Facilities Engineer

Job Summary : #src / main.py

import os

import json

import re

import PyPDF2

import docx

import requests

import pdfplumber

import docx

import hashlib

import io

import zipfile

import xml.etree.ElementTree as ET

import openpyxl

import hashlib

import unicodedata

import regex

from typing import Pattern

from flask import Flask, jsonify, request

from google.cloud import firestore

from flask_cors import CORS

from datetime import datetime, timezone

from dotenv import load_dotenv

import firebase_admin

from firebase_admin import credentials, firestore, storage

from bs4 import BeautifulSoup # for URL text extraction

import PyPDF2

import unicodedata, regex, uuid

from crew_ai.document_crew import run_analysis

import asyncio

from concurrent.futures import ThreadPoolExecutor

import uuid

from urllib.parse import unquote

from threading import Thread

from itertools import zip_longest

from crewai import Crew

from crew_ai.agent import capture

from crew_ai.document_task import capture_task

from crew_ai.document_task import extract_orgname # adjust import if needed

import regex

import unicodedata

from typing import Union

import unicodedata

# Load environment variables

load_dotenv()

# Initialize Flask app

app = Flask(__name__) # MUST come first

# Configure CORS

allowed_origins = os.getenv("ALLOWED_ORIGINS", "

• ") # e.g. "http : / / localhost : 3000"

CORS(app, resources={r" /

service_account_info = json.loads(os.getenv("FIREBASE_SERVICE_ACCOUNT"))

cred = credentials.Certificate(service_account_info)

firebase_admin.initialize_app(cred, {

"storageBucket" : service_account_info.get("storageBucket")

})

db = firestore.client()

bucket = storage.bucket()

print(f" Firebase initialized with bucket : {bucket.name}")

# Azure OpenAI Setup

AZURE_API_KEY = os.getenv("AZURE_OPENAI_API_KEY")

AZURE_API_BASE = os.getenv("AZURE_OPENAI_API_BASE")

AZURE_API_VERSION = os.getenv("AZURE_OPENAI_API_VERSION")

AZURE_DEPLOYMENT_NAME = os.getenv("AZURE_OPENAI_DEPLOYMENT_NAME")

AZURE_API_URL = f"{AZURE_API_BASE} / openai / deployments / {AZURE_DEPLOYMENT_NAME} / chat / completions?api-version={AZURE_API_VERSION}"

def extract_pdf_text(content) :

text = ""

try :

with pdfplumber.open(io.BytesIO(content)) as pdf :

for page in pdf.pages :

page_text = page.extract_text()

if page_text :

text += page_text + "

except Exception as e :

print(" PDF parsing error : " + str(e))

return text

def extract_docx_text(content) :

text = ""

try :

doc = docx.Document(io.BytesIO(content))

text = "

".join([p.text for p in doc.paragraphs])

for table in doc.tables :

for row in table.rows :

text += "

" + " ".join(cell.text for cell in row.cells)

except Exception as e :

print(" DOCX parsing error : " + str(e))

return text

def extract_txt_text(content) :

try :

return content.decode("utf-8", errors="ignore")

except Exception as e :

print(" TXT decoding error : " + str(e))

return ""

def extract_text_by_type(filename, content) :

filename_lower = filename.lower()

if filename_lower.endswith(".pdf") :

return extract_pdf_text(content)

if filename_lower.endswith(".docx") :

return extract_docx_text(content)

if filename_lower.endswith(".txt") :

return extract_txt_text(content)

if filename_lower.endswith(".pptx") :

return extract_pptx_text(content)

if filename_lower.endswith(".xlsx") :

return extract_xlsx_text(content)

print(f" Unsupported file type : {filename}")

return ""

def get_files_from_firebase_folder(folder_name) :

"""

Reads all files from Firebase Storage folder, extracts text, and returns as a single string.

"""

texts = []

try :

blobs = list(bucket.list_blobs(prefix=f"Upload / {folder_name} / "))

print(f"