webscout 8.3.2__py3-none-any.whl → 8.3.4__py3-none-any.whl

webscout/Provider/scira_chat.py

@@ -17,19 +17,74 @@ class SciraAI(Provider):
     A class to interact with the Scira AI chat API.
     """
 
-    AVAILABLE_MODELS = {
-        "scira-default": "Grok3-mini", # thinking model
-        "scira-grok-3": "Grok3",
-        "scira-anthropic": "Claude 4 Sonnet",
-        "scira-anthropic-thinking": "Claude 4 Sonnet Thinking", # thinking model
-        "scira-vision" : "Grok2-Vision", # vision model
-        "scira-4o": "GPT4o",
-        "scira-qwq": "QWQ-32B",
-        "scira-o4-mini": "o4-mini",
-        "scira-google": "gemini 2.5 flash Thinking", # thinking model
-        "scira-google-pro": "gemini 2.5 pro",
-        "scira-llama-4": "llama 4 Maverick",
+    # Model mapping: actual model names to Scira API format
+    MODEL_MAPPING = {
+        "grok-3-mini": "scira-default",
+        "grok-3-mini-fast": "scira-x-fast-mini",
+        "grok-3-fast": "scira-x-fast",
+        "gpt-4.1-nano": "scira-nano",
+        "grok-3": "scira-grok-3",
+        "grok-4": "scira-grok-4",
+        "grok-2-vision-1212": "scira-vision",
+        "grok-2-latest": "scira-g2",
+        "gpt-4o-mini": "scira-4o-mini",
+        "o4-mini-2025-04-16": "scira-o4-mini",
+        "o3": "scira-o3",
+        "qwen/qwen3-32b": "scira-qwen-32b",
+        "qwen3-30b-a3b": "scira-qwen-30b",
+        "deepseek-v3-0324": "scira-deepseek-v3",
+        "claude-3-5-haiku-20241022": "scira-haiku",
+        "mistral-small-latest": "scira-mistral",
+        "gemini-2.5-flash-lite-preview-06-17": "scira-google-lite",
+        "gemini-2.5-flash": "scira-google",
+        "gemini-2.5-pro": "scira-google-pro",
+        "claude-sonnet-4-20250514": "scira-anthropic",
+        "claude-sonnet-4-20250514-thinking": "scira-anthropic-thinking",
+        "claude-4-opus-20250514": "scira-opus",
+        "claude-4-opus-20250514-pro": "scira-opus-pro",
+        "meta-llama/llama-4-maverick-17b-128e-instruct": "scira-llama-4",
     }
+    
+    # Reverse mapping: Scira format to actual model names
+    SCIRA_TO_MODEL = {v: k for k, v in MODEL_MAPPING.items()}
+    # Add special cases for aliases and duplicate mappings
+    SCIRA_TO_MODEL["scira-anthropic-thinking"] = "claude-sonnet-4-20250514"
+    SCIRA_TO_MODEL["scira-opus-pro"] = "claude-4-opus-20250514"
+    SCIRA_TO_MODEL["scira-x-fast"] = "grok-3-fast"
+    SCIRA_TO_MODEL["scira-x-fast-mini"] = "grok-3-mini-fast"
+    SCIRA_TO_MODEL["scira-nano"] = "gpt-4.1-nano"
+    SCIRA_TO_MODEL["scira-qwen-32b"] = "qwen/qwen3-32b"
+    SCIRA_TO_MODEL["scira-qwen-30b"] = "qwen3-30b-a3b"
+    SCIRA_TO_MODEL["scira-deepseek-v3"] = "deepseek-v3-0324"
+    SCIRA_TO_MODEL["scira-grok-4"] = "grok-4"
+    MODEL_MAPPING["claude-4-opus-20250514-pro"] = "scira-opus-pro"
+    # Available models list (actual model names + scira aliases)
+    AVAILABLE_MODELS = list(MODEL_MAPPING.keys()) + list(SCIRA_TO_MODEL.keys())
+    
+    @classmethod
+    def _resolve_model(cls, model: str) -> str:
+        """
+        Resolve a model name to its Scira API format.
+        
+        Args:
+            model: Either an actual model name or a Scira alias
+            
+        Returns:
+            The Scira API format model name
+            
+        Raises:
+            ValueError: If the model is not supported
+        """
+        # If it's already a Scira format, return as-is
+        if model in cls.SCIRA_TO_MODEL:
+            return model
+            
+        # If it's an actual model name, convert to Scira format
+        if model in cls.MODEL_MAPPING:
+            return cls.MODEL_MAPPING[model]
+            
+        # Model not found
+        raise ValueError(f"Invalid model: {model}. Choose from: {cls.AVAILABLE_MODELS}")
 
     def __init__(
         self,
@@ -42,7 +97,7 @@ class SciraAI(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        model: str = "scira-default",
+        model: str = "grok-3-mini",
         chat_id: str = None,
         user_id: str = None,
         browser: str = "chrome",
@@ -67,9 +122,9 @@ class SciraAI(Provider):
             system_prompt (str): System prompt for the AI.
 
         """
-        if model not in self.AVAILABLE_MODELS:
-            raise ValueError(f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}")
-
+        # Resolve the model to Scira format
+        self.model = self._resolve_model(model)
+        
         self.url = "https://scira.ai/api/search"
 
         # Initialize LitAgent for user agent generation
@@ -103,7 +158,6 @@ class SciraAI(Provider):
         self.max_tokens_to_sample = max_tokens
         self.timeout = timeout
         self.last_response = {}
-        self.model = model
         self.chat_id = chat_id or str(uuid.uuid4())
         self.user_id = user_id or f"user_{str(uuid.uuid4())[:8].upper()}"
 
@@ -154,14 +208,20 @@ class SciraAI(Provider):
         return self.fingerprint
 
     @staticmethod
-    def _scira_extractor(chunk: Union[str, Dict[str, Any]]) -> Optional[str]:
-        """Extracts content from the Scira stream format '0:"..."'."""
+    def _scira_extractor(chunk: Union[str, Dict[str, Any]]) -> Optional[dict]:
+        """Extracts g and 0 chunks from the Scira stream format.
+        Returns a dict: {"g": [g1, g2, ...], "0": zero} if present.
+        """
         if isinstance(chunk, str):
-            match = re.search(r'0:"(.*?)"(?=,|$)', chunk) # Look for 0:"...", possibly followed by comma or end of string
-            if match:
-                # Decode potential unicode escapes like \u00e9 and handle escaped quotes/backslashes
-                content = match.group(1).encode().decode('unicode_escape')
-                return content.replace('\\\\', '\\').replace('\\"', '"')
+            g_matches = re.findall(r'g:"(.*?)"', chunk)
+            zero_match = re.search(r'0:"(.*?)"(?=,|$)', chunk)
+            result = {}
+            if g_matches:
+                result["g"] = [g.encode().decode('unicode_escape').replace('\\', '\\').replace('\\"', '"') for g in g_matches]
+            if zero_match:
+                result["0"] = zero_match.group(1).encode().decode('unicode_escape').replace('\\', '\\').replace('\\"', '"')
+            if result:
+                return result
         return None
 
     def ask(
@@ -169,7 +229,9 @@ class SciraAI(Provider):
         prompt: str,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> Dict[str, Any]: # Note: Stream parameter removed as API doesn't seem to support it
+        stream: bool = True,  # Default to True, always stream
+        raw: bool = False,    # Added raw parameter
+    ) -> Union[Dict[str, Any], Any]:
         conversation_prompt = self.conversation.gen_complete_prompt(prompt)
         if optimizer:
             if optimizer in self.__available_optimizers:
@@ -194,107 +256,198 @@ class SciraAI(Provider):
             "timezone": "Asia/Calcutta"
         }
 
-        try:
-            # Use curl_cffi post with impersonate
-            response = self.session.post(
-                self.url,
-                json=payload,
-                timeout=self.timeout,
-                impersonate="chrome120" # Add impersonate
-            )
-            if response.status_code != 200:
-                # Try to get response content for better error messages
-                try: # Use try-except for reading response content
-                    error_content = response.text
-                except:
-                    error_content = "<could not read response content>"
-
-                if response.status_code in [403, 429]:
-                    print(f"Received status code {response.status_code}, refreshing identity...")
-                    self.refresh_identity()
-                    response = self.session.post(
-                        self.url, json=payload, timeout=self.timeout,
-                        impersonate="chrome120" # Add impersonate to retry
-                    )
-                    if not response.ok:
+        def for_stream():
+            try:
+                response = self.session.post(
+                    self.url,
+                    json=payload,
+                    timeout=self.timeout,
+                    impersonate="chrome120",
+                    stream=True
+                )
+                if response.status_code != 200:
+                    try:
+                        error_content = response.text
+                    except:
+                        error_content = "<could not read response content>"
+
+                    if response.status_code in [403, 429]:
+                        print(f"Received status code {response.status_code}, refreshing identity...")
+                        self.refresh_identity()
+                        response = self.session.post(
+                            self.url, json=payload, timeout=self.timeout,
+                            impersonate="chrome120", stream=True
+                        )
+                        if not response.ok:
+                            raise exceptions.FailedToGenerateResponseError(
+                                f"Failed to generate response after identity refresh - ({response.status_code}, {response.reason}) - {error_content}"
+                            )
+                        print("Identity refreshed successfully.")
+                    else:
                         raise exceptions.FailedToGenerateResponseError(
-                            f"Failed to generate response after identity refresh - ({response.status_code}, {response.reason}) - {error_content}"
+                            f"Request failed with status code {response.status_code}. Response: {error_content}"
                         )
-                    print("Identity refreshed successfully.")
-                else:
-                    raise exceptions.FailedToGenerateResponseError(
-                        f"Request failed with status code {response.status_code}. Response: {error_content}"
-                    )
-
-            response_text_raw = response.text # Get raw response text
-
-            # Process the text using sanitize_stream line by line
-            processed_stream = sanitize_stream(
-                data=response_text_raw.splitlines(), # Split into lines
-                intro_value=None, # No simple prefix
-                to_json=False,    # Content is not JSON
-                content_extractor=self._scira_extractor # Use the specific extractor
-            )
 
-            # Aggregate the results from the generator
+                processed_stream = sanitize_stream(
+                    data=response.iter_content(chunk_size=None),
+                    intro_value=None,
+                    to_json=False,
+                    content_extractor=self._scira_extractor,
+                    raw=raw
+                )
+
+                streaming_response = ""
+                in_think = False
+                for content in processed_stream:
+                    if content is None:
+                        continue
+                    if isinstance(content, dict):
+                        # Handle g chunks
+                        g_chunks = content.get("g", [])
+                        zero_chunk = content.get("0")
+                        if g_chunks:
+                            if not in_think:
+                                if raw:
+                                    yield "<think>\n\n"
+                                else:
+                                    yield "<think>\n\n"
+                                in_think = True
+                            for g in g_chunks:
+                                if raw:
+                                    yield g
+                                else:
+                                    yield dict(text=g)
+                        if zero_chunk is not None:
+                            if in_think:
+                                if raw:
+                                    yield "</think>\n\n"
+                                else:
+                                    yield "</think>\n\n"
+                                in_think = False
+                            if raw:
+                                yield zero_chunk
+                            else:
+                                streaming_response += zero_chunk
+                                yield dict(text=zero_chunk)
+                    else:
+                        # fallback for old string/list logic
+                        if raw:
+                            yield content
+                        else:
+                            if content and isinstance(content, str):
+                                streaming_response += content
+                                yield dict(text=content)
+                if not raw:
+                    self.last_response = {"text": streaming_response}
+                    self.conversation.update_chat_history(prompt, streaming_response)
+            except CurlError as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}") from e
+            except Exception as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed: {e}")
+
+        def for_non_stream():
+            # Always use streaming logic, but aggregate the result
             full_response = ""
-            for content in processed_stream:
-                if content and isinstance(content, str):
-                    full_response += content
-
-            self.last_response = {"text": full_response}
-            self.conversation.update_chat_history(prompt, full_response)
-            return {"text": full_response}
-        except CurlError as e: # Catch CurlError
-            raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}") from e
-        except Exception as e:
-            raise exceptions.FailedToGenerateResponseError(f"Request failed: {e}")
+            for chunk in for_stream():
+                if raw:
+                    if isinstance(chunk, str):
+                        full_response += chunk
+                else:
+                    if isinstance(chunk, dict) and "text" in chunk:
+                        full_response += chunk["text"]
+            if not raw:
+                self.last_response = {"text": full_response}
+                self.conversation.update_chat_history(prompt, full_response)
+                return {"text": full_response}
+            else:
+                return full_response
+
+        return for_stream() if stream else for_non_stream()
 
     def chat(
         self,
         prompt: str,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> str:
-        return self.get_message(
-            self.ask(
-                prompt, optimizer=optimizer, conversationally=conversationally
+        stream: bool = True,  # Default to True, always stream
+        raw: bool = False,    # Added raw parameter
+    ) -> Any:
+        def for_stream():
+            for response in self.ask(
+                prompt, optimizer=optimizer, conversationally=conversationally, stream=True, raw=raw
+            ):
+                if raw:
+                    yield response
+                else:
+                    if isinstance(response, dict):
+                        yield self.get_message(response)
+                    else:
+                        # For <think> and </think> tags (strings), yield as is
+                        yield response
+        def for_non_stream():
+            result = self.ask(
+                prompt, optimizer=optimizer, conversationally=conversationally, stream=False, raw=raw
             )
-        )
+            if raw:
+                return result
+            else:
+                if isinstance(result, dict):
+                    return self.get_message(result)
+                else:
+                    return result
+        return for_stream() if stream else for_non_stream()
 
     def get_message(self, response: dict) -> str:
+        """
+        Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        # Extractor handles formatting
-        return response.get("text", "").replace('\\n', '\n').replace('\\n\\n', '\n\n')
+        return response.get("text", "")
 
 if __name__ == "__main__":
-    print("-" * 100)
+    print("-" * 80)
     print(f"{'Model':<50} {'Status':<10} {'Response'}")
-    print("-" * 100)
-
-    test_prompt = "Say 'Hello' in one word"
-
-    # Test each model
+    print("-" * 80)
+    
+    # Test all available models
+    working = 0
+    total = len(SciraAI.AVAILABLE_MODELS)
+    
     for model in SciraAI.AVAILABLE_MODELS:
-        print(f"\rTesting {model}...", end="")
-
         try:
-            test_ai = SciraAI(model=model, timeout=120)  # Increased timeout
-            response = test_ai.chat(test_prompt)
-
-            if response and len(response.strip()) > 0:
+            test_ai = SciraAI(model=model, timeout=60)
+            # Test stream first
+            response_stream = test_ai.chat("Say 'Hello' in one word", stream=True)
+            response_text = ""
+            print(f"\r{model:<50} {'Streaming...':<10}", end="", flush=True)
+            for chunk in response_stream:
+                response_text += chunk
+                # Optional: print chunks as they arrive for visual feedback
+                # print(chunk, end="", flush=True) 
+            
+            if response_text and len(response_text.strip()) > 0:
                 status = "✓"
                 # Clean and truncate response
-                clean_text = response.strip().encode('utf-8', errors='ignore').decode('utf-8')
+                clean_text = response_text.strip() # Already decoded in get_message
                 display_text = clean_text[:50] + "..." if len(clean_text) > 50 else clean_text
             else:
-                status = "✗"
-                display_text = "Empty or invalid response"
-
+                status = "✗ (Stream)"
+                display_text = "Empty or invalid stream response"
             print(f"\r{model:<50} {status:<10} {display_text}")
+            
+            # Optional: Add non-stream test if needed, but stream test covers basic functionality
+            # print(f"\r{model:<50} {'Non-Stream...':<10}", end="", flush=True)
+            # response_non_stream = test_ai.chat("Say 'Hi' again", stream=False)
+            # if not response_non_stream or len(response_non_stream.strip()) == 0:
+            #      print(f"\r{model:<50} {'✗ (Non-Stream)':<10} Empty non-stream response")
+
+
         except Exception as e:
-            error_msg = str(e)
-            # Truncate very long error messages
-            if len(error_msg) > 100:
-                error_msg = error_msg[:97] + "..."
-            print(f"\r{model:<50} {'✗':<10} Error: {error_msg}")
+            print(f"\r{model:<50} {'✗':<10} {str(e)}")
+

webscout/Provider/scnet.py

@@ -185,24 +185,24 @@ class SCNet(Provider):
         stream: bool = False,
         optimizer: Optional[str] = None,
         conversationally: bool = False,
+        raw: bool = False,  # Added raw parameter
     ) -> Union[str, Generator[str, None, None]]:
         def for_stream_chat():
-            # ask() yields dicts or strings when streaming
-            gen = self.ask(
-                prompt, stream=True, raw=False, # Ensure ask yields dicts
-                optimizer=optimizer, conversationally=conversationally
-            )
-            for response_dict in gen:
-                yield self.get_message(response_dict) # get_message expects dict
-
+            for response in self.ask(
+                prompt, stream=True, raw=raw, optimizer=optimizer, conversationally=conversationally
+            ):
+                if raw:
+                    yield response
+                else:
+                    yield self.get_message(response)
         def for_non_stream_chat():
-            # ask() returns dict or str when not streaming
             response_data = self.ask(
-                prompt, stream=False, raw=False, # Ensure ask returns dict
-                optimizer=optimizer, conversationally=conversationally
+                prompt, stream=False, raw=raw, optimizer=optimizer, conversationally=conversationally
             )
-            return self.get_message(response_data) # get_message expects dict
-
+            if raw:
+                return response_data if isinstance(response_data, str) else self.get_message(response_data)
+            else:
+                return self.get_message(response_data)
         return for_stream_chat() if stream else for_non_stream_chat()
 
     def get_message(self, response: dict) -> str:

webscout/Provider/searchchat.py

@@ -232,6 +232,7 @@ class SearchChatAI(Provider):
         stream: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
+        raw: bool = False,  # Added raw parameter
     ) -> Union[str, Generator[str, None, None]]:
         """
         Chat with the API.
@@ -246,22 +247,21 @@ class SearchChatAI(Provider):
             Either a string response or a generator for streaming
         """
         def for_stream_chat():
-            # ask() yields dicts or strings when streaming
-            gen = self.ask(
-                prompt, stream=True, raw=False, # Ensure ask yields dicts
-                optimizer=optimizer, conversationally=conversationally
-            )
-            for response_dict in gen:
-                yield self.get_message(response_dict) # get_message expects dict
-                
+            for response in self.ask(
+                prompt, stream=True, raw=raw, optimizer=optimizer, conversationally=conversationally
+            ):
+                if raw:
+                    yield response
+                else:
+                    yield self.get_message(response)
         def for_non_stream_chat():
-             # ask() returns dict or str when not streaming
             response_data = self.ask(
-                prompt, stream=False, raw=False, # Ensure ask returns dict
-                optimizer=optimizer, conversationally=conversationally
+                prompt, stream=False, raw=raw, optimizer=optimizer, conversationally=conversationally
             )
-            return self.get_message(response_data) # get_message expects dict
-            
+            if raw:
+                return response_data if isinstance(response_data, str) else self.get_message(response_data)
+            else:
+                return self.get_message(response_data)
         return for_stream_chat() if stream else for_non_stream_chat()
 
     def get_message(self, response: dict) -> str:

webscout/Provider/sonus.py

@@ -208,23 +208,24 @@ class SonusAI(Provider):
         optimizer: str = None,
         conversationally: bool = False,
         reasoning: bool = False,
+        raw: bool = False,  # Added raw parameter
     ) -> Union[str, Generator[str, None, None]]:
         def for_stream_chat():
-            # ask() yields dicts when raw=False
-            for response_dict in self.ask(
-                prompt, stream=True, raw=False, # Ensure ask yields dicts
-                optimizer=optimizer, conversationally=conversationally, reasoning=reasoning
+            for response in self.ask(
+                prompt, stream=True, raw=raw, optimizer=optimizer, conversationally=conversationally, reasoning=reasoning
             ):
-                yield self.get_message(response_dict)
-        
+                if raw:
+                    yield response
+                else:
+                    yield self.get_message(response)
         def for_non_stream_chat():
-             # ask() returns dict or str when raw=False/True
             response_data = self.ask(
-                prompt, stream=False, raw=False, # Ensure ask returns dict
-                optimizer=optimizer, conversationally=conversationally, reasoning=reasoning
+                prompt, stream=False, raw=raw, optimizer=optimizer, conversationally=conversationally, reasoning=reasoning
             )
-            return self.get_message(response_data) # get_message expects dict
-
+            if raw:
+                return response_data if isinstance(response_data, str) else self.get_message(response_data)
+            else:
+                return self.get_message(response_data)
         return for_stream_chat() if stream else for_non_stream_chat()
 
     def get_message(self, response: dict) -> str:

webscout/Provider/toolbaz.py

@@ -24,6 +24,7 @@ class Toolbaz(Provider):
     AVAILABLE_MODELS = [
         "gemini-2.5-flash",
         "gemini-2.0-flash-thinking",
+        "sonar",
         "gemini-2.0-flash",
         "gemini-1.5-flash",
         "o3-mini",
@@ -223,14 +224,22 @@ class Toolbaz(Provider):
                     intro_value=None, # No simple prefix
                     to_json=False,    # Content is text
                     content_extractor=self._toolbaz_extractor, # Use the tag remover
-                    yield_raw_on_error=True # Yield even if extractor somehow fails (though unlikely for regex)
+                    yield_raw_on_error=True, # Yield even if extractor somehow fails (though unlikely for regex)
+                    raw=raw
                 )
 
                 for content_chunk in processed_stream:
                     # content_chunk is the string with tags removed
-                    if content_chunk and isinstance(content_chunk, str):
-                        streaming_text += content_chunk
-                        yield {"text": content_chunk} if not raw else content_chunk
+                    if isinstance(content_chunk, bytes):
+                        content_chunk = content_chunk.decode('utf-8', errors='ignore')
+                    if content_chunk is None:
+                        continue
+                    if raw:
+                        yield content_chunk
+                    else:
+                        if content_chunk and isinstance(content_chunk, str):
+                            streaming_text += content_chunk
+                            yield {"text": content_chunk}
 
                 self.last_response = {"text": streaming_text}
                 self.conversation.update_chat_history(prompt, streaming_text)
@@ -274,28 +283,36 @@ class Toolbaz(Provider):
         stream: bool = False,
         optimizer: Optional[str] = None,
         conversationally: bool = False,
+        raw: bool = False,  # Added raw parameter
     ) -> Union[str, Generator[str, None, None]]:
         """Generates a response from the Toolbaz API."""
         def for_stream_chat():
             # ask() yields dicts when raw=False
-            for response_dict in self.ask(
+            for response in self.ask(
                 prompt,
                 stream=True,
-                raw=False, # Ensure ask yields dicts
+                raw=raw,
                 optimizer=optimizer,
                 conversationally=conversationally
             ):
-                yield self.get_message(response_dict)
+                if raw:
+                    yield response
+                else:
+                    yield self.get_message(response)
 
         def for_non_stream_chat():
             # ask() returns a dict when stream=False
             response_dict = self.ask(
                 prompt,
                 stream=False,
+                raw=raw,
                 optimizer=optimizer,
                 conversationally=conversationally,
             )
-            return self.get_message(response_dict)
+            if raw:
+                return response_dict
+            else:
+                return self.get_message(response_dict)
 
         return for_stream_chat() if stream else for_non_stream_chat()