Merge pull request #23 from BananeChocolat/patch-2

Modification du menu (osef des allergenes <3)

src/cogs/restauration.py

@@ -4,21 +4,23 @@
 
 from __future__ import annotations
 
-import io
 import json
+import re
 from os import path
 from typing import TYPE_CHECKING
-from zipfile import ZipFile
 
+import discord
 import httpx
-from bs4 import BeautifulSoup, Tag
-from discord import File, HTTPException, TextChannel
+from bs4 import BeautifulSoup
+from discord import HTTPException, TextChannel, app_commands
 from discord.ext import tasks
 from discord.ext.commands import Cog  # pyright: ignore[reportMissingTypeStubs]
 
 if TYPE_CHECKING:
     from bot import MP2IBot
 
+
+IMAGES_REGEX = re.compile(r"https://lycee-kleber.com.fr/wp-content/uploads/\d{4}/\d{2}/([^.]+).jpg")
 RESTAURATION_PATH = "./data/restauration.json"
 
 
@@ -29,6 +31,7 @@ def __init__(self, bot: MP2IBot) -> None:
         self.already_posted: list[str] = self.read_restauration_file()
 
     async def cog_load(self) -> None:
+        await self.check_menu()
         self.check_menu.start()
 
     async def cog_unload(self) -> None:
@@ -61,88 +64,59 @@ def read_restauration_file(self) -> list[str]:
         with open(RESTAURATION_PATH, "r") as f:
             return json.load(f)
 
-    async def get_menu_imgs(self) -> dict[str, io.BytesIO]:
-        """
-        This function will get the images from the page "restaurant-scolaire".
-        Something noticeable is that, it doesn't search for each images links in the page, but rather get the whole
-        page downloaded in a zip, using an export method of mbn.
-
-        The zip structure looks like :
-        .
-        ├── xxx/
-        │   ├── photo_dddd-ddddd.jpeg
-        │   ├── photo_dddd-ddddd.jpeg
-        │   ├── v_photo_dddd-ddddd.jpeg
-        │   └── v_photo_dddd-ddddd.jpeg
-        └── Restaurant scolaire_xxx.html
-
-        So, by checking the file name in xxx/, we can check if the image was already sent.
-        Btw, the v_* files are just the photos in a very very low quality. So we ignore them.
-
-        By doing this, we are able to get all images in one single request.
-        But we *need* to fetch them all each time, ig.
-        Something possible is to compare the ID_METATAG, to avoid redundant requests.
-        Otherwise, we can look at each single photos in the page 1 by 1, to not make extra request if already fetched.
-        But this implies to multiply the minimal amount of request needed.
+    async def get_imgs(self) -> tuple[tuple[str, ...], tuple[str, ...]]:
+        """Scrap the website to get the menu and allergenes images.
+
+        Returns:
+            A tuple with fr:MENUs, and a second tuple with fr:ALLERGENES.
         """
         async with httpx.AsyncClient() as client:
-            result = await client.get("https://lyc-kleber.monbureaunumerique.fr/l-etablissement/restaurant-scolaire/")
+            result = await client.get("https://lycee-kleber.com.fr/restauration")
             page = result.text
 
-            scrap = BeautifulSoup(page, "html.parser")
-
-            if not isinstance(tag := scrap.find("input", id="ID_METATAG"), Tag):
-                raise Exception("Could not find the tag")
-            if not isinstance(tag_id := tag.get("value"), str):
-                raise Exception("Could not find the tag id")
-
-            print(tag_id)  # temp : just check if the tag change when file changes.
-
-            result = await client.get(
-                f"https://lyc-kleber.monbureaunumerique.fr/exportContent?&ACTION=EXPORTER&exportType=FICHE&ID_FICHE={tag_id}"
-            )
-            zip_buffer = io.BytesIO(result.read())
-            zip_obj = ZipFile(zip_buffer, "r")
-
-            imgs_buffers: dict[str, io.BytesIO] = {}
-            imgs = [obj for obj in zip_obj.infolist() if obj.filename.endswith("jpeg") and obj.file_size > 50_000]
-
-            for img in imgs:
-                buffer = io.BytesIO(zip_obj.read(img.filename))
-                buffer.seek(0)
-                imgs_buffers[img.filename] = buffer
+        scrap = BeautifulSoup(page, "html.parser")
+        element = scrap.find_all("a", href=lambda r: bool(IMAGES_REGEX.match(r)))
+        links: list[str] = [e.get("href") for e in element]
 
-            return imgs_buffers
+        menus: tuple[str, ...] = tuple(
+            l for l in links if (m := IMAGES_REGEX.match(l)) and m.group(1).lower().startswith("menu")
+        )
+        allergenes: tuple[str, ...] = tuple(
+            l for l in links if (m := IMAGES_REGEX.match(l)) and m.group(1).lower().startswith("allergenes")
+        )
+        return menus, allergenes
 
-    async def post_menu(self, imgs: dict[str, io.BytesIO]) -> None:
-        """
-        This function will post the new images on Discord.
+    @tasks.loop(minutes=60)
+    async def check_menu(self) -> None:
+        """Post the menu if there is a new one, checked every hour."""
+        menus, _ = await self.get_imgs()
+        menus = [m for m in menus if m not in self.already_posted]  # filter with only new ones.
+        for img_link in menus:
+            self.add_restauration_file(img_link)
 
-        It will look for each channel named "menu-cantine" in all guilds the bot have.
-        Then it will just post all the images in one single message for all of those.
-        """
         channels: list[TextChannel] = [
             ch for ch in self.bot.get_all_channels() if isinstance(ch, TextChannel) and ch.name == "menu-cantine"
         ]
+
         for channel in channels:
-            files = [File(buffer, filename=filename) for filename, buffer in imgs.items()]
             try:
-                await channel.send(files=files)
+                await channel.send("\n".join(menus))
             except HTTPException:
                 pass
-            [buffer.seek(0) for buffer in imgs.values()]
 
-    @tasks.loop(minutes=30)
-    async def check_menu(self) -> None:
-        try:
-            menus = await self.get_menu_imgs()
-        except Exception:
-            return
-
-        menus = {filename: image_file for filename, image_file in menus.items() if filename not in self.already_posted}
-        for filename in menus:
-            self.add_restauration_file(filename)
-        await self.post_menu(menus)
+    @app_commands.command(name="allergenes", description="Affiche les allergènes du menu du jour.")
+    async def allergen(self, inter: discord.Interaction):
+        _, allergens = await self.get_imgs()
+        bn = "\n"
+        await inter.response.send_message(
+            (
+                "Voici les allergènes du menu du jour :\n"
+                f"{bn.join(allergens)}"
+                "\n\nS'ils ne sont pas à jour, c'est que le lycée ne les a pas publié.\n"
+                "Attention : les allergènes sont susceptibles d'être modifiés, merci de se référer au panneau"
+                " d'affichage à la restauration scolaire."
+            )
+        )
 
 
 async def setup(bot: MP2IBot):