Katahdin - A language where syntax and semantics are mutable at runtime | Hacker News Reader

Katahdin - A language where syntax and semantics are mutable at runtime

chrisseaton.com

31 points_b8r018 comments

I hope my master's project is something interesting like this.

However, my first thought upon reading the title of the link was about the comments that were going to appear here: either "lisp already does this" or "this will never be accepted as a mainstream language because it's too chaotic to the project."

Section 4.2.3 answered the question I was searching for: what happens in this hybrid model when an "interior" language fails to parse? It's all fun and games to strap on a new mini langage to your source file, but it's less fun and games when you misplace a paren. I wondered if the entire block threw a parse exception or if the whole file failed to load. It appears based on the skim-read that it's the later.

p.s. why is "270 version control commits" an achievement?

Means he used version control. (should be ordinary and expected but is quite rare for grad student code, in my experience)

This was a Computer Science dissertation (by a good friend of mine). It scratched an itch, furthered someones education and checked the box of a required component of a degree course. I don't think it was looking to change the world. I don't think anyone is writing apps in Kat.

Why did he go into military service?

I don't mean to pry, but was he forced/drafted, or just a personal choice? Seems unfortunate to me.

It's a huge privilege to command Her Majesty's soldiers, and something that one can only do when young, so I went for it. Also, the leadership training and maturity that you gain from the Royal Military Academy Sandhurst are second to none in the world.

Thanks for the clarification Chris; my worst fear would have been you trading a few years in exchange for your university tuition. But seems like that wasn't the case.

Enjoy your journey, don't forget to keep hacking :-)

Cheers!

How was the name chosen? As someone from Maine I expected the author to be local and was surprised to see he was from the UK.

Languages such as OCaml already have the ability to do this at compile time. What's the benefit to allowing this at runtime? (Which effectively restricts this language to being interpreted, and severely limits the use of external tools such as code refactorers and smart editors...)

Common Lisp does this and it's not restricted to being interpreted, you just have to include compiler, linker etc. as part of the language runtime.

So if I throw a bunch of Python code into a bunch of C strings and have main() set up a Python interpreter to eval those strings, I have compiled a Python program?

That seems like quite a copout.

What are you talking about? That is nothing like the topic of the article or what the parent comment is talking about.

Many Common Lisp runtimes compile things that you "eval" into native machine code when the eval statement runs. That is what the parent is referring to.

Many other languages have different "phases" of compilation. An example is Perl 5's BEGIN, which is Perl code that runs at compile time. This means you can write Perl to affect the compilation of the Perl program.

(Why would you do this? Here's an example. When loading the module List::Util, you want to be able to use its "first" function as though it was declared in your module. You can do this because List::Util can run some code at compile-time that installs the "first" subroutine into your namespace. Then as compilation (and runtime) continues, you will magically have that "first" subroutine in your namespace, and the program will compile properly.

If there was no ability to run code at compile time, then importing stuff from a module into the caller would have to be a language feature. Making the action of running code at compile time (and installing subroutines at runtime) makes it possible to implement this without help from the language author.)

I am strongly reminded of Perl 6 and Lisp.

I guess my question is, "what does Katahdin provide that Perl 6 doesn't" (leaving aside the .NET integration)?

I guess my question is, "what does Katahdin provide that Perl 6 doesn't"

The author with his degree?

Well, yes. :-) But I'm considering the language from me, a user.

> I guess my question is, "what does Katahdin provide that Perl 6 doesn't"

A language that doesn't look like line noise?

I like Perl. :P

Makes me wonder how it affects performance when done at runtime

How is this better than macros and read-macros?

Reply on news.ycombinator.com