Spaces:
Sleeping
Sleeping
Update app/listing_scraper.py
Browse files- app/listing_scraper.py +12 -27
app/listing_scraper.py
CHANGED
|
@@ -61,27 +61,18 @@ class ListingScraper:
|
|
| 61 |
def _get_products_on_page(self) -> List[Dict[str, Any]]:
|
| 62 |
page = self.page
|
| 63 |
products: List[Dict[str, Any]] = []
|
| 64 |
-
logger.debug("Extracting products from listing
|
| 65 |
|
| 66 |
try:
|
| 67 |
-
#
|
| 68 |
-
rows = page.locator("div.
|
| 69 |
-
logger.debug(f"Found {len(rows)} potential product
|
| 70 |
|
| 71 |
for idx, row in enumerate(rows):
|
| 72 |
try:
|
| 73 |
-
# Find title and link inside
|
| 74 |
-
title_elem =
|
| 75 |
-
|
| 76 |
-
try:
|
| 77 |
-
candidate = row.locator(t_sel).first
|
| 78 |
-
if candidate and candidate.count() > 0:
|
| 79 |
-
title_elem = candidate
|
| 80 |
-
break
|
| 81 |
-
except Exception:
|
| 82 |
-
continue
|
| 83 |
-
|
| 84 |
-
if not title_elem:
|
| 85 |
continue
|
| 86 |
|
| 87 |
title = (title_elem.inner_text(timeout=300) or "").strip()
|
|
@@ -90,21 +81,15 @@ class ListingScraper:
|
|
| 90 |
if not title or not link or "dvd_view" not in link.lower():
|
| 91 |
continue
|
| 92 |
|
| 93 |
-
# Extract price text container
|
| 94 |
price = None
|
| 95 |
try:
|
| 96 |
row_text = row.inner_text(timeout=300)
|
| 97 |
p_match = re.search(r"Lowest\s+Price\s*:\s*\$?\s*([\d,.]+)", row_text, re.IGNORECASE)
|
| 98 |
if p_match:
|
| 99 |
price = float(p_match.group(1).replace(",", ""))
|
| 100 |
-
else:
|
| 101 |
-
# Fallback price element search
|
| 102 |
-
price_elem = row.locator("div.price, span.price, strong").first
|
| 103 |
-
if price_elem.count() > 0:
|
| 104 |
-
price_text = price_elem.inner_text(timeout=300).strip()
|
| 105 |
-
price = _parse_price(price_text)
|
| 106 |
except Exception as e:
|
| 107 |
-
logger.debug(f"
|
| 108 |
|
| 109 |
if price is None:
|
| 110 |
self.skipped_unknown_price += 1
|
|
@@ -117,7 +102,7 @@ class ListingScraper:
|
|
| 117 |
|
| 118 |
products.append({"url": link, "title": title, "price": price})
|
| 119 |
except Exception as e:
|
| 120 |
-
logger.debug(f"
|
| 121 |
continue
|
| 122 |
|
| 123 |
logger.info(f"Extracted {len(products)} qualifying product(s) from page")
|
|
@@ -129,7 +114,7 @@ class ListingScraper:
|
|
| 129 |
def _resolve_next_page_url(self, current_url: str) -> Optional[str]:
|
| 130 |
page = self.page
|
| 131 |
|
| 132 |
-
# Method 1: Check pagination
|
| 133 |
try:
|
| 134 |
next_selectors = [
|
| 135 |
"ul.pagination li.active + li a",
|
|
@@ -149,7 +134,7 @@ class ListingScraper:
|
|
| 149 |
except Exception:
|
| 150 |
pass
|
| 151 |
|
| 152 |
-
# Method 2: Fallback query parameter page increment
|
| 153 |
try:
|
| 154 |
parsed = urlparse(current_url)
|
| 155 |
qs = parse_qs(parsed.query)
|
|
|
|
| 61 |
def _get_products_on_page(self) -> List[Dict[str, Any]]:
|
| 62 |
page = self.page
|
| 63 |
products: List[Dict[str, Any]] = []
|
| 64 |
+
logger.debug("Extracting products from new listing card layout")
|
| 65 |
|
| 66 |
try:
|
| 67 |
+
# Target the new Tailwind min-w-0 product container blocks
|
| 68 |
+
rows = page.locator("div.min-w-0").all()
|
| 69 |
+
logger.debug(f"Found {len(rows)} potential product cards")
|
| 70 |
|
| 71 |
for idx, row in enumerate(rows):
|
| 72 |
try:
|
| 73 |
+
# Find title and link inside h3 > a
|
| 74 |
+
title_elem = row.locator("h3 a, a[href*='dvd_view']").first
|
| 75 |
+
if not title_elem or title_elem.count() == 0:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
continue
|
| 77 |
|
| 78 |
title = (title_elem.inner_text(timeout=300) or "").strip()
|
|
|
|
| 81 |
if not title or not link or "dvd_view" not in link.lower():
|
| 82 |
continue
|
| 83 |
|
| 84 |
+
# Extract price text from the container block
|
| 85 |
price = None
|
| 86 |
try:
|
| 87 |
row_text = row.inner_text(timeout=300)
|
| 88 |
p_match = re.search(r"Lowest\s+Price\s*:\s*\$?\s*([\d,.]+)", row_text, re.IGNORECASE)
|
| 89 |
if p_match:
|
| 90 |
price = float(p_match.group(1).replace(",", ""))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
except Exception as e:
|
| 92 |
+
logger.debug(f"Card {idx}: Could not extract price: {e}")
|
| 93 |
|
| 94 |
if price is None:
|
| 95 |
self.skipped_unknown_price += 1
|
|
|
|
| 102 |
|
| 103 |
products.append({"url": link, "title": title, "price": price})
|
| 104 |
except Exception as e:
|
| 105 |
+
logger.debug(f"Card {idx}: Error processing card: {e}")
|
| 106 |
continue
|
| 107 |
|
| 108 |
logger.info(f"Extracted {len(products)} qualifying product(s) from page")
|
|
|
|
| 114 |
def _resolve_next_page_url(self, current_url: str) -> Optional[str]:
|
| 115 |
page = self.page
|
| 116 |
|
| 117 |
+
# Method 1: Check pagination buttons/links
|
| 118 |
try:
|
| 119 |
next_selectors = [
|
| 120 |
"ul.pagination li.active + li a",
|
|
|
|
| 134 |
except Exception:
|
| 135 |
pass
|
| 136 |
|
| 137 |
+
# Method 2: Fallback query parameter page increment
|
| 138 |
try:
|
| 139 |
parsed = urlparse(current_url)
|
| 140 |
qs = parse_qs(parsed.query)
|