Announce: syntax across languages
Pascal Rigaux
What's this for?
* Language Designers:
Looking for operator or function names? Well have a look at the
following and
remember using existing one may ease the transition :)
* Language Users:
You know one language and want to find the corresponding operator or
function
in another language
* Language lovers:
Want to know the various ways people invented for
commenting/assigning/...?
CHANGES:
- the page has now much more entries
- one can check the missing entries per language at
http://merd.net/pixel/language-study/syntax-across-languages-per-language/
Contributions, suggestions and fixes are welcome.
--
Pixel
programming languages addict http://merd.net/pixel/language-study/
--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
David Feuer
Labeling the (==) and (/=) operators as deep equality does not seem quite
right. Their meaning _can be_ that of deep equality, but it can also be very
different. I think I have seen them called notional equality.
David Feuer
David Feuer
deal with the very large semantic differences between languages. It seems to
be designed mainly for Java-like OO languages, and can't deal with other
things too well. For example, Haskell has no built-in notion of control flow
whatsoever, no built-in (or standard) notion of exceptions, no concept of a
"statement", no built-in concept of a mutable variable, rather different
lexical rules than "syntax across languages" seems prepared for, a very, very
different idea of a class and method, which is really not OO at all, etc.
etc.
David Feuer
Pascal Rigaux
(this is about Haskell)
AFAIK:
- the deep vs shallow problem doesn't really occur in non mutable data
structures. And I think that's the main semantic difference between
various
equality testing.
- the fact that you can give an implementation of Eq *not* computing
deep
equality is no problem for me. The typical semantic *is* deep equality,
even
if the language doesn't ensure it.
Pascal Rigaux
> I think this "syntax across languages" thing is kind of silly, if it can't
> deal with the very large semantic differences between languages.
I agree it's kind of hard to mix everything together... but i'm trying
anyhow.
Either tell me how to enhance it or ignore it ;)
> It seems to be designed mainly for Java-like OO languages,
i think the Java camp will say the opposite, having a look at the many
functions needing an anonymous function[1]
eg: find filter foldl any all
> and can't deal with other
> things too well. For example, Haskell has no built-in notion of control flow
> whatsoever, no built-in (or standard) notion of exceptions, no concept of a
> "statement", no built-in concept of a mutable variable, rather different
> lexical rules than "syntax across languages" seems prepared for, a very, very
> different idea of a class and method, which is really not OO at all, etc.
AFAIK there are many things in haskell that looks alike things in other
languages. I agree i've not put any special monad entries (>> >>= ...)
since
haskell is the only one having this. I try to keep things as general as
possible (i didn't put the prolog's cut syntax...)
As for Haskell and OO, i see a nice parallel between mix-ins, java
interfaces
and type classes (one could say it has multi-methods :)
[1] well even if its do-able using inner classes or methods, it's not
really
practical, that's why you don't find those functions in Java.
Johan Redestig
acually, the variable identifier regexp for java is not correct. java
accepts $ in really obscure ways :)