The first post in this blog was published on 25th of August 2008. This place was always living on a shared PHP/MySQL hosting. It was driven by a custom made PHP CMS for over 10 years. It is time to move forward and follow my actual language of choice - JavaScript. I spent some time recently migrating it to Zeit and this article is showing my process.

I'm using Git and GitHub in particular a lot. And when I say a lot I really mean all the time. Recently I tracked a week of work and found that 62% of my working time goes into code reviews. Sometimes I'm checking out a branch locally and trying stuff but really most of my time goes into github.com. I spent some time analyzing why the code review process is so time consuming for me. I identified couple of reasons, made a tool and changed some of my habits. In this article I'll show you how I improved my code review speed and lower the time to 38%.

The hooks API is a wonderful idea. There are some slick patterns involved that push the React development to a more functional approach. I'm interested in trying that new API and decided to use it for my latest project. However, after a couple of days, it looked like I can't build my app only with hooks. I needed something else. And that's mainly because each hook works on a local component level. I can't really transfer state or exchange reducers between the components. That's why I created Jolly Roger. It has similar helpers but works on a global app level.

I had an interesting bug in my React application. It happened that the problem was in the fact that I was using forEach instead of for.

The bug that I was encounter was

Recently I spent some time working on my own JavaScript playground called Demoit. Something like CodeSandbox, JSBin or Codepen. I already blogged about why I did it but decided to write down some implementation details. Everything happens at runtime in the browser so it is pretty interesting project.

I had some time during the weekend and decided to work on my slides for an event at the end of the month. I reached the part where I have to do a live coding session and I was wondering what tool to use. At the end I created my own called Demoit. This is a short article explaining how it works.

If you use React you probably know about the so called hooks. They were officially announced at this year's ReactConf by Sophie Alpert and Dan Abramov. Their presentation could be seen here. I, same as many others got intrigued by this new feature. A little bit confused by if I like them or not but kind of excited. This article pretty much sums up my thinkings around React hooks and aims to give a balanced opinion.

If your users deal with images (photos, product images, etc.) it will be great if they have the ability to apply some basic editing after upload, e.g. adjust brightness and contrast for sharper view, crop, add some text, etc. However, if image editing is not the primary feature of your product, the profit from investment in such secondary feature won't cover the costs. Integrating a 3rd party image editor appears to be a wiser choice.

Long long time ago in a kingdom far far away there was an app. The app was supported by the well known React and Redux families but there was a problem. It was damn slow. People started complaining and the app had to do something. It had to deliver its content quickly so it provides better user experience. Then the server-side rendering was born.

Years ago when Facebook announced their JSX syntax we had a wave of comments how this was against some of the well established good practices. The main point of most people was that it violates the separation of concerns. They said that React and its JSX are mixing HTML, CSS and JavaScript which were suppose to be separated.

In this article we will see how React and its ecosystem has quite good separation of concerns. We will prove that markup, styles and logic may live in the same JavaScript land and still be separated.

Recently we had an interesting task at work. The user makes a selection of items and clicks a button. Then for every selected item we have to make a request to our API. The thing is that the user may click as many items as he/she wants. In order to speed up the process we decided to handle four requests in parallel and once some of them is finished we pull the next one. If you ask why exactly four requests in parallel read this paper.

Redux is a library that acts as a state container and helps managing your application data flow. It was introduced back in 2015 at ReactEurope conference (video) by Dan Abramov. It is similar to Flux architecture and has a lot in common with it. In this section we will create a small counter app using Redux alongside React.

On my machine I have four applications always open - VSCode, Chrome, iTerm and Slack. I spend most of my time in Chrome and VSCode. My editor is full with awesome extensions and I feel pretty good there. What I am doing for the browser is making sure that I have fewer tabs open and install only extensions that I really use. One thing though I can achieve so far. I can't find the perfect new tab extension.

(If you are lazy jump to this section to learn what is this article all about.)

Remember The Island movie from 2005. I watched it again these days and I realized that the old sci-fi titles start with something really interesting. It is funny how they create a world of flying vehicles and say something like "The year is 2019 …". Well, we are 2018 and the public transport is still on the ground. We still can't clone people (I hope so) or travel in a giant spaceship which looks like a fully-functional city.

If you are writing React applications you probably know about higher order components or render props (which by the way I think is kind of a form of higher order component pattern). In both cases we have a component that encapsulates logic and passes props down to children. Recently at work we came to the idea that we may push this further and represent some functionalities which are out of React in the same fashion - with a single tag in our components tree.

Sometimes when you learn something new you get really excited. Excited to that level so you want to teach it to someone. That is the case with the concept which I found a couple of months ago. It is an implementation of the command pattern using generators. Or the well known saga used in the redux-saga library. In this article we will see how the idea makes our asynchronous code simpler and easy to read. We will also implement it ourself using generators.

This article is about Stent - a Redux-liked library that creates and manages state machines. Stent implements some of the Redux’s core ideas and in fact looks a lot like it. At the end of this post we will see that both libraries have a lot in common. Stent is just using state machines under the hood and eliminates some of the boilerplate that comes with Redux’s workflow.

Recently I started questioning the state management in React applications. I’ve made some really interesting conclusions and in this article I’ll show you that what we call a state management may not be exactly about managing state.

If you write JavaScript today you probably use some sort of a transpilation tool. A tool that reads your hipster code and convert it to code that works in the browser. In this article we are going to see what is actually send to the browser and how exactly libraries like Babel polyfill some of the ES6 features.

There are lots of things which I like in React. Mostly the fact that it teaches interesting patterns. One of my favorites ones is higher-order component. In this article we’ll do a couple of experiments and will see how powerful this approach could be.

(This article is part of React in patterns series.)

React is highly composable. And the API that enables that is props.children. It gives us the power to create a placeholder that is later filled with content from the outside.

React is probably one of the best choices for building UI. Good design, support and community. However, there are cases where we want to use an external service or we want to integrate something completely different. We all know that React works heavily with the actual DOM and basically controls what’s rendered on the screen. That’s why integrating of third-party components could be tricky. In this article we will see how to mix React and jQuery’s UI plugin and do it safely. read more comments

When we start using React we very soon also start asking questions. Where I’m suppose to put my data, how to communicate changes or how to manage state? The answers of this questions are very often matter of context and sometimes just practice and experience with the library. However, there is a pattern which is used widely and helps organizing React based applications - splitting the components into presentational and containers.

Long time I was searching for a good front-end framework. Framework that will help me write scalable and easy to maintain UI. Even though React is just a library for rendering it comes with so many benefits that I can easily say “I found it”. And like every thing that I use a lot I started seeing some patterns. Techniques that are applied over and over again and I see in the code of other developers. It’s time that I start documenting, discussing and sharing these patterns.

I’m writing code for approximately ten years now and at some point the word “writing” became more important then “code”. I found out that it is easy to generate code but it is difficult to write code. Writing means creating something meaningful. Something that other human beings will read. The code is indeed sent to machines and they do understand ugly code. Not the same for humans though. An important part of our job is to make sure that our code is clear for other teammates.

The setup here is available at GitHub here.

Half an year ago I published A modern React starter pack based on webpack. The starter provides the basic tooling around React. However, I noticed that very often I need even less stuff than that. That’s usually when I want to hack something quickly. In this blog post we’ll see what’s the bare minimum to work with React.

If you are front-end developer dealing with single page applications you probably know that JavaScript ecosystem is not perfect at all. A few things may go wrong and break your build. In this article I’ll go through those features. Features which are, by my humble opinion, problematic.

One of the things that I really like at work is the fact that we spend time in research tickets. We have unknowns and we make a research to find out more information on particular topic. We then base our decisions on the results. Recently we had to decide whether to use an <iframe> for a third party widget development. I feel that the collected information may be handy to someone else so I decided to write it down here. read more comments

Last week I landed on an article by Manuel Wieser called Dominant Colors for Lazy-Loading Images and I found the topic really interesting. It’s about lazy-loading of images. Something that Medium is using.

You probably heard about Kik, NPM and left-pad saga this week. Shortly, a company Kik asked a developer Azer Koçulu to give the ownership on a NPM module. The module name matches the name of the company. The developer refuses and the company reaches the registry (NPM). The module was transfered to the company based on a NPM policy. The developer then decided to remove all his modules from the registry. The bad thing is that one of these modules left-pad is a dependency of many other modules. As a result of the un-publishing all the packages that depend on left-pad can not be built. Some really popular tools like Babel and React started getting broken builds.

Destructuring is one of my favorite ES6(ES2015) features. It just shortens my code and helps me be more explicit with my statements. Let’s see it in action.

You have no idea how fun is to transpiler JavaScript. I’m digging into that last few weeks and there is a step where I have to transform an AST into valid code. There is one tree node which I really like - sequence expression.

I like reading code of other developers. It’s a nice way to learn patterns, techniques and small tricks. Recently I found something about Array.prototype.length which caught my attention.

Well, I kind of lied in the title of this article. Of course that it is the end of the function. Once we call return everything else after that is simply not executed. Ops … I did it again, I kind of lied again.

(This blog post is inspired by A modern JavaScript router in 100 lines and heavily related to Navigo router project)

If you build single page applications you probably know that one of the must-have parts is the router. The bit that knows how to tweak the content of the address bar and notifies the rest of the system for URL changes. In this article we will discuss the various aspects of the routing in the browser.

Two months ago I published a starter pack for React based on webpack. Today I found out that I need almost the same thing but without the React bit. This simplifies the setup but there are still some tricky parts. So, I made a brand new repository webpack-library-starter and placed all the stuff that we need for creating a JavaScript library.

I love using functions like map, filter or reduce. They are an important part of my arsenal and I simply can’t stop using them. Today I had to solve an interesting problem and I ended up using reduce.

The new ES6 specification of JavaScript simply works better for me. It helps expressing my ideas in a shorter and cleaner way. So, I want to write ES6 everywhere. Even while I’m testing my applications.

The Flux implementation discussed in this article is available here github.com/krasimir/fluxiny. Feel free to use it in a browser directly or as a npm dependency.

I’m obsessed by making my code simpler. I didn’t say smaller because having less code doesn’t mean that is simple and easy to work with. I believe that big part of the problems in the software industry come from the unnecessary complexity. Complexity which is a result of our own abstractions. You know, we (the programmers) like to abstract. We like placing things in black boxes and hope that these boxes work together.

I hope you know about currying. If not then please read this book. It’s basically a process of calling a function with less parameters than it expects. Ok, not exactly calling the function but prepare another function that will run the original one. Some people call the returned function higher-order factory function. Really powerful concept.

GitHub’s pull requests are an important part of my/our development process. That’s why I was thinking about creating a template that will enforce the standards.

While I was open sourcing react-place I noticed that there is some complexity around preparing the component for releasing. I decided to document the process here so I have a solid resource next time. You may be surprised but writing the working jsx file doesn’t mean that the component is ready for publishing and is usable for other developers.

Checkout React webpack starter in GitHub.

You know how crazy is the JavaScript world nowadays. There are new frameworks, libraries and tools coming every day. Frequently I’m exploring some of these goodies. I got a week long holiday. I promised to myself that I’ll not code, read or watch about code. Well, it’s stronger than me. React werbpack starter is the result of my no-programming week.

When CommonJS was announced we all thought “Finally something that will organize our code”. However, there are some cons that we should be aware of. It’s not only unicorns and rainbows. In this article we will see how a simple require line makes our code difficult to test.

You probably know the famous quote that the code should be written for humans to understand and accidentally for computers to execute. Writing code that compiles is easy. Writing readable code is completely different thing. Working in a team is like sharing the kitchen with your roommates. You all should care for the dishes and keep clean. And it is not only because of the others but because of you. You don’t want your dinner in a mixed place and dirty dishes right.

You’ve probably heard of Browserify. It’s a nice npm module for bundling your JavaScript for a client-side usage. It lets you use similar to Node.js modular system but for the code running in the browser. I had few issues testing modules in an app that uses Browserify. So I had to learn how it works and probably hack it in order to solve my problem.

So it's been a few months since I published something here. It’s not because I’m lazy (that’s true though) but because I was working on my second book. Now the book is almost finished and I’ll start actively blogging again. In this article we’ll see how to use the local storage of the browser as a communication channel.

I don’t think that we have to discuss the importance of having tests. That’s not something that we add to the project in addition. That’s something that the project is based on. And because testing is so important we have bunch of tools in our disposal. We all know (I hope) how to test our back-end code. However, once we move to the front-end is a bit different. At the moment we are working on a big single page application and testing is one of our main focuses. In this article you will see how to unit test our client-side JavaScript.

You've probably heard about ToDoMVC project. It's same ToDo application made with different frameworks. It's interesting how the same problem is solved by different programmers following different concepts. This article is about making the ToDoMVC application with AbsurdJS.

The browsers nowadays are smart. They optimize everything and help us to produce better applications. They process our code as fast as possible and even on mobile devices deliver a pleasant experience. However, at the same time, it is possible to write buggy code and make the browser freezing. We are not talking about slowing down the rendering. We are talking about no rendering at all. There are cases where we want so much from the browser that it just can’t handle it.

This blog has 460 posts. In some of them, I need to show video clips demonstrating some feature or browser behaviour. I found that it was much better to do that in an animated Gif. However, they play automatically which is kinda annoying. Imagine that we have ten Gifs on the page and while you are reading they all blink or show moving content. It's like having a page full with banners.

Hi guys, you would be pleased to know that I have teamed up with Packt Publishing to organize a Giveaway of the new book - Getting Started with Grunt: The JavaScript Task Runner. And three lucky winners stand a chance to win 3 digital copies of this book. Keep reading to find out how you can be one of the Lucky Winners.

The off-canvas menu is pretty popular nowadays. You know, the little button in the right upper corner which opens the navigation with a slide-in effect. That's what we are going to build in this article. The interesting things is that we are not going to use CSS neither HTML, only JavaScript.

The this keyword is always important in JavaScript. I'll admit that it took me some time to actually understand how it works. You know, sometimes JavaScript is all about the scope. Where you are and what you have an access to. This article is about the bind function. Something which I use very often.

If you believe in responsive design you probably use a lot of media queries. That's a CSS feature which gives us the power to add something like if statements. We are able to apply rules only if the current page meets certain conditions. In this article I'll show you how I use media queries in JavaScript.

If you build web applications you probably communicate with the DOM a lot. Accessing and manipulating DOM elements is the thing which we do almost every day. Very often we collect information from different controls, we need to set values, change the content of div or span tags. Of course there are million libraries which handle these actions. The most popular jQuery, is de factor a standard. However, sometimes you need something smaller. In this article we will build our own class for managing DOM elements.

This blog post is part of series about AbsurdJS. I'll continue filling the library with small and self organized black boxes. In the last article we talked about creating a JavaScript router. This time we will go through the process of making Ajax requests.

(The router here is now placed in its own project - Navigo. There is also another article Deep dive into client-side routing which you may find interesting)

Nowadays the popular single page applications are everywhere. Having such application means that you need a solid routing mechanism. Frameworks like Emberjs are truly build on top of a Router class. I'm still not sure that this is a concept which I like, but I'm absolutely sure that AbsurdJS should have a build-in Router. And, as with everything in this little library, it should be small, simple class. Let's see how such a module may look like.

This seems like a trivial task. However, it is not. There are few popular solutions, but they have a big problem and if you use them a lot you will spot the issue. read more comments

I like the quote that the programming is all about managing complexity. Maybe you've heard that the computer world is a giant construction of abstractions. We simply wrap things and produce new tools over and over again. Just think for a minute. The languages which you use have build-in functionalities and they are probably abstracted functions of other low level operations. It's the same with JavaScript. read more comments

I was surprised by the good feedback for JavaScript template engine in just 20 lines and decided to blog for another small utility function which I'm using often. While we are talking about JavaScript in the browser, most of the operations are asynchronous. We are dealing with callbacks all the time and sometimes we end up with awesome callback hell.

I'm still working on my JavaScript based preprocessor - AbsurdJS. It started as a CSS preprocessor, but later it was expanded to CSS/HTML preprocessor. Shortly, it allows JavaScript to CSS/HTML conversion. Of course, because it generates HTML it was normal to act as a template engine. I.e. somehow to fill the markup with data. read more comments

AbsurdJS became one of my favorite projects. I spent a good amount of time developing it and it's getting more and more interesting. I also received some positive feedback, so I think it is time to write a bit more about the module and explain how it actually works and what exactly is made for.

As javascript developers we all have to write a lot of for loops. Before a couple of months I saw an alternative syntax, which I really liked. It saves some time and it looks much nicer. However, one of my colleagues reported that this fancy syntax is not working as it should.

While I worked on AbsurdJS I needed a function which accepts numerous objects and combines their properties. I.e. something like the _.extend method of UnderscoreJS. What I did is actually to use one more dependency just because of such method. John-David Dalton nicely pointed out my mistake and added a simple function which solved the problem. I change it a bit and now it acts as UnderscoreJS's version.

I'm currently working on an animation-heavy web application. There are long chains of CSS transitions/animations, which I have to do. I wrote this little library, because I needed something lightweight with minimalistic API. I think that it deserves its own repository and I created one.

I'm very happy visiting Stackoverflow more often then Facebook these days. Today I landed on this question. I'm not sure that I understand the real context of the situation, but I found something helpful, which may be useful for you. What if we have an object and want to use it as a base for JavaScript class (i.e. function). The idea is to create instances and basically extend the original object.

I just noticed that the most popular repository in my GitHub account is EventBus. There is already an article about it, but it's a little bit outdated. I needed to clarify few things, so here is a new one.

I just answer on that question in StackOverflow. I think that this is a common quetion so it worths writing about it.

I just read this very interesting article published on know.cujojs.com. It's about aspect oriented programming. At the beginning I thought "Oh, that's kinda cool", but when I start thinking about how I'll use it in practice I notice something bad. The blog post itself is well written and informative, but doesn't include any cons of the used technique.

The good old Internet Explorer. It's full with surprises. I just fixed a bug happening only in IE. Everything works just great in every other browser.

I'm writing a lot of JavaScript these days. Sometimes it's really crazy what you can do with that language. I have bunch of tools in my arsenal and I wanted to document them somehow. Writing an article is a good option. By tools I mean patterns, piece of logic or just syntax shortcuts. However, this article is not exactly about design patterns, but it uses some of them. It's just about few ideas, which I came into during my daily scripting.

I created a simple JavaScript library for running functions in a sequence. The story behind this is published here. Today I made few interesting changes which deserve a new blog post. read more comments

While I'm working on Auxilio I end up in a sitatuation where I have to call few javascript functions in a sequence. It's an interesting how this could be solved and I'll be happy if you share your opinion for such problem.

Don't get me wrong, Jasmine is a wonderful framework. I'm using it for testing JavaScript in both places - front-end (in browser) and back-end (Nodejs). However, the client side has some disadvantages, which I just fixed. I hope that someday these changes will be moved to the official version (or at least I'll make a pull request very soon). read more comments

Checkout Revealing the magic: How to properly convert HTML string to a DOM element article.

I'm working on a chrome extension which accepts user text, work with it and output the result in a div. I used jQuery in the beginning, for several tasks, but later I decided to remove it and deal with pure javascript/DOM manipulations. The output div was filled by using innerHTML property and everything was ok. But, at some point, I sent few buttons to the user and he has to interact with them. However, once I update the div's content the event listeners are detached and the buttons became non-functional. read more comments