Starting the Rift Programming Language

During my first semester of my Masters Degree I had two particular courses that I found quite interesting. One was called Advance Topics of Compilers and the other was Advance Topics of Distributed systems. The objective of the first course was the study of a more complex compiler design than the one given in a previous course of *Compilers*, while the second one was the study of distributed systems, programming for distributed systems, among other things.

After these two courses started I decided to develop a project that would be used for both courses. That project would be a programming language which would have a concurrency paradigm.

Languages

At first I started exploring other languages that had concurrency paradigms. Among others, I explored the Go Programming Language, which ended up being one of the main sources of inspiration for this project.

Go is a language that makes it simple to use threads, known by goroutines. To start a new thread the go command is used. To see it working visit this link.

go say('World')
say('Hello')

What the code above does is to execute the say() function twice, only the first function runs in a different goroutine than the second one. When executing this function you may see Hello followed by World or vice versa.

This is such a simple way to start a new thread. Just calling a function with go before the name of the function. This was something I had to had on my language.

One other thing Go had that was very interesting was Channels. Channels are a very simple way to send some value from a goroutine to another. See this example. So to send and receive, respectively, something from a Channel you do the following:

c <- 10
x := <- x

In the code above the c variable would be a channel and the x variable would be some integer variable. The first line is an example of sending the number 10 to channel c. In another thread you can simple call the second line (assuming c would be the same channel for both threads) to receive something from the channel.

Multithreading

Meanwhile back at the course about distributed systems I was exploring different ways to program algorithms that used threads. Among a few technologies I used the POSIX Threads standard, or pthreads. This standard defined an API that can be used with the C programming language to create multithreading programs.

I made some experiments with pthreads for a while by implementing a few algorithms.

At this moment I started thinking how I would implement the compiler in terms out outputted code. In this project I just couldn’t output any low level code like x86 assembly because I had never worked with x86 before. And to implement a compiler that turns a high level language with multithreading features into x86 in such limited time would have been just impossible. Well, not impossible, just very hard with a high probability of failure involved, all with limited time.

So I decided to just output C code. The outputted code would be C99 using pthreads.

Rift

Having those things I went to develop a concept for the language. That concept would be what the language would look like, what would it allow, what data types did it had, how the multithreading feature would look like, if it had classes, among many other things.

I came up with this end result.

The language also needed a name. Because this would have the notion of parallelism I started thinking about division. So I went to a synonymous dictionary and looked for synonymous of Division. The name Rift showed up. It sounded cool. And like this the development of the Rift Programming Language began!

Tags: ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: