Fix build docker image

docker/Dockerfile

@@ -33,7 +33,7 @@ COPY docker/supervisor-api.conf /etc/supervisor/conf.d/api.conf
 COPY docker/background.png /usr/share/images/fluxbox/ubuntu-light.png
 
 # Add user, fix permissions
-RUN chown "${SEL_UID}:${SEL_GID}" $HOME/.local
+RUN chown "${SEL_UID}:${SEL_GID}" $HOME/.local /opt/venv/share
 
 # Switch user
 USER $SEL_UID

g4f/Provider/needs_auth/Grok.py

@@ -3,14 +3,16 @@ from __future__ import annotations
 import os
 import json
 import time
+import nodriver
+import asyncio
 from typing import Dict, Any, AsyncIterator
 
-from ...typing import Messages, Cookies, AsyncResult
+from ...typing import Messages, AsyncResult
 from ...providers.response import JsonConversation, Reasoning, ImagePreview, ImageResponse, TitleGeneration, AuthResult, RequestLogin
-from ...requests import StreamSession, get_args_from_nodriver, DEFAULT_HEADERS
+from ...requests import StreamSession, get_nodriver, DEFAULT_HEADERS
 from ...requests.raise_for_status import raise_for_status
 from ..base_provider import AsyncAuthedProvider, ProviderModelMixin
-from ..helper import format_prompt, get_cookies, get_last_user_message
+from ..helper import format_prompt, get_last_user_message
 
 class Conversation(JsonConversation):
     def __init__(self,
@@ -33,30 +35,40 @@ class Grok(AsyncAuthedProvider, ProviderModelMixin):
     model_aliases = {"grok-3-r1": "grok-3-thinking"}
 
     @classmethod
-    async def on_auth_async(cls, cookies: Cookies = None, proxy: str = None, **kwargs) -> AsyncIterator:
+    async def on_auth_async(cls, proxy: str = None, **kwargs) -> AsyncIterator:
-        if cookies is None:
+        auth_result = AuthResult(headers=DEFAULT_HEADERS, impersonate="chrome")
-            cookies = get_cookies(cls.cookie_domain, False, True, False)
+        auth_result.headers["referer"] = cls.url + "/"
-        if cookies is not None and "sso" in cookies:
+        browser, stop_browser = await get_nodriver(proxy=proxy)
-            yield AuthResult(
-                cookies=cookies,
-                impersonate="chrome",
-                proxy=proxy,
-                headers=DEFAULT_HEADERS
-            )
-            return
         yield RequestLogin(cls.__name__, os.environ.get("G4F_LOGIN_URL") or "")
-        yield AuthResult(
+        try:
-            **await get_args_from_nodriver(
+            page = browser.main_tab
-                cls.url,
+            has_headers = False
-                proxy=proxy,
+            def on_request(event: nodriver.cdp.network.RequestWillBeSent, page=None):
-                wait_for='[href="/chat#private"]'
+                nonlocal has_headers
-            )
+                if event.request.url.startswith(cls.conversation_url + "/new"):
-        )
+                    for key, value in event.request.headers.items():
+                        auth_result.headers[key.lower()] = value
+                    has_headers = True
+            await page.send(nodriver.cdp.network.enable())
+            page.add_handler(nodriver.cdp.network.RequestWillBeSent, on_request)
+            page = await browser.get(cls.url)
+            auth_result.headers["user-agent"] = await page.evaluate("window.navigator.userAgent", return_by_value=True)
+            while True:
+                if has_headers:
+                    break
+                await asyncio.sleep(1)
+            auth_result.cookies = {}
+            for c in await page.send(nodriver.cdp.network.get_cookies([cls.url])):
+                auth_result.cookies[c.name] = c.value
+            await page.close()
+        finally:
+            stop_browser()
+        yield auth_result
 
     @classmethod
     async def _prepare_payload(cls, model: str, message: str) -> Dict[str, Any]:
         return {
-            "temporary": False,
+            "temporary": True,
             "modelName": "grok-latest" if model == "grok-2" else "grok-3",
             "message": message,
             "fileAttachments": [],
@@ -83,7 +95,6 @@ class Grok(AsyncAuthedProvider, ProviderModelMixin):
         model: str,
         messages: Messages,
         auth_result: AuthResult,
-        return_conversation: bool = True,
         conversation: Conversation = None,
         **kwargs
     ) -> AsyncResult:
@@ -99,7 +110,6 @@ class Grok(AsyncAuthedProvider, ProviderModelMixin):
                 url = f"{cls.conversation_url}/{conversation_id}/responses"
             async with session.post(url, json=payload) as response:
                 await raise_for_status(response)
-
                 thinking_duration = None
                 async for line in response.iter_lines():
                     if line:
@@ -133,9 +143,7 @@ class Grok(AsyncAuthedProvider, ProviderModelMixin):
                             title = result.get("title", {}).get("newTitle", "")
                             if title:
                                 yield TitleGeneration(title)
-                            
-
                         except json.JSONDecodeError:
                             continue
-                if return_conversation and conversation_id is not None:
+                # if conversation_id is not None:
-                    yield Conversation(conversation_id)
+                #     yield Conversation(conversation_id)

g4f/client/__init__.py

@@ -41,7 +41,7 @@ except NameError:
 def add_chunk(content, chunk):
     if content == "" and isinstance(chunk, (MediaResponse, AudioResponse)):
         content = chunk
-    else:
+    elif not isinstance(chunk, Reasoning):
         content = str(content) + str(chunk)
     return content
 
@@ -65,6 +65,7 @@ def iter_response(
     stop: Optional[list[str]] = None
 ) -> ChatCompletionResponseType:
     content = ""
+    reasoning_content = []
     finish_reason = None
     tool_calls = None
     usage = None
@@ -92,9 +93,8 @@ def iter_response(
         elif isinstance(chunk, ProviderInfo):
             provider = chunk
             continue
-        elif isinstance(chunk, BaseConversation):
+        elif isinstance(chunk, Reasoning):
-            yield chunk
+            reasoning_content.append(chunk)
-            continue
         elif isinstance(chunk, HiddenResponse):
             continue
         elif isinstance(chunk, Exception):
@@ -141,7 +141,8 @@ def iter_response(
         chat_completion = ChatCompletion.model_construct(
             content, finish_reason, completion_id, int(time.time()), usage=usage,
             **filter_none(tool_calls=[ToolCallModel.model_construct(**tool_call) for tool_call in tool_calls]) if tool_calls is not None else {},
-            conversation=None if conversation is None else conversation.get_dict()
+            conversation=None if conversation is None else conversation.get_dict(),
+            reasoning_content=reasoning_content if reasoning_content else None
         )
     if provider is not None:
         chat_completion.provider = provider.name
@@ -168,6 +169,7 @@ async def async_iter_response(
     stop: Optional[list[str]] = None
 ) -> AsyncChatCompletionResponseType:
     content = ""
+    reasoning_content = []
     finish_reason = None
     completion_id = ''.join(random.choices(string.ascii_letters + string.digits, k=28))
     idx = 0
@@ -193,6 +195,8 @@ async def async_iter_response(
             elif isinstance(chunk, ProviderInfo):
                 provider = chunk
                 continue
+            elif isinstance(chunk, Reasoning) and not stream:
+                reasoning_content.append(chunk)
             elif isinstance(chunk, HiddenResponse):
                 continue
             elif isinstance(chunk, Exception):
@@ -241,7 +245,8 @@ async def async_iter_response(
                 **filter_none(
                     tool_calls=[ToolCallModel.model_construct(**tool_call) for tool_call in tool_calls]
                 ) if tool_calls is not None else {},
-                conversation=conversation
+                conversation=conversation,
+                reasoning_content=reasoning_content if reasoning_content else None
             )
         if provider is not None:
             chat_completion.provider = provider.name

g4f/client/stubs.py

@@ -7,6 +7,7 @@ from time import time
 from ..image import extract_data_uri
 from ..image.copy_images import get_media_dir
 from ..client.helper import filter_markdown
+from ..providers.response import Reasoning
 from .helper import filter_none
 
 try:
@@ -132,16 +133,21 @@ class ResponseMessageContent(BaseModel):
 class ChatCompletionMessage(BaseModel):
     role: str
     content: str
+    reasoning_content: list[Reasoning] = None
     tool_calls: list[ToolCallModel] = None
 
     @classmethod
-    def model_construct(cls, content: str, tool_calls: list = None):
+    def model_construct(cls, content: str, reasoning_content: list[Reasoning] = None, tool_calls: list = None):
-        return super().model_construct(role="assistant", content=content, **filter_none(tool_calls=tool_calls))
+        return super().model_construct(role="assistant", content=content, **filter_none(tool_calls=tool_calls, reasoning_content=reasoning_content))
 
     @field_serializer('content')
     def serialize_content(self, content: str):
         return str(content)
 
+    @field_serializer('reasoning_content')
+    def serialize_reasoning_content(self, reasoning_content: list):
+        return "".join([str(content) for content in reasoning_content]) if reasoning_content else None
+
     def save(self, filepath: str, allowed_types = None):
         if hasattr(self.content, "data"):
             os.rename(self.content.data.replace("/media", get_media_dir()), filepath)
@@ -184,7 +190,8 @@ class ChatCompletion(BaseModel):
         created: int = None,
         tool_calls: list[ToolCallModel] = None,
         usage: UsageModel = None,
-        conversation: dict = None
+        conversation: dict = None,
+        reasoning_content: list[Reasoning] = None
     ):
         return super().model_construct(
             id=f"chatcmpl-{completion_id}" if completion_id else None,
@@ -193,7 +200,7 @@ class ChatCompletion(BaseModel):
             model=None,
             provider=None,
             choices=[ChatCompletionChoice.model_construct(
-                ChatCompletionMessage.model_construct(content, tool_calls),
+                ChatCompletionMessage.model_construct(content, reasoning_content, tool_calls),
                 finish_reason,
             )],
             **filter_none(usage=usage, conversation=conversation)
@@ -245,16 +252,20 @@ class ClientResponse(BaseModel):
 class ChatCompletionDelta(BaseModel):
     role: str
     content: Optional[str]
+    reasoning_content: Optional[str] = None
 
     @classmethod
     def model_construct(cls, content: Optional[str]):
+        if isinstance(content, Reasoning):
+            return super().model_construct(role="reasoning", content=content, reasoning_content=str(content))
         return super().model_construct(role="assistant", content=content)
 
     @field_serializer('content')
     def serialize_content(self, content: Optional[str]):
         if content is None:
             return ""
-
+        if isinstance(content, Reasoning):
+            return None
         return str(content)
 
 class ChatCompletionDeltaChoice(BaseModel):

g4f/tools/files.py

@@ -89,11 +89,14 @@ def secure_filename(filename: str) -> str:
     # Keep letters, numbers, basic punctuation and all Unicode chars
     filename = re.sub(
         r'[^\w.,_+-]+',
-        '_', 
+        '_',
-        unquote(filename).strip(), 
+        unquote(filename).strip(),
         flags=re.UNICODE
     )
-    filename = filename[:100].strip(".,_-+")
+    max_length = 100
+    if len(filename.encode()) > len(filename):
+        max_length = 50
+    filename = filename[:max_length].strip(".,_+-")
     return filename
 
 def supports_filename(filename: str):

new.py

@@ -0,0 +1,215 @@
+import json
+import uuid
+import g4f.debug
+import requests
+from g4f.client import Client
+
+def upload_and_process(files_or_urls, bucket_id=None):
+    if bucket_id is None:
+        bucket_id = str(uuid.uuid4())
+
+    if isinstance(files_or_urls, list):  # URLs
+        files = {'files': ('downloads.json', json.dumps(
+            files_or_urls), 'application/json')}
+    elif isinstance(files_or_urls, dict):  # Files
+        files = files_or_urls
+    else:
+        raise ValueError(
+            "files_or_urls must be a list of URLs or a dictionary of files")
+
+    upload_response = requests.post(
+        f'http://localhost:8080/backend-api/v2/files/{bucket_id}', files=files)
+
+    if upload_response.status_code == 200:
+        upload_data = upload_response.json()
+        print(f"Upload successful. Bucket ID: {upload_data['bucket_id']}")
+    else:
+        print(
+            f"Upload failed: {upload_response.status_code} - {upload_response.text}")
+
+    response = requests.get(
+        f'http://localhost:8080/backend-api/v2/files/{bucket_id}', stream=True, headers={'Accept': 'text/event-stream'})
+    for line in response.iter_lines():
+        if line:
+            line = line.decode('utf-8')
+            if line.startswith('data:'):
+                try:
+                    data = json.loads(line[5:])  # remove data: prefix
+                    if "action" in data:
+                        print(f"SSE Event: {data}")
+                    elif "error" in data:
+                        print(f"Error: {data['error']['message']}")
+                    else:
+                        # Assuming it's file content
+                        print(f"File data received: {data}")
+                except json.JSONDecodeError as e:
+                    print(f"Error decoding JSON: {e}")
+            else:
+                print(f"Unhandled SSE event: {line}")
+    response.close()
+    return bucket_id
+
+
+# Example with URLs
+
+# Enable debug mode
+g4f.debug.logging = True
+
+client = Client()
+
+# Upload example file
+files = {'files': ('demo.docx', open('demo.docx', 'rb'))}
+bucket_id = upload_and_process(files)
+
+# Send request with file:
+response = client.chat.completions.create(
+    model='gpt-4o',
+    messages=[{"role": "user", "content": [
+        {"type": "text", "text": "Discribe this file."},
+        {"bucket_id": bucket_id}
+    ]}],
+)
+print(response.choices[0].message.content)
+exit()
+# import asyncio
+# from g4f.client import AsyncClient
+# import g4f.Provider
+
+# async def main():
+#     client = AsyncClient(provider=g4f.Provider.MarkItDown)
+
+#     # Transcribe a audio file
+#     with open("audio.wav", "rb") as audio_file:
+#         response = await client.chat.completions.create("", media=[audio_file])
+#         print(response.choices[0].message.content)
+
+# if __name__ == "__main__":
+#     asyncio.run(main())
+
+#exit()
+import requests
+
+# Open the audio file in binary mode
+with open('demo.docx', 'rb') as audio_file:
+    # Make the POST request
+    response = requests.post('http://localhost:8080/api/markitdown', files={'file': audio_file})
+
+    # Check the response and print the transcription
+    if response.status_code == 200:
+        data = response.json()
+        print(data['text'])
+    else:
+        print(f"Error: {response.status_code}, {response.text}")
+exit()
+
+# from openai import OpenAI
+# client = OpenAI(base_url="http://localhost:8080/v1", api_key="secret")
+
+# with open("audio.wav", "rb") as file:
+#     transcript = client.audio.transcriptions.create(
+#         model="",
+#         extra_body={"provider": "MarkItDown"},
+#         file=file
+#     )
+# print(transcript.text)
+
+exit()
+import asyncio
+import time
+from g4f import AsyncClient
+from g4f.Provider import PollinationsAI
+async def test():
+    client = AsyncClient()
+    response = client.chat.completions.create("guten tag", stream=True, provider=HarProvider)
+    async for chunk in response:
+        if chunk.choices[0].finish_reason == "stop":
+            break
+        print(chunk.choices[0].delta.content, end="", flush=True)
+    print()
+
+asyncio.run(test())
+time.sleep(1)
+exit()
+
+client = Client(provider=PollinationsAI)
+response = client.media.generate("Hello", model="hypnosis-tracy")
+response.data[0].save("hypnosis.mp3")
+
+client = Client(provider=Gemini)
+response = client.media.generate("Hello", model="gemini-audio")
+response.data[0].save("gemini.ogx")
+
+client = Client(provider=EdgeTTS)
+response = client.media.generate("Hello", audio={"locale": "en-US"})
+response.data[0].save("edge-tts.mp3")
+
+exit()
+import requests
+import uuid
+import json
+
+def upload_and_process(files_or_urls, bucket_id=None):
+    if bucket_id is None:
+        bucket_id = str(uuid.uuid4())
+    
+    if isinstance(files_or_urls, list): #URLs
+        files = {'files': ('downloads.json', json.dumps(files_or_urls), 'application/json')}
+    elif isinstance(files_or_urls, dict): #Files
+        files = files_or_urls
+    else:
+        raise ValueError("files_or_urls must be a list of URLs or a dictionary of files")
+
+    upload_response = requests.post(f'http://localhost:8080/v1/files/{bucket_id}', files=files)
+
+    if upload_response.status_code == 200:
+        upload_data = upload_response.json()
+        print(f"Upload successful. Bucket ID: {upload_data['bucket_id']}")
+    else:
+        print(f"Upload failed: {upload_response.status_code} - {upload_response.text}")
+
+    response = requests.get(f'http://localhost:8080/v1/files/{bucket_id}', stream=True, headers={'Accept': 'text/event-stream'})
+    for line in response.iter_lines():
+      if line:
+          line = line.decode('utf-8')
+          if line.startswith('data:'):
+              try:
+                  data = json.loads(line[5:]) #remove data: prefix
+                  if "action" in data:
+                      print(f"SSE Event: {data}")
+                  elif "error" in data:
+                      print(f"Error: {data['error']['message']}")
+                  else:
+                      print(f"File data received: {data}") #Assuming it's file content
+              except json.JSONDecodeError as e:
+                  print(f"Error decoding JSON: {e}")
+          else:
+              print(f"Unhandled SSE event: {line}")
+    response.close()
+    return bucket_id
+
+# Example with URLs
+#Example with files
+#files = {'files': open('document.pdf', 'rb'), 'files': open('data.json', 'rb')}
+#bucket_id = upload_and_process(files)
+import asyncio
+from g4f.client import Client
+
+import g4f.debug
+g4f.debug.logging = True
+
+client = Client()
+
+
+files = {'files': ('demo.docx', open('demo.docx', 'rb'))}
+bucket_id = upload_and_process(files)
+
+
+response = client.chat.completions.create(
+    [{"role": "user", "content": [
+        {"type": "text", "text": "Discribe this file."},
+        {"bucket_id": bucket_id}
+    ]}],
+    "o1",
+)
+print(response.choices[0].message.content)
+