{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain_community.document_loaders import PyPDFLoader\n",
    "loader = PyPDFLoader(\"/content/Example SOP (1) (1).pdf\")\n",
    "docs = loader.load()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'SOPsResponse' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[1], line 22\u001b[0m\n\u001b[1;32m     19\u001b[0m \u001b[38;5;28;01mclass\u001b[39;00m \u001b[38;5;21;01mRoles_response\u001b[39;00m(BaseModel):\n\u001b[1;32m     20\u001b[0m     roles: \u001b[38;5;28mlist\u001b[39m[\u001b[38;5;28mstr\u001b[39m]\n\u001b[0;32m---> 22\u001b[0m \u001b[38;5;28;01mclass\u001b[39;00m \u001b[38;5;21;01mSopGenerator\u001b[39;00m:\n\u001b[1;32m     23\u001b[0m     \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__init__\u001b[39m(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m     24\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mapi_key \u001b[38;5;241m=\u001b[39m os\u001b[38;5;241m.\u001b[39mgetenv(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mOPENAI_API_KEY\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "Cell \u001b[0;32mIn[1], line 89\u001b[0m, in \u001b[0;36mSopGenerator\u001b[0;34m()\u001b[0m\n\u001b[1;32m     60\u001b[0m   response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mclient\u001b[38;5;241m.\u001b[39mbeta\u001b[38;5;241m.\u001b[39mchat\u001b[38;5;241m.\u001b[39mcompletions\u001b[38;5;241m.\u001b[39mparse(\n\u001b[1;32m     61\u001b[0m             model\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel,\n\u001b[1;32m     62\u001b[0m             messages\u001b[38;5;241m=\u001b[39m[\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     83\u001b[0m     temperature\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0.1\u001b[39m\n\u001b[1;32m     84\u001b[0m     )\n\u001b[1;32m     86\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m json\u001b[38;5;241m.\u001b[39mloads(response\u001b[38;5;241m.\u001b[39mchoices[\u001b[38;5;241m0\u001b[39m]\u001b[38;5;241m.\u001b[39mmessage\u001b[38;5;241m.\u001b[39mcontent)\n\u001b[0;32m---> 89\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mgenerate_sops\u001b[39m(\u001b[38;5;28mself\u001b[39m, roles, docs_text) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m \u001b[43mSOPsResponse\u001b[49m:\n\u001b[1;32m     90\u001b[0m     roles_sops_all \u001b[38;5;241m=\u001b[39m {}\n\u001b[1;32m     92\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m role \u001b[38;5;129;01min\u001b[39;00m roles:\n",
      "\u001b[0;31mNameError\u001b[0m: name 'SOPsResponse' is not defined"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import json\n",
    "from openai import OpenAI\n",
    "from pydantic import BaseModel, Field\n",
    "from typing import List, Dict, Optional\n",
    "\n",
    "class SOPs(BaseModel):\n",
    "    must: Optional[List[str]] = Field(default_factory=list)\n",
    "    shall: Optional[List[str]] = Field(default_factory=list)\n",
    "    will: Optional[List[str]] = Field(default_factory=list)\n",
    "\n",
    "class RoleSOPs(BaseModel):\n",
    "    sops: SOPs\n",
    "\n",
    "class SOPsFound(BaseModel):\n",
    "    message:str\n",
    "    status:bool\n",
    "\n",
    "class Roles_response(BaseModel):\n",
    "    roles: list[str]\n",
    "    \n",
    "class SOPsResponse(BaseModel):\n",
    "    roles_sops: Dict[str, SOPs] = Field(default_factory=dict)\n",
    "class SopGenerator:\n",
    "    def __init__(self):\n",
    "        self.api_key = os.getenv(\"OPENAI_API_KEY\")\n",
    "        self.client = OpenAI(api_key=self.api_key)\n",
    "        self.model = \"gpt-4o-mini\"\n",
    "\n",
    "    def get_roles(self, docs):\n",
    "        # Extract the text content from the Document objects\n",
    "        docs_text = [doc.page_content for doc in docs] \n",
    "        response = self.client.beta.chat.completions.parse(\n",
    "        model=self.model,\n",
    "        messages=[\n",
    "            {\n",
    "                \"role\": \"system\",\n",
    "                \"content\": '''Suppose you are a role/postion extractor from a company document ,\n",
    "                 you extract the roles as a list e.g[\"finacial analyist,\"data scientist]... etc\n",
    "                 if no roles are found return and empty list''',\n",
    "            },\n",
    "            {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": [\n",
    "                    {\n",
    "                        \"type\": \"text\", # Changed from \"document chunk\" to \"text\"\n",
    "                        \"text\": text\n",
    "                    } for text in docs_text\n",
    "                ]\n",
    "            }\n",
    "        ],\n",
    "        response_format=Roles_response,\n",
    "        max_tokens=1024,\n",
    "        temperature=0.1\n",
    "        )\n",
    "    \n",
    "        return json.loads(response.choices[0].message.content)\n",
    "    \n",
    "    def check_role_sop(self, role, docs):\n",
    "\n",
    "      docs_text = [doc.page_content for doc in docs] \n",
    "      response = self.client.beta.chat.completions.parse(\n",
    "                model=self.model,\n",
    "                messages=[\n",
    "                    {\n",
    "                        \"role\": \"system\",\n",
    "                        \"content\": f'''Tour roles is to check if the sops for the provided roles: {roles} are found in the document\n",
    "                        i.e You are validating the document can provide the sops \n",
    "                        you retur status = True if yes and a proper message and status=False and a proper message if no\n",
    "                        keep the message short e.g \"sops found for the roles: roles..\n",
    "                        '''\n",
    "                    },\n",
    "                    {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": [\n",
    "                    {\n",
    "                        \"type\": \"text\", # Changed from \"document chunk\" to \"text\"\n",
    "                        \"text\": text\n",
    "                    } for text in docs_text\n",
    "                ]\n",
    "            }\n",
    "        ],\n",
    "        response_format=SOPsFound,\n",
    "        max_tokens=1024,\n",
    "        temperature=0.1\n",
    "        )\n",
    "      \n",
    "      return json.loads(response.choices[0].message.content)\n",
    "\n",
    "\n",
    "    def generate_sops(self, roles, docs_text) -> SOPsResponse:\n",
    "        roles_sops_all = {}\n",
    "\n",
    "        for role in roles:\n",
    "            docs_text = [doc.page_content for doc in docs] \n",
    "            response = self.client.beta.chat.completions.parse(\n",
    "                model=self.model,\n",
    "                messages=[\n",
    "                    {\n",
    "                        \"role\": \"system\",\n",
    "                        \"content\": f'''You are a Standard Operating Procedure (SOP) extractor. Your task is to find SOPs for the role \"{role}\" in the provided text. '\n",
    "                            'SOPs should be categorized under \"must\", \"shall\", and \"will\". '\n",
    "                            'If no SOPs are found for the role, return an empty list for each category'''\n",
    "                    },\n",
    "                    {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": [\n",
    "                    {\n",
    "                        \"type\": \"text\", # Changed from \"document chunk\" to \"text\"\n",
    "                        \"text\": text\n",
    "                    } for text in docs_text\n",
    "                ]\n",
    "            }\n",
    "        ],\n",
    "        response_format=RoleSOPs,\n",
    "        max_tokens=1024,\n",
    "        temperature=0.1\n",
    "        )\n",
    "            role_sop = json.loads(response.choices[0].message.content)\n",
    "            roles_sops_all[role] = role_sop\n",
    "            \n",
    "        return roles_sops_all\n",
    "           \n",
    "# Example usage:\n",
    "service = SopGenerator()\n",
    "roles = [\"Devops engineers\"]\n",
    "sops_response = service.check_role_sop(roles,docs)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "erp",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}