hh-shipments 0.3.19 → 0.3.21

package/package.json

@@ -18,7 +18,7 @@
     "test": "node test/run-tests.js && node test/run-carriers-tests.js"
   },
   "type": "commonjs",
-  "version": "0.3.19",
+  "version": "0.3.21",
   "bin": {
     "hh-shipments": "server.js"
   },

package/public/app.js

@@ -168,9 +168,50 @@ function loadTesseract() {
   return tesseractLoading;
 }
 
+// Tesseract reads dark-text-on-light far better than the reverse. Chat
+// screenshots (dark mode) come out as white-on-black, so detect dark images
+// and invert them before OCR. Light images (labels) pass through untouched.
+async function maybeInvertForOcr(file) {
+  try {
+    const bmp = await createImageBitmap(file);
+    const w = bmp.width, h = bmp.height;
+    const canvas = document.createElement('canvas');
+    // sample at reduced size for speed
+    const sw = Math.min(200, w), sh = Math.min(200, h);
+    canvas.width = sw; canvas.height = sh;
+    const ctx = canvas.getContext('2d', { willReadFrequently: true });
+    ctx.drawImage(bmp, 0, 0, sw, sh);
+    const px = ctx.getImageData(0, 0, sw, sh).data;
+    let sum = 0, n = 0;
+    for (let i = 0; i < px.length; i += 16) {
+      sum += 0.299 * px[i] + 0.587 * px[i + 1] + 0.114 * px[i + 2];
+      n++;
+    }
+    const avg = sum / n;
+    bmp.close();
+    if (avg > 110) return file; // already light — leave it alone
+    // dark image: invert to dark-on-light
+    const c2 = document.createElement('canvas');
+    c2.width = w; c2.height = h;
+    const x2 = c2.getContext('2d');
+    const full = await createImageBitmap(file);
+    x2.drawImage(full, 0, 0);
+    full.close();
+    const img = x2.getImageData(0, 0, w, h);
+    const d = img.data;
+    for (let i = 0; i < d.length; i += 4) {
+      d[i] = 255 - d[i]; d[i + 1] = 255 - d[i + 1]; d[i + 2] = 255 - d[i + 2];
+    }
+    x2.putImageData(img, 0, 0);
+    const blob = await new Promise((res) => c2.toBlob(res, 'image/jpeg', 0.92));
+    return blob ? new File([blob], 'inv.jpg', { type: 'image/jpeg' }) : file;
+  } catch (e) { return file; }
+}
+
 async function ocrFile(file, statusEl) {
   await loadTesseract();
   statusEl.textContent = 'Reading screenshot…';
+  file = await maybeInvertForOcr(file);
   const { data } = await Tesseract.recognize(file, 'eng', { logger: (m) => { if (m.status === 'recognizing text') statusEl.textContent = `Reading… ${Math.round(m.progress * 100)}%`; } });
   statusEl.textContent = '';
   return data.text || '';

package/public/index.html

@@ -4,7 +4,7 @@
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, viewport-fit=cover">
 <meta name="theme-color" content="#0a84ff">
-<meta name="app-version" content="0.3.19">
+<meta name="app-version" content="0.3.21">
 <title>HH Shipments</title>
 <link rel="manifest" href="/manifest.webmanifest">
 <link rel="apple-touch-icon" href="/icon-192.png">

package/src/extract.js

@@ -90,25 +90,23 @@
       }
     }
     // Chat screenshots (Messenger): the other person's name is the header at
-    // the very top ("Abdullah Sohail"), above the message bubbles. Take the
-    // first title-case name in the top lines — never Hunter's own name /
-    // address bubble, never app or chat UI words.
+    // the very top ("Abdullah Sohail"), above the message bubbles. OCR wraps
+    // it in garbage ("{® Abdullah Sohail LC Be"), so find the name *within*
+    // the top lines instead of requiring a clean line. Never Hunter's own
+    // name/address bubble, never app or chat UI words.
     if (!out.seller_name) {
       var cheadBlock = /\b(UPS|FEDEX|USPS|DHL|GROUND|EXPRESS|PRIORITY|BILLING|TRACKING|POSTAGE|LABEL|SHIPMENT|SHIP\b|SCREENSHOT|READING|FINDING|UPLOADING|EXTRACT|PASTE|DETAILS|SAVE|TAP|CLIPBOARD|NUMBER|BACK|HOME|ARCHIVE|SETTINGS|HUNTER|HILLS|THANK|THANKS|YOU|DELIVERED|SEEN|TYPING|AFTERMARKET|DISPLAY)\b/i;
       var clines = String(text).split(/\r?\n/);
       var checked = 0;
-      for (var ci = 0; ci < clines.length && checked < 6; ci++) {
-        var ch = clines[ci].trim();
-        if (!ch || ch.length > 40) continue;
+      for (var ci = 0; ci < clines.length && checked < 8; ci++) {
+        var ch = clines[ci];
+        if (!ch || !ch.trim() || ch.length > 80) continue;
         checked++;
         if (/\d/.test(ch)) continue;
-        var cw = ch.split(/\s+/);
-        if (cw.length < 2 || cw.length > 3) continue;
-        var cok = true;
-        for (var cj = 0; cj < cw.length; cj++) if (!/^[A-Z][a-z]{2,}$/.test(cw[cj])) { cok = false; break; }
-        if (!cok) continue;
-        if (cheadBlock.test(ch)) continue;
-        out.seller_name = ch;
+        var nm = ch.match(/\b([A-Z][a-z]{2,}(?:\s+[A-Z][a-z]{2,}){1,2})\b/);
+        if (!nm) continue;
+        if (cheadBlock.test(nm[1])) continue;
+        out.seller_name = nm[1];
         break;
       }
     }