From 25895eb63799a1c38a40f03a48e85cdd8e9dc4b9 Mon Sep 17 00:00:00 2001 From: Heiner Lohaus Date: Wed, 3 Jan 2024 16:02:06 +0100 Subject: Add create images to Bing --- g4f/Provider/Bestim.py | 43 +++--- g4f/Provider/Bing.py | 304 +++++++------------------------------ g4f/Provider/__init__.py | 1 + g4f/Provider/bing/conversation.py | 43 ++++++ g4f/Provider/bing/create_images.py | 146 ++++++++++++++++++ g4f/Provider/bing/upload_image.py | 182 ++++++++++++++++++++++ g4f/Provider/create_images.py | 71 +++++++++ g4f/__init__.py | 8 + g4f/gui/client/html/index.html | 5 + g4f/gui/client/js/chat.v1.js | 26 +++- g4f/gui/server/backend.py | 13 +- g4f/requests.py | 5 +- g4f/webdriver.py | 3 + requirements.txt | 1 + 14 files changed, 568 insertions(+), 283 deletions(-) create mode 100644 g4f/Provider/bing/conversation.py create mode 100644 g4f/Provider/bing/create_images.py create mode 100644 g4f/Provider/bing/upload_image.py create mode 100644 g4f/Provider/create_images.py diff --git a/g4f/Provider/Bestim.py b/g4f/Provider/Bestim.py index 312655b8..be95b48a 100644 --- a/g4f/Provider/Bestim.py +++ b/g4f/Provider/Bestim.py @@ -1,42 +1,35 @@ from __future__ import annotations -from ..typing import Messages, List, Dict +from ..typing import Messages from .base_provider import BaseProvider, CreateResult +from ..requests import get_session_from_browser from uuid import uuid4 import requests -def format_prompt(messages) -> List[Dict[str, str]]: - - return [{"id": str(uuid4()), "content": '\n'.join(f'{m["role"]}: {m["content"]}' for m in messages), "from": "you"}] - class Bestim(BaseProvider): url = "https://chatgpt.bestim.org" supports_gpt_35_turbo = True supports_message_history = True - working = True + working = False supports_stream = True - @staticmethod + @classmethod def create_completion( + cls, model: str, messages: Messages, stream: bool, proxy: str = None, **kwargs ) -> CreateResult: - + session = get_session_from_browser(cls.url, proxy=proxy) headers = { - 'POST': '/chat/send2/ HTTP/3', - 'Host': 'chatgpt.bestim.org', 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:121.0) Gecko/20100101 Firefox/121.0', 'Accept': 'application/json, text/event-stream', 'Accept-Language': 'en-US,en;q=0.5', 'Accept-Encoding': 'gzip, deflate, br', 'Referer': 'https://chatgpt.bestim.org/chat/', - 'Content-Type': 'application/json', - 'Content-Length': '109', 'Origin': 'https://chatgpt.bestim.org', - 'Cookie': 'NpZAER=qKkRHguMIOraVbJAWpoyzGLFjZwYlm; qKkRHguMIOraVbJAWpoyzGLFjZwYlm=8ebb5ae1561bde05354de5979b52c6e1-1704058188-1704058188; NpZAER_hits=2; _csrf-front=fcf20965823c0a152ae8f9cdf15b23022bb26cdc6bf32a9d4c8bfe78dcc6b807a%3A2%3A%7Bi%3A0%3Bs%3A11%3A%22_csrf-front%22%3Bi%3A1%3Bs%3A32%3A%22a5wP6azsc7dxV8rmwAXaNsl8XS1yvW5V%22%3B%7D', 'Alt-Used': 'chatgpt.bestim.org', 'Connection': 'keep-alive', 'Sec-Fetch-Dest': 'empty', @@ -44,27 +37,25 @@ class Bestim(BaseProvider): 'Sec-Fetch-Site': 'same-origin', 'TE': 'trailers' } - data = { - - "messagesHistory": format_prompt(messages), + "messagesHistory": [{ + "id": str(uuid4()), + "content": m["content"], + "from": "you" if m["role"] == "user" else "bot" + } for m in messages], "type": "chat", } - - response = requests.post( + response = session.post( url="https://chatgpt.bestim.org/chat/send2/", headers=headers, json=data, - proxies={"https": proxy} + proxies={"https": proxy}, + stream=True ) - response.raise_for_status() - - for chunk in response.iter_lines(): - - if b"event: trylimit" not in chunk: - - yield chunk.decode().removeprefix("data: ") + for line in response.iter_lines(): + if not line.startswith(b"event: trylimit"): + yield line.decode().removeprefix("data: ") diff --git a/g4f/Provider/Bing.py b/g4f/Provider/Bing.py index e03b413f..57d352a9 100644 --- a/g4f/Provider/Bing.py +++ b/g4f/Provider/Bing.py @@ -1,35 +1,25 @@ from __future__ import annotations -import string import random import json import os -import re -import io -import base64 -import numpy as np import uuid -import urllib.parse import time -from PIL import Image -from aiohttp import ClientSession, ClientTimeout -from ..typing import AsyncResult, Messages +from urllib import parse +from aiohttp import ClientSession, ClientTimeout + +from ..typing import AsyncResult, Messages from .base_provider import AsyncGeneratorProvider +from ..webdriver import get_browser, get_driver_cookies +from .bing.upload_image import upload_image +from .bing.create_images import create_images, format_images_markdown, wait_for_login +from .bing.conversation import Conversation, create_conversation, delete_conversation class Tones(): creative = "Creative" balanced = "Balanced" precise = "Precise" -default_cookies = { - 'SRCHD' : 'AF=NOFORM', - 'PPLState' : '1', - 'KievRPSSecAuth': '', - 'SUID' : '', - 'SRCHUSR' : '', - 'SRCHHPGUSR' : f'HV={int(time.time())}', -} - class Bing(AsyncGeneratorProvider): url = "https://bing.com/chat" working = True @@ -55,9 +45,9 @@ class Bing(AsyncGeneratorProvider): context = create_context(messages[:-1]) if not cookies: - cookies = default_cookies + cookies = Defaults.cookies else: - for key, value in default_cookies.items(): + for key, value in Defaults.cookies.items(): if key not in cookies: cookies[key] = value @@ -71,106 +61,6 @@ def create_context(messages: Messages): for message in messages ) -class Conversation(): - def __init__(self, conversationId: str, clientId: str, conversationSignature: str, imageInfo: dict=None) -> None: - self.conversationId = conversationId - self.clientId = clientId - self.conversationSignature = conversationSignature - self.imageInfo = imageInfo - -async def create_conversation(session: ClientSession, tone: str, image: str = None, proxy: str = None) -> Conversation: - url = 'https://www.bing.com/turing/conversation/create?bundleVersion=1.1199.4' - async with session.get(url, proxy=proxy) as response: - data = await response.json() - - conversationId = data.get('conversationId') - clientId = data.get('clientId') - conversationSignature = response.headers.get('X-Sydney-Encryptedconversationsignature') - - if not conversationId or not clientId or not conversationSignature: - raise Exception('Failed to create conversation.') - conversation = Conversation(conversationId, clientId, conversationSignature, None) - if isinstance(image,str): - try: - config = { - "visualSearch": { - "maxImagePixels": 360000, - "imageCompressionRate": 0.7, - "enableFaceBlurDebug": 0, - } - } - is_data_uri_an_image(image) - img_binary_data = extract_data_uri(image) - is_accepted_format(img_binary_data) - img = Image.open(io.BytesIO(img_binary_data)) - width, height = img.size - max_image_pixels = config['visualSearch']['maxImagePixels'] - compression_rate = config['visualSearch']['imageCompressionRate'] - - if max_image_pixels / (width * height) < 1: - new_width = int(width * np.sqrt(max_image_pixels / (width * height))) - new_height = int(height * np.sqrt(max_image_pixels / (width * height))) - else: - new_width = width - new_height = height - try: - orientation = get_orientation(img) - except Exception: - orientation = None - new_img = process_image(orientation, img, new_width, new_height) - new_img_binary_data = compress_image_to_base64(new_img, compression_rate) - data, boundary = build_image_upload_api_payload(new_img_binary_data, conversation, tone) - headers = session.headers.copy() - headers["content-type"] = f'multipart/form-data; boundary={boundary}' - headers["referer"] = 'https://www.bing.com/search?q=Bing+AI&showconv=1&FORM=hpcodx' - headers["origin"] = 'https://www.bing.com' - async with session.post("https://www.bing.com/images/kblob", data=data, headers=headers, proxy=proxy) as image_upload_response: - if image_upload_response.status != 200: - raise Exception("Failed to upload image.") - - image_info = await image_upload_response.json() - if not image_info.get('blobId'): - raise Exception("Failed to parse image info.") - result = {'bcid': image_info.get('blobId', "")} - result['blurredBcid'] = image_info.get('processedBlobId', "") - if result['blurredBcid'] != "": - result["imageUrl"] = "https://www.bing.com/images/blob?bcid=" + result['blurredBcid'] - elif result['bcid'] != "": - result["imageUrl"] = "https://www.bing.com/images/blob?bcid=" + result['bcid'] - result['originalImageUrl'] = ( - "https://www.bing.com/images/blob?bcid=" - + result['blurredBcid'] - if config['visualSearch']["enableFaceBlurDebug"] - else "https://www.bing.com/images/blob?bcid=" - + result['bcid'] - ) - conversation.imageInfo = result - except Exception as e: - print(f"An error happened while trying to send image: {str(e)}") - return conversation - -async def list_conversations(session: ClientSession) -> list: - url = "https://www.bing.com/turing/conversation/chats" - async with session.get(url) as response: - response = await response.json() - return response["chats"] - -async def delete_conversation(session: ClientSession, conversation: Conversation, proxy: str = None) -> list: - url = "https://sydney.bing.com/sydney/DeleteSingleConversation" - json = { - "conversationId": conversation.conversationId, - "conversationSignature": conversation.conversationSignature, - "participant": {"id": conversation.clientId}, - "source": "cib", - "optionsSets": ["autosave"] - } - async with session.post(url, json=json, proxy=proxy) as response: - try: - response = await response.json() - return response["result"]["value"] == "Success" - except: - return False - class Defaults: delimiter = "\x1e" ip_address = f"13.{random.randint(104, 107)}.{random.randint(0, 255)}.{random.randint(0, 255)}" @@ -264,123 +154,28 @@ class Defaults: 'eredirecturl', 'nojbfedge' ] + + cookies = { + 'SRCHD' : 'AF=NOFORM', + 'PPLState' : '1', + 'KievRPSSecAuth': '', + 'SUID' : '', + 'SRCHUSR' : '', + 'SRCHHPGUSR' : f'HV={int(time.time())}', + } def format_message(msg: dict) -> str: return json.dumps(msg, ensure_ascii=False) + Defaults.delimiter -def build_image_upload_api_payload(image_bin: str, conversation: Conversation, tone: str): - payload = { - 'invokedSkills': ["ImageById"], - 'subscriptionId': "Bing.Chat.Multimodal", - 'invokedSkillsRequestData': { - 'enableFaceBlur': True - }, - 'convoData': { - 'convoid': "", - 'convotone': tone - } - } - knowledge_request = { - 'imageInfo': {}, - 'knowledgeRequest': payload - } - boundary="----WebKitFormBoundary" + ''.join(random.choices(string.ascii_letters + string.digits, k=16)) - data = ( - f'--{boundary}' - + '\r\nContent-Disposition: form-data; name="knowledgeRequest"\r\n\r\n' - + json.dumps(knowledge_request, ensure_ascii=False) - + "\r\n--" - + boundary - + '\r\nContent-Disposition: form-data; name="imageBase64"\r\n\r\n' - + image_bin - + "\r\n--" - + boundary - + "--\r\n" - ) - return data, boundary - -def is_data_uri_an_image(data_uri: str): - try: - # Check if the data URI starts with 'data:image' and contains an image format (e.g., jpeg, png, gif) - if not re.match(r'data:image/(\w+);base64,', data_uri): - raise ValueError("Invalid data URI image.") - # Extract the image format from the data URI - image_format = re.match(r'data:image/(\w+);base64,', data_uri).group(1) - # Check if the image format is one of the allowed formats (jpg, jpeg, png, gif) - if image_format.lower() not in ['jpeg', 'jpg', 'png', 'gif']: - raise ValueError("Invalid image format (from mime file type).") - except Exception as e: - raise e - -def is_accepted_format(binary_data: bytes) -> bool: - try: - check = False - if binary_data.startswith(b'\xFF\xD8\xFF'): - check = True # It's a JPEG image - elif binary_data.startswith(b'\x89PNG\r\n\x1a\n'): - check = True # It's a PNG image - elif binary_data.startswith(b'GIF87a') or binary_data.startswith(b'GIF89a'): - check = True # It's a GIF image - elif binary_data.startswith(b'\x89JFIF') or binary_data.startswith(b'JFIF\x00'): - check = True # It's a JPEG image - elif binary_data.startswith(b'\xFF\xD8'): - check = True # It's a JPEG image - elif binary_data.startswith(b'RIFF') and binary_data[8:12] == b'WEBP': - check = True # It's a WebP image - # else we raise ValueError - if not check: - raise ValueError("Invalid image format (from magic code).") - except Exception as e: - raise e - -def extract_data_uri(data_uri: str) -> bytes: - try: - data = data_uri.split(",")[1] - data = base64.b64decode(data) - return data - except Exception as e: - raise e - -def get_orientation(data: bytes) -> int: - try: - if data[:2] != b'\xFF\xD8': - raise Exception('NotJpeg') - with Image.open(data) as img: - exif_data = img._getexif() - if exif_data is not None: - orientation = exif_data.get(274) # 274 corresponds to the orientation tag in EXIF - if orientation is not None: - return orientation - except Exception: - pass - -def process_image(orientation: int, img: Image.Image, new_width: int, new_height: int) -> Image.Image: - try: - # Initialize the canvas - new_img = Image.new("RGB", (new_width, new_height), color="#FFFFFF") - if orientation: - if orientation > 4: - img = img.transpose(Image.FLIP_LEFT_RIGHT) - if orientation in [3, 4]: - img = img.transpose(Image.ROTATE_180) - if orientation in [5, 6]: - img = img.transpose(Image.ROTATE_270) - if orientation in [7, 8]: - img = img.transpose(Image.ROTATE_90) - new_img.paste(img, (0, 0)) - return new_img - except Exception as e: - raise e - -def compress_image_to_base64(img, compression_rate) -> str: - try: - output_buffer = io.BytesIO() - img.save(output_buffer, format="JPEG", quality=int(compression_rate * 100)) - return base64.b64encode(output_buffer.getvalue()).decode('utf-8') - except Exception as e: - raise e - -def create_message(conversation: Conversation, prompt: str, tone: str, context: str = None, web_search: bool = False, gpt4_turbo: bool = False) -> str: +def create_message( + conversation: Conversation, + prompt: str, + tone: str, + context: str = None, + image_info: dict = None, + web_search: bool = False, + gpt4_turbo: bool = False +) -> str: options_sets = Defaults.optionsSets if tone == Tones.creative: options_sets.append("h3imaginative") @@ -429,9 +224,9 @@ def create_message(conversation: Conversation, prompt: str, tone: str, context: 'target': 'chat', 'type': 4 } - if conversation.imageInfo != None and "imageUrl" in conversation.imageInfo and "originalImageUrl" in conversation.imageInfo: - struct['arguments'][0]['message']['originalImageUrl'] = conversation.imageInfo['originalImageUrl'] - struct['arguments'][0]['message']['imageUrl'] = conversation.imageInfo['imageUrl'] + if image_info and "imageUrl" in image_info and "originalImageUrl" in image_info: + struct['arguments'][0]['message']['originalImageUrl'] = image_info['originalImageUrl'] + struct['arguments'][0]['message']['imageUrl'] = image_info['imageUrl'] struct['arguments'][0]['experienceType'] = None struct['arguments'][0]['attachedFileInfo'] = {"fileName": None, "fileType": None} if context: @@ -454,17 +249,23 @@ async def stream_generate( web_search: bool = False, gpt4_turbo: bool = False ): + headers = Defaults.headers + if cookies: + headers["Cookie"] = "; ".join(f"{k}={v}" for k, v in cookies.items()) async with ClientSession( - timeout=ClientTimeout(total=900), - headers=Defaults.headers if not cookies else {**Defaults.headers, "Cookie": "; ".join(f"{k}={v}" for k, v in cookies.items())}, - ) as session: - conversation = await create_conversation(session, tone, image, proxy) + timeout=ClientTimeout(total=900), + headers=headers + ) as session: + conversation = await create_conversation(session, proxy) + image_info = None + if image: + image_info = await upload_image(session, image, tone, proxy) try: async with session.ws_connect('wss://sydney.bing.com/sydney/ChatHub', autoping=False, params={'sec_access_token': conversation.conversationSignature}, proxy=proxy) as wss: await wss.send_str(format_message({'protocol': 'json', 'version': 1})) await wss.receive(timeout=900) - await wss.send_str(create_message(conversation, prompt, tone, context, web_search, gpt4_turbo)) + await wss.send_str(create_message(conversation, prompt, tone, context, image_info, web_search, gpt4_turbo)) response_txt = '' returned_text = '' @@ -488,9 +289,11 @@ async def stream_generate( inline_txt = card['inlines'][0].get('text') response_txt += inline_txt + '\n' elif message.get('contentType') == "IMAGE": - query = urllib.parse.quote(message.get('text')) - url = f"\nhttps://www.bing.com/images/create?q={query}" - response_txt += url + prompt = message.get('text') + try: + response_txt += format_images_markdown(await create_images(session, prompt, proxy), prompt) + except: + response_txt += f"\nhttps://www.bing.com/images/create?q={parse.quote(prompt)}" final = True if response_txt.startswith(returned_text): new = response_txt[len(returned_text):] @@ -500,7 +303,18 @@ async def stream_generate( elif response.get('type') == 2: result = response['item']['result'] if result.get('error'): - raise Exception(f"{result['value']}: {result['message']}") + if result["value"] == "CaptchaChallenge": + driver = get_browser(proxy=proxy) + try: + for chunk in wait_for_login(driver): + yield chunk + cookies = get_driver_cookies(driver) + finally: + driver.quit() + async for chunk in stream_generate(prompt, tone, image, context, proxy, cookies, web_search, gpt4_turbo): + yield chunk + else: + raise Exception(f"{result['value']}: {result['message']}") return finally: await delete_conversation(session, conversation, proxy) diff --git a/g4f/Provider/__init__.py b/g4f/Provider/__init__.py index 9703983a..ed1bc5bb 100644 --- a/g4f/Provider/__init__.py +++ b/g4f/Provider/__init__.py @@ -3,6 +3,7 @@ from __future__ import annotations from ..base_provider import BaseProvider, ProviderType from .retry_provider import RetryProvider from .base_provider import AsyncProvider, AsyncGeneratorProvider +from .create_images import CreateImagesProvider from .deprecated import * from .needs_auth import * from .unfinished import * diff --git a/g4f/Provider/bing/conversation.py b/g4f/Provider/bing/conversation.py new file mode 100644 index 00000000..8b33853c --- /dev/null +++ b/g4f/Provider/bing/conversation.py @@ -0,0 +1,43 @@ +from aiohttp import ClientSession + + +class Conversation(): + def __init__(self, conversationId: str, clientId: str, conversationSignature: str) -> None: + self.conversationId = conversationId + self.clientId = clientId + self.conversationSignature = conversationSignature + +async def create_conversation(session: ClientSession, proxy: str = None) -> Conversation: + url = 'https://www.bing.com/turing/conversation/create?bundleVersion=1.1199.4' + async with session.get(url, proxy=proxy) as response: + data = await response.json() + + conversationId = data.get('conversationId') + clientId = data.get('clientId') + conversationSignature = response.headers.get('X-Sydney-Encryptedconversationsignature') + + if not conversationId or not clientId or not conversationSignature: + raise Exception('Failed to create conversation.') + return Conversation(conversationId, clientId, conversationSignature) + +async def list_conversations(session: ClientSession) -> list: + url = "https://www.bing.com/turing/conversation/chats" + async with session.get(url) as response: + response = await response.json() + return response["chats"] + +async def delete_conversation(session: ClientSession, conversation: Conversation, proxy: str = None) -> list: + url = "https://sydney.bing.com/sydney/DeleteSingleConversation" + json = { + "conversationId": conversation.conversationId, + "conversationSignature": conversation.conversationSignature, + "participant": {"id": conversation.clientId}, + "source": "cib", + "optionsSets": ["autosave"] + } + async with session.post(url, json=json, proxy=proxy) as response: + try: + response = await response.json() + return response["result"]["value"] == "Success" + except: + return False diff --git a/g4f/Provider/bing/create_images.py b/g4f/Provider/bing/create_images.py new file mode 100644 index 00000000..c868a6f0 --- /dev/null +++ b/g4f/Provider/bing/create_images.py @@ -0,0 +1,146 @@ + + +import asyncio +import time, json, os +from aiohttp import ClientSession +from bs4 import BeautifulSoup +from urllib.parse import quote +from typing import Generator + +from ...webdriver import WebDriver, get_driver_cookies, get_browser +from ...Provider.helper import get_event_loop + +BING_URL = "https://www.bing.com" + +def wait_for_login(driver: WebDriver, timeout: int = 1200): + driver.get(f"{BING_URL}/") + value = driver.get_cookie("_U") + if value: + return + login_url = os.environ.get("G4F_LOGIN_URL") + if login_url: + yield f"Please login: [Bing]({login_url})\n\n" + start_time = time.time() + while True: + if time.time() - start_time > timeout: + raise RuntimeError("Timeout error") + value = driver.get_cookie("_U") + if value: + return + time.sleep(0.1) + +def create_session(cookies: dict): + headers = { + "accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7", + "accept-encoding": "gzip, deflate, br", + "accept-language": "en-US,en;q=0.9,zh-CN;q=0.8,zh-TW;q=0.7,zh;q=0.6", + "content-type": "application/x-www-form-urlencoded", + "referrer-policy": "origin-when-cross-origin", + "referrer": "https://www.bing.com/images/create/", + "origin": "https://www.bing.com", + "user-agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36 Edg/111.0.1661.54", + "sec-ch-ua": "\"Microsoft Edge\";v=\"111\", \"Not(A:Brand\";v=\"8\", \"Chromium\";v=\"111\"", + "sec-ch-ua-mobile": "?0", + "sec-fetch-dest": "document", + "sec-fetch-mode": "navigate", + "sec-fetch-site": "same-origin", + "sec-fetch-user": "?1", + "upgrade-insecure-requests": "1", + } + if cookies: + headers["cookie"] = "; ".join(f"{k}={v}" for k, v in cookies.items()) + return ClientSession(headers=headers) + +async def create_images(session: ClientSession, prompt: str, proxy: str = None, timeout: int = 200): + url_encoded_prompt = quote(prompt) + payload = f"q={url_encoded_prompt}&rt=4&FORM=GENCRE" + url = f"{BING_URL}/images/create?q={url_encoded_prompt}&rt=4&FORM=GENCRE" + async with session.post( + url, + allow_redirects=False, + data=payload, + timeout=timeout, + ) as response: + response.raise_for_status() + errors = [ + "this prompt is being reviewed", + "this prompt has been blocked", + "we're working hard to offer image creator in more languages" + ] + text = (await response.text()).lower() + for error in errors: + if error in text: + raise RuntimeError(f"Create images failed: {error}") + if response.status != 302: + url = f"{BING_URL}/images/create?q={url_encoded_prompt}&rt=3&FORM=GENCRE" + async with session.post(url, allow_redirects=False, proxy=proxy, timeout=timeout) as response: + if response.status != 302: + raise RuntimeError(f"Create images failed. Status Code: {response.status}") + + redirect_url = response.headers["Location"].replace("&nfy=1", "") + redirect_url = f"{BING_URL}{redirect_url}" + request_id = redirect_url.split("id=")[1] + async with session.get(redirect_url) as response: + response.raise_for_status() + + polling_url = f"{BING_URL}/images/create/async/results/{request_id}?q={url_encoded_prompt}" + start_time = time.time() + while True: + if time.time() - start_time > timeout: + raise RuntimeError("Timeout error") + async with session.get(polling_url) as response: + if response.status != 200: + raise RuntimeError(f"Polling images faild. Status Code: {response.status}") + text = await response.text() + if not text: + await asyncio.sleep(1) + else: + break + + error = None + try: + error = json.loads(text).get("errorMessage") + except: + pass + if error == "Pending": + raise RuntimeError("Prompt is been blocked") + elif error: + raise RuntimeError(f"Error: {error}") + + return get_images(text) + +def format_images_markdown(images: list, prompt: str) -> str: + images = [f"[![#{idx+1} {prompt}]({image}?w=200&h=200)]({image})" for idx, image in enumerate(images)] + return f"\n\n\n\n" + ("\n".join(images)) + "\n\n\n" + +def get_images(text: str) -> list: + html_soup = BeautifulSoup(text, "html.parser") + tags = html_soup.find_all("img") + image_links = [img["src"] for img in tags if "mimg" in img["class"]] + images = [link.split("?w=")[0] for link in image_links] + bad_images = [ + "https://r.bing.com/rp/in-2zU3AJUdkgFe7ZKv19yPBHVs.png", + "https://r.bing.com/rp/TX9QuO3WzcCJz1uaaSwQAz39Kb0.jpg", + ] + if any(im in bad_images for im in images): + raise RuntimeError("Bad images found") + if not images: + raise RuntimeError("No images found") + return images + +def create_completion(prompt: str, proxy: str = None) -> Generator: + loop = get_event_loop() + driver = get_browser(proxy=proxy) + async def run_session(): + cookies = get_driver_cookies(driver) + session = create_session(cookies) + try: + return await create_images(session, prompt, proxy) + finally: + await session.close() + try: + yield from wait_for_login(driver) + images = loop.run_until_complete(run_session()) + yield format_images_markdown(images, prompt) + finally: + driver.quit() \ No newline at end of file diff --git a/g4f/Provider/bing/upload_image.py b/g4f/Provider/bing/upload_image.py new file mode 100644 index 00000000..2788edfd --- /dev/null +++ b/g4f/Provider/bing/upload_image.py @@ -0,0 +1,182 @@ +from __future__ import annotations + +import string +import random +import json +import re +import io +import base64 +import numpy as np +from PIL import Image +from aiohttp import ClientSession + +async def upload_image( + session: ClientSession, + image: str, + tone: str, + proxy: str = None +): + try: + image_config = { + "maxImagePixels": 360000, + "imageCompressionRate": 0.7, + "enableFaceBlurDebug": 0, + } + is_data_uri_an_image(image) + img_binary_data = extract_data_uri(image) + is_accepted_format(img_binary_data) + img = Image.open(io.BytesIO(img_binary_data)) + width, height = img.size + max_image_pixels = image_config['maxImagePixels'] + if max_image_pixels / (width * height) < 1: + new_width = int(width * np.sqrt(max_image_pixels / (width * height))) + new_height = int(height * np.sqrt(max_image_pixels / (width * height))) + else: + new_width = width + new_height = height + try: + orientation = get_orientation(img) + except Exception: + orientation = None + new_img = process_image(orientation, img, new_width, new_height) + new_img_binary_data = compress_image_to_base64(new_img, image_config['imageCompressionRate']) + data, boundary = build_image_upload_api_payload(new_img_binary_data, tone) + headers = session.headers.copy() + headers["content-type"] = f'multipart/form-data; boundary={boundary}' + headers["referer"] = 'https://www.bing.com/search?q=Bing+AI&showconv=1&FORM=hpcodx' + headers["origin"] = 'https://www.bing.com' + async with session.post("https://www.bing.com/images/kblob", data=data, headers=headers, proxy=proxy) as response: + if response.status != 200: + raise RuntimeError("Failed to upload image.") + image_info = await response.json() + if not image_info.get('blobId'): + raise RuntimeError("Failed to parse image info.") + result = {'bcid': image_info.get('blobId', "")} + result['blurredBcid'] = image_info.get('processedBlobId', "") + if result['blurredBcid'] != "": + result["imageUrl"] = "https://www.bing.com/images/blob?bcid=" + result['blurredBcid'] + elif result['bcid'] != "": + result["imageUrl"] = "https://www.bing.com/images/blob?bcid=" + result['bcid'] + result['originalImageUrl'] = ( + "https://www.bing.com/images/blob?bcid=" + + result['blurredBcid'] + if image_config["enableFaceBlurDebug"] + else "https://www.bing.com/images/blob?bcid=" + + result['bcid'] + ) + return result + except Exception as e: + raise RuntimeError(f"Add image failed: {e}") + + +def build_image_upload_api_payload(image_bin: str, tone: str): + payload = { + 'invokedSkills': ["ImageById"], + 'subscriptionId': "Bing.Chat.Multimodal", + 'invokedSkillsRequestData': { + 'enableFaceBlur': True + }, + 'convoData': { + 'convoid': "", + 'convotone': tone + } + } + knowledge_request = { + 'imageInfo': {}, + 'knowledgeRequest': payload + } + boundary="----WebKitFormBoundary" + ''.join(random.choices(string.ascii_letters + string.digits, k=16)) + data = ( + f'--{boundary}' + + '\r\nContent-Disposition: form-data; name="knowledgeRequest"\r\n\r\n' + + json.dumps(knowledge_request, ensure_ascii=False) + + "\r\n--" + + boundary + + '\r\nContent-Disposition: form-data; name="imageBase64"\r\n\r\n' + + image_bin + + "\r\n--" + + boundary + + "--\r\n" + ) + return data, boundary + +def is_data_uri_an_image(data_uri: str): + try: + # Check if the data URI starts with 'data:image' and contains an image format (e.g., jpeg, png, gif) + if not re.match(r'data:image/(\w+);base64,', data_uri): + raise ValueError("Invalid data URI image.") + # Extract the image format from the data URI + image_format = re.match(r'data:image/(\w+);base64,', data_uri).group(1) + # Check if the image format is one of the allowed formats (jpg, jpeg, png, gif) + if image_format.lower() not in ['jpeg', 'jpg', 'png', 'gif']: + raise ValueError("Invalid image format (from mime file type).") + except Exception as e: + raise e + +def is_accepted_format(binary_data: bytes) -> bool: + try: + check = False + if binary_data.startswith(b'\xFF\xD8\xFF'): + check = True # It's a JPEG image + elif binary_data.startswith(b'\x89PNG\r\n\x1a\n'): + check = True # It's a PNG image + elif binary_data.startswith(b'GIF87a') or binary_data.startswith(b'GIF89a'): + check = True # It's a GIF image + elif binary_data.startswith(b'\x89JFIF') or binary_data.startswith(b'JFIF\x00'): + check = True # It's a JPEG image + elif binary_data.startswith(b'\xFF\xD8'): + check = True # It's a JPEG image + elif binary_data.startswith(b'RIFF') and binary_data[8:12] == b'WEBP': + check = True # It's a WebP image + # else we raise ValueError + if not check: + raise ValueError("Invalid image format (from magic code).") + except Exception as e: + raise e + +def extract_data_uri(data_uri: str) -> bytes: + try: + data = data_uri.split(",")[1] + data = base64.b64decode(data) + return data + except Exception as e: + raise e + +def get_orientation(data: bytes) -> int: + try: + if data[:2] != b'\xFF\xD8': + raise Exception('NotJpeg') + with Image.open(data) as img: + exif_data = img._getexif() + if exif_data is not None: + orientation = exif_data.get(274) # 274 corresponds to the orientation tag in EXIF + if orientation is not None: + return orientation + except Exception: + pass + +def process_image(orientation: int, img: Image.Image, new_width: int, new_height: int) -> Image.Image: + try: + # Initialize the canvas + new_img = Image.new("RGB", (new_width, new_height), color="#FFFFFF") + if orientation: + if orientation > 4: + img = img.transpose(Image.FLIP_LEFT_RIGHT) + if orientation in [3, 4]: + img = img.transpose(Image.ROTATE_180) + if orientation in [5, 6]: + img = img.transpose(Image.ROTATE_270) + if orientation in [7, 8]: + img = img.transpose(Image.ROTATE_90) + new_img.paste(img, (0, 0)) + return new_img + except Exception as e: + raise e + +def compress_image_to_base64(img, compression_rate) -> str: + try: + output_buffer = io.BytesIO() + img.save(output_buffer, format="JPEG", quality=int(compression_rate * 100)) + return base64.b64encode(output_buffer.getvalue()).decode('utf-8') + except Exception as e: + raise e \ No newline at end of file diff --git a/g4f/Provider/create_images.py b/g4f/Provider/create_images.py new file mode 100644 index 00000000..852ba01a --- /dev/null +++ b/g4f/Provider/create_images.py @@ -0,0 +1,71 @@ +from __future__ import annotations + +import re +from ..typing import CreateResult, Messages +from ..base_provider import BaseProvider, ProviderType +from .. import debug + +system_message = """ +You can generate custom images with the DALL-E 3 image generator. +To generate a image with a prompt, do this: + +Don't use images with data uri. It is important to use a prompt instead. + +""" + +class CreateImagesProvider(BaseProvider): + def __init__( + self, + provider: ProviderType, + create_images: callable, + system_message: str = system_message + ) -> None: + self.provider = provider + self.create_images = create_images + self.system_message = system_message + self. __name__ = provider.__name__ + if hasattr(provider, "url"): + self.url = provider.url + self.working = provider.working + self.supports_stream = provider.supports_stream + + def create_completion( + self, + model: str, + messages: Messages, + stream: bool = False, + **kwargs + ) -> CreateResult: + messages.insert(0, {"role": "system", "content": self.system_message}) + image_placeholder = "" + for chunk in self.provider.create_completion(model, messages, stream, **kwargs): + if image_placeholder or "<" in chunk: + image_placeholder += chunk + if ">" in image_placeholder: + result = re.search(r' str: + messages.insert(0, {"role": "system", "content": self.system_message}) + response = await self.provider.create_async(model, messages, **kwargs) + result = re.search(r'', response) + if result: + search = result.group(0) + prompt = result.group(1) + images = "".join([*self.create_images(prompt)]) + return response.replace(search, images) + return response \ No newline at end of file diff --git a/g4f/__init__.py b/g4f/__init__.py index 699dc238..dc7808f9 100644 --- a/g4f/__init__.py +++ b/g4f/__init__.py @@ -68,6 +68,7 @@ class ChatCompletion: ignored : list[str] = None, ignore_working: bool = False, ignore_stream_and_auth: bool = False, + patch_provider: callable = None, **kwargs) -> Union[CreateResult, str]: model, provider = get_model_and_provider(model, provider, stream, ignored, ignore_working, ignore_stream_and_auth) @@ -83,6 +84,9 @@ class ChatCompletion: if proxy: kwargs['proxy'] = proxy + if patch_provider: + provider = patch_provider(provider) + result = provider.create_completion(model, messages, stream, **kwargs) return result if stream else ''.join(result) @@ -92,6 +96,7 @@ class ChatCompletion: provider : Union[ProviderType, str, None] = None, stream : bool = False, ignored : list[str] = None, + patch_provider: callable = None, **kwargs) -> Union[AsyncResult, str]: model, provider = get_model_and_provider(model, provider, False, ignored) @@ -101,6 +106,9 @@ class ChatCompletion: return provider.create_async_generator(model, messages, **kwargs) raise StreamNotSupportedError(f'{provider.__name__} does not support "stream" argument in "create_async"') + if patch_provider: + provider = patch_provider(provider) + return provider.create_async(model, messages, **kwargs) class Completion: diff --git a/g4f/gui/client/html/index.html b/g4f/gui/client/html/index.html index da7aeefb..b47f2a37 100644 --- a/g4f/gui/client/html/index.html +++ b/g4f/gui/client/html/index.html @@ -124,6 +124,11 @@ Web Access +
+ + + Image Generator +