warp-beacon 2.6.80__tar.gz → 2.6.82__tar.gz

{warp_beacon-2.6.80/warp_beacon.egg-info → warp_beacon-2.6.82}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: warp_beacon
-Version: 2.6.80
+Version: 2.6.82
 Summary: Telegram bot for expanding external media links
 Home-page: https://github.com/sb0y/warp_beacon
 Author: Andrey Bagrintsev

warp_beacon-2.6.82/warp_beacon/__version__.py

@@ -0,0 +1,2 @@
+__version__ = "2.6.82"
+

{warp_beacon-2.6.80 → warp_beacon-2.6.82}/warp_beacon/scraper/instagram/wb_instagrapi.py

@@ -1,6 +1,6 @@
 import logging
 from typing import Callable
-from urllib.parse import urlparse
+from urllib.parse import urlparse, parse_qs, urlencode, urlunparse
 from pathlib import Path
 import requests
 
@@ -14,20 +14,59 @@ class WBClient(Client):
 	def __init__(self) -> None:
 		super().__init__()
 		self.progress_callback = None
+		self.session = requests.Session()
+		# may be I should remove '"Sec-Fetch-*", "Upgrade-Insecure-Requests", "DNT"' ?
+		self.session.headers.update({
+			"User-Agent": (
+				"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+				"(KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36"
+			),
+			"Accept": (
+				"text/html,application/xhtml+xml,application/xml;"
+				"q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8"
+			),
+			"Accept-Language": "en-US,en;q=0.9",
+			"Accept-Encoding": "gzip, deflate, br",
+			"Referer": "https://www.instagram.com/",
+			"Connection": "keep-alive",
+			"Sec-Fetch-Site": "same-origin",
+			"Sec-Fetch-Mode": "navigate",
+			"Sec-Fetch-User": "?1",
+			"Sec-Fetch-Dest": "document",
+			"Upgrade-Insecure-Requests": "1",
+			"DNT": "1",
+		})
+		self.essential_params = {"oe", "oh", "_nc_ht", "_nc_cat", "_nc_oc", "_nc_ohc", "_nc_gid"}
 
 	def set_progress_callback(self, callback: Callable[[int | None, int, Path], None]) -> None:
 		if not callback or not callable(callback):
 			raise TypeError("Progress callback must be callable")
 		self.progress_callback = callback
 
+	def sanitize_instagram_url(self, url: str) -> str:
+		parsed = urlparse(url)
+		query = parse_qs(parsed.query)
+		filtered_query = {k: v for k, v in query.items() if k in self.essential_params}
+		new_query = urlencode(filtered_query, doseq=True)
+		return urlunparse(parsed._replace(query=new_query))
+
 	def video_download_by_url(self, url: str, filename: str = "", folder: Path = "") -> Path:
+		url = self.sanitize_instagram_url(url)
 		fname = urlparse(url).path.rsplit("/", 1)[1]
 		filename = f"{filename}.{fname.rsplit('.', 1)[1]}" if filename else fname
 		path = Path(folder or Path.cwd()) / filename
 
 		logging.info("Downloading video from '%s' to '%s'", url, path)
 
-		response = requests.get(url, stream=True, timeout=self.request_timeout)
+		prepared = self.session.prepare_request(requests.Request("GET", url))
+		logging.info("Prepared headers: %s", prepared.headers)
+		response = self.session.send(
+			prepared,
+			stream=True,
+			verify=False,
+			proxies=self.public.proxies,
+			timeout=self.request_timeout
+		)
 		response.raise_for_status()
 		logging.info("Response headers: %s", response.headers)
 
@@ -64,13 +103,23 @@ class WBClient(Client):
 	def photo_download_by_url(
 		self, url: str, filename: str = "", folder: Path = ""
 	) -> Path:
+		url = self.sanitize_instagram_url(url)
 		fname = urlparse(url).path.rsplit("/", 1)[1]
 		filename = f"{filename}.{(filename, fname.rsplit('.', 1)[1]) if filename else fname}"
 		path = Path(folder) / filename
 
 		logging.info("Downloading photo from '%s' to '%s'", url, path)
-
-		response = requests.get(url, stream=True, timeout=self.request_timeout)
+		logging.info("[Downloader] Using proxies: %s", self.public.proxies)
+
+		prepared = self.session.prepare_request(requests.Request("GET", url))
+		logging.info("Prepared headers: %s", prepared.headers)
+		response = self.session.send(
+			prepared,
+			stream=True,
+			verify=False,
+			proxies=self.public.proxies,
+			timeout=self.request_timeout
+		)
 		response.raise_for_status()
 		logging.info("Response headers: %s", response.headers)
 

{warp_beacon-2.6.80 → warp_beacon-2.6.82/warp_beacon.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: warp_beacon
-Version: 2.6.80
+Version: 2.6.82
 Summary: Telegram bot for expanding external media links
 Home-page: https://github.com/sb0y/warp_beacon
 Author: Andrey Bagrintsev

warp_beacon-2.6.80/warp_beacon/__version__.py

@@ -1,2 +0,0 @@
-__version__ = "2.6.80"
-