Update ocr_utils.py

ocr_utils.py

@@ -1,5 +1,4 @@
 import io, re
-from typing import List, Dict, Tuple, Any
 from PIL import Image, ImageDraw
 
 try:
@@ -9,16 +8,10 @@ except Exception:
     pytesseract = None
     TessOutput = None
 
-# regex for price detection
-PRICE_RE = re.compile(r"(\$\s*\d{1,4}(?:\.\d{2})?|\b\d{1,4}\.\d{2}\b)")
-
 def ocr_image(image_bytes: bytes):
     """
-    Returns (full_text, tokens, size).
-    - full_text: the whole OCR text as one string
-    - tokens: list of {'text','conf','box'}
-    - size: (width, height)
-    Always returns 3 values, even if OCR fails.
+    Always return exactly three values: (text, tokens, size)
+    even if OCR fails or is missing.
     """
     try:
         img = Image.open(io.BytesIO(image_bytes)).convert("RGB")
@@ -34,60 +27,46 @@ def ocr_image(image_bytes: bytes):
         return "", [], img.size
 
     tokens = []
-    for i, txt in enumerate(data["text"]):
-        if not txt:
+    for i, word in enumerate(data["text"]):
+        if not word:
             continue
         try:
             conf = float(data.get("conf", ["-1"])[i])
-        except:
+        except Exception:
             conf = -1.0
-        x, y, w, h = data["left"][i], data["top"][i], data["width"][i], data["height"][i]
-        tokens.append({"text": txt, "conf": conf, "box": (x, y, w, h)})
+        box = (
+            data["left"][i],
+            data["top"][i],
+            data["width"][i],
+            data["height"][i],
+        )
+        tokens.append({"text": word, "conf": conf, "box": box})
 
-    full = " ".join(t["text"] for t in tokens)
-    return full, tokens, img.size
+    full_text = " ".join(t["text"] for t in tokens)
+    return full_text, tokens, img.size
 
 
-def guess_price(tokens: List[Dict[str,Any]]):
-    """Pick the smallest plausible price (typical for tags)."""
-    best = None
+def guess_price(tokens):
+    """Find first $xx.xx or similar pattern."""
+    price_pattern = re.compile(r"\$?\d{1,4}(?:[.,]\d{2})?")
     for t in tokens:
-        m = PRICE_RE.search(t["text"].replace(",", ""))
+        m = price_pattern.search(t["text"])
         if m:
-            raw = m.group(0).replace("$", "").strip()
+            val = m.group(0)
             try:
-                val = float(raw)
-                if 0.5 <= val <= 1000:
-                    if best is None or val < best[0]:
-                        best = (val, t["box"])
-            except:
+                return float(val.replace("$", "").replace(",", "")), t["box"]
+            except Exception:
                 pass
-    return best
+    return None, None
 
 
-def guess_title(full_text: str) -> str:
-    """Return a short readable title."""
-    text = (full_text or "").strip()
-    if not text:
-        return ""
-    words = [w for w in re.findall(r"[A-Za-z][A-Za-z0-9\-]{2,}", text) if not w.isdigit()]
-    return " ".join(words[:4]) if words else "Item"
+def guess_title(text: str) -> str:
+    """Simplify long OCR text into a short title."""
+    words = re.findall(r"[A-Za-z0-9\-]{3,}", text)
+    return " ".join(words[:5]) if words else "Item"
 
 
-def annotate_price_box(image_bytes: bytes, box: Tuple[int,int,int,int], label: str = None) -> bytes:
-    """Draw a red box around the detected price and optional label."""
+def annotate_price_box(image_bytes: bytes, box, label: str = None):
+    """Draw red rectangle around detected price."""
     try:
-        img = Image.open(io.BytesIO(image_bytes)).convert("RGB")
-        draw = ImageDraw.Draw(img)
-        x, y, w, h = box
-        draw.rectangle([x, y, x + w, y + h], outline=(255, 0, 0), width=3)
-        if label:
-            tw = max(50, int(len(label) * 8))
-            y0 = max(0, y - 22)
-            draw.rectangle([x, y0, x + tw, y0 + 22], fill=(255, 0, 0))
-            draw.text((x + 4, y0 + 4), label, fill=(255, 255, 255))
-        out = io.BytesIO()
-        img.save(out, format="PNG")
-        return out.getvalue()
-    except Exception:
-        return image_bytes
+        img = Image.open(io.By