DEV Community: Paul

JOINING TABLES WITH PANDAS. WHAT IS IT? WHY DO YOU EVEN NEED IT?

Paul — Mon, 08 Dec 2025 10:01:36 +0000

Within this resource, we’re going to explore joining tables using a popular python library called Pandas.

WHY JOIN TABLES?

There are several ways to join datasets, but the bigger question is: why bother joining tables at all?

Why not simply work with the necessary table or data frame and avoid the extra hassle?

Well, let’s imagine it this way,

As a data expert, you’re given two datasets: one containing customer information and the other detailing store transactions.

Your task is to identify customers who spent over $999 bucks in the past week to qualify for a Black Friday discount.

Keep in mind that these are separate, existing datasets.

While working directly with the transactions dataset might help identify transactions over $999 bucks, the real challenge lies in identifying who made these transactions.

What are their names and contact details?

These questions can only be answered once you bring the customers dataset into the mix.

See how it works?

Joining datasets is just as important as putting the puzzle pieces together—without it, you’re limited!

But let’s pause for a moment to tackle the one super question you’ve been waiting for.

_Why am I getting these datasets one by one—like they’re dropping clues for a treasure hunt I didn’t sign up for?

Did the data gods see me toss that banana peel into the wrong bin at the park and decide I must suffer for my sins?

Why not just serve me a combined dataset straight from the source? Is that too much to ask?_

I hear you, my friend, and I’m here to help you understand why.

In most cases, databases are designed to store various datasets with different relationship patterns—much like how the customers dataset connects to the transactions dataset.

Apart from scalability, other reasons for practicing dataset separation include:

Access control for sensitive data
Query efficiency
Logical organization
Data ownership
Compliance and regulation amongst others.

TYPES OF JOINING TABLES

Alright, before we wrap things up, let’s quickly go over a few join types you’re likely to come across.

Inner Join
Left Join
Right Join
Outer Join

Don't panic, they're interesting and fun to work with.

Inner Join is one of the simplest and most popular join types, where datasets are merged based on columns that are common to both datasets.

Let’s revisit our customers and transactions datasets:

Customers Dataset
Customer_ID Name      Contact
101         Alice     alice@example.com
102         Bob       bob@example.com
103         Charlie   charlie@example.com
104         Diana     diana@example.com
105         Eve       eve@example.com

Transactions Dataset
Transaction_ID  Customer_ID Amount  Date
201             101         1200    2024-12-01
202             103         1500    2024-12-02
203             104         999     2024-12-03
204             106         750     2024-12-04
205             107         1800    2024-12-05`

As you can see, both datasets contain various diverse columns ranging from customer names to contact and amount to date of transaction respectively, but…at least one column shares it’s presence on both dataset making it a common factor.

And that’s the Customer_ID column.

This commonality helps us perform joins effectively.

So inner join in this case is simply merging both tables on the Customer_ID column, excluding all rows whose customer_ID have no matching values on both tables.

I’ll show you:

import pandas as pd

inner_join_result = pd.merge(customers_df, transactions_df, on="Customer_ID", how="inner")

As demonstrated in the code snippet, performing an inner join requires importing the pandas library and using its merge() method. This method combines the datasets by specifying the on parameter, which indicates the common column to merge on (in this case, Customer_ID). The how="inner" argument specifies the join type which is "inner".

The results of the above inner join would be:

Customer_ID Name    Contact             Transaction_ID  Amount  Date
101         Alice   alice@example.com   201             1200    2024-12-01
103         Charlie charlie@example.com 202             1500    2024-12-02
104         Diana   diana@example.com   203             999 2024-12-03

Let's analyse the result for better understanding:

Alice (101) appears in both datasets, so her information and similar rows such as Charlie(103) and Diana(104) are included.
Eve (105) and Bob(102) are in the Customers dataset but not in Transactions, so they are excluded.
Customer_ID 106 and 107 from the Transactions dataset are excluded because they do not match any rows in the Customers dataset.

In summary, only rows with matching customer_id is the output of our inner join.

On the other hand, rows whose Customer_ID not found on the other merging dataset are completely excluded form the resulting dataset.

how is this useful and real life application

It’s important

Too much to take in?

I’ll explain.

If you wish to explore other types of join with me such as anti-join, semi-join, self-join and many more, do give a like and read here:

Till we talk again.

ciao.

Now go join those Datasets with conviction

Blockchain Debunked! Part 2. Understanding the Different Blockchain Types.

Paul — Mon, 27 Jan 2025 22:51:11 +0000

In the previous article, we only walked through what blockchain is and why it has grown so popular and special over time. Here in this article, we are going to outline types of blockchains and just a little bit about blockchain architecture.

Types of Blockchain

Actually, there are diverse types of blockchains existing but we will focus on the four major and important ones in no particular order, their benefits, drawbacks, and ideal uses:

i. Public Blockchain: Just as the name goes it's a blockchain system where anyone anywhere can participate. Example: Solana, Ethereum and Bitcoin are all public blockchains.

Key Features:

Open to anyone
No central authority
High transparency
Secured by consensus mechanisms like proof of work(PoW) or proof of stake(PoS)

Drawbacks:

Scalability: The wide access to public blockchain prevents its ability to handle large volumes of transactions and growing user base. The more the user base grows, the slower it becomes.
Privacy: Its open nature of increases the exposure of users sensitive information to the public.

Use case: Popular cryptocurrencies such as Bitcoin and Ethereum use a public blockchain for decentralized payments and other crypto-related activities such as trading tokenized assets.

ii. Private Blockchain: This blockchain is the opposite of a public blockchain in sense that it restricts participation to a select few members. It's permissioned and often controlled by a single entity.

Think of it like a party strictly on invite.

Key Features:

Restricted access (only invited members)
Highly centralized
Increased privacy

Benefits:

Compared to public blockchain, there's increase in privacy since it's not open to the public.
It's privacy nature in turn accounts for security.
More scalable(faster) since there are less users on the blockchain.

Drawbacks:

Lacks Transparency. Sometimes this is for the greater good with intention to tighten security.
Centralization.
Expensive to maintain.

Use case:
Some healthcare systems use this blockchain to manage patient records securely, certain finance institutions use it for cross-border transactions and companies like Walmart use a private blockchain to track goods from production to delivery.

iii. Hybrid Blockchain: It's a combination of both public and private blockchain at customized level. Example Aergo Enterprise and Swisscoin.

Key Features:

Allows customizable level of access control.
Designed to reduce the setbacks of both private and public blockchains and leveraging on their benefits.

Benefits:

Unlike public blockchain, Hybrid Blockchain is more scalable and yet manages to possess the extra cool features of a public blockchain alongside private's as well.
Only selected information can be made public or private depending on the data that need to be shared.

Drawback:

Isn't completely transparent since it has a little bit of the centralized nature of private blockchains.

Use case: Enterprises adopt a hybrid blockchain system to enable them customize their transparency to both internal stake holders and external users. Example XinFin use a hybrid blockchain that has Ethereum's public blockchain system and Quorum for the private use.

That's all about the core blockchain types and I hope you enjoyed your read.

In the next article, we're going to cover Blockchain Architecture. If you want to explore the internal make-up of blockchains, kindly look out for the next article.

Thanks for stopping by. See you soon.

Blockchain Debunked!

Paul — Tue, 07 Jan 2025 08:01:33 +0000

As a web3 enthusiast, I know you've come across the word "Blockchain" on several occasions.

In fact, while you stay consistent in the world of web3 and crypto, you will encounter multiple mentions of the same word almost everyday. Maybe in your sleep too, like me...

But what really is this term and why is it special? That's exactly what we will cover here. Now...

What is Blockchain?

Simple, it’s a series of concrete blocks tied together by a gigantic caterpillar chain. Yup, that’s blockchain.

Oh, wait…that's absolute nonsense. My bad I didn’t realize I was already on camera. Let’s try that again.

Have you ever taken records of your transactions?

Bagged that Nike Air killer wear at $599 while returning from the park. Tipped the waitress $20 for a smooth service, and managed to get everything on paper, nicely documented.

That's what banks do. They keep records of every bit of financial transaction in what's called a financial ledger.

Now, imagine taking that ledger, making it digital, and sharing it with everyone involved—except it’s super secure, and no one can mess with past records. That’s the idea behind blockchain.

In short: It's a record.

In this case, a blockchain is a record of transactions made in the world of cryptocurrency. Practically, imagine when Peter withdraws some USDC token from his crypto wallet to go touch grass or..transfers a percentage of his Bitcoin to my wallet lol.

Each of these transactions from withdrawal to transfer are stored in a blockchain, permanently. But here's the fun part: it's stored as a block - which brings us to a quick review of the words "block" and "chain"

Block: is the form the data takes when stored. Think of it like a digital box of information.

Chain: is what links all the blocks together forming an unbreakable connection

So, whenever a transaction is executed...be it sending crypto to a friend, paying for coffee, or hypnotizing Peter into a Bitcoin donation, it takes on a block-like form, gets stored, and links up with other pre-existing blocks via chains.

Hope that wasn't too much to take in?

Essentially, that's what a blockchain is: a chain of blocks. Not too hard right? So now we get it. A blockchain stores crypto transactions and yada yada.

But wait, what then makes a chain of blocks special?

Well, a lot makes a blockchain special, and to mention a few:
➟ Decentralization
➟ Transparency
➟ Security
➟ Immutability

Don't sweat it. We gonna dig deeper into these features in a bit. But first? sip some water.

Alright let's dive in.

Decentralization: In blockchain, instead of trusting one central authority (like a bank or exchange), control is spread out among multiple systems, or nodes. That way, no single party has full control, making it more secure and resistant to failure.

Transparency: In the blockchain world, everything is open for anyone to see. Every transaction that happens is recorded on the blockchain and can be viewed by anyone in the network.

It’s like the most honest record keeping ever where no wrong dealings can hide.

Security: Security on the blockchain is top-tier. Every transaction is encrypted and linked to previous ones in a way that’s nearly impossible to break.

It’s like trying to break into a vault that’s been guarded by a thousand lasers and an army of K-9s.

Immutability: Once something’s written on the blockchain, it’s there for good. You can’t go back and change it.

This is huge for ensuring that no one can alter the history of transactions, making blockchain the most honest record-keeping system.

To complete it's divinity, let's compare the traditional banking system the blockchain system side-by-side.

Now, don’t get me wrong, the fiat/traditional system has its perks, and it has been the backbone of financial systems for centuries. But when we really fact-check it against blockchain’s powers, some of its flaws start to show such as:

➟ Zero transparency(banks control everything)
➟ Centralization
➟ Zero immutability(transactions can be altered)
➟ Poor security and many more.

We could dive deeper into the comparison but hey, I'm afraid I can't. Scared the CBN president might show up in the middle of my sleep for sparking an anti-fiat movement of web3 cultists.

Maybe I'll make that deeper comparison on the next article after I've signed my will(oh gosh, now I've gotta get a lawyer lol). But yea, that's it.

A basic, edible explanation of "Blockchain".

What did I miss? Think it covers the essentials?

"Tech is just a Huge Mistake..." -- Part 1.

Paul — Sat, 14 Dec 2024 12:54:03 +0000

Hi there,

Welcome to my humble aboard.

I believe you’re packed up for this interesting read.

Lock in as we ride through!

Back in 2022, it was that time of the year when streets were lit up with flashy lights, nicely decorated trees, and snow raining down helplessly.

It is a special evening I dare not forget.

A battle for the crown. And the best at getting wiped or the wiper.

I & Abel, a childhood friend., was positioned for war.

Sweat enveloped my hands like a glove as I punched the buttons, seeking redemption.

Abel was an intermediate software engineer at the time, crushing it so hard in the industry (tech) - that anyone would wish for something similar.

“Paul get into tech. Let’s start with Python”…Abel interrupted while I received the trash of a lifetime.

Back-to-back losses and currently at another 5:2 loss over the play-station.

“Stop it!. Don’t need your sympathy. Winning the next game”…I replied.

Abel: “Serious, bro. See…I know AI & Web3 is the future. So python would be a cool kickoff”.

…First time someone carefully pitched the tech industry to me. Tech? How would I squeeze such time into two jobs? Two damn demanding 9-5s.

“Would this transition be my haven, risky, or my doom?”… I thought, while preparing the perfect question in the short response.

“Um…that would be nice but python you said?”…” What’s that mate?”…” Isn’t a python a reptile? Bro c’mon”.

Laughing uncontrollably, he struggled to reach for his iPhone 13 on the sofa.

“Whatsapp. Sent you a link”…he said a few seconds later.

Hurriedly looked it up in curiosity…

“100 Days of Python - Angela Yu”.

“That’s our entry. Get the course on Udemy & get to work.” He added.

Two years later, around January 2024. Paul had emerged as a professional backend developer, heavy on Python.

But that’s not the catch…

After adopting the traditional onboarding track. Learn HTML, CSS, Python, and framework. Build awesome stuff and voila, Tech Bro.

Backend Engineer!.

Next? Finding jobs…

Everything got so tiring at this stage. Numerous applications (maybe not enough)…

Till I bagged two contracts via a friend's referral.

Quite short-lived gigs. One in E-Commerce and the other in Ed-Tech. And I’m grateful they got my hands dirty.

The latter did me dirty, stretching me thick and thin for the best. And that was it!

Back to square one again. The annoying job-hunt circle.

“Am I supposed to be here?”

“Python stack isn’t recognized in your locality. Your two tech years are so f**cked”

“Tech is one hella mistake. Wake up!”

Depressing thoughts firing at me daily.

Sad & angry. I kept questioning my abilities and this high-income skill as they call it.

“So what to do now?” I asked myself.

To be continued….