Lag Effect

README.md

@@ -4,8 +4,7 @@
 [![PyPI version](https://badge.fury.io/py/mockllm.svg)](https://badge.fury.io/py/mockllm)
 [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 
-A FastAPI-based mock LLM server that mimics OpenAI and Anthropic API formats. Instead of calling actual language models,
-it uses predefined responses from a YAML configuration file. 
+A FastAPI-based mock LLM server that mimics OpenAI and Anthropic API formats. Instead of calling actual language models, it uses predefined responses from a YAML configuration file. 
 
 This is made for when you want a deterministic response for testing or development purposes.
 
@@ -17,8 +16,6 @@ Check out the [CodeGate](https://github.com/stacklok/codegate) project when you'
 - Streaming support (character-by-character response streaming)
 - Configurable responses via YAML file
 - Hot-reloading of response configurations
-- JSON logging
-- Error handling
 - Mock token counting
 
 ## Installation
@@ -128,10 +125,11 @@ curl -X POST http://localhost:8000/v1/messages \
 
 ### Response Configuration
 
-Responses are configured in `responses.yml`. The file has two main sections:
+Responses are configured in `responses.yml`. The file has three main sections:
 
 1. `responses`: Maps input prompts to predefined responses
 2. `defaults`: Contains default configurations like the unknown response message
+3. `settings`: Contains server behavior settings like network lag simulation
 
 Example `responses.yml`:
 ```yaml
@@ -141,71 +139,39 @@ responses:
 
 defaults:
   unknown_response: "I don't know the answer to that. This is a mock response."
-```
-
-### Hot Reloading
-
-The server automatically detects changes to `responses.yml` and reloads the configuration without requiring a restart.
-
-## Development
-
-The project uses Poetry for dependency management and includes a Makefile to help with common development tasks:
-
-```bash
-# Set up development environment
-make setup
-
-# Run all checks (setup, lint, test)
-make all
 
-# Run tests
-make test
-
-# Format code
-make format
-
-# Run all linting and type checking
-make lint
-
-# Clean up build artifacts
-make clean
-
-# See all available commands
-make help
+settings:
+  lag_enabled: true
+  lag_factor: 10  # Higher values = faster responses (10 = fast, 1 = slow)
 ```
 
-### Development Commands
+### Network Lag Simulation
 
-- `make setup`: Install all development dependencies with Poetry
-- `make test`: Run the test suite
-- `make format`: Format code with black and isort
-- `make lint`: Run all code quality checks (format, lint, type)
-- `make build`: Build the package with Poetry
-- `make clean`: Remove build artifacts and cache files
-- `make install-dev`: Install package with development dependencies
+The server can simulate network latency for more realistic testing scenarios. This is controlled by two settings:
 
-For more details on available commands, run `make help`.
+- `lag_enabled`: When true, enables artificial network lag
+- `lag_factor`: Controls the speed of responses
+  - Higher values (e.g., 10) result in faster responses
+  - Lower values (e.g., 1) result in slower responses
+  - Affects both streaming and non-streaming responses
 
-## Error Handling
+For streaming responses, the lag is applied per-character with slight random variations to simulate realistic network conditions.
 
-The server includes comprehensive error handling:
-
-- Invalid requests return 400 status codes with descriptive messages
-- Server errors return 500 status codes with error details
-- All errors are logged using JSON format
+### Hot Reloading
 
-## Logging
+The server automatically detects changes to `responses.yml` and reloads the configuration without restarting the server.
 
-The server uses JSON-formatted logging for:
+## Testing
 
-- Incoming request details
-- Response configuration loading
-- Error messages and stack traces
+To run the tests:
+```bash
+poetry run pytest
+```
 
 ## Contributing
 
-Contributions are welcome! Please feel free to submit a Pull Request.
+Contributions are welcome! Please open an issue or submit a PR.
 
 ## License
 
-This project is licensed under the Apache License, Version 2.0 - see the [LICENSE](LICENSE) file for details.
+This project is licensed under the [Apache 2.0 License](LICENSE).

example.responses.yml

@@ -5,4 +5,8 @@ responses:
   "what is the meaning of life?": "According to this mock response, the meaning of life is to write better mock servers."
 
 defaults:
-  unknown_response: "I don't know the answer to that. This is a mock response."
+  unknown_response: "I don't know the answer to that. This is a mock response."
+
+settings:
+  lag_enabled: true
+  lag_factor: 10  # Higher values = faster responses (10 = fast, 1 = slow)

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "mockllm"
-version = "0.0.6"
+version = "0.0.7"
 description = "A mock server that mimics OpenAI and Anthropic API formats for testing"
 authors = ["Luke Hinds <[email protected]>"]
 license = "Apache-2.0"

src/mockllm/__init__.py

@@ -2,4 +2,4 @@
 Mock LLM Server - You will do what I tell you!
 """
 
-__version__ = "0.0.6"
+__version__ = "0.0.7"

src/mockllm/config.py

@@ -1,6 +1,8 @@
+import asyncio
 import logging
+import random
 from pathlib import Path
-from typing import Dict, Generator, Optional
+from typing import AsyncGenerator, Dict, Generator, Optional
 
 import yaml
 from fastapi import HTTPException
@@ -20,6 +22,8 @@ def __init__(self, yaml_path: str = "responses.yml"):
         self.last_modified = 0
         self.responses: Dict[str, str] = {}
         self.default_response = "I don't know the answer to that."
+        self.lag_enabled = False
+        self.lag_factor = 10
         self.load_responses()
 
     def load_responses(self) -> None:
@@ -33,6 +37,9 @@ def load_responses(self) -> None:
                     self.default_response = data.get("defaults", {}).get(
                         "unknown_response", self.default_response
                     )
+                    settings = data.get("settings", {})
+                    self.lag_enabled = settings.get("lag_enabled", False)
+                    self.lag_factor = settings.get("lag_factor", 10)
                 self.last_modified = int(current_mtime)
                 logger.info(
                     f"Loaded {len(self.responses)} responses from {self.yaml_path}"
@@ -62,3 +69,35 @@ def get_streaming_response(
             # Yield response character by character
             for char in response:
                 yield char
+
+    async def get_response_with_lag(self, prompt: str) -> str:
+        """Get response with artificial lag for non-streaming responses."""
+        response = self.get_response(prompt)
+        if self.lag_enabled:
+            # Base delay on response length and lag factor
+            delay = len(response) / (self.lag_factor * 10)
+            await asyncio.sleep(delay)
+        return response
+
+    async def get_streaming_response_with_lag(
+        self, prompt: str, chunk_size: Optional[int] = None
+    ) -> AsyncGenerator[str, None]:
+        """Generator that yields response content with artificial lag."""
+        response = self.get_response(prompt)
+
+        if chunk_size:
+            for i in range(0, len(response), chunk_size):
+                chunk = response[i : i + chunk_size]
+                if self.lag_enabled:
+                    delay = len(chunk) / (self.lag_factor * 10)
+                    await asyncio.sleep(delay)
+                yield chunk
+        else:
+            for char in response:
+                if self.lag_enabled:
+                    # Add random variation to character delay
+                    base_delay = 1 / (self.lag_factor * 10)
+                    variation = random.uniform(-0.5, 0.5) * base_delay
+                    delay = max(0, base_delay + variation)
+                    await asyncio.sleep(delay)
+                yield char

src/mockllm/server.py

@@ -30,15 +30,14 @@
 
 async def openai_stream_response(content: str, model: str) -> AsyncGenerator[str, None]:
     """Generate OpenAI-style streaming response in SSE format."""
-    # Send the first message with role
     first_chunk = OpenAIStreamResponse(
         model=model,
         choices=[OpenAIStreamChoice(delta=OpenAIDeltaMessage(role="assistant"))],
     )
     yield f"data: {first_chunk.model_dump_json()}\n\n"
 
-    # Stream the content character by character
-    for chunk in response_config.get_streaming_response(content):
+    # Stream the content character by character with lag
+    async for chunk in response_config.get_streaming_response_with_lag(content):
         chunk_response = OpenAIStreamResponse(
             model=model,
             choices=[OpenAIStreamChoice(delta=OpenAIDeltaMessage(content=chunk))],
@@ -58,7 +57,7 @@ async def anthropic_stream_response(
     content: str, model: str
 ) -> AsyncGenerator[str, None]:
     """Generate Anthropic-style streaming response in SSE format."""
-    for chunk in response_config.get_streaming_response(content):
+    async for chunk in response_config.get_streaming_response_with_lag(content):
         stream_response = AnthropicStreamResponse(
             delta=AnthropicStreamDelta(delta={"text": chunk})
         )
@@ -98,7 +97,9 @@ async def openai_chat_completion(
                 media_type="text/event-stream",
             )
 
-        response_content = response_config.get_response(last_message.content)
+        response_content = await response_config.get_response_with_lag(
+            last_message.content
+        )
 
         # Calculate mock token counts
         prompt_tokens = len(str(request.messages).split())
@@ -159,7 +160,9 @@ async def anthropic_chat_completion(
                 media_type="text/event-stream",
             )
 
-        response_content = response_config.get_response(last_message.content)
+        response_content = await response_config.get_response_with_lag(
+            last_message.content
+        )
 
         # Calculate mock token counts
         prompt_tokens = len(str(request.messages).split())