7effrey89

using Azure;
using Azure.AI.Translation.Document;

/// <summary>
/// This program demonstrates how to use Azure's Document Translation service to translate documents
/// stored in an Azure Blob Storage container from one language to another.
/// It authenticates using a custom endpoint and resource key, then submits a translation job,
/// waits for completion, and prints the status and results for each document.
/// https://learn.microsoft.com/en-us/azure/ai-services/translator/document-translation/how-to-guides/use-rest-api-programmatically?tabs=csharp
/// https://learn.microsoft.com/en-us/dotnet/api/overview/azure/ai.translation.document-readme?view=azure-dotnet

7effrey89

# Script to create searchable PDF from scan PDF or images using Azure Form Recognizer / Document Intelligence
#
# USAGE:
#   python Untitled-1.py <input_file> [-o <output_file>]
#
# ARGUMENTS:
#   <input_file>   Path to your PDF or image file (jpg, jpeg, tif, tiff, bmp, png).
#   -o <output_file> (optional) Name for the output searchable PDF.
#                    If omitted, output will be <input_file>.ocr.pdf
#

7effrey89

from azure.storage.filedatalake import DataLakeServiceClient
from azure.identity import ClientSecretCredential

# Service principal credentials
app = "<app Registration Name>"    # App Registration Name
tenant_id = '<tenant_id>'          # Directory (tenant) ID
client_id = '<client_id>'          # Application (client) ID
client_secret = '<client_secret>'  # Client secret

# Storage account name

7effrey89

trigger:
  branches:
    include:
      - main

variables:
  resourceGroupName: '<Name of Resource Group>' # My Resource Group Name
  dataFactoryName: '<Name of ADF instance>' # My Data Factory Name
  pipelineName: '<Name of Pipeline>' # My Data Factory Pipeline Name

7effrey89

trigger:
  branches:
    include:
      - main

variables:
  pWorkspaceId: '<GUID OF FABRIC WORKSPACE>' #My Fabric Workspace ID
  pPipelineId: '<GUID OF FABRIC PIPELINE IN WORKSPACE>' #My Fabric Pipeline ID

stages:

7effrey89

# %%
!pip install mltable

# %%
# AML to query delta table in lakehouse to pandas dataframe
from mltable import from_delta_lake

#url = "https://onelake.dfs.fabric.microsoft.com/ws_ramboll_data/bronze.Lakehouse/Tables/taxidatacsv"
abfs = "abfss://ws_ramboll_data@onelake.dfs.fabric.microsoft.com/bronze.Lakehouse/Tables/taxidatacsv"
df2 = from_delta_lake(abfs).to_pandas_dataframe()

7effrey89

7effrey89

import os
import requests
import json

################################## Facts ######################################################
#GPT costs for GPT-4o and GPT-4o-Mini for Sweden Central per 1000 tokens
#https://azure.microsoft.com/en-us/pricing/calculator
GPT4O_GLOBAL_DEPLOYMENT_INPUT = 0.0025
GPT4O_GLOBAL_DEPLOYMENT_OUTPUT = 0.0100

7effrey89

# Pre-requisites:
# 1. Create a service princple in Azure AD
# 2. Create a new client secret for it (note down the client-id, tenant-id, and secret-value)
# 3. Go to the Azure portal and find the Translator service among the Azure AI Services.
# 4. In the Access Control (IAM) tab, assign the service principle the "Cognitive Services User" role
# 5. In the Resource management tab, enable a System assigned Identity. 
# 6. under Azure Role Assignment, choose scope: Storage, Resource: your blob storage (where your documents reside), Role: "Storage Blob Data Contributor" 
# 7- Run the code below to get the service to translate all documents in one container to another container

from azure.core.credentials import AccessToken, AzureKeyCredential, TokenCredential

7effrey89

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.edge.service import Service
from selenium.webdriver.edge.options import Options as EdgeOptions
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC

#selenium 4.23.1 was used for this code
#edge browser: Version 127.0.2651.105 (Official build) (64-bit)
#edge driver: Version 127.0.2651.105 (Stable Channel) (64-bit)