botrun-flow-lang 5.12.263__py3-none-any.whl → 6.2.21__py3-none-any.whl

botrun_flow_lang/langgraph_agents/agents/util/youtube_util.py

@@ -1,90 +1,90 @@
-from urllib.parse import urlparse, parse_qs
-from dotenv import load_dotenv
-import re
-import os
-
-load_dotenv()
-
-
-def get_video_id(url: str) -> str:
-    """
-    Extract video ID from various YouTube URL formats.
-    Supports:
-    - Standard watch URLs (youtube.com/watch?v=...)
-    - Shortened URLs (youtu.be/...)
-    - Embed URLs (youtube.com/embed/...)
-    """
-    # Try parsing as standard URL first
-    parsed_url = urlparse(url)
-
-    # Handle youtu.be URLs
-    if parsed_url.netloc == "youtu.be":
-        return parsed_url.path.lstrip("/")
-
-    # Handle standard youtube.com URLs
-    if parsed_url.netloc in ["www.youtube.com", "youtube.com"]:
-        if parsed_url.path == "/watch":
-            # Standard watch URL
-            return parse_qs(parsed_url.query).get("v", [""])[0]
-        elif "/embed/" in parsed_url.path:
-            # Embed URL
-            return parsed_url.path.split("/embed/")[-1]
-
-    # Try extracting video ID using regex as fallback
-    video_id_match = re.search(r"(?:v=|\/)([0-9A-Za-z_-]{11}).*", url)
-    if video_id_match:
-        return video_id_match.group(1)
-
-    raise ValueError(f"Could not extract video ID from URL: {url}")
-
-
-def get_youtube_summary(url: str, prompt: str = None) -> str:
-    # print("[get_youtube_summary]url============>", url)
-    # print("[get_youtube_summary]prompt============>", prompt)
-    from google import genai
-    from google.genai.types import HttpOptions, Part
-    from google.oauth2 import service_account
-    from google.genai import types
-
-    if prompt is None:
-        prompt = "Write a short and engaging blog post based on this video."
-
-    try:
-        credentials = service_account.Credentials.from_service_account_file(
-            os.getenv("GOOGLE_APPLICATION_CREDENTIALS_FOR_FASTAPI"),
-            scopes=["https://www.googleapis.com/auth/cloud-platform"],
-        )
-        client = genai.Client(
-            # http_options=HttpOptions(api_version="v1"),
-            credentials=credentials,
-            project="scoop-386004",
-            location="us-central1",
-        )
-        model_id = "gemini-2.5-flash"
-        # model_id = "gemini-2.5-pro"
-
-        response = client.models.generate_content(
-            model=model_id,
-            contents=[
-                types.Content(
-                    role="user",
-                    parts=[
-                        types.Part(
-                            file_data=types.FileData(
-                                mime_type="video/mp4", file_uri=url
-                            )
-                        ),
-                        types.Part(text=prompt),
-                    ],
-                )
-            ],
-        )
-
-        # print(response.text)
-        return response.text
-    except Exception as e:
-        import traceback
-
-        traceback.print_exc()
-        print(f"Error getting YouTube summary: {e}")
-        return f"Error: Failed to get YouTube summary: {e}"
+from urllib.parse import urlparse, parse_qs
+from dotenv import load_dotenv
+import re
+import os
+
+load_dotenv()
+
+
+def get_video_id(url: str) -> str:
+    """
+    Extract video ID from various YouTube URL formats.
+    Supports:
+    - Standard watch URLs (youtube.com/watch?v=...)
+    - Shortened URLs (youtu.be/...)
+    - Embed URLs (youtube.com/embed/...)
+    """
+    # Try parsing as standard URL first
+    parsed_url = urlparse(url)
+
+    # Handle youtu.be URLs
+    if parsed_url.netloc == "youtu.be":
+        return parsed_url.path.lstrip("/")
+
+    # Handle standard youtube.com URLs
+    if parsed_url.netloc in ["www.youtube.com", "youtube.com"]:
+        if parsed_url.path == "/watch":
+            # Standard watch URL
+            return parse_qs(parsed_url.query).get("v", [""])[0]
+        elif "/embed/" in parsed_url.path:
+            # Embed URL
+            return parsed_url.path.split("/embed/")[-1]
+
+    # Try extracting video ID using regex as fallback
+    video_id_match = re.search(r"(?:v=|\/)([0-9A-Za-z_-]{11}).*", url)
+    if video_id_match:
+        return video_id_match.group(1)
+
+    raise ValueError(f"Could not extract video ID from URL: {url}")
+
+
+def get_youtube_summary(url: str, prompt: str = None) -> str:
+    # print("[get_youtube_summary]url============>", url)
+    # print("[get_youtube_summary]prompt============>", prompt)
+    from google import genai
+    from google.genai.types import HttpOptions, Part
+    from google.oauth2 import service_account
+    from google.genai import types
+
+    if prompt is None:
+        prompt = "Write a short and engaging blog post based on this video."
+
+    try:
+        credentials = service_account.Credentials.from_service_account_file(
+            os.getenv("GOOGLE_APPLICATION_CREDENTIALS_FOR_FASTAPI"),
+            scopes=["https://www.googleapis.com/auth/cloud-platform"],
+        )
+        client = genai.Client(
+            # http_options=HttpOptions(api_version="v1"),
+            credentials=credentials,
+            project="scoop-386004",
+            location="us-central1",
+        )
+        model_id = "gemini-2.5-flash"
+        # model_id = "gemini-2.5-pro"
+
+        response = client.models.generate_content(
+            model=model_id,
+            contents=[
+                types.Content(
+                    role="user",
+                    parts=[
+                        types.Part(
+                            file_data=types.FileData(
+                                mime_type="video/mp4", file_uri=url
+                            )
+                        ),
+                        types.Part(text=prompt),
+                    ],
+                )
+            ],
+        )
+
+        # print(response.text)
+        return response.text
+    except Exception as e:
+        import traceback
+
+        traceback.print_exc()
+        print(f"Error getting YouTube summary: {e}")
+        return f"Error: Failed to get YouTube summary: {e}"

botrun_flow_lang/langgraph_agents/cache/langgraph_botrun_cache.py

@@ -1,197 +1,197 @@
-"""
-Botrun ID-based cache for LangGraph React Agent graphs.
-Provides session isolation with parameter validation for graph caching.
-"""
-
-import hashlib
-import json
-import time
-from typing import Dict, Any, Optional
-from botrun_flow_lang.utils.botrun_logger import get_default_botrun_logger
-
-logger = get_default_botrun_logger()
-
-
-class LangGraphBotrunCache:
-    """
-    Botrun ID-based cache for LangGraph React Agent graphs.
-
-    Cache structure:
-    {
-        "botrun_id_1": {
-            "graph": <graph_instance>,
-            "params_hash": "abc123...",
-            "created_at": <timestamp>
-        },
-        "botrun_id_2": {
-            "graph": <graph_instance>,
-            "params_hash": "def456...",
-            "created_at": <timestamp>
-        }
-    }
-    """
-
-    def __init__(self):
-        self._cache: Dict[str, Dict[str, Any]] = {}
-
-    def get_params_hash(
-        self,
-        system_prompt: str,
-        botrun_flow_lang_url: str,
-        user_id: str,
-        model_name: str,
-        lang: str,
-        mcp_config: Optional[Dict[str, Any]] = None,
-    ) -> str:
-        """
-        Generate MD5 hash from parameters for cache validation.
-
-        Args:
-            system_prompt: The system prompt to use for the agent
-            botrun_flow_lang_url: URL for botrun flow lang
-            user_id: User ID
-            model_name: Model name to use
-            lang: Language code (e.g., "en", "zh-TW")
-            mcp_config: MCP servers configuration dict
-
-        Returns:
-            str: MD5 hash of parameters
-        """
-        # Create a stable representation of all parameters
-        cache_data = {
-            "system_prompt": system_prompt,
-            "botrun_flow_lang_url": botrun_flow_lang_url,
-            "user_id": user_id,
-            "model_name": model_name,
-            "lang": lang,
-            "mcp_config": mcp_config or {},
-        }
-
-        # Convert to JSON string for consistent hashing
-        cache_str = json.dumps(cache_data, sort_keys=True)
-
-        # Generate MD5 hash
-        return hashlib.md5(cache_str.encode("utf-8")).hexdigest()
-
-    def get_cached_graph(self, botrun_id: str, params_hash: str) -> Optional[Any]:
-        """
-        Get cached graph if botrun_id exists and params_hash matches.
-
-        Args:
-            botrun_id: The botrun ID to look up
-            params_hash: Expected parameter hash for validation
-
-        Returns:
-            Cached graph instance if found and valid, None otherwise
-        """
-        if botrun_id not in self._cache:
-            logger.debug(f"Botrun ID {botrun_id} not found in cache")
-            return None
-
-        cache_entry = self._cache[botrun_id]
-        cached_hash = cache_entry.get("params_hash")
-
-        if cached_hash != params_hash:
-            logger.info(
-                f"Parameter hash mismatch for botrun_id {botrun_id}. "
-                f"Cached: {cached_hash[:8]}..., Current: {params_hash[:8]}... "
-                f"Cache will be invalidated."
-            )
-            # Clear cache for this botrun_id since parameters changed
-            self.clear_botrun_cache(botrun_id)
-            return None
-
-        logger.info(f"Cache hit for botrun_id {botrun_id}")
-        return cache_entry.get("graph")
-
-    def cache_graph(self, botrun_id: str, params_hash: str, graph: Any):
-        """
-        Cache graph for specific botrun_id with parameter hash.
-
-        Args:
-            botrun_id: The botrun ID to cache for
-            params_hash: Parameter hash for validation
-            graph: Graph instance to cache
-        """
-        self._cache[botrun_id] = {
-            "graph": graph,
-            "params_hash": params_hash,
-            "created_at": time.time(),
-        }
-        logger.info(
-            f"Graph cached for botrun_id {botrun_id} with hash {params_hash[:8]}..."
-        )
-
-    def clear_botrun_cache(self, botrun_id: str):
-        """
-        Clear cache for specific botrun_id.
-
-        Args:
-            botrun_id: The botrun ID to clear cache for
-        """
-        if botrun_id in self._cache:
-            del self._cache[botrun_id]
-            logger.info(f"Cache cleared for botrun_id {botrun_id}")
-
-    def cleanup_old_cache(self, max_age_hours: int = 24):
-        """
-        Remove old cache entries to prevent memory buildup.
-
-        Args:
-            max_age_hours: Maximum age in hours before cache entry is removed
-        """
-        current_time = time.time()
-        max_age_seconds = max_age_hours * 3600
-        botrun_ids_to_remove = []
-
-        for botrun_id, cache_entry in self._cache.items():
-            created_at = cache_entry.get("created_at", 0)
-            if current_time - created_at > max_age_seconds:
-                botrun_ids_to_remove.append(botrun_id)
-
-        for botrun_id in botrun_ids_to_remove:
-            del self._cache[botrun_id]
-            logger.info(f"Removed old cache entry for botrun_id {botrun_id}")
-
-        if botrun_ids_to_remove:
-            logger.info(f"Cleaned up {len(botrun_ids_to_remove)} old cache entries")
-
-    def get_cache_stats(self) -> Dict[str, Any]:
-        """
-        Get cache statistics for monitoring.
-
-        Returns:
-            Dict with cache statistics
-        """
-        current_time = time.time()
-        total_entries = len(self._cache)
-
-        if total_entries == 0:
-            return {
-                "total_entries": 0,
-                "oldest_entry_age_hours": 0,
-                "newest_entry_age_hours": 0,
-                "average_age_hours": 0,
-            }
-
-        ages = []
-        for cache_entry in self._cache.values():
-            created_at = cache_entry.get("created_at", current_time)
-            age_hours = (current_time - created_at) / 3600
-            ages.append(age_hours)
-
-        return {
-            "total_entries": total_entries,
-            "oldest_entry_age_hours": max(ages),
-            "newest_entry_age_hours": min(ages),
-            "average_age_hours": sum(ages) / len(ages),
-        }
-
-
-# Global cache instance
-_global_cache = LangGraphBotrunCache()
-
-
-def get_botrun_cache() -> LangGraphBotrunCache:
-    """Get the global botrun cache instance."""
-    return _global_cache
+"""
+Botrun ID-based cache for LangGraph React Agent graphs.
+Provides session isolation with parameter validation for graph caching.
+"""
+
+import hashlib
+import json
+import time
+from typing import Dict, Any, Optional
+from botrun_flow_lang.utils.botrun_logger import get_default_botrun_logger
+
+logger = get_default_botrun_logger()
+
+
+class LangGraphBotrunCache:
+    """
+    Botrun ID-based cache for LangGraph React Agent graphs.
+
+    Cache structure:
+    {
+        "botrun_id_1": {
+            "graph": <graph_instance>,
+            "params_hash": "abc123...",
+            "created_at": <timestamp>
+        },
+        "botrun_id_2": {
+            "graph": <graph_instance>,
+            "params_hash": "def456...",
+            "created_at": <timestamp>
+        }
+    }
+    """
+
+    def __init__(self):
+        self._cache: Dict[str, Dict[str, Any]] = {}
+
+    def get_params_hash(
+        self,
+        system_prompt: str,
+        botrun_flow_lang_url: str,
+        user_id: str,
+        model_name: str,
+        lang: str,
+        mcp_config: Optional[Dict[str, Any]] = None,
+    ) -> str:
+        """
+        Generate MD5 hash from parameters for cache validation.
+
+        Args:
+            system_prompt: The system prompt to use for the agent
+            botrun_flow_lang_url: URL for botrun flow lang
+            user_id: User ID
+            model_name: Model name to use
+            lang: Language code (e.g., "en", "zh-TW")
+            mcp_config: MCP servers configuration dict
+
+        Returns:
+            str: MD5 hash of parameters
+        """
+        # Create a stable representation of all parameters
+        cache_data = {
+            "system_prompt": system_prompt,
+            "botrun_flow_lang_url": botrun_flow_lang_url,
+            "user_id": user_id,
+            "model_name": model_name,
+            "lang": lang,
+            "mcp_config": mcp_config or {},
+        }
+
+        # Convert to JSON string for consistent hashing
+        cache_str = json.dumps(cache_data, sort_keys=True)
+
+        # Generate MD5 hash
+        return hashlib.md5(cache_str.encode("utf-8")).hexdigest()
+
+    def get_cached_graph(self, botrun_id: str, params_hash: str) -> Optional[Any]:
+        """
+        Get cached graph if botrun_id exists and params_hash matches.
+
+        Args:
+            botrun_id: The botrun ID to look up
+            params_hash: Expected parameter hash for validation
+
+        Returns:
+            Cached graph instance if found and valid, None otherwise
+        """
+        if botrun_id not in self._cache:
+            logger.debug(f"Botrun ID {botrun_id} not found in cache")
+            return None
+
+        cache_entry = self._cache[botrun_id]
+        cached_hash = cache_entry.get("params_hash")
+
+        if cached_hash != params_hash:
+            logger.info(
+                f"Parameter hash mismatch for botrun_id {botrun_id}. "
+                f"Cached: {cached_hash[:8]}..., Current: {params_hash[:8]}... "
+                f"Cache will be invalidated."
+            )
+            # Clear cache for this botrun_id since parameters changed
+            self.clear_botrun_cache(botrun_id)
+            return None
+
+        logger.info(f"Cache hit for botrun_id {botrun_id}")
+        return cache_entry.get("graph")
+
+    def cache_graph(self, botrun_id: str, params_hash: str, graph: Any):
+        """
+        Cache graph for specific botrun_id with parameter hash.
+
+        Args:
+            botrun_id: The botrun ID to cache for
+            params_hash: Parameter hash for validation
+            graph: Graph instance to cache
+        """
+        self._cache[botrun_id] = {
+            "graph": graph,
+            "params_hash": params_hash,
+            "created_at": time.time(),
+        }
+        logger.info(
+            f"Graph cached for botrun_id {botrun_id} with hash {params_hash[:8]}..."
+        )
+
+    def clear_botrun_cache(self, botrun_id: str):
+        """
+        Clear cache for specific botrun_id.
+
+        Args:
+            botrun_id: The botrun ID to clear cache for
+        """
+        if botrun_id in self._cache:
+            del self._cache[botrun_id]
+            logger.info(f"Cache cleared for botrun_id {botrun_id}")
+
+    def cleanup_old_cache(self, max_age_hours: int = 24):
+        """
+        Remove old cache entries to prevent memory buildup.
+
+        Args:
+            max_age_hours: Maximum age in hours before cache entry is removed
+        """
+        current_time = time.time()
+        max_age_seconds = max_age_hours * 3600
+        botrun_ids_to_remove = []
+
+        for botrun_id, cache_entry in self._cache.items():
+            created_at = cache_entry.get("created_at", 0)
+            if current_time - created_at > max_age_seconds:
+                botrun_ids_to_remove.append(botrun_id)
+
+        for botrun_id in botrun_ids_to_remove:
+            del self._cache[botrun_id]
+            logger.info(f"Removed old cache entry for botrun_id {botrun_id}")
+
+        if botrun_ids_to_remove:
+            logger.info(f"Cleaned up {len(botrun_ids_to_remove)} old cache entries")
+
+    def get_cache_stats(self) -> Dict[str, Any]:
+        """
+        Get cache statistics for monitoring.
+
+        Returns:
+            Dict with cache statistics
+        """
+        current_time = time.time()
+        total_entries = len(self._cache)
+
+        if total_entries == 0:
+            return {
+                "total_entries": 0,
+                "oldest_entry_age_hours": 0,
+                "newest_entry_age_hours": 0,
+                "average_age_hours": 0,
+            }
+
+        ages = []
+        for cache_entry in self._cache.values():
+            created_at = cache_entry.get("created_at", current_time)
+            age_hours = (current_time - created_at) / 3600
+            ages.append(age_hours)
+
+        return {
+            "total_entries": total_entries,
+            "oldest_entry_age_hours": max(ages),
+            "newest_entry_age_hours": min(ages),
+            "average_age_hours": sum(ages) / len(ages),
+        }
+
+
+# Global cache instance
+_global_cache = LangGraphBotrunCache()
+
+
+def get_botrun_cache() -> LangGraphBotrunCache:
+    """Get the global botrun cache instance."""
+    return _global_cache

botrun_flow_lang/llm_agent/llm_agent.py

@@ -1,19 +1,19 @@
-from pydantic import BaseModel
-from typing import Optional
-
-
-class LlmAgent(BaseModel):
-    """
-    @include_in_history: 是否將這次的回答加入 history
-    @max_system_prompt_length: 如果 system prompt 超過這個長度，會整個被捨棄
-    """
-
-    name: str = ""
-    print_output: bool = True
-    print_plotly: bool = True
-    output: Optional[str] = None
-    model: str = "openai/gpt-4o-2024-08-06"
-    system_prompt: str
-    gen_image: bool = False
-    include_in_history: bool = True
-    max_system_prompt_length: Optional[int] = None
+from pydantic import BaseModel
+from typing import Optional
+
+
+class LlmAgent(BaseModel):
+    """
+    @include_in_history: 是否將這次的回答加入 history
+    @max_system_prompt_length: 如果 system prompt 超過這個長度，會整個被捨棄
+    """
+
+    name: str = ""
+    print_output: bool = True
+    print_plotly: bool = True
+    output: Optional[str] = None
+    model: str = "openai/gpt-4o-2024-08-06"
+    system_prompt: str
+    gen_image: bool = False
+    include_in_history: bool = True
+    max_system_prompt_length: Optional[int] = None