp(['eval', tab,
-      `!!document.querySelector('button[aria-label="Copy"]')`
+      `!!document.querySelector('${S.copyButton}')`
     ]).catch(() => 'false');
     if (found === 'true') return;
   }
@@ -102,16 +105,17 @@ async function waitForCopyButton(tab) {
 
 async function extractAnswer(tab) {
   // Click copy button → our interceptor captures the text.
-  await cdp(['eval', tab, `document.querySelector('button[aria-label="Copy"]')?.click()`]);
+  await cdp(['eval', tab, `document.querySelector('${S.copyButton}')?.click()`]);
   await new Promise(r => setTimeout(r, 400));
 
   const answer = await cdp(['eval', tab, `window.__geminiClipboard || ''`]);
   if (!answer) throw new Error('Clipboard interceptor returned empty text');
 
   // Click "Sources" button to open the sidebar with proper source cards
+  const sourceExcludeFilter = S.sourcesExclude.map(e => `!s.url.includes('${e}')`).join(' && ');
   await cdp(['eval', tab, `
     (function() {
-      var btn = document.querySelector('button.legacy-sources-sidebar-button, button.mdс-button--outline');
+      var btn = document.querySelector('${S.sourcesSidebarButton}');
       if (!btn) btn = Array.from(document.querySelectorAll('button')).find(b => b.innerText?.trim() === 'Sources');
       if (btn) { btn.click(); return 'clicked'; }
       return 'not-found';
@@ -131,17 +135,17 @@ async function extractAnswer(tab) {
         var container = sourceHeading.closest('.container') || sourceHeading.parentElement;
         var links = Array.from(container.querySelectorAll('a[href^="http"]'))
           .map(a => ({ url: a.href.split('#')[0], title: a.innerText?.trim().split('\\n')[0] || '' }))
-          .filter(s => s.url && !s.url.includes('gemini.google') && !s.url.includes('gstatic') && !s.url.includes('google.com/search'))
+          .filter(s => s.url && ${sourceExcludeFilter})
           .filter((v, i, arr) => arr.findIndex(x => x.url === v.url) === i)
           .slice(0, 8);
         return JSON.stringify(links);
       }
       // Fallback: inline source cards with aria-labels
-      var cards = Array.from(document.querySelectorAll('button[aria-label*="citation from"]'));
+      var cards = Array.from(document.querySelectorAll('${S.citationButtonPattern}'));
       if (cards.length) {
         return JSON.stringify(cards.map(b => {
           var label = b.getAttribute('aria-label') || '';
-          var name = label.match(/from\\s+(.+?)\\.\\s/)?.[1] || label;
+          var name = label.match(${S.citationNameRegex})?.[1] || label;
           return { url: '', title: name };
         }));
       }
@@ -188,7 +192,7 @@ async function main() {
     // Wait for input to be ready
     const deadline = Date.now() + 10000;
     while (Date.now() < deadline) {
-      const ready = await cdp(['eval', tab, `!!document.querySelector('rich-textarea .ql-editor')`]).catch(() => 'false');
+      const ready = await cdp(['eval', tab, `!!document.querySelector('${S.input}')`]).catch(() => 'false');
       if (ready === 'true') break;
       await new Promise(r => setTimeout(r, 400));
     }
@@ -198,7 +202,7 @@ async function main() {
     await typeIntoGemini(tab, query);
     await new Promise(r => setTimeout(r, 400));
 
-    await cdp(['eval', tab, `document.querySelector('button[aria-label*="Send"]')?.click()`]);
+    await cdp(['eval', tab, `document.querySelector('${S.sendButton}')?.click()`]);
 
     await waitForCopyButton(tab);
 

package/extractors/google-ai.mjs

@@ -14,6 +14,7 @@ import { tmpdir } from 'os';
 import { join, dirname } from 'path';
 import { fileURLToPath } from 'url';
 import { dismissConsent, handleVerification } from './consent.mjs';
+import { SELECTORS } from './selectors.mjs';
 
 const __dir = dirname(fileURLToPath(import.meta.url));
 const CDP = join(__dir, '..', 'cdp.mjs');
@@ -24,6 +25,8 @@ const STREAM_STABLE_ROUNDS = 3;
 const STREAM_TIMEOUT = 45000;
 const MIN_ANSWER_LENGTH = 50;
 
+const S = SELECTORS.google;
+
 // ---------------------------------------------------------------------------
 
 function cdp(args, timeoutMs = 30000) {
@@ -66,7 +69,7 @@ async function waitForStreamComplete(tab) {
     await new Promise(r => setTimeout(r, STREAM_POLL_INTERVAL));
 
     const lenStr = await cdp(['eval', tab,
-      `(document.querySelector('.pWvJNd')?.innerText?.length || 0) + ''`
+      `(document.querySelector('${S.answerContainer}')?.innerText?.length || 0) + ''`
     ]).catch(() => '0');
 
     const len = parseInt(lenStr) || 0;
@@ -85,14 +88,15 @@ async function waitForStreamComplete(tab) {
 }
 
 async function extractAnswer(tab) {
+  const excludeFilter = S.sourceExclude.map(e => `!a.href.includes('${e}')`).join(' && ');
   const raw = await cdp(['eval', tab, `
     (function() {
-      var el = document.querySelector('.pWvJNd');
+      var el = document.querySelector('${S.answerContainer}');
       if (!el) return JSON.stringify({ answer: '', sources: [] });
       var answer = el.innerText.trim();
-      var sources = Array.from(document.querySelectorAll('a[href^="http"]'))
-        .filter(a => !a.href.includes('google.') && !a.href.includes('gstatic') && !a.href.includes('googleapis'))
-        .map(a => ({ url: a.href.split('#')[0], title: (a.closest('[data-snhf]')?.querySelector('h3, [role=heading]')?.innerText || a.innerText?.trim().split('\\n')[0] || '').slice(0, 100) }))
+      var sources = Array.from(document.querySelectorAll('${S.sourceLink}'))
+        .filter(a => ${excludeFilter})
+        .map(a => ({ url: a.href.split('#')[0], title: (a.closest('${S.sourceHeadingParent}')?.querySelector('h3, [role=heading]')?.innerText || a.innerText?.trim().split('\\n')[0] || '').slice(0, 100) }))
         .filter(s => s.url && s.url.length > 10)
         .filter((v, i, arr) => arr.findIndex(x => x.url === v.url) === i)
         .slice(0, 10);

package/extractors/perplexity.mjs

@@ -14,6 +14,7 @@ import { tmpdir } from 'os';
 import { join, dirname } from 'path';
 import { fileURLToPath } from 'url';
 import { dismissConsent } from './consent.mjs';
+import { SELECTORS } from './selectors.mjs';
 
 const __dir = dirname(fileURLToPath(import.meta.url));
 const CDP = join(__dir, '..', 'cdp.mjs');
@@ -22,6 +23,8 @@ const PAGES_CACHE = `${tmpdir().replace(/\\/g, '/')}/cdp-pages.json`;
 const COPY_POLL_INTERVAL = 600;
 const COPY_TIMEOUT = 30000;
 
+const S = SELECTORS.perplexity;
+
 // ---------------------------------------------------------------------------
 
 function cdp(args, timeoutMs = 30000) {
@@ -87,7 +90,7 @@ async function waitForCopyButton(tab) {
   while (Date.now() < deadline) {
     await new Promise(r => setTimeout(r, COPY_POLL_INTERVAL));
     const found = await cdp(['eval', tab,
-      `!!document.querySelector('button[aria-label="Copy"]')`
+      `!!document.querySelector('${S.copyButton}')`
     ]).catch(() => 'false');
     if (found === 'true') return;
   }
@@ -95,7 +98,7 @@ async function waitForCopyButton(tab) {
 }
 
 async function extractAnswer(tab) {
-  await cdp(['eval', tab, `document.querySelector('button[aria-label="Copy"]')?.click()`]);
+  await cdp(['eval', tab, `document.querySelector('${S.copyButton}')?.click()`]);
   await new Promise(r => setTimeout(r, 400));
 
   const answer = await cdp(['eval', tab, `window.__pplxClipboard || ''`]);
@@ -103,8 +106,8 @@ async function extractAnswer(tab) {
 
   const raw = await cdp(['eval', tab, `
     (function() {
-      var sources = Array.from(document.querySelectorAll('[data-pplx-citation-url]'))
-        .map(el => ({ url: el.getAttribute('data-pplx-citation-url'), title: el.querySelector('a')?.innerText?.trim() || '' }))
+      var sources = Array.from(document.querySelectorAll('${S.sourceItem}'))
+        .map(el => ({ url: el.getAttribute('data-pplx-citation-url'), title: el.querySelector('${S.sourceLink}')?.innerText?.trim() || '' }))
         .filter(s => s.url)
         .filter((v, i, arr) => arr.findIndex(x => x.url === v.url) === i)
         .slice(0, 10);
@@ -144,23 +147,23 @@ async function main() {
     await cdp(['nav', tab, 'https://www.perplexity.ai/'], 35000);
     await dismissConsent(tab, cdp);
 
-    // Wait for React app to mount #ask-input (up to 8s)
+    // Wait for React app to mount input (up to 8s)
     const deadline = Date.now() + 8000;
     while (Date.now() < deadline) {
-      const found = await cdp(['eval', tab, `!!document.querySelector('#ask-input')`]).catch(() => 'false');
+      const found = await cdp(['eval', tab, `!!document.querySelector('${S.input}')`]).catch(() => 'false');
       if (found === 'true') break;
       await new Promise(r => setTimeout(r, 400));
     }
     await new Promise(r => setTimeout(r, 300));
 
     await injectClipboardInterceptor(tab);
-    await cdp(['click', tab, '#ask-input']);
+    await cdp(['click', tab, S.input]);
     await new Promise(r => setTimeout(r, 400));
     await cdp(['type', tab, query]);
     await new Promise(r => setTimeout(r, 400));
     // Submit with Enter (most reliable across Chrome instances)
     await cdp(['eval', tab,
-      `document.querySelector('#ask-input')?.dispatchEvent(new KeyboardEvent('keydown',{key:'Enter',bubbles:true,keyCode:13})), 'ok'`
+      `document.querySelector('${S.input}')?.dispatchEvent(new KeyboardEvent('keydown',{key:'Enter',bubbles:true,keyCode:13})), 'ok'`
     ]);
 
     await waitForCopyButton(tab);

package/extractors/selectors.mjs

@@ -0,0 +1,52 @@
+// extractors/selectors.mjs
+// Centralized CSS selectors for all engines.
+// Update selectors here when a site changes its UI.
+
+export const SELECTORS = {
+  // ──────────────────────────────────────────────
+  // Perplexity (perplexity.ai)
+  // ──────────────────────────────────────────────
+  perplexity: {
+    input: '#ask-input',
+    copyButton: 'button[aria-label="Copy"]',
+    sourceItem: '[data-pplx-citation-url]',
+    sourceLink: 'a',
+    consent: '#onetrust-accept-btn-handler',
+  },
+
+  // ──────────────────────────────────────────────
+  // Bing Copilot (copilot.microsoft.com)
+  // ──────────────────────────────────────────────
+  bing: {
+    input: '#userInput',
+    copyButton: 'button[data-testid="copy-ai-message-button"]',
+    sourceLink: 'a[href^="http"][target="_blank"]',
+    sourceExclude: 'copilot.microsoft.com',
+    consent: '#onetrust-accept-btn-handler',
+  },
+
+  // ──────────────────────────────────────────────
+  // Google AI Mode (google.com/search?udm=50)
+  // ──────────────────────────────────────────────
+  google: {
+    answerContainer: '.pWvJNd',
+    sourceLink: 'a[href^="http"]',
+    sourceExclude: ['google.', 'gstatic', 'googleapis'],
+    sourceHeadingParent: '[data-snhf]',
+    consent: '#L2AGLb, button[jsname="b3VHJd"], .tHlp8d',
+  },
+
+  // ──────────────────────────────────────────────
+  // Gemini (gemini.google.com/app)
+  // ──────────────────────────────────────────────
+  gemini: {
+    input: 'rich-textarea .ql-editor',
+    copyButton: 'button[aria-label="Copy"]',
+    sendButton: 'button[aria-label*="Send"]',
+    sourcesSidebarButton: 'button.legacy-sources-sidebar-button',
+    sourcesExclude: ['gemini.google', 'gstatic', 'google.com/search'],
+    citationButtonPattern: 'button[aria-label*="citation from"]',
+    // For parsing citation aria-labels: "View source details for citation from {name}. Opens side panel."
+    citationNameRegex: /from\s+(.+?)\.\s/,
+  },
+};

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@apmantza/greedysearch-pi",
-  "version": "1.1.5",
-  "description": "Pi extension: search Perplexity, Bing Copilot, and Google AI in parallel with optional Gemini synthesis — grounded AI answers, not just links",
+  "version": "1.1.7",
+  "description": "Pi extension: browser-automation tool that searches Perplexity, Bing Copilot, and Google AI in parallel, extracts answers and sources via CDP, with optional Gemini synthesis — grounded AI answers from real browser interactions.",
   "type": "module",
   "keywords": [
     "pi-package"