{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "53ba4306",
   "metadata": {},
   "source": [
    "AWL Human-in-the-Loop Workflow Demo - [Open this notebook in JupyterLite](https://repolab.github.io/jupyterlite-playground/lab/index.html?fromURL=https://raw.githubusercontent.com/OO-LD/awl-python/refs/heads/main/examples/human_in_the_loop_async.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aba1eb16",
   "metadata": {},
   "source": [
    "Install required packages. In a pyodide/jupyterlite environment this may take ~30 seconds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6427c868",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "if sys.platform == 'emscripten':\n",
    "    # see: https://github.com/pyodide/pyodide/issues/5834\n",
    "    import micropip\n",
    "    micropip.uninstall(['typing-extensions'])\n",
    "%pip install 'bokeh==3.7.3' 'jupyter_bokeh' 'panel==1.7.5' 'awl[hitl]'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44cc4e76",
   "metadata": {},
   "source": [
    "Define data models and a workflow. Functions / steps decorated with hitl (human in the loop) will prompt the user for input by an auto-generated UI (see also [OO-LD UI Demo](https://github.com/OO-LD/oold-python/blob/main/examples/linked_data_editor.ipynb)).\n",
    "\n",
    "*Note: async workflow only required for pyodide/jupyterlite. In a regular jupyter or python environment you can also use sync code, see [here](https://github.com/OO-LD/awl-python/blob/main/examples/human_in_the_loop.ipynb)* "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1705973e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from enum import Enum\n",
    "import json\n",
    "from typing import Any, Optional\n",
    "from oold.model import LinkedBaseModel\n",
    "from pydantic import Field\n",
    "from awl.hitl import entry_point, hitl\n",
    "\n",
    "class MachineParams(LinkedBaseModel):\n",
    "    \"\"\"Parameters transferred to the machine\"\"\"\n",
    "    model_config = {\n",
    "        \"json_schema_extra\": {\n",
    "            \"required\": [\"param1\"],\n",
    "        },\n",
    "    }\n",
    "    param1: int = Field(50, ge=0, le=100)\n",
    "\n",
    "class Quality(str, Enum):\n",
    "    good = \"Good\"\n",
    "    bad = \"Bad\"\n",
    "    worse = \"Worse\"\n",
    "\n",
    "class ProcessDocumentation(LinkedBaseModel):\n",
    "    \"\"\"Visual result inspection\"\"\"\n",
    "    quality: Quality\n",
    "    \"\"\"Good is defined as...\"\"\"\n",
    "\n",
    "class YesNoEnum(Enum):\n",
    "    yes = True\n",
    "    no = False\n",
    "\n",
    "class YesNo(LinkedBaseModel):\n",
    "    answer: bool # YesNoEnum\n",
    "\n",
    "@hitl\n",
    "async def set_machine_params(params: MachineParams):\n",
    "    # transfer params to machine\n",
    "    return params\n",
    "\n",
    "@hitl\n",
    "async def document_result(params: ProcessDocumentation):\n",
    "    # validate documentation\n",
    "    return params\n",
    "\n",
    "@hitl\n",
    "async def document_more(params: YesNo) -> YesNo:\n",
    "    return params\n",
    "\n",
    "def archive_data(params: Any):\n",
    "    # store documentation in database\n",
    "    pass\n",
    "\n",
    "@entry_point(gui=True, jupyter=True)\n",
    "async def workflow():\n",
    "    print(\"Enter\")\n",
    "    machine_params = await set_machine_params() # prompts user\n",
    "    do_document_more = True\n",
    "    while(do_document_more):\n",
    "        result_evaluation = await document_result() # prompts user\n",
    "        do_document_more = (await document_more()).answer\n",
    "        \n",
    "    print(\"Machine parameters: \", machine_params)\n",
    "    print(\"Result evaluation: \", result_evaluation)\n",
    "    archive_data(machine_params) # runs automatically\n",
    "    archive_data(result_evaluation)  # runs automatically\n",
    "    \n",
    "await workflow()\n",
    "print(\"Done\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}