Building an Azure Function using Python (Crossover between Reality Stone & Time Stone in Python Verse)

Hi Guys!

Today, we’ll be discussing a preview features from Microsoft Azure. Building an Azure function using Python on it’s Linux/Ubuntu VM. Since this is a preview feature, we cannot implement this to production till now. However, my example definitely has more detailed steps & complete code guide compared to whatever available over the internet.

In this post, I will take one of my old posts & enhance it as per this post. Hence, I’ll post those modified scripts. However, I won’t discuss the logic in details as most of these scripts have cosmetic changes to cater to this requirement.

In this post, we’ll only show Ubuntu run & there won’t be Windows or MAC comparison.

Initial Environment Preparation:

  1. Set-up new virtual machine on Azure.
  2. Set-up Azure function environments on that server.

Set-up new virtual machine on Azure:

I’m not going into the details of how to create Ubuntu VM on Microsoft Azure. You can refer the steps in more information here.

After successful creation, the VM will look like this –

Azure VM - Ubuntu

Detailed information you can get after clicking this hyperlink over the name of the VM.

Azure-VM Basic Details

You have to open port 7071 for application testing from the local using postman.

You can get it from the network option under VM as follows –

Network-Configuration

Make sure that you are restricting these ports to specific network & not open to ALL traffic.

So, your VM is ready now.

To update Azure CLI, you need to use the following commands –

sudo apt-get update && sudo apt-get install –only-upgrade -y azure-cli

Set-up Azure function environments on that server:

To set-up the environment, you don’t have to go for Python installation as by default Ubuntu in Microsoft Azure comes up with desired Python version, i.e., Python3.6. However, to run the python application, you need to install the following app –

  1. Microsoft SDK. You will get the details from this link.
  2. Installing node-js. You will get the details from this link.
  3. You need to install a docker. However, as per Microsoft official version, this is not required. But, you can create a Docker container to distribute the python function in Azure application. I would say you can install this just in case if you want to continue with this approach. You will get the details over here. If you want to know details about the Docker. And, how you want to integrate python application. You can refer to this link.
  4. Your desired python packages. In this case, we’ll be modifying this post – “Encryption/Decryption, JSON, API, Flask Framework in Python (Crossover between Reality Stone & Time Stone in Python Verse).” We’ll be modifying a couple of lines only to cater to this functionality & deploying the same as an Azure function.
  5. Creating an Azure function template on Ubuntu. The essential detail you’ll get it from here. However, over there, it was not shown in detailed steps of python packages & how you can add all the dependencies to publish it in details. It was an excellent post to start-up your knowledge.

Let’s see these components status & very brief details –

Microsoft SDK:

To check the dot net version. You need to type the following commands in Ubuntu –

dotnet –info

And, the output will look like this –

DotNet-Version

Node-Js:

Following is the way to verify your node-js version & details –

node -v

npm -v

And, the output looks like this –

Node-Js

Docker:

Following is the way to test your docker version –

docker -v

And, the output will look like this –

Docker-Version

Python Packages:

Following are the python packages that we need to run & publish that in Azure cloud as an Azure function –

pip freeze | grep -v “pkg-resources” > requirements.txt

And, the output is –

Requirements

You must be wondered that why have I used this grep commands here. I’ve witnessed that on many occassion in Microsoft Azure’s Linux VM it produces one broken package called resource=0.0.0, which will terminate the deployment process. Hence, this is very crucial to eliminate those broken packages.

Now, we’re ready for our python scripts. But, before that, let’s see the directory structure over here –

Win_Vs_Ubuntu-Cloud

Creating an Azure Function Template on Ubuntu: 

Before we post our python scripts, we’ll create these following components, which is essential for our Python-based Azure function –

  • Creating a group:

              Creating a group either through Azure CLI or using a docker, you can proceed. The commands for Azure CLI is as follows –

az group create –name “rndWestUSGrp” –location westus

It is advisable to use double quotes for parameters value. Otherwise, you might land-up getting the following error – “Error: “resourceGroupName” should satisfy the constraint – “Pattern”: /^[-w._]+$/“.

I’m sure. You don’t want to face that again. And, here is the output –

CreateDeploymentGroup

Note that, here I haven’t used the double-quotes. But, to avoid any unforeseen issues – you should use double-quotes. You can refer the docker command from the above link, which I’ve shared earlier.

Now, you need to create one storage account where the metadata information of your function will be stored. You will create that as follows –

az storage account create –name cryptpy2019 –location westus –resource-group rndWestUSGrp –sku Standard_LRS

And, the output will look like this –

AccountCreate_1

Great. Now, we’ll create a virtual environment for Python3.6.

python3.6 -m venv .env
source .env/bin/activate

Python-VM

Now, we’ll create a local function project.

func init encPro

And, the output you will get is as follows –

Local-Function

Inside this directory, you’ll see the following files –

Local-Function-Details

You need to edit the host.json with these default lines –

{
 “version”: “2.0”,
 “extensionBundle”: {
                                       “id”: “Microsoft.Azure.Functions.ExtensionBundle”,
                                       “version”: “[1.*, 2.0.0)”
                                     }
}

And, the final content of these two files (excluding the requirements.txt) will look like this –

Configuration

Finally, we’ll create the template function by this following command –

func new

This will follow with steps finish it. You need to choose Python as your programing language. You need to choose an HTTP trigger template. Once you created that successfully, you’ll see the following files –

func_New

Note that, our initial function name is -> getVal.

By default, Azure will generate some default code inside the __init__.py. The details of those two files can be found here.

Since we’re ready with our environment setup. We can now discuss our Python scripts –

1. clsConfigServer.py (This script contains all the parameters of the server.)

###########################################
#### Written By: SATYAKI DE        ########
#### Written On: 10-Feb-2019       ########
####                               ########
#### Objective: Parameter File     ########
###########################################

import os
import platform as pl

# Checking with O/S system
os_det = pl.system()

class clsConfigServer(object):
    Curr_Path = os.path.dirname(os.path.realpath(__file__))

    if os_det == "Windows":
        config = {
            'FILE': 'acct_addr_20180112.csv',
            'SRC_FILE_PATH': Curr_Path + '\\' + 'src_file\\',
            'PROFILE_FILE_PATH': Curr_Path + '\\' + 'profile\\',
            'HOST_IP_ADDR': '0.0.0.0',
            'DEF_SALT': 'iooquzKtqLwUwXG3rModqj_fIl409vemWg9PekcKh2o=',
            'ACCT_NBR_SALT': 'iooquzKtqLwUwXG3rModqj_fIlpp1vemWg9PekcKh2o=',
            'NAME_SALT': 'iooquzKtqLwUwXG3rModqj_fIlpp1026Wg9PekcKh2o=',
            'PHONE_SALT': 'iooquzKtqLwUwXG3rMM0F5_fIlpp1026Wg9PekcKh2o=',
            'EMAIL_SALT': 'iooquzKtqLwU0653rMM0F5_fIlpp1026Wg9PekcKh2o='
        }
    else:
        config = {
            'FILE': 'acct_addr_20180112.csv',
            'SRC_FILE_PATH': Curr_Path + '/' + 'src_file/',
            'PROFILE_FILE_PATH': Curr_Path + '/' + 'profile/',
            'HOST_IP_ADDR': '0.0.0.0',
            'DEF_SALT': 'iooquzKtqLwUwXG3rModqj_fIl409vemWg9PekcKh2o=',
            'ACCT_NBR_SALT': 'iooquzKtqLwUwXG3rModqj_fIlpp1vemWg9PekcKh2o=',
            'NAME_SALT': 'iooquzKtqLwUwXG3rModqj_fIlpp1026Wg9PekcKh2o=',
            'PHONE_SALT': 'iooquzKtqLwUwXG3rMM0F5_fIlpp1026Wg9PekcKh2o=',
            'EMAIL_SALT': 'iooquzKtqLwU0653rMM0F5_fIlpp1026Wg9PekcKh2o='
        }

2. clsEnDec.py (This script is a lighter version of encryption & decryption of our previously discussed scenario. Hence, we won’t discuss in details. You can refer my earlier post to understand the logic of this script.)

###########################################
#### Written By: SATYAKI DE        ########
#### Written On: 25-Jan-2019       ########
#### Package Cryptography needs to ########
#### install in order to run this  ########
#### script.                       ########
####                               ########
#### Objective: This script will   ########
#### encrypt/decrypt based on the  ########
#### hidden supplied salt value.   ########
###########################################

from cryptography.fernet import Fernet
import logging

from getVal.clsConfigServer import clsConfigServer as csf

class clsEnDec(object):

    def __init__(self):
        # Calculating Key
        self.token = str(csf.config['DEF_SALT'])

    def encrypt_str(self, data, token):
        try:
            # Capturing the Salt Information
            t1 = self.token
            t2 = token

            if t2 == '':
                salt = t1
            else:
                salt = t2

            logging.info("Encrypting the value!")

            # Checking Individual Types inside the Dataframe
            cipher = Fernet(salt)
            encr_val = str(cipher.encrypt(bytes(data,'utf8'))).replace("b'","").replace("'","")

            strV1 = "Encrypted value:: " + str(encr_val)
            logging.info(strV1)

            return encr_val

        except Exception as e:
            x = str(e)
            print(x)
            encr_val = ''

            return encr_val

    def decrypt_str(self, data, token):
        try:
            # Capturing the Salt Information
            t1 = self.token
            t2 = token

            if t2 == '':
                salt = t1
            else:
                salt = t2

            logging.info("Decrypting the value!")

            # Checking Individual Types inside the Dataframe
            cipher = Fernet(salt)
            decr_val = str(cipher.decrypt(bytes(data,'utf8'))).replace("b'","").replace("'","")

            strV2 = "Decrypted value:: " + str(decr_val)
            logging.info(strV2)

            return decr_val

        except Exception as e:
            x = str(e)
            print(x)
            decr_val = ''

            return decr_val

3. clsFlask.py (This is the main server script that will the encrypt/decrypt class from our previous scenario. This script will capture the requested JSON from the client, who posted from the clients like another python script or third-party tools like Postman.)

###########################################
#### Written By: SATYAKI DE            ####
#### Written On: 25-Jan-2019           ####
#### Package Flask package needs to    ####
#### install in order to run this      ####
#### script.                           ####
####                                   ####
#### Objective: This script will       ####
#### encrypt/decrypt based on the      ####
#### supplied salt value. Also,        ####
#### this will capture the individual  ####
#### element & stored them into JSON   ####
#### variables using flask framework.  ####
###########################################

from getVal.clsConfigServer import clsConfigServer as csf
from getVal.clsEnDec import clsEnDecAuth

getVal = clsEnDec()

import logging

class clsFlask(object):
    def __init__(self):
        self.xtoken = str(csf.config['DEF_SALT'])

    def getEncryptProcess(self, dGroup, input_data, dTemplate):
        try:
            # It is sending default salt value
            xtoken = self.xtoken

            # Capturing the individual element
            dGroup = dGroup
            input_data = input_data
            dTemplate = dTemplate

            # This will check the mandatory json elements
            if ((dGroup != '') & (dTemplate != '')):

                # Based on the Group & Element it will fetch the salt
                # Based on the specific salt it will encrypt the data
                if ((dGroup == 'GrDet') & (dTemplate == 'subGrAcct_Nbr')):
                    xtoken = str(csf.config['ACCT_NBR_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.encrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrName')):
                    xtoken = str(csf.config['NAME_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.encrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrPhone')):
                    xtoken = str(csf.config['PHONE_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.encrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrEmail')):
                    xtoken = str(csf.config['EMAIL_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.encrypt_str(input_data, xtoken)
                else:
                    ret_val = ''
            else:
                ret_val = ''

            # Return value
            return ret_val

        except Exception as e:
            ret_val = ''
            # Return the valid json Error Response
            return ret_val

    def getDecryptProcess(self, dGroup, input_data, dTemplate):
        try:
            xtoken = self.xtoken

            # Capturing the individual element
            dGroup = dGroup
            input_data = input_data
            dTemplate = dTemplate

            # This will check the mandatory json elements
            if ((dGroup != '') & (dTemplate != '')):

                # Based on the Group & Element it will fetch the salt
                # Based on the specific salt it will decrypt the data
                if ((dGroup == 'GrDet') & (dTemplate == 'subGrAcct_Nbr')):
                    xtoken = str(csf.config['ACCT_NBR_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.decrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrName')):
                    xtoken = str(csf.config['NAME_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.decrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrPhone')):
                    xtoken = str(csf.config['PHONE_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.decrypt_str(input_data, xtoken)
                elif ((dGroup == 'GrDet') & (dTemplate == 'subGrEmail')):
                    xtoken = str(csf.config['EMAIL_SALT'])

                    strV1 = "xtoken: " + str(xtoken)
                    logging.info(strV1)
                    strV2 = "Flask Input Data: " + str(input_data)
                    logging.info(strV2)

                    #x = cen.clsEnDecAuth()
                    ret_val = getVal.decrypt_str(input_data, xtoken)
                else:
                    ret_val = ''
            else:
                ret_val = ''

            # Return value
            return ret_val

        except Exception as e:
            ret_val = ''
            # Return the valid Error Response
            return ret_val

4. __init__.py (This autogenerated script contains the primary calling methods of encryption & decryption based on the element header & values after enhanced as per the functionality.)

###########################################
#### Written By: SATYAKI DE            ####
#### Written On: 08-Jun-2019           ####
#### Package Flask package needs to    ####
#### install in order to run this      ####
#### script.                           ####
####                                   ####
#### Objective: Main Calling scripts.  ####
#### This is an autogenrate scripts.   ####
#### However, to meet the functionality####
#### we've enhanced as per our logic.  ####
###########################################
__all__ = ['clsFlask']

import logging
import azure.functions as func
import json

from getVal.clsFlask import clsFlask

getVal = clsFlask()

def main(req: func.HttpRequest) -> func.HttpResponse:
    logging.info('Python Encryption function processed a request.')

    str_val = 'Input Payload:: ' + str(req.get_json())
    str_1 = str(req.get_json())

    logging.info(str_val)

    ret_val = {}
    DataIn = ''
    dGroup = ''
    dTemplate = ''
    flg = ''

    if (str_1 != ''):
        try:
            req_body = req.get_json()
            dGroup = req_body.get('dataGroup')

            try:
                DataIn = req_body.get('data')
                strV15 = 'If Part:: ' + str(DataIn)

                logging.info(strV15)

                if ((DataIn == '') | (DataIn == None)):
                    raise ValueError

                flg = 'Y'
            except ValueError:
                DataIn = req_body.get('edata')
                strV15 = 'Else Part:: ' + str(DataIn)
                logging.info(strV15)
                flg = 'N'
            except:
                DataIn = req_body.get('edata')
                strV15 = 'Else Part:: ' + str(DataIn)
                logging.info(strV15)
                flg = 'N'

            dTemplate = req_body.get('dataTemplate')

        except ValueError:
            pass

    strV5 = "Encrypt Decrypt Flag:: " + flg
    logging.info(strV5)

    if (flg == 'Y'):

        if ((DataIn != '') & ((dGroup != '') & (dTemplate != ''))):

            logging.info("Encryption Started!")
            ret_val = getVal.getEncryptProcess(dGroup, DataIn, dTemplate)
            strVal2 = 'Return Payload:: ' + str(ret_val)
            logging.info(strVal2)

            xval = json.dumps(ret_val)

            return func.HttpResponse(xval)
        else:
            return func.HttpResponse(
                 "Please pass a data in the request body",
                 status_code=400
            )
    else:

        if ((DataIn != '') & ((dGroup != '') & (dTemplate != ''))):

            logging.info("Decryption Started!")
            ret_val2 = getVal.getDecryptProcess(dGroup, DataIn, dTemplate)
            strVal3 = 'Return Payload:: ' + str(ret_val)
            logging.info(strVal3)

            xval1 = json.dumps(ret_val2)

            return func.HttpResponse(xval1)
        else:
            return func.HttpResponse(
                "Please pass a data in the request body",
                status_code=400
            )

In this script, based on the value of an flg variable, we’re calling our encryption or decryption methods. And, the value of the flg variable is set based on the following logic –

try:
    DataIn = req_body.get('data')
    strV15 = 'If Part:: ' + str(DataIn)

    logging.info(strV15)

    if ((DataIn == '') | (DataIn == None)):
        raise ValueError

    flg = 'Y'
except ValueError:
    DataIn = req_body.get('edata')
    strV15 = 'Else Part:: ' + str(DataIn)
    logging.info(strV15)
    flg = 'N'
except:
    DataIn = req_body.get('edata')
    strV15 = 'Else Part:: ' + str(DataIn)
    logging.info(strV15)
    flg = 'N'

So, if the application gets the “data” element then – it will consider the data needs to be encrypted; otherwise, it will go for decryption. And, based on that – it is setting the value.

Now, we’re ready to locally run our application –

func host start

And, the output will look like this –

StartingAzureFunction-Python
StartingAzureFunction-Python 2

Let’s test it from postman –

Encrypt:

Postman-Encrypt

Decrypt:

Postman-Decrypt

Great. Now, we’re ready to publish this application to Azure cloud.

As in our earlier steps, we’ve already built our storage account for the metadata. Please scroll to top to view that again. Now, using that information, we’ll make the function app with a more meaningful name –

az functionapp create –resource-group rndWestUSGrp –os-type Linux \
–consumption-plan-location westus –runtime python \
–name getEncryptDecrypt –storage-account cryptpy2019

CreatingFunctionPython

Let’s publish the function –

sudo func azure functionapp publish “getEncryptDecrypt” –build-native-deps

On many occassion, without the use of “–build-native-deps” might leads to failure. Hence, I’ve added that to avoid such scenarios.

Publishing-Function

Now, we need to test our first published complex Azure function with Python through postman –

Encrypt:

PubishedFuncPostmanEncrypt

Decrypt:

PubishedFuncPostmanDecrypt

Wonderful! So, it is working.

You can see the function under the Azure portal –

Deployed-Function

Let’s see some other important features of this function –

Monitor: You can monitor two ways. One is by clicking the monitor options you will get the individual requests level details & also get to see the log information over here –

Function-Monitor-Details-1

Clicking Application Insights will give you another level of detailed logs, which can be very useful for debugging. We’ll touch this at the end of this post with a very brief discussion.

Function-Monitor-Details-3.JPG

As you can see, clicking individual lines will show the details further.

Let’s quickly check the application insights –

Application-Insights-1

Application Insights will give you a SQL like an interface where you can get the log details of all your requests.

Application-Insights-2

You can expand the individual details for further information.

Application-Insights-3

You can change the parameter name & other details & click the run button to get all the log details for your debugging purpose.

So, finally, we’ve achieved our goal. This is relatively long posts. But, I’m sure this will help you to create your first python-based function on the Azure platform.

Hope, you will like this approach. Let me know your comment on the same.

I’ll bring some more exciting topic in the coming days from the Python verse.

Till then, Happy Avenging! 😀

Note: All the data posted here are representational data & available over the internet.

Improvement of Pandas data processing performance using Multi-threading with the Queue (Another crossover of Space Stone, Reality Stone & Power Stone)

Today, we’ll discuss how to improve your panda’s data processing power using Multi-threading. Note that, we are not going to use any third party python package. Also, we’ll be using a couple of python scripts, which we’ve already discussed in our previous posts. Hence, this time, I won’t post them here.

Please refer the following scripts –

a. callClient.py
b. callRunServer.py
c. clsConfigServer.py
d. clsEnDec.py
e. clsFlask.py
f. clsL.py
g. clsParam.py
h. clsSerial.py
i. clsWeb.py

Please find the above scripts described here with details.

So, today, we’ll be looking into how the multi-threading really helps the application to gain some performance over others.

Let’s go through our existing old sample files –

Sample Data

And, we’ve four columns that are applicable for encryption. This file contains 10K records. That means the application will make 40K calls to the server for a different kind of encryption for each column.

Now, if you are going with the serial approach, which I’ve already discussed here, will take significant time for data processing. However, if we could club a few rows as one block & in this way we can create multiple blocks out of our data csv like this –

Data_Blocks

As you can see that blocks are marked with a different color. So, now if you send each block of data in parallel & send the data for encryption. Ideally, you will be able to process data much faster than the usual serial process. And, this what we would be looking for with the help of python’s multi-threading & queue. Without the queue, this program won’t be possible as the queue maintains the data & process integrity.

One more thing we would like to explain here. Whenever this application is sending the block of data. It will be posting that packed into a (key, value) dictionary randomly. Key will be the thread name. The reason, we’re not expecting data after process might arrive in some random order wrapped with the dictionary as well. Once the application received all the dictionary with dataframe with encrypted/decrypted data, the data will be rearranged based on the key & then joined back with the rest of the data.

Let’s see one sample way of sending & receiving random thread –

Data Packing

The left-hand side, the application is splitting the recordset into small chunks of a group. Once, those group created, using python multi-threading the application is now pushing them into the queue for the producer to produce the encrypted/decrypted value. Similar way, after processing the application will push the final product into the queue for consuming the final output.

This is the pictorial representation of dictionary ordering based on the key-value & then the application will extract the entire data to form the target csv file.

Final_Data_Sort

Let’s explore the script –

1. clsParallel.py (This script will consume the split csv files & send the data blocks in the form of the dictionary using multi-threading to the API for encryption in parallel. Hence, the name comes into the picture.)

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
import pandas as p
import clsWeb as cw
import datetime
from clsParam import clsParam as cf
import threading
from queue import Queue
import gc
import signal
import time
import os

# Declaring Global Variable
q = Queue()
m = Queue()
tLock = threading.Lock()
threads = []

fin_dict = {}
fin_dict_1 = {}
stopping = threading.Event()

# Disbling Warnings
def warn(*args, **kwargs):
    pass
import warnings
warnings.warn = warn

class clsParallel(object):
    def __init__(self):
        self.path = cf.config['PATH']
        self.EncryptMode = str(cf.config['ENCRYPT_MODE'])
        self.DecryptMode = str(cf.config['DECRYPT_MODE'])
        self.num_worker_threads = int(cf.config['NUM_OF_THREAD'])
        

    # Lookup Methods for Encryption
    def encrypt_acctNbr(self, row):
        # Declaring Local Variable
        en_AcctNbr = ''
        json_source_str = ''

        # Capturing essential values
        EncryptMode = self.EncryptMode
        lkp_acctNbr = row['Acct_Nbr']
        str_acct_nbr = str(lkp_acctNbr)
        fil_acct_nbr = str_acct_nbr.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_acct_nbr + '","dataTemplate":"subGrAcct_Nbr"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_acct_nbr)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            en_AcctNbr = x.getResponse(EncryptMode)
        else:
            en_AcctNbr = ''

        return en_AcctNbr

    def encrypt_Name(self, row):
        # Declaring Local Variable
        en_AcctName = ''

        # Capturing essential values
        EncryptMode = self.EncryptMode
        lkp_acctName = row['Name']
        str_acct_name = str(lkp_acctName)
        fil_acct_name = str_acct_name.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_acct_name + '","dataTemplate":"subGrName"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_acct_name)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            en_AcctName = x.getResponse(EncryptMode)
        else:
            en_AcctName = ''

        return en_AcctName

    def encrypt_Phone(self, row):
        # Declaring Local Variable
        en_Phone = ''

        # Capturing essential values
        EncryptMode = self.EncryptMode
        lkp_phone = row['Phone']
        str_phone = str(lkp_phone)
        fil_phone = str_phone.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_phone + '","dataTemplate":"subGrPhone"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_phone)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            en_Phone = x.getResponse(EncryptMode)
        else:
            en_Phone = ''

        return en_Phone

    def encrypt_Email(self, row):
        # Declaring Local Variable
        en_Email = ''

        # Capturing essential values
        EncryptMode = self.EncryptMode
        lkp_email = row['Email']
        str_email = str(lkp_email)
        fil_email = str_email.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_email + '","dataTemplate":"subGrEmail"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_email)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            en_Email = x.getResponse(EncryptMode)
        else:
            en_Email = ''

        return en_Email

    # Lookup Methods for Decryption
    def decrypt_acctNbr(self, row):
        # Declaring Local Variable
        de_AcctNbr = ''
        json_source_str = ''

        # Capturing essential values
        EncryptMode = self.DecryptMode
        lkp_acctNbr = row['Acct_Nbr']
        str_acct_nbr = str(lkp_acctNbr)
        fil_acct_nbr = str_acct_nbr.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_acct_nbr + '","dataTemplate":"subGrAcct_Nbr"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_acct_nbr)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            de_AcctNbr = x.getResponse(EncryptMode)
        else:
            de_AcctNbr = ''

        return de_AcctNbr

    def decrypt_Name(self, row):
        # Declaring Local Variable
        de_AcctName = ''

        # Capturing essential values
        EncryptMode = self.DecryptMode
        lkp_acctName = row['Name']
        str_acct_name = str(lkp_acctName)
        fil_acct_name = str_acct_name.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_acct_name + '","dataTemplate":"subGrName"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_acct_name)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            de_AcctName = x.getResponse(EncryptMode)
        else:
            de_AcctName = ''

        return de_AcctName

    def decrypt_Phone(self, row):
        # Declaring Local Variable
        de_Phone = ''

        # Capturing essential values
        EncryptMode = self.DecryptMode
        lkp_phone = row['Phone']
        str_phone = str(lkp_phone)
        fil_phone = str_phone.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_phone + '","dataTemplate":"subGrPhone"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_phone)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            de_Phone = x.getResponse(EncryptMode)
        else:
            de_Phone = ''

        return de_Phone

    def decrypt_Email(self, row):
        # Declaring Local Variable
        de_Email = ''

        # Capturing essential values
        EncryptMode = self.DecryptMode
        lkp_email = row['Email']
        str_email = str(lkp_email)
        fil_email = str_email.strip()

        # Forming JSON String for this field
        json_source_str = '{"dataGroup":"GrDet","data":"' + fil_email + '","dataTemplate":"subGrEmail"}'

        # Identifying Length of the field
        len_acct_nbr = len(fil_email)

        # This will trigger the service if it has valid data
        if len_acct_nbr > 0:
            x = cw.clsWeb(json_source_str)
            de_Email = x.getResponse(EncryptMode)
        else:
            de_Email = ''

        return de_Email

    def getEncrypt(self, df_dict):
        try:
            df_input = p.DataFrame()
            df_fin = p.DataFrame()

            # Assigning Target File Basic Name
            for k, v in df_dict.items():
                Thread_Name = k
                df_input = v

            # Checking total count of rows
            count_row = int(df_input.shape[0])
            # print('Part number of records to process:: ', count_row)

            if count_row > 0:

                # Deriving rows
                df_input['Encrypt_Acct_Nbr'] = df_input.apply(lambda row: self.encrypt_acctNbr(row), axis=1)
                df_input['Encrypt_Name'] = df_input.apply(lambda row: self.encrypt_Name(row), axis=1)
                df_input['Encrypt_Phone'] = df_input.apply(lambda row: self.encrypt_Phone(row), axis=1)
                df_input['Encrypt_Email'] = df_input.apply(lambda row: self.encrypt_Email(row), axis=1)

                # Dropping original columns
                df_input.drop(['Acct_Nbr', 'Name', 'Phone', 'Email'], axis=1, inplace=True)

                # Renaming new columns with the old column names
                df_input.rename(columns={'Encrypt_Acct_Nbr':'Acct_Nbr'}, inplace=True)
                df_input.rename(columns={'Encrypt_Name': 'Name'}, inplace=True)
                df_input.rename(columns={'Encrypt_Phone': 'Phone'}, inplace=True)
                df_input.rename(columns={'Encrypt_Email': 'Email'}, inplace=True)

                # New Column List Orders
                column_order = ['Acct_Nbr', 'Name', 'Acct_Addr_1', 'Acct_Addr_2', 'Phone', 'Email', 'Serial_No']
                df_fin = df_input.reindex(column_order, axis=1)

                fin_dict[Thread_Name] = df_fin

            return 0
        except Exception as e:
            df_error = p.DataFrame({'Acct_Nbr':str(e), 'Name':'', 'Acct_Addr_1':'', 'Acct_Addr_2':'', 'Phone':'', 'Email':'', 'Serial_No':''})
            fin_dict[Thread_Name] = df_error

            return 1

    def getEncryptWQ(self):
        item_dict = {}
        item = ''

        while True:
            try:
                #item_dict = q.get()
                item_dict = q.get_nowait()

                for k, v in item_dict.items():
                    # Assigning Target File Basic Name
                    item = str(k)

                if ((item == 'TEND') | (item == '')):
                    break

                if ((item != 'TEND') | (item != '')):
                    self.getEncrypt(item_dict)

                q.task_done()
            except Exception:
                break

    def getEncryptParallel(self, df_payload):
        start_pos = 0
        end_pos = 0
        l_dict = {}
        c_dict = {}
        min_val_list = {}
        cnt = 0
        num_worker_threads = self.num_worker_threads
        split_df = p.DataFrame()
        df_ret = p.DataFrame()

        # Assigning Target File Basic Name
        df_input = df_payload

        # Checking total count of rows
        count_row = df_input.shape[0]
        print('Total number of records to process:: ', count_row)

        interval = int(count_row / num_worker_threads) + 1
        actual_worker_task = int(count_row / interval) + 1

        for i in range(actual_worker_task):
            t = threading.Thread(target=self.getEncryptWQ)
            t.start()
            threads.append(t)
            name = str(t.getName())

            if ((start_pos + interval) < count_row):
                end_pos = start_pos + interval
            else:
                end_pos = start_pos + (count_row - start_pos)

            split_df = df_input.iloc[start_pos:end_pos]
            l_dict[name] = split_df

            if ((start_pos > count_row) | (start_pos == count_row)):
                break
            else:
                start_pos = start_pos + interval

            q.put(l_dict)
            cnt += 1

        # block until all tasks are done
        q.join()

        # stop workers
        for i in range(actual_worker_task):
            c_dict['TEND'] = p.DataFrame()
            q.put(c_dict)

        for t in threads:
            t.join()

        for k, v in fin_dict.items():
            min_val_list[int(k.replace('Thread-',''))] = v

        min_val = min(min_val_list, key=int)

        for k, v in sorted(fin_dict.items(), key=lambda k:int(k[0].replace('Thread-',''))):
            if int(k.replace('Thread-','')) == min_val:
                df_ret = fin_dict[k]
            else:
                d_frames = [df_ret, fin_dict[k]]
                df_ret = p.concat(d_frames)

        # Releasing Memory
        del[[split_df]]
        gc.collect()

        return df_ret

    def getDecrypt(self, df_encrypted_dict):
        try:
            df_input = p.DataFrame()
            df_fin = p.DataFrame()

            # Assigning Target File Basic Name
            for k, v in df_encrypted_dict.items():
                Thread_Name = k
                df_input = v

            # Checking total count of rows
            count_row = int(df_input.shape[0])

            if count_row > 0:

                # Deriving rows
                df_input['Decrypt_Acct_Nbr'] = df_input.apply(lambda row: self.decrypt_acctNbr(row), axis=1)
                df_input['Decrypt_Name'] = df_input.apply(lambda row: self.decrypt_Name(row), axis=1)
                df_input['Decrypt_Phone'] = df_input.apply(lambda row: self.decrypt_Phone(row), axis=1)
                df_input['Decrypt_Email'] = df_input.apply(lambda row: self.decrypt_Email(row), axis=1)

                # Dropping original columns
                df_input.drop(['Acct_Nbr', 'Name', 'Phone', 'Email'], axis=1, inplace=True)

                # Renaming new columns with the old column names
                df_input.rename(columns={'Decrypt_Acct_Nbr':'Acct_Nbr'}, inplace=True)
                df_input.rename(columns={'Decrypt_Name': 'Name'}, inplace=True)
                df_input.rename(columns={'Decrypt_Phone': 'Phone'}, inplace=True)
                df_input.rename(columns={'Decrypt_Email': 'Email'}, inplace=True)

                # New Column List Orders
                column_order = ['Acct_Nbr', 'Name', 'Acct_Addr_1', 'Acct_Addr_2', 'Phone', 'Email']
                df_fin = df_input.reindex(column_order, axis=1)

                fin_dict_1[Thread_Name] = df_fin

            return 0

        except Exception as e:
            df_error = p.DataFrame({'Acct_Nbr': str(e), 'Name': '', 'Acct_Addr_1': '', 'Acct_Addr_2': '', 'Phone': '', 'Email': ''})
            fin_dict_1[Thread_Name] = df_error

            return 1

    def getDecryptWQ(self):
        item_dict = {}
        item = ''

        while True:
            try:
                #item_dict = q.get()
                item_dict = m.get_nowait()

                for k, v in item_dict.items():
                    # Assigning Target File Basic Name
                    item = str(k)

                if ((item == 'TEND') | (item == '')):
                    return True
                    #break

                if ((item != 'TEND') | (item != '')):
                    self.getDecrypt(item_dict)

                m.task_done()
            except Exception:
                break


    def getDecryptParallel(self, df_payload):
        start_pos = 0
        end_pos = 0
        l_dict_1 = {}
        c_dict_1 = {}
        cnt = 0
        num_worker_threads = self.num_worker_threads
        split_df = p.DataFrame()
        df_ret_1 = p.DataFrame()

        min_val_list = {}

        # Assigning Target File Basic Name
        df_input_1 = df_payload

        # Checking total count of rows
        count_row = df_input_1.shape[0]
        print('Total number of records to process:: ', count_row)

        interval = int(count_row / num_worker_threads) + 1
        actual_worker_task = int(count_row / interval) + 1

        for i in range(actual_worker_task):
            t_1 = threading.Thread(target=self.getDecryptWQ)
            t_1.start()
            threads.append(t_1)
            name = str(t_1.getName())

            if ((start_pos + interval) < count_row):
                end_pos = start_pos + interval
            else:
                end_pos = start_pos + (count_row - start_pos)

            split_df = df_input_1.iloc[start_pos:end_pos]
            l_dict_1[name] = split_df

            if ((start_pos > count_row) | (start_pos == count_row)):
                break
            else:
                start_pos = start_pos + interval

            m.put(l_dict_1)
            cnt += 1

        # block until all tasks are done
        m.join()

        # stop workers
        for i in range(actual_worker_task):
            c_dict_1['TEND'] = p.DataFrame()
            m.put(c_dict_1)

        for t_1 in threads:
            t_1.join()

        for k, v in fin_dict_1.items():
            min_val_list[int(k.replace('Thread-',''))] = v

        min_val = min(min_val_list, key=int)

        for k, v in sorted(fin_dict_1.items(), key=lambda k:int(k[0].replace('Thread-',''))):
            if int(k.replace('Thread-','')) == min_val:
                df_ret_1 = fin_dict_1[k]
            else:
                d_frames = [df_ret_1, fin_dict_1[k]]
                df_ret_1 = p.concat(d_frames)

        # Releasing Memory
        del[[split_df]]
        gc.collect()

        return df_ret_1

Let’s explain the key snippet from the code. For your information, we’re not going to describe all the encryption methods such as –

# Encryption Method
encrypt_acctNbr

encrypt_Name
encrypt_Phone
encrypt_Email

# Decryption Method
decrypt_acctNbr
decrypt_Name
decrypt_Phone
decrypt_Email

As we’ve already described the logic of these methods in our previous post.

# Checking total count of rows
count_row = df_input.shape[0]
print('Total number of records to process:: ', count_row)

interval = int(count_row / num_worker_threads) + 1
actual_worker_task = int(count_row / interval) + 1

Fetching the total number of rows from the dataframe. Based on the row count, the application will derive the actual number of threads that will be used for parallelism.

for i in range(actual_worker_task):
    t = threading.Thread(target=self.getEncryptWQ)
    t.start()
    threads.append(t)
    name = str(t.getName())

    if ((start_pos + interval) < count_row):
        end_pos = start_pos + interval
    else:
        end_pos = start_pos + (count_row - start_pos)

    split_df = df_input.iloc[start_pos:end_pos]
    l_dict[name] = split_df

    if ((start_pos > count_row) | (start_pos == count_row)):
        break
    else:
        start_pos = start_pos + interval

    q.put(l_dict)
    cnt += 1

Here, the application is splitting the data into multiple groups of smaller data packs & then combining them into (key, value) dictionary & finally placed them into the individual queue.

# block until all tasks are done
q.join()

This will join the queue process. This will ensure that queues are free after consuming the data.

# stop workers
for i in range(actual_worker_task):
    c_dict['TEND'] = p.DataFrame()
    q.put(c_dict)

for t in threads:
    t.join()

The above lines are essential. As this will help the process to identify that no more data are left to send at the queue. And, the main thread will wait until all the threads are done.

for k, v in fin_dict.items():
    min_val_list[int(k.replace('Thread-',''))] = v

min_val = min(min_val_list, key=int)

Once, all the jobs are done. The application will find the minimum thread value & based on that we can sequence all the data chunks as explained in our previous image & finally clubbed them together to form the complete csv.

for k, v in sorted(fin_dict.items(), key=lambda k:int(k[0].replace('Thread-',''))):
    if int(k.replace('Thread-','')) == min_val:
        df_ret = fin_dict[k]
    else:
        d_frames = [df_ret, fin_dict[k]]
        df_ret = p.concat(d_frames)

As already explained, using the starting point of our data dictionary element, the application is clubbing the data back to the main csv.

Next method, which we’ll be explaining is –

getEncryptWQ

Please find the key lines –

while True:
    try:
        #item_dict = q.get()
        item_dict = q.get_nowait()

        for k, v in item_dict.items():
            # Assigning Target File Basic Name
            item = str(k)

        if ((item == 'TEND') | (item == '')):
            break

        if ((item != 'TEND') | (item != '')):
            self.getEncrypt(item_dict)

        q.task_done()
    except Exception:
        break

This method will consume the data & processing it for encryption or decryption. This will continue to do the work until or unless it receives the key value as TEND or the queue is empty.

Let’s compare the statistics between Windows & MAC.

Let’s see the file structure first –

Windows (16 GB – Core 2) Vs Mac (10 GB – Core 2):

Win_Vs_MAC

Windows (16 GB – Core 2):

Performance_Stats_Windows

Mac (10 GB – Core 2):

Performance_Stats_MAC

Find the complete directory from both the machine.
Windows (16 GB – Core 2):

Win_Files

Mac (10 GB – Core 2):

MAC_Files

Here is the final output –

Sample_OUTPut

So, we’ve achieved our target goal.

Let me know – how do you like this post. Please share your suggestion & comments.

I’ll be back with another installment from the Python verse.

Till then – Happy Avenging!

Pandas, Numpy, Encryption/Decryption, Hidden Files In Python (Crossover between Space Stone, Reality Stone & Mind Stone of Python-Verse)

So, here we come up with another crossover of Space Stone, Reality Stone & Mind Stone of Python-Verse. It is indeed exciting & I cannot wait to explore that part further. Today, in this post, we’ll see how one application can integrate all these key ingredients in Python to serve the purpose. Our key focus will be involving popular packages like Pandas, Numpy & Popular Encryption-Decryption techniques, which include some hidden files as well.

So, our objective here is to proceed with the encryption & decryption technique. But, there is a catch. We need to store some salt or tokenized value inside a hidden file. Our application will extract the salt value from it & then based on that it will perform Encrypt/Decrypt on the data.

Why do we need this approach?

The answer is simple. On many occasions, we don’t want to store our right credentials in configuration files. Also, we don’t want to keep our keys to open to other developers. There are many ways you can achieve this kind of security.  Today, I’ll be showing a different approach to make the same.

Let’s explore.

As usual, I’ll provide the solution, which is tested in Windows & MAC & provide the script. Also, I’ll explain the critical lines of those scripts to understand it from a layman point of view. And, I won’t explain any script, which I’ve already explained in my earlier post. So, you have to refer my old post for that.

To encrypt & decrypt, we need the following files, which contains credentials in a csv. Please find the sample data –

Config_orig.csv

Orig_File

Please see the file, which will be hidden by the application process.

Token_Salt_File

As you can see, this column contains the salt, which will be used in our Encryption/Decryption.

1. clsL.py (This script will create the csv files or any intermediate debug csv file after the corresponding process. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
###########################################
#### Written By: SATYAKI DE        ########
#### Written On: 25-Jan-2019       ########
####                               ########
#### Objective: Log File           ########
###########################################
import pandas as p
import platform as pl
from clsParam import clsParam as cf

class clsL(object):
    def __init__(self):
        self.path = cf.config['PATH']

    def logr(self, Filename, Ind, df, subdir=None):
        try:
            x = p.DataFrame()
            x = df
            sd = subdir

            os_det = pl.system()

            if sd == None:
                if os_det == "Windows":
                    fullFileName = self.path + '\\' + Filename
                else:
                    fullFileName = self.path + '/' + Filename
            else:
                if os_det == "Windows":
                    fullFileName = self.path + '\\' + sd + "\\" + Filename
                else:
                    fullFileName = self.path + '/' + sd + "/" + Filename

            if Ind == 'Y':
                x.to_csv(fullFileName, index=False)

            return 0

        except Exception as e:
            y = str(e)
            print(y)
            return 3

2. clsParam.py (This is the script that will be used as a parameter file & will be used in other python scripts.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
###########################################
#### Written By: SATYAKI DE        ########
#### Written On: 25-Jan-2019       ########
#### Objective: Parameter File     ########
###########################################

import os
import platform as pl

class clsParam(object):

    config = {
        'FILENAME' : 'test.amca',
        'OSX_MOD_FILE_NM': '.test.amca',
        'CURR_PATH': os.path.dirname(os.path.realpath(__file__)),
        'NORMAL_FLAG': 32,
        'HIDDEN_FLAG': 34,
        'OS_DET': pl.system()
    }

 

3. clsWinHide.py (This script contains the core logic of hiding/unhiding a file under Windows OS. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
###########################################
#### Written By: SATYAKI DE          ######
#### Written On: 25-Jan-2019         ######
####                                 ######
#### This script will hide or Unhide ######
#### Files in Windows.               ######
###########################################

import win32file
import win32con
from clsParam import clsParam as cp

class clsWinHide(object):
    def __init__(self):
        self.path = cp.config['CURR_PATH']
        self.FileName = cp.config['FILENAME']
        self.normal_file_flag = cp.config['NORMAL_FLAG']

    def doit(self):
        try:
            path = self.path
            FileName = self.FileName

            FileNameWithPath = path + '\\' + FileName
            flags = win32file.GetFileAttributesW(FileNameWithPath)
            win32file.SetFileAttributes(FileNameWithPath,win32con.FILE_ATTRIBUTE_HIDDEN | flags)

            return 0
        except Exception as e:
            x = str(e)
            print(x)

            return 1

    def undoit(self):
        try:
            path = self.path
            FileName = self.FileName
            normal_file_flag = self.normal_file_flag

            FileNameWithPath = path + '\\' + FileName
            win32file.SetFileAttributes(FileNameWithPath,win32con.FILE_ATTRIBUTE_NORMAL | int(normal_file_flag))

            return 0
        except Exception as e:
            x = str(e)
            print(x)

            return 1

Key lines that we would like to explore are as follows –

def doit()

flags = win32file.GetFileAttributesW(FileNameWithPath)
win32file.SetFileAttributes(FileNameWithPath,win32con.FILE_ATTRIBUTE_HIDDEN | flags)

The above two lines under doit() functions are changing the file attributes in Windows OS to the hidden mode by assigning the FILE_ATTRIBUTE_HIDDEN property.

def undoit()

normal_file_flag = self.normal_file_flag

FileNameWithPath = path + '\\' + FileName
win32file.SetFileAttributes(FileNameWithPath,win32con.FILE_ATTRIBUTE_NORMAL | int(normal_file_flag))

As the script suggested, the application is setting the file attribute of a hidden file to FILE_ATTRIBUTE_NORMAL & set the correct flag from parameters, which leads to the file appears as a normal windows file.

4. clsOSXHide.py (This script contains the core logic of hiding/unhiding a file under OSX, i.e., MAC OS. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
###########################################
#### Written By: SATYAKI DE           #####
#### Written On: 25-Jan-2019          #####
####                                  #####
#### Objective: This script will hide #####
#### or Unhide the file in OSX.       #####
###########################################

import os
from clsParam import clsParam as cp

class clsOSXHide(object):
    def __init__(self):
        self.path = cp.config['CURR_PATH']
        self.FileName = cp.config['FILENAME']
        self.OSX_Mod_FileName = cp.config['OSX_MOD_FILE_NM']
        self.normal_file_flag = cp.config['NORMAL_FLAG']

    def doit(self):
        try:
            path = self.path
            FileName = self.FileName

            FileNameWithPath = path + '/' + FileName
            os.rename(FileNameWithPath, os.path.join(os.path.dirname(FileNameWithPath),'.'
                                                     + os.path.basename(FileNameWithPath)))

            return 0
        except Exception as e:
            x = str(e)
            print(x)

            return 1

    def undoit(self):
        try:
            path = self.path
            FileName = self.FileName
            OSX_Mod_FileName = self.OSX_Mod_FileName

            FileNameWithPath = path + '/' + FileName
            os.rename(OSX_Mod_FileName, FileNameWithPath)

            return 0
        except Exception as e:
            x = str(e)
            print(x)

            return 1

The key lines that we’ll be exploring here are as follows –

def doit()

FileNameWithPath = path + '/' + FileName
os.rename(FileNameWithPath, os.path.join(os.path.dirname(FileNameWithPath),'.'
                                         + os.path.basename(FileNameWithPath)))

In MAC or Linux, any file starts with ‘.’ will be considered as a hidden file. Hence, we’re changing the file type by doing this manipulation.

def undoit()

OSX_Mod_FileName = self.OSX_Mod_FileName

FileNameWithPath = path + '/' + FileName
os.rename(OSX_Mod_FileName, FileNameWithPath)

In this case, our application simply renaming a file with its the original file to get the file as a normal file.

Let’s understand that in Linux or MAC, you have a lot of other ways to restrict any files as it has much more granular level access control.  But, I thought, why not take a slightly different & fun way to achieve the same. After all, we’re building an Infinity War for Python verse. A little bit of fun will certainly make some sense. 🙂

5. clsProcess.py (This script will invoke any of the hide scripts, i.e. clsWinHide.py or clsOSXHide.py based on the OS platform. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
###########################################
#### Written By: SATYAKI DE          ######
#### Written On: 25-Jan-2019         ######
####                                 ######
#### Objective: Based on the OS, this######
#### script calls the actual script. ######
###########################################

from clsParam import clsParam as cp

plat_det = cp.config['OS_DET']

# Based on the platform
# Application is loading subprocess
# in order to avoid library missing
# case against cross platform

if plat_det == "Windows":
    import clsWinHide as win
else:
    import clsOSXHide as osx

# End of conditional class load

class clsProcess(object):
    def __init__(self):
        self.os_det = plat_det

    def doit(self):
        try:

            os_det = self.os_det
            print("OS Info: ", os_det)

            if os_det == "Windows":
                win_doit = win.clsWinHide()
                ret_val = win_doit.doit()
            else:
                osx_doit = osx.clsOSXHide()
                ret_val = osx_doit.doit()

            return ret_val
        except Exception as e:
            x = str(e)
            print(x)

            return 1

    def undoit(self):
        try:

            os_det = self.os_det
            print("OS Info: ", os_det)

            if os_det == "Windows":
                win_doit = win.clsWinHide()
                ret_val = win_doit.undoit()
            else:
                osx_doit = osx.clsOSXHide()
                ret_val = osx_doit.undoit()

            return ret_val
        except Exception as e:
            x = str(e)
            print(x)

            return 1

Key lines to explores are as follows –

from clsParam import clsParam as cp

plat_det = cp.config['OS_DET']

# Based on the platform
# Application is loading subprocess
# in order to avoid library missing
# case against cross platform

if plat_det == "Windows":
    import clsWinHide as win
else:
    import clsOSXHide as osx

This step is very essential to run the same python scripts in both the environments, e.g. in this case like MAC & Windows.

So, based on the platform details, which the application is getting from the clsParam class, it is loading the specific class to the application. And why it is so important.

Under Windows OS, this will work if you load both the class. But, under MAC, this will fail as the first program will try to load all the libraries & it may happen that the pywin32/pypiwin32 package might not available under MAC. Anyway, you are not even using that package. So, this conditional class loading is significant.

os_det = self.os_det
print("OS Info: ", os_det)

if os_det == "Windows":
    win_doit = win.clsWinHide()
    ret_val = win_doit.doit()
else:
    osx_doit = osx.clsOSXHide()
    ret_val = osx_doit.doit()

As you can see that, based on the OS, it is invoking the correct function of that corresponding class.

6. clsEnDec.py (This script will read the credentials from a csv file & then based on the salt captured from the hidden file, it will either encrypt or decrypt the content. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
###########################################
#### Written By: SATYAKI DE        ########
#### Written On: 25-Jan-2019       ########
#### Package Cryptography needs to ########
#### install in order to run this  ########
#### script.                       ########
####                               ########
#### Objective: This script will   ########
#### encrypt/decrypt based on the  ########
#### hidden supplied salt value.   ########
###########################################

import pandas as p
from cryptography.fernet import Fernet

class clsEnDec(object):

    def __init__(self, token):
        # Calculating Key
        self.token = token

    def encrypt_str(self):
        try:
            # Capturing the Salt Information
            salt = self.token
            # Fetching the content of lookup file
            df_orig = p.read_csv('Config_orig.csv', index_col=False)

            # Checking Individual Types inside the Dataframe
            cipher = Fernet(salt)

            df_orig['User'] = df_orig['User'].apply(lambda x1: cipher.encrypt(bytes(x1,'utf8')))
            df_orig['Pwd'] = df_orig['Pwd'].apply(lambda x2: cipher.encrypt(bytes(x2,'utf8')))

            # Writing to the File
            df_orig.to_csv('Encrypt_Config.csv', index=False)

            return 0
        except Exception as e:
            x = str(e)
            print(x)
            return 1

    def decrypt_str(self):
        try:
            # Capturing the Salt Information
            salt = self.token
            # Checking Individual Types inside the Dataframe
            cipher = Fernet(salt)

            # Fetching the Encrypted csv file
            df_orig = p.read_csv('Encrypt_Config.csv', index_col=False)

            df_orig['User'] = df_orig['User'].apply(lambda x1: str(cipher.decrypt(bytes(x1[2:-1],'utf8'))).replace("b'","").replace("'",""))
            df_orig['Pwd'] = df_orig['Pwd'].apply(lambda x2: str(cipher.decrypt(bytes(x2[2:-1],'utf8'))).replace("b'","").replace("'",""))

            # Writing to the file
            df_orig.to_csv('Decrypt_Config.csv', index=False)

            return 0
        except Exception as e:
            x = str(e)
            print(x)
            return 1

Key lines from this script are as follows –

def encrypt_str()

# Checking Individual Types inside the Dataframe
cipher = Fernet(salt)

df_orig['User'] = df_orig['User'].apply(lambda x1: cipher.encrypt(bytes(x1,'utf8')))
df_orig['Pwd'] = df_orig['Pwd'].apply(lambda x2: cipher.encrypt(bytes(x2,'utf8')))

So, once you captured the salt from that hidden file, the application is capturing that value over here. And, based on that both the field will be encrypted. But, note that cryptography package is required for this. And, you need to pass bytes value to work this thing. Hence, we’ve used bytes() function over here.

def decrypt_str()

cipher = Fernet(salt)

# Fetching the Encrypted csv file
df_orig = p.read_csv('Encrypt_Config.csv', index_col=False)

df_orig['User'] = df_orig['User'].apply(lambda x1: str(cipher.decrypt(bytes(x1[2:-1],'utf8'))).replace("b'","").replace("'",""))
df_orig['Pwd'] = df_orig['Pwd'].apply(lambda x2: str(cipher.decrypt(bytes(x2[2:-1],'utf8'))).replace("b'","").replace("'",""))

Again, in this step, our application is extracting the salt & then it retrieves the encrypted values of corresponding fields & applies the decryption logic on top of it. Note that, since we need to pass bytes value to get it to work. Hence, your output will be appended with (b’xxxxx’). To strip that, we’ve used the replace() functions. You can use regular expression using pattern matching as well.

7. callEnDec.py (This script will create the split csv files or final merge file after the corresponding process. However, this can be used as normal verbose debug logging as well. Hence, the name comes into the picture.)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
###########################################
#### Written By: SATYAKI DE           #####
#### Written On: 25-Jan-2019          #####
####                                  #####
#### Objective: Main calling function #####
###########################################

import clsEnDec as ed
import clsProcess as h
from clsParam import clsParam as cp
import time as t
import pandas as p

def main():
    print("")
    print("#" * 60)
    print("Calling (Encryption/Decryption) Package!!")
    print("#" * 60)
    print("")

    # Unhiding the file
    x = h.clsProcess()
    ret_val_unhide = x.undoit()

    if ret_val_unhide == 0:
        print("Successfully Unhide the file!")
    else:
        print("Unsuccessful to Unhide the file!")

    # To See the Unhide file
    t.sleep(10)

    print("*" * 60)
    print("Proceeding with Encryption...")
    print("*" * 60)

    # Getting Salt Value from the hidden files
    # by temporarily making it available
    FileName = cp.config['FILENAME']
    df = p.read_csv(FileName, index_col=False)
    salt = str(df.iloc[0]['Token_Salt'])
    print("-" * 60)
    print("Salt: ", salt)
    print("-" * 60)

    # Calling the Encryption Method
    x = ed.clsEnDec(salt)
    ret_val = x.encrypt_str()

    if ret_val == 0:
        print("Encryption Successful!")
    else:
        print("Encryption Failure!")

    print("")
    print("*" * 60)
    print("Checking Decryption Now...")
    print("*" * 60)

    # Calling the Decryption Method
    ret_val1 = x.decrypt_str()

    if ret_val1 == 0:
        print("Decryption Successful!")
    else:
        print("Decryption Failure!")

    # Hiding the salt file
    x = h.clsProcess()
    ret_val_hide = x.doit()

    if ret_val_hide == 0:
        print("Successfully Hide the file!")
    else:
        print("Unsuccessful to Hide the file!")

    print("*" * 60)
    print("Operation Done!")
    print("*" * 60)

if __name__ == '__main__':
    main()

And, here comes the final calling methods.

The key lines that we would like to discuss –

# Getting Salt Value from the hidden files
# by temporarily making it available
FileName = cp.config['FILENAME']
df = p.read_csv(FileName, index_col=False)
salt = str(df.iloc[0]['Token_Salt'])

As I’ve shown that, we have our hidden files that contain only 1 row & 1 column. To extract the specific value we’ve used iloc with the row number as 0 along with the column name, i.e. Token_Salt.

Now, let’s see how it runs –

Windows (64 bit):

Win_Run

Mac (32 bit):

MAC_Run

So, from the screenshot, we can see our desired output & you can calculate the aggregated value based on our sample provided in the previous screenshot.

Let’s check the Encrypted & Decrypted values –

Encrypted Values (Encrypt_Config.csv):

Encrypted_File

Decrypted Values (Decrypt_Config.csv):

Decrypted_File

So, finally, we’ve achieved our target.

I hope this will give you some more idea about more insights into the Python verse. Let me know – how do you think about this post.

Till then – Happy Avenging!