fix: update base url to use environment variables

examples/anthropic/javelin_anthropic_univ_endpoint.py

@@ -18,7 +18,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration for Bedrock
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
 )
 client = JavelinClient(config)

examples/anthropic/openai_compatible_univ_anthropic.py

@@ -15,7 +15,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     timeout=120,
 )

examples/azure-openai/azure_general_route.py

@@ -18,7 +18,7 @@ def init_azure_client_sync():
         javelin_headers = {"x-api-key": javelin_api_key}
         client = AzureOpenAI(
             api_key=llm_api_key,
-            base_url="https://api-dev.javelin.live/v1/query/azure-openai",
+            base_url=f"{os.getenv('JAVELIN_BASE_URL')}/v1/query/azure-openai",
             default_headers=javelin_headers,
             api_version="2024-02-15-preview"
         )

examples/azure-openai/azure_openai_javelin_stream_&_non-stream.js

@@ -4,12 +4,11 @@ import { Stream } from 'openai/streaming.mjs';
 const javelinApiKey = ""; // javelin api key here
 const llmApiKey = ""; // llm api key
 
-const javelinBaseUrl = 'https://api.javelin.live/v1/query';
 
 async function getCompletion() {
   try {
     const routeName = 'AzureOpenAIRoute';
-    const url = "https://api.javelin.live/v1/query/AzureOpenAIRoute";
+    const url = `${process.env.JAVELIN_BASE_URL}/v1/query/${routeName}`;
 
     const response = await axios.post(
       url,

examples/azure-openai/javelin_azureopenai_univ_endpoint.py

@@ -14,7 +14,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     default_headers={
         "Content-Type": "application/json",

examples/azure-openai/langchain_azure_universal.py

@@ -13,9 +13,7 @@
 load_dotenv()
 azure_openai_api_key = os.getenv("AZURE_OPENAI_API_KEY")
 javelin_api_key = os.getenv("JAVELIN_API_KEY")
-base_url = os.getenv(
-    "JAVELIN_BASE_URL", "https://api.javelin.live"
-)  # Default to generic base URL
+base_url = os.getenv("JAVELIN_BASE_URL")
 
 # The name of your Azure deployment (e.g., "gpt-4")
 # or whatever you’ve set in Azure. Must also match x-javelin-model if

examples/azure-openai/langchain_chatmodel_example.py

@@ -4,9 +4,10 @@
 dotenv.load_dotenv()
 
 from langchain_openai import AzureChatOpenAI
-
+url = os.path.join(os.getenv("JAVELIN_BASE_URL"), "v1")
+print(url)
 model = AzureChatOpenAI(
-    azure_endpoint="https://api-dev.javelin.live/v1",
+    azure_endpoint=url,
     azure_deployment="gpt35",
     openai_api_version="2023-03-15-preview",
     extra_headers={"x-javelin-route": "azureopenai_univ", "x-api-key": os.environ.get("JAVELIN_API_KEY")}

examples/azure-openai/openai_compatible_univ_azure.py

@@ -17,7 +17,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     timeout=120,
 )

examples/bedrock/bedrock_client_universal.py

@@ -23,7 +23,8 @@ def init_bedrock():
 
     # Initialize Javelin Client (if you want the route registered)
     config = JavelinConfig(
-        javelin_api_key=os.getenv("JAVELIN_API_KEY") # Replace with your Javelin API key
+        javelin_api_key=os.getenv("JAVELIN_API_KEY"), # Replace with your Javelin API key
+        base_url=os.getenv("JAVELIN_BASE_URL")
     )
     javelin_client = JavelinClient(config)
 

examples/bedrock/bedrock_general_route.py

@@ -56,7 +56,7 @@ def get_bedrock_client():
         client = boto3.client(
             service_name="bedrock-runtime",
             region_name="us-east-1",
-            endpoint_url="https://api-dev.javelin.live/v1/",
+            endpoint_url=os.path.join(os.getenv("JAVELIN_BASE_URL"), "v1"),
             aws_access_key_id=aws_access_key_id,
             aws_secret_access_key=aws_secret_access_key
         )

examples/bedrock/javelin_bedrock_univ_endpoint.py

@@ -14,7 +14,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration for Bedrock
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
 )
 client = JavelinClient(config)

examples/bedrock/langchain-bedrock-universal.py

@@ -27,7 +27,8 @@ def init_bedrock():
 
     # Initialize Javelin client
     config = JavelinConfig(
-        javelin_api_key=os.getenv("JAVELIN_API_KEY")  # add your Javelin API key here
+        javelin_api_key=os.getenv("JAVELIN_API_KEY"),  # add your Javelin API key here
+        base_url=os.getenv("JAVELIN_BASE_URL")
     )
     javelin_client = JavelinClient(config)
 

examples/bedrock/openai_compatible_univ_bedrock.py

@@ -12,7 +12,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     timeout=120,
 )

examples/gemini/document_processing.py

@@ -26,6 +26,7 @@ def initialize_javelin_client():
     javelin_api_key = os.getenv("JAVELIN_API_KEY")
     config = JavelinConfig(
         javelin_api_key=javelin_api_key,
+        base_url=os.getenv("JAVELIN_BASE_URL")
     )
     return JavelinClient(config)
 

examples/gemini/gemini-universal.py

@@ -39,7 +39,7 @@ def init_gemini_client():
     # Javelin configuration
 
     javelin_api_key = os.getenv("JAVELIN_API_KEY")  # define your javelin api key here
-    config = JavelinConfig(javelin_api_key=javelin_api_key)
+    config = JavelinConfig(javelin_api_key=javelin_api_key, base_url=os.getenv("JAVELIN_BASE_URL"))
     client = JavelinClient(config)
     rout_name = "google_univ"  # define your universal route name here
     # Register the Gemini client with Javelin

examples/gemini/javelin_gemini_univ_endpoint.py

@@ -14,7 +14,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     llm_api_key=os.getenv("OPENAI_API_KEY"),
 )

examples/gemini/langchain_chatmodel_example.py

@@ -5,7 +5,7 @@
 
 from langchain.chat_models import init_chat_model
 
-model = init_chat_model("gemini-1.5-flash", model_provider="openai", base_url="https://api-dev.javelin.live/v1",
+model = init_chat_model("gemini-1.5-flash", model_provider="openai", base_url=f"{os.getenv('JAVELIN_BASE_URL')}/v1",
 extra_headers={"x-javelin-route": "google_univ", "x-api-key": os.environ.get("JAVELIN_API_KEY"), "Authorization": f"Bearer {os.environ.get('GEMINI_API_KEY')}"})
 
 print(model.invoke("write a poem about a cat"))

examples/gemini/openai_compatible_univ_gemini.py

@@ -17,7 +17,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     llm_api_key=os.getenv("OPENAI_API_KEY"),
     timeout=120,

examples/gemini/strawberry.py

@@ -47,7 +47,7 @@ def openai_chat_completions():
 def initialize_javelin_client():
     javelin_api_key = os.getenv("JAVELIN_API_KEY")
     config = JavelinConfig(
-        base_url="https://api-dev.javelin.live",
+        base_url=os.getenv("JAVELIN_BASE_URL"),
         javelin_api_key=javelin_api_key,
     )
     return JavelinClient(config)

examples/mistral/langchain_chatmodel_example.py

@@ -5,7 +5,7 @@
 
 from langchain.chat_models import init_chat_model
 
-model = init_chat_model("mistral-large-latest", model_provider="openai", base_url="https://api-dev.javelin.live/v1",
+model = init_chat_model("mistral-large-latest", model_provider="openai", base_url=f"{os.getenv('JAVELIN_BASE_URL')}/v1",
 extra_headers={"x-javelin-route": "mistral_univ", "x-api-key": os.environ.get("JAVELIN_API_KEY"), "Authorization": f"Bearer {os.environ.get('MISTRAL_API_KEY')}"})
 
 print(model.invoke("write a poem about a cat"))

examples/openai/javelin_openai_univ_endpoint.py

@@ -14,7 +14,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     llm_api_key=os.getenv("OPENAI_API_KEY"),
 )

examples/openai/langchain-openai-universal.py

@@ -20,7 +20,7 @@
 MODEL_NAME_CHAT = "gpt-3.5-turbo"  # For chat
 MODEL_NAME_EMBED = "text-embedding-ada-002"
 ROUTE_NAME = "openai_univ"
-BASE_URL = os.getenv("JAVELIN_BASE_URL", "https://api.javelin.live")  # Default base URL
+BASE_URL = os.getenv("JAVELIN_BASE_URL")  # Default base URL
 
 
 def init_chat_llm_non_streaming():

examples/openai/langchain_callback_example.py

@@ -0,0 +1,61 @@
+import dotenv
+import os
+from typing import Any, Dict, List
+
+from langchain_core.callbacks import BaseCallbackHandler
+from langchain_core.messages import BaseMessage
+from langchain.chat_models import init_chat_model
+
+dotenv.load_dotenv()
+
+class HeaderCallbackHandler(BaseCallbackHandler):
+    """Custom callback handler that modifies the headers on chat model start."""
+
+    def __init__(self):
+        self.api_key = os.environ.get("JAVELIN_API_KEY")
+
+    def on_chain_start(
+        self, serialized: Dict[str, Any], inputs: Dict[str, Any], **kwargs: Any
+    ) -> Any:
+        """Run when chain starts running."""
+        print("Chain started")
+        print(serialized, inputs, kwargs)
+
+    def on_chat_model_start(
+        self, serialized: Dict[str, Any], messages: List[List[BaseMessage]], **kwargs: Any
+    ) -> Any:
+        """Run when Chat Model starts running."""
+        # The serialized dict contains the model configuration
+        print(self.__super().on_chat_model_start(serialized, messages, **kwargs))
+        if "kwargs" in serialized:
+            # Add or update the headers in the model kwargs
+            if "model_kwargs" not in serialized["kwargs"]:
+                serialized["kwargs"]["model_kwargs"] = {}
+            if "extra_headers" not in serialized["kwargs"]["model_kwargs"]:
+                serialized["kwargs"]["model_kwargs"]["extra_headers"] = {}
+
+            # Determine the route based on the model provider
+            provider = serialized.get("name", "").lower()
+            route = "azureopenai_univ" if "azure" in provider else "openai_univ"
+
+            headers = {
+                "x-javelin-route": route,
+                "x-api-key": self.api_key
+            }
+            serialized["kwargs"]["model_kwargs"]["extra_headers"].update(headers)
+            print(f"Modified headers to: {headers}")
+
+# Initialize the callback handler
+callback_handler = HeaderCallbackHandler()
+
+# Initialize the chat model with the callback handler
+model = init_chat_model(
+    "gpt-4o-mini", 
+    model_provider="openai",
+    base_url="http://127.0.0.1:8000/v1",
+    extra_headers={"x-javelin-route": "openai_univ", "x-api-key": os.environ.get("JAVELIN_API_KEY")},
+    callbacks=[callback_handler]  # Add our custom callback handler
+)
+
+# Test the model
+print(model.invoke("Hello, world!")) 

examples/openai/langchain_chatmodel_example.py

@@ -5,7 +5,7 @@
 
 from langchain.chat_models import init_chat_model
 
-model = init_chat_model("gpt-4o-mini", model_provider="openai", base_url="https://api-dev.javelin.live/v1",
+model = init_chat_model("gpt-4o-mini", model_provider="openai", base_url=f"{os.getenv('JAVELIN_BASE_URL')}/v1",
 extra_headers={"x-javelin-route": "openai_univ", "x-api-key": os.environ.get("JAVELIN_API_KEY")})
 
 print(model.invoke("Hello, world!"))

examples/openai/o1-03_function-calling.py

@@ -19,7 +19,7 @@ def init_openai_client():
 
 def init_javelin_client(openai_client, route_name="openai_univ"):
     javelin_api_key = os.getenv("JAVELIN_API_KEY")
-    config = JavelinConfig(javelin_api_key=javelin_api_key)
+    config = JavelinConfig(javelin_api_key=javelin_api_key, base_url=os.getenv("JAVELIN_BASE_URL"))
     client = JavelinClient(config)
     client.register_openai(openai_client, route_name=route_name)
     return client

examples/openai/openai-azure-fun_calling.ipynb

@@ -128,7 +128,7 @@
     "\n",
     "def init_javelin_client(openai_client, route_name=\"openai_univ\"):\n",
     "    javelin_api_key = os.getenv(\"JAVELIN_API_KEY\")\n",
-    "    config = JavelinConfig(javelin_api_key=javelin_api_key)\n",
+    "    config = JavelinConfig(javelin_api_key=javelin_api_key, base_url=os.getenv(\"JAVELIN_BASE_URL\"))\n",
     "    client = JavelinClient(config)\n",
     "    client.register_openai(openai_client, route_name=route_name)\n",
     "    return client\n",

examples/openai/openai-universal.py

@@ -39,6 +39,7 @@ def init_javelin_client_sync(openai_client):
         )  # define your javelin api key here
         config = JavelinConfig(
             javelin_api_key=javelin_api_key,
+            base_url=os.getenv("JAVELIN_BASE_URL"),
         )
         client = JavelinClient(config)
         rout_name = "openai_univ"  # define your universal route name here
@@ -118,7 +119,7 @@ def init_javelin_client_async(openai_async_client):
     """Initialize JavelinClient for async usage and register the OpenAI route."""
     try:
         javelin_api_key = os.getenv("JAVELIN_API_KEY")  # add your javelin api key here
-        config = JavelinConfig(javelin_api_key=javelin_api_key)
+        config = JavelinConfig(javelin_api_key=javelin_api_key, base_url=os.getenv("JAVELIN_BASE_URL"))
         client = JavelinClient(config)
         client.register_openai(openai_async_client, route_name="openai_univ")
         return client

examples/openai/openai_compatible_univ.py

@@ -17,7 +17,7 @@ def print_response(provider: str, response: Dict[str, Any]) -> None:
 
 # Setup client configuration
 config = JavelinConfig(
-    base_url="https://api-dev.javelin.live",
+    base_url=os.getenv("JAVELIN_BASE_URL"),
     javelin_api_key=os.getenv("JAVELIN_API_KEY"),
     llm_api_key=os.getenv("OPENAI_API_KEY"),
     timeout=120,

examples/openai/openai_embedding_example.py

@@ -37,7 +37,7 @@
 javelin_headers = {"x-api-key": javelin_api_key, "x-javelin-route": "myusers"}
 
 llm = ChatOpenAI(
-    openai_api_base="https://api.javelin.live/v1/query",
+    openai_api_base=f"{os.getenv('JAVELIN_BASE_URL')}/v1/query",
     openai_api_key=llm_api_key,
     model_kwargs={"extra_headers": javelin_headers},
 )

examples/openai/openai_general_route.py

@@ -21,7 +21,7 @@ def init_sync_openai_client():
         print(f"[DEBUG] Synchronous OpenAI client key: {openai_api_key}")
         return OpenAI(
             api_key=openai_api_key,
-            base_url="https://api-dev.javelin.live/v1/query/openai",
+            base_url=f"{os.getenv('JAVELIN_BASE_URL')}/v1/query/openai",
             default_headers=javelin_headers
         )
     except Exception as e:

examples/openai/openai_javelin_stream&non-stream.js

@@ -3,7 +3,7 @@ import OpenAI from "openai";
 
 const openai_client = new OpenAI({
   apiKey: "",  // add your api key
-  baseURL: "https://api.javelin.live/v1/query",
+  baseURL: `${process.env.JAVELIN_BASE_URL}/v1/query`,
   defaultHeaders: {
     "x-api-key": "", // add here javelin api key
     "x-javelin-route": "OpenAIInspect",

examples/rag/javelin_rag_embeddings_demo.ipynb

@@ -108,15 +108,15 @@
     "# Initialize Azure OpenAI client for embeddings\n",
     "azure_openai_client = AzureOpenAI(\n",
     "    api_key=llm_api_key,\n",
-    "    base_url=\"https://api.javelin.live/query\",\n",
+    "    base_url=os.getenv(\"JAVELIN_BASE_URL\"),\n",
     "    default_headers=javelin_headers_embeddings,\n",
     "    api_version=\"2023-05-15\",\n",
     ")\n",
     "\n",
     "# Initialize LLM\n",
     "llm = AzureChatOpenAI(\n",
     "    api_key=llm_api_key,\n",
-    "    azure_endpoint=\"https://api.javelin.live/query/azureopenai\",\n",
+    "    azure_endpoint=f\"{os.getenv('JAVELIN_BASE_URL')}/query/azureopenai\",\n",
     "    azure_deployment=\"gpt35\",\n",
     "    openai_api_version=\"2024-02-15-preview\",\n",
     "    model_kwargs={\"extra_headers\": javelin_headers_llm}\n",

examples/rag/rag_implemetation_javelin.ipynb

@@ -530,7 +530,7 @@
    "outputs": [],
    "source": [
     "config = JavelinConfig(\n",
-    "    base_url=\"https://api-dev.javelin.live\",\n",
+    "    base_url=os.getenv(\"JAVELIN_BASE_URL\"),\n",
     "    javelin_api_key=os.getenv(\"JAVELIN_API_KEY\"),\n",
     "    llm_api_key=os.getenv(\"OPENAI_API_KEY\"),\n",
     ")\n",

examples/route_examples/aexample.py

@@ -183,7 +183,7 @@ async def main():
 
     try:
         config = JavelinConfig(
-            base_url="https://api.javelin.live",
+            base_url=os.getenv("JAVELIN_BASE_URL"),
             javelin_api_key=javelin_api_key,
             javelin_virtualapikey=javelin_virtualapikey,
             llm_api_key=llm_api_key,

examples/route_examples/drop_in_replacement.py

@@ -115,6 +115,7 @@ def main():
 
     try:
         config = JavelinConfig(
+            base_url=os.getenv("JAVELIN_BASE_URL"),
             javelin_api_key=javelin_api_key,
             javelin_virtualapikey=javelin_virtualapikey,
             llm_api_key=llm_api_key,