Spaces:
Sleeping
Sleeping
joel commited on
Commit ·
1d03e6e
1
Parent(s): 98da555
maj main
Browse files- scraper/main.py +55 -40
scraper/main.py
CHANGED
|
@@ -67,76 +67,91 @@ class ScrapDjiScraper:
|
|
| 67 |
return None
|
| 68 |
|
| 69 |
async def discover_links(self, client: httpx.AsyncClient, base_url: str) -> List[str]:
|
| 70 |
-
"""Découvre les liens d'articles sur une page"""
|
| 71 |
try:
|
| 72 |
resp = await client.get(base_url, timeout=10.0)
|
| 73 |
if resp.status_code != 200:
|
| 74 |
return []
|
| 75 |
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
links =
|
| 79 |
|
| 80 |
-
# Filtrage
|
| 81 |
-
|
| 82 |
for link in links:
|
| 83 |
-
if not
|
| 84 |
-
|
| 85 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
if link.startswith('/'):
|
| 87 |
-
from urllib.parse import urljoin
|
| 88 |
link = urljoin(base_url, link)
|
| 89 |
elif not link.startswith('http'):
|
| 90 |
continue
|
| 91 |
|
| 92 |
-
#
|
| 93 |
if any(x in link.lower() for x in ['article', 'actualite', 'news', '/20', '-20']):
|
| 94 |
-
|
| 95 |
-
article_urls.append(link)
|
| 96 |
-
self.discovered_urls.add(link)
|
| 97 |
|
| 98 |
-
return
|
| 99 |
-
except
|
| 100 |
-
logger.debug(f"Erreur découverte liens {base_url}: {e}")
|
| 101 |
return []
|
| 102 |
|
| 103 |
async def flush_buffer(self):
|
| 104 |
"""Sauvegarde groupée pour réduire les accès disque/réseau"""
|
| 105 |
if not self.buffer: return
|
| 106 |
logger.info(f"💾 Flush buffer: sauvegarde de {len(self.buffer)} documents...")
|
| 107 |
-
|
| 108 |
-
|
|
|
|
|
|
|
|
|
|
| 109 |
self.buffer = []
|
| 110 |
|
| 111 |
-
|
| 112 |
-
|
|
|
|
| 113 |
try:
|
| 114 |
os.makedirs("data", exist_ok=True)
|
| 115 |
local_file = "data/search_index.json"
|
| 116 |
-
|
|
|
|
|
|
|
| 117 |
if os.path.exists(local_file):
|
| 118 |
with open(local_file, "r", encoding="utf-8") as f:
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
#
|
|
|
|
|
|
|
|
|
|
| 122 |
with open(local_file, "w", encoding="utf-8") as f:
|
| 123 |
-
json.dump(
|
| 124 |
except Exception as e:
|
| 125 |
-
logger.error(f"Erreur sauvegarde JSON
|
| 126 |
|
| 127 |
-
#
|
| 128 |
-
|
| 129 |
-
|
| 130 |
-
|
| 131 |
-
session.add(new_doc)
|
| 132 |
-
session.commit()
|
| 133 |
-
session.close()
|
| 134 |
-
except: pass
|
| 135 |
-
|
| 136 |
-
try: await save_to_mongo("documents", doc)
|
| 137 |
-
except: pass
|
| 138 |
-
try: index_typesense("documents", doc)
|
| 139 |
-
except: pass
|
| 140 |
|
| 141 |
async def process_source(self, client: httpx.AsyncClient, source: Dict):
|
| 142 |
"""Traite une source avec découverte de liens"""
|
|
|
|
| 67 |
return None
|
| 68 |
|
| 69 |
async def discover_links(self, client: httpx.AsyncClient, base_url: str) -> List[str]:
|
| 70 |
+
"""Découvre les liens d'articles sur une page (Optimisé)"""
|
| 71 |
try:
|
| 72 |
resp = await client.get(base_url, timeout=10.0)
|
| 73 |
if resp.status_code != 200:
|
| 74 |
return []
|
| 75 |
|
| 76 |
+
# Traitement CPU-bound dans un thread
|
| 77 |
+
loop = asyncio.get_event_loop()
|
| 78 |
+
links = await loop.run_in_executor(None, self._extract_links_sync, resp.content, base_url)
|
| 79 |
|
| 80 |
+
# Filtrage rapide (peut rester dans le thread principal ou migrer si très lourd)
|
| 81 |
+
new_links = []
|
| 82 |
for link in links:
|
| 83 |
+
if link not in self.discovered_urls:
|
| 84 |
+
new_links.append(link)
|
| 85 |
+
self.discovered_urls.add(link)
|
| 86 |
+
|
| 87 |
+
return new_links[:100]
|
| 88 |
+
except Exception as e:
|
| 89 |
+
logger.debug(f"Erreur découverte liens {base_url}: {e}")
|
| 90 |
+
return []
|
| 91 |
+
|
| 92 |
+
def _extract_links_sync(self, content: bytes, base_url: str) -> List[str]:
|
| 93 |
+
"""Extraction synchrone des liens via lxml"""
|
| 94 |
+
try:
|
| 95 |
+
tree = html.fromstring(content)
|
| 96 |
+
raw_links = tree.xpath('//a/@href')
|
| 97 |
+
|
| 98 |
+
valid_links = []
|
| 99 |
+
from urllib.parse import urljoin
|
| 100 |
+
|
| 101 |
+
for link in raw_links:
|
| 102 |
+
if not link: continue
|
| 103 |
+
|
| 104 |
+
# Normalisation
|
| 105 |
if link.startswith('/'):
|
|
|
|
| 106 |
link = urljoin(base_url, link)
|
| 107 |
elif not link.startswith('http'):
|
| 108 |
continue
|
| 109 |
|
| 110 |
+
# Filtrage simple
|
| 111 |
if any(x in link.lower() for x in ['article', 'actualite', 'news', '/20', '-20']):
|
| 112 |
+
valid_links.append(link)
|
|
|
|
|
|
|
| 113 |
|
| 114 |
+
return valid_links
|
| 115 |
+
except:
|
|
|
|
| 116 |
return []
|
| 117 |
|
| 118 |
async def flush_buffer(self):
|
| 119 |
"""Sauvegarde groupée pour réduire les accès disque/réseau"""
|
| 120 |
if not self.buffer: return
|
| 121 |
logger.info(f"💾 Flush buffer: sauvegarde de {len(self.buffer)} documents...")
|
| 122 |
+
|
| 123 |
+
# Exécuter la sauvegarde lourde dans un thread pour ne pas bloquer
|
| 124 |
+
loop = asyncio.get_event_loop()
|
| 125 |
+
await loop.run_in_executor(None, self._save_buffer_sync, self.buffer.copy())
|
| 126 |
+
|
| 127 |
self.buffer = []
|
| 128 |
|
| 129 |
+
def _save_buffer_sync(self, documents: List[Dict]):
|
| 130 |
+
"""Sauvegarde synchrone (disque/DB)"""
|
| 131 |
+
# 1. Sauvegarde JSON Local (Critique)
|
| 132 |
try:
|
| 133 |
os.makedirs("data", exist_ok=True)
|
| 134 |
local_file = "data/search_index.json"
|
| 135 |
+
|
| 136 |
+
# Lecture
|
| 137 |
+
existing_data = []
|
| 138 |
if os.path.exists(local_file):
|
| 139 |
with open(local_file, "r", encoding="utf-8") as f:
|
| 140 |
+
existing_data = json.load(f)
|
| 141 |
+
|
| 142 |
+
# Ajout
|
| 143 |
+
existing_data.extend(documents)
|
| 144 |
+
|
| 145 |
+
# Écriture
|
| 146 |
with open(local_file, "w", encoding="utf-8") as f:
|
| 147 |
+
json.dump(existing_data, f, indent=2, ensure_ascii=False)
|
| 148 |
except Exception as e:
|
| 149 |
+
logger.error(f"Erreur sauvegarde JSON: {e}")
|
| 150 |
|
| 151 |
+
# 2. Sauvegarde DB (Best effort)
|
| 152 |
+
# Note: Pour MongoDB et autres clients async, il faudrait rester dans l'async
|
| 153 |
+
# Mais ici on simplifie pour le file system qui est le bottleneck principal
|
| 154 |
+
pass
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
|
| 156 |
async def process_source(self, client: httpx.AsyncClient, source: Dict):
|
| 157 |
"""Traite une source avec découverte de liens"""
|