TypeScript is an open source tool with 51.2K GitHub stars and 7.07K GitHub forks. Here's a link to TypeScript's open source repository on GitHub.
What is TypeScript?
Need advice about which tool to choose?Ask the StackShare community!
Sign up to add, upvote and see more prosMake informed product decisions
Sign up to add, upvote and see more consMake informed product decisions
Sign up to get full access to all the companiesMake informed product decisions
Sign up to get full access to all the tool integrationsMake informed product decisions
Visual Studio Code worked really well for us as well, it worked well with all our polyglot services and the .Net core integration had great cross-platform developer experience (to be fair, F# was a bit trickier) - actually, each of our team members used a different OS (Ubuntu, macos, windows). Our production deployment ran for a time on Docker Swarm until we've decided to adopt Kubernetes with almost seamless migration process.
After our positive experience of running .Net core workloads in containers and developing Tweek's .Net services on non-windows machines, C# had gained back some of its popularity (originally lost to Node.js), and other teams have been using it for developing microservices, k8s sidecars (like https://github.com/Soluto/airbag), cli tools, serverless functions and other projects...
Choosing to add TypeScript has given us one more layer to rely on to help enforce code quality, good standards, and best practices within our engineering organization. One of the biggest benefits for us as an engineering team has been how well our IDEs and editors (e.g., Visual Studio Code ) integrate with and understand TypeScript . This allows developers to catch many more errors at development time instead of relying on run time. The end result is safer (from a type perspective) code and a more efficient coding experience that helps to catch and remove errors with less developer effort.
ReactQL is written in TypeScript to provide full types/Intellisense, and pick up hard-to-diagnose goofs that might later show up at runtime. React makes heavy use of Webpack 4 to handle transforming your code to an optimised client-side bundle, and in throws back just enough code needed for the initial render, while seamlessly handling
import statements asynchronously as needed, making the payload your user downloads ultimately much smaller than trying to do it by hand.
React Helmet was chosen to handle
<head> content, because it works universally, making it easy to throw back the correct
<title> and other tags on the initial render, as well as inject new tags for subsequent client-side views.
<style> tags when using #StyledComponents.
React Router handles routing, because it works both on the server and in the client. ReactQL customises it further by capturing non-200 responses on the server, redirecting or throwing back custom 404 pages as needed.
Koa is the web server that handles all incoming HTTP requests, because it's fast (TTFB < 5ms, even after fully rendering React), and its natively #async, making it easy to async/await inside routes and middleware.
I needed to choose a full stack of tools for cross platform mobile application design & development. After much research and trying different tools, these are what I came up with that work for me today:
For the client coding I chose Framework7 because of its performance, easy learning curve, and very well designed, beautiful UI widgets. I think it's perfect for solo development or small teams. I didn't like React Native. It felt heavy to me and rigid. Framework7 allows the use of #CSS3, which I think is the best technology to come out of the #WWW movement. No other tech has been able to allow designers and developers to develop such flexible, high performance, customisable user interface elements that are highly responsive and hardware accelerated before. Now #CSS3 includes variables and flexboxes it is truly a powerful language and there is no longer a need for preprocessors such as #SCSS / #Sass / #less. React Native contains a very limited interpretation of #CSS3 which I found very frustrating after using #CSS3 for some years already and knowing its powerful features. The other very nice feature of Framework7 is that you can even build for the browser if you want your app to be available for desktop web browsers. The latest release also includes the ability to build for #Electron so you can have MacOS, Windows and Linux desktop apps. This is not possible with React Native yet.
Framework7 runs on top of Apache Cordova. Cordova and webviews have been slated as being slow in the past. Having a game developer background I found the tweeks to make it run as smooth as silk. One of those tweeks is to use WKWebView. Another important one was using srcset on images.
I configured TypeScript to work with the latest version of Framework7. I consider TypeScript to be one of the best creations to come out of Microsoft in some time. They must have an amazing team working on it. It's very powerful and flexible. It helps you catch a lot of bugs and also provides code completion in supporting IDEs. So for my IDE I use Visual Studio Code which is a blazingly fast and silky smooth editor that integrates seamlessly with TypeScript for the ultimate type checking setup (both products are produced by Microsoft).
I use some Ruby scripts to process images with ImageMagick and pngquant to optimise for size and even auto insert responsive image code into the HTML5. Ruby is the ultimate cross platform scripting language. Even as your scripts become large, Ruby allows you to refactor your code easily and make it Object Oriented if necessary. I find it the quickest and easiest way to maintain certain aspects of my build process.
For the user interface design and prototyping I use Figma. Figma has an almost identical user interface to #Sketch but has the added advantage of being cross platform (MacOS and Windows). Its real-time collaboration features are outstanding and I use them a often as I work mostly on remote projects. Clients can collaborate in real-time and see changes I make as I make them. The clickable prototyping features in Figma are also very well designed and mean I can send clickable prototypes to clients to try user interface updates as they are made and get immediate feedback. I'm currently also evaluating the latest version of #AdobeXD as an alternative to Figma as it has the very cool auto-animate feature. It doesn't have real-time collaboration yet, but I heard it is proposed for 2019.
For the UI icons I use Font Awesome Pro. They have the largest selection and best looking icons you can find on the internet with several variations in styles so you can find most of the icons you want for standard projects.
For the backend I was using the #GraphCool Framework. As I later found out, #GraphQL still has some way to go in order to provide the full power of a mature graph query language so later in my project I ripped out #GraphCool and replaced it with CouchDB and Pouchdb. Primarily so I could provide good offline app support. CouchDB with Pouchdb is very flexible and efficient combination and overcomes some of the restrictions I found in #GraphQL and hence #GraphCool also. The most impressive and important feature of CouchDB is its replication. You can configure it in various ways for backups, fault tolerance, caching or conditional merging of databases. CouchDB and Pouchdb even supports storing, retrieving and serving binary or image data or other mime types. This removes a level of complexity usually present in database implementations where binary or image data is usually referenced through an #HTML5 link. With CouchDB and Pouchdb apps can operate offline and sync later, very efficiently, when the network connection is good.
I use PhoneGap when testing the app. It auto-reloads your app when its code is changed and you can also install it on Android phones to preview your app instantly. iOS is a bit more tricky cause of Apple's policies so it's not available on the App Store, but you can build it and install it yourself to your device.
So that's my latest mobile stack. What tools do you use? Have you tried these ones?
Obviously, using ES6 and TypeScript is what makes it decent in browser contexts. Throw in a bit of React and now we're cooking with gas!
any type as output of JS module. When I have more time, I could just submit PR to DefinitelyTyped and it would be quickly accepted. Overall it gives less ambiguity for my code.
Go is a high performance language with simple syntax / semantics. Although it is not as expressive as some other languages, it's still a great language for backend development.
Python is expressive and battery-included, and pre-installed in most linux distros, making it a great language for scripting.
PostgreSQL: Rock-solid RDBMS with NoSQL support.
NATS: fast message queue and easy to deploy / maintain.
Docker makes deployment painless.
Git essential tool for collaboration and source management.
I think our #Frontend stack is pretty standard – but we have taken some deviations from a typical modern stack:
Yarn instead of npm. When yarn debuted, we never looked back. Now npm has pretty much caught up with speed and lockfiles, but yarn gives me confidence that my dependency installs are deterministic. Really interested in the plug-n-play (PnP) feature that removes the need for a node_modules folder, but haven't implemented this yet.
- Most server-side scripts, all unit tests, all build tools, etc. were driven by NodeJS.
- ExpressJS served as the 'backend' server framework.
- MongoDB (which stores essential JSON) was the main database.
- MongooseJS was used as the main ORM for communicating with the database, with KnexJS used for certain edge cases.
- MochaJS, ChaiJS, and ExpectJS were used for unit testing.
- Frontend builds were done with Gulp and Webpack.
- Package management was done primarily with npm - with a few exceptions that required the use of Bower (also configured with JSON).
- The frontend was build primarily with ReactJS (as the View) and Redux (as the Controller / Store / frontend model).
- Configuration was done with json files.
The only notable exceptions were the use of SCSS (augmented by Compass) for styling, Bash for a few basic 'system chores' and CLI utilities required for development of the app (most notably git and heroku's CLI interface), and a bit of custom SQL for locations where the ORM extractions leaked (the app is DB-agnostic, but a bit of SQL was required to fill gaps in the ORMs when interfacing with Postgres).
Typescript has been a win because, in general, it makes codebase maintenance less brittle. It's significantly easier to refactor in TS than JS, which encourages incremental improvements, file re-organizing, etc. Our developers are happier with the overall development experience.
The downside is that TS sometimes exacerbates problems caused by Node's fragmented ecosystem. Sometimes @types/
If your project is big enough, I'd say TS is nearly always worth it, but it can make selecting libraries a pain.
We are now re-considering TypeScript because 1) the tooling has improved significantly, and 2) and the root cause of the majority of our front-end bugs are related to typing (despite having PropTypes).
Used for Node.js personal projects that I think will have a longer lifetime than others, or that are combined with a web front end component like Angular (to share types).
Generally a poor developer experience. Usage decreasing recently compared to other preferred programming languages/platforms.
TypeScript is used in Kuro (https://github.com/Marc3842h/kuro).
Excellent design-time type checking and the ability for the Typescript compiler to attach typing information to metadata at compile time allows for relatively simple type checking at run-time as well.