guezoloic/millesima-ai-engine
1
0
Fork 0
mirror of https://github.com/guezoloic/millesima-ai-engine.git synced 2026-03-28 18:03:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

ajout: ajout fonction json

Browse Source
This commit is contained in:
Loïc GUEZO
2026-02-04 21:01:48 +01:00
parent e6b04fc740
commit 469a351f2a
2 changed files with 20 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
main.py
View File

@@ -3,30 +3,6 @@ from typing import Any
from bs4 import BeautifulSoup from bs4 import BeautifulSoup
import json import json
url = "louis-latour-aloxe-corton-1er-cru-les-chaillots-2018.html"
# response = requests.get(url)
# soup = BeautifulSoup(response.text, 'html.parser')
class MillesimaSoup(BeautifulSoup):
def __init__(self, markup="", features="html.parser", *args, **kwargs):
super().__init__(markup, features, *args, **kwargs)
self._json_data = self._extract_json_data()
def _extract_json_data(self) -> dict[str, Any]:
script = self.find("script", id="__NEXT_DATA__")
if script and script.string:
try:
data: dict[str, Any] = json.loads(script.string)
for element in ['props', 'pageProps', 'initialReduxState', 'product', 'content']:
data.get(element)
return data
except json.decoder.JSONDecodeError:
return {}
return {}
class Scraper: class Scraper:
""" """
@@ -47,7 +23,7 @@ class Scraper:
# TCP et d'avoir toujours une connexion constante avec le server # TCP et d'avoir toujours une connexion constante avec le server
self._session: requests.Session = requests.Session() self._session: requests.Session = requests.Session()
self._url: str = "https://www.millesima.fr/" self._url: str = "https://www.millesima.fr/"
self._soup = None self._soup = self.getsoup()
def _request(self, subdir: str, use_cache: bool = True) -> requests.Response | requests.HTTPError: def _request(self, subdir: str, use_cache: bool = True) -> requests.Response | requests.HTTPError:
""" """
@@ -71,11 +47,11 @@ class Scraper:
if self._response.url == target_url: if self._response.url == target_url:
return self._response return self._response
self._response: requests.Response = self._session.get(target_url, timeout=10) self._response: requests.Response = self._session.get(
target_url, timeout=10)
self._response.raise_for_status() self._response.raise_for_status()
return self._response return self._response
def getsoup(self, subdir: str = "/") -> BeautifulSoup: def getsoup(self, subdir: str = "/") -> BeautifulSoup:
""" """
@@ -85,25 +61,19 @@ class Scraper:
:return: Description :return: Description
:rtype: BeautifulSoup :rtype: BeautifulSoup
""" """
self._request(subdir) if subdir != None:
self._soup = BeautifulSoup(self._response.text, "html.parser") self._request(subdir)
self._soup = BeautifulSoup(self._response.text, "html.parser")
return self._soup return self._soup
def get_json_data(self):
print(Scraper().getsoup(url)) script = self._soup.find("script", id="__NEXT_DATA__")
if script and script.string:
# # On cible la balise magique try:
# script_tag = soup.find('script', id='__NEXT_DATA__') data: dict[str, Any] = json.loads(script.string)
# print(script_tag) for element in ['props', 'pageProps', 'initialReduxState', 'product', 'content']:
data.get(element)
# if script_tag: return data
# # On transforme le texte en vrai dictionnaire Python except json.decoder.JSONDecodeError:
# data = json.loads(script_tag.string) pass
# # Navigation dans l'objet (Next.js structure toujours comme ça) return {}
# product_info = data['props']['pageProps']['initialReduxState']['product']['content']
# print(f"Vin : {product_info['productName']}")
# print(f"Prix HT : {product_info['items'][0]['htPrice']} €")
# print(f"Stock : {product_info['items'][0]['stock']}")

4
test_main.py
View File

@@ -1,4 +1,6 @@
from main import * from main import *
scraper = Scraper()
def test_soup(): def test_soup():
assert getsoup("https://example.com").find('h1').text == "Example Domain" assert scraper.getsoup().find('h1').text[3:12] == "MILLESIMA"