speedy-utils 1.1.20__tar.gz → 1.1.22__tar.gz

speedy_utils-1.1.22/.github/copilot-instructions.md

@@ -0,0 +1,67 @@
+## Coding Standards
+
+- **Naming**
+  - snake_case for functions & variables
+  - PascalCase for classes
+- **Formatting**
+  - 4-space indentation
+  - Single quotes for strings
+  - f-strings for interpolation
+- **Typing & Docs**
+  - Add type hints to all public functions & methods
+  - Keep docstrings minimal; comment only non-obvious logic
+- **Comparisons**
+  - Use `is` / `is not` when comparing with `None`
+
+## Tooling Assumptions
+
+- Editor: **VS Code**
+- Pylance setting: `"python.analysis.typeCheckingMode": "basic"`
+  - Code must satisfy **basic** type checking (no strict-mode warnings).
+  - Prefer clear, straightforward typing over complex generics.
+
+## Preferred Libraries
+
+- Logging: `loguru`
+- Database: `Peewee` + **PostgreSQL** (or **SQLite** locally)
+- Web API: `FastAPI`
+- Pydantic: `FastAPI`
+- Testing: `pytest`
+- Async: `asyncio` + `aiohttp`
+- PydanticV2: `pydantic` (v2.x)
+
+## Project Architecture
+
+- **speedy_utils**: General utility library for caching, parallel processing, file I/O, data manipulation. See `src/speedy_utils/__init__.py` for public API.
+- **llm_utils**: LLM-specific utilities including chat formatting, language model tasks, vector caching, and VLLM server management. See `src/llm_utils/__init__.py`.
+- Data flows: Function caching with disk/memory backends, parallel execution via threads/processes, auto-detect file serialization (JSON/JSONL/Pickle).
+
+## Developer Workflows
+
+- **Build/Install**: `uv pip install .` or `pip install -e .` for development.
+- **Test**: `python test.py` (runs unittest suite) or `pytest` for individual tests.
+- **Lint/Format**: `ruff check .` and `ruff format .` (configured in `ruff.toml`).
+- **Type Check**: `mypy` (strict settings in `pyproject.toml`).
+- **Debug**: Use `ipdb` for interactive debugging, `debugpy` for VS Code.
+- **Scripts**: `mpython` for multi-process Python execution, `svllm` for VLLM server, `svllm-lb` for load balancer.
+
+## Integration Points
+
+- **External APIs**: OpenAI API via `llm_utils.lm.openai_memoize`, VLLM servers for local inference.
+- **Distributed**: Ray backend for parallel processing (install with `[ray]` extra).
+- **Async**: `aiohttp` for async HTTP, `asyncio` for concurrency.
+- **Dependencies**: Managed via `pyproject.toml`, install with `uv` preferred.
+
+## When user provide a list of problems:
+
+Directly go over the list and edit the code to fix the problems.
+
+## Common problems when generated code to avoid:
+
+_Stick to these choices unless explicitly instructed otherwise._
+
+---
+
+## Rules for Code Generation
+
+- Each line should less than 88 characters.

{speedy_utils-1.1.20 → speedy_utils-1.1.22}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: speedy-utils
-Version: 1.1.20
+Version: 1.1.22
 Summary: Fast and easy-to-use package for data science
 Project-URL: Homepage, https://github.com/anhvth/speedy
 Project-URL: Repository, https://github.com/anhvth/speedy

speedy_utils-1.1.22/notebooks/test_multi_thread.ipynb

@@ -0,0 +1,82 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "4c7c0614",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from speedy_utils import *"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3725a7bb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def sleep_and_count(n):\n",
+    "    thread_id = n\n",
+    "    n = 10\n",
+    "    for i in range(n):\n",
+    "        time.sleep(2)\n",
+    "        print(f'[Thread {thread_id}] Count: {i+1}/{n}')\n",
+    "    return True\n",
+    "\n",
+    "inputs = list(range(10))\n",
+    "results = multi_thread(sleep_and_count, inputs, workers=4)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ace1dc5c",
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "NameError",
+     "evalue": "name 'results' is not defined",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[31m---------------------------------------------------------------------------\u001b[39m",
+      "\u001b[31mNameError\u001b[39m                                 Traceback (most recent call last)",
+      "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[15]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[32m----> \u001b[39m\u001b[32m1\u001b[39m \u001b[43mresults\u001b[49m\n",
+      "\u001b[31mNameError\u001b[39m: name 'results' is not defined"
+     ]
+    }
+   ],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b98f317b",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.13.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

{speedy_utils-1.1.20 → speedy_utils-1.1.22}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "speedy-utils"
-version = "1.1.20"
+version = "1.1.22"
 description = "Fast and easy-to-use package for data science"
 authors = [{ name = "AnhVTH", email = "anhvth.226@gmail.com" }]
 readme = "README.md"

{speedy_utils-1.1.20 → speedy_utils-1.1.22}/src/llm_utils/__init__.py

@@ -6,6 +6,15 @@ from llm_utils.lm.base_prompt_builder import BasePromptBuilder
 
 LLM = LLMTask
 
+# Convenience functions for killing VLLM servers
+def kill_all_vllm() -> int:
+    """Kill all tracked VLLM server processes. Returns number of processes killed."""
+    return LLMTask.kill_all_vllm()
+
+def kill_vllm_on_port(port: int) -> bool:
+    """Kill VLLM server on specific port. Returns True if server was killed."""
+    return LLMTask.kill_vllm_on_port(port)
+
 from .chat_format import (
     build_chatml_input,
     display_chat_messages_as_html,
@@ -35,5 +44,7 @@ __all__ = [
     "get_model_name",
     "VectorCache",
     "BasePromptBuilder",
-    "LLM"
+    "LLM",
+    "kill_all_vllm",
+    "kill_vllm_on_port"
 ]