Sarvesh Gupta

How to Create a REST API with Python and FastAPI Framework

2021-01-03T17:21:00.014+05:30

Welcome! Let's take a look at "How to create a REST API with Python and FastAPI framework"🚀.

API is essential for building a multi-layer application structure, so choosing the right framework for your API can make the performance of your application robust.

FastAPI is a modern, fast (high-performance), a web framework for building APIs with Python 3.6+ based on standard Python type hints (source). It is written in Python with Starlette and Pydantic as a base. Big bull companies like Microsoft, Uber, Netflix, etc. have already started using FastAPI.

One of the reasons why I like FastAPI is because of the automatic interactive documentation. It will auto-generate Swagger documentation (not only Swagger but also Redoc) for you without any trouble of writing additional code.

Let's Begin Coding...

Prerequisites:

Python 3.6+ (and how to use it😂)
pip (if already installed then ignore)

Install FastAPI from pypi using pip, you also need to install an ASGI server.

pip install fastapi

pip install uvicorn[standard]

If you are familiar with the flask framework then it is very much similar to it, and if you are not then no worries. Once both are installed create a file app.py with the below code.

Let's run app.py with the below command

uvicorn app:app --reload

Adding --reload to the line will enable auto-refresh to your API whenever you save new changes. You can replace the left side app with your file/module name. For example, if you create a file as main.py then your command would be uvicorn main:app --reload.

Your API is running by now, and you can access it by going to http://localhost:8000/ from your browser. To see the documentation page of your API go to Swagger - http://localhost:8000/docs or Redoc - http://localhost:8000/redoc.

Understanding the above Code Example

After creating an instance you have defined data which is a list of dictionaries (i.e. fruits) to perform CRUD (Create Read Update and Delete) operations on it (Note: you can define your own data model or even use the database directly).

You have created an API that:

Receives HTTP GET request in path / and /fruits , returns welcome note and list of fruits respectively.
Receives HTTP GET request in path /find/fruits/{fruit_id} and finds fruit with given fruit ID in URL. path parameter fruit_id should be an integer as you have defined in the function argument.

Why end here let's add more methods to the API

The above code has a body (BaseModel class) with Pydantic for the PUT and POST method. Now save this code and your API should auto-refresh, you can perform all the CRUD operations.

You can now try integrating the Postgres (or any other) database using SQAlchemy and make it a production-ready REST API but that would be a topic for the next blog.

Thank you for visiting my blog😊! If you enjoyed this article, you can subscribe to my blog for the latest updates. Let me know your comments about this article.

What is Lambda (Anonymous Function) in Python and How to use it with Filter, Map & Reduce

2020-10-20T23:31:00.011+05:30

Anonymous function means that function which does not have a name, so now we can simply say function without having a name is lambda function. To create a normal function in python we need to use the def keyword and while creating the lambda function we need to use the lambda keyword. Hope you have got a brief idea of what is lambda in Python.

Normal Function:

Lambda Function:

Output:

Output from normal function: 25

Output from lambda function: 25

Here above-defined both function is returning the same thing but using lambda function we will be removing the uses of keywords like def, return and also we are reducing the number of lines, You can use any number of argument as you want but the expression should be only one.

Lambda function does not require a return statement because it returns the expression itself directly. Where ever you need one expression with any number of arguments you can use the lambda function without even assigning the variable at all.

Use Lambda with Filter, Map & Reduce

Filter:

It is an inbuilt function in python, As the name suggests it takes the input (Iterable data) and returns the filtered data. Whenever you have a list or any other iterable and you want filtered data from that you can use the filter function it would be the best choice.

filter(function, list)

Now here you can see filter function takes two argument - list and function, List is your data that you want to filter, and function where you will define your operations. Here in the place of the function argument, we are going to use the lambda function. Now let's see in below code.

Using Filter with Lambda:

Output:

Even numbers: [6, 12, 18]

Explanation:

In the above example code, list_data is the list of any random integer and then I am passing it the to filter function. Now I have created the lambda function that will take list data one by one and match with our condition (even % 2 == 0) if the condition becomes is true then it will put the returned value inside a new list. And finally, we are printing the list i.e. even_num.

Map

It is an inbuilt function in python, It takes the input (Iterable data) and performs some operation as you define and returns final data. So if you want to perform some operation on all the elements of a list or any other iterable data you can use the map function.

map(function, list)

Now here you can see map function takes two arguments - list and function, List is your data that you want to perform some operation on it and function where you will define your operations. Here in the place of the function argument, we are going to use the lambda function. Now let's see in below code.

Using Map with Lambda:

Output:

Double numbers: [6, 10, 18, 12, 26, 24, 36, 14]

Explanation:

In the above example code, list_data is the list of any random integer and then I am passing it the to map function. Now I have created the lambda function that will take list data one by one and perform the operation (i.e. multiplying by 2) to each element of a list and then it will return the final list with having all the elements multiplied by 2.

Reduce

As the name suggests it will reduce your list to a single value. So if you have a chuck of value and you want to add all the values or multiply all the values you can choose this cool function to handle that for you.

reduce(function, list)

Now here you can see filter function takes two argument - list and function, List is your data that you want to reduce it into a single value and function where you will define your operations. Here in the place of the function argument, we are going to use the lambda function.

Now here let me tell you one thing otherwise you will get an error while executing this code, reduce is not an inbuilt function it belongs to a module name as functools. Now from this module you can import reduce now let's see the example.

Using Reduce with Lambda:

Output:

Addition result: 73

Explanation:

In the above example code, list_data is the list of any random integer and then I am passing it the to map function. Now I have created the lambda function that will take list data and add all the elements one by one and then returns one single final value.

If you see in this code you will not find a list while creating a reduce because reduce only returns a single value.

Thank you for visiting my blog😊! If you enjoyed this article, you can subscribe to my blog for the latest updates. Let me know your comments about this article.

What is Cryptography and the Process of Encryption/Decryption with Python (AES - CBC Mode)

2020-06-14T22:31:00.003+05:30

In the IT industry, Cryptography is one of the most important techniques to secure sensitive data. According to Wikipedia Cryptography is about constructing and analyzing protocols that prevent third parties or the public from reading private messages. What this exactly means let's understand, now assume that you have a secret message that you do not want to share with any other except the receiver. Of course, you are sending the message to the receiver so he only will receive that message but what if someone manages to hack your message. In this case, if your message will be in a normal text he can easily read it, so to avoid this type of risk we encrypt our message before sending it. So that even if someone manages to hack our message he can not read the message. As per my understanding, the Encryption/Decryption or Encoding/Decoding of the message is called Cryptography.

There are multiple ways to Encrypt and Decrypt the message, I will be covering here AES (Advanced Encryption Standard) encryption and decryption using Python. AES encryption and description technique use the key to encrypt the message and the same key will be used to decrypt the message. Enough talk let's start with some coding stuff...

The Github link for this blog you can find Here

Encrypt/Decrypt Messages

We will use Python's pycryptodom library, it is open source and available on PyPi so you can easily download using pip by hitting the below command.

pip install pycryptodom

Now the library is installed so let's write a code to encrypt our message using a key, we will also decrypt the message using the same key in this code.

I am using CBC Mode (Cyphertext Block Chaining Mode) in this mode, each Block (explained below) gets XOR-ed with the previous ciphertext block before encryption. There are multiple modes in AES you can find in the Docs.

In this code, I am generating a random key using the Random method from a pycryptodom library. You can use your own key to encrypt your message. Remember that AES uses a fixed-size key (16, 24, or 32 bytes only) for encryption.

The Initialization Vector is a piece of data unpredictable to adversaries, this argument is required to pass while creating the AES object.

I have created two functions encrypt_msg and decrypt_msg. In these functions, I have created an object of AES and passed key, mode, and iv as an argument. Now before moving to the next line let understand what is Block? Since the method encrypt() and decrypt() from CBC Mode of AES expects a fixed size of data at a single time in order to encrypt the message, AES divides the message into 16 bytes of each Block. If your message is less or greater than 16 bytes you have to add a character according to maintain 16 bytes of blocks.

My message in the code is "I Love Making Blog" and the length of this string is 18 bytes including spaces. The first block is properly filled here with 16 bytes but the second block has less character so we are filling it with "*" in order to complete 16 bytes of the Block.

BLOCK-1

I		L	o	v	e		M	a	k	i	n	g		B	l
1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16

BLOCK-2

o	g	*	*	*	*	*	*	*	*	*	*	*	*	*	*
1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16

Now let's understand with one more example suppose you have less than 16 bytes string "I Love Java", here the length of the string is 12 bytes so in order to maintain 16 bytes of Block we are adding 5 more characters in the string.

I		L	o	v	e		J	a	v	a	*	*	*	*	*
1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16

I hope you understood what is Block. Now in order to maintain 16 bytes of each block we are using pad and unpad module from Crypto.Util.Padding. In a pad module, we have to pass our message and number of block size which 16 bytes equivalent of AES.block_size. Once after padding the message we are ready to pass the message to encrypt() method of CBC cipher and finally return the encrypted message from encrpt_msg() function.

The decrypt_msg() function works the same as the encrypt_msg function, here we are unpadding the message and decrypting it in order to achieve the original plain text message. Still, our message will not be in plain text, it will be in binary form so to convert binary to a plain text we are using the decode method from Python and returning the decrypted message.

The function encrypt_msg() (and likewise decrypt_msg()) from above code expect string message to encrypt and decrypt the message.

We are going to understand how AES works in CBC Mode. In the below diagram, Blocks data are nothing but 16 bytes message inputs that provided with pad to encrypt the data. Now in CBC Mode, it encrypts the first block which is Block-1 as it is with a provided key after this it passes the encrypted data to the next Block. In the next block which is Block-2, encrypted data from Block-1 get XOR with Block-2 data then encrypted, and so on.

Encrypt/Decrypt Files

In the above code, I have used Encryption/Decryption on files, it uses the same concept as we have done in the message. Feel free to ask questions in the comments section below.

Thank you for visiting my blog😊! If you enjoyed this article, you can subscribe to my blog for the latest updates. Let me know your comments about this article.