News for hackers. Stuff that matters.
Re:Perl Cannot Be Parsed: A Formal Proof (2008) by fijal on Sunday May 26, 2013 @09:13AM (#5770994)
Er. This proof is seriously flawed, starting with "what is Perl"? How do we define a language that's not parseable? Does it only mean that you can't know what the program will do until you run it? That's true for Ruby, Python, pretty much everything that has any level of meta-programming. Also, it lacks any formal definition of what is to be proved. Statement: "Perl is unparseable" is not very well defined. "Perl" is the first problem, the second problem is "unparseable" - does it mean you can't write a grammar that does apply to all valid perl programs and does not to all invalid ones? We come back to what's "valid" and "invalid" program.

Obviously if the statement is "you can't write any useful linting tools", which seems to be what the author is after, than fine, but you have to define "useful".

Re:Perl Cannot Be Parsed: A Formal Proof (2008) by chrismorgan on Sunday May 26, 2013 @08:19PM (#5772891)
Ruby cannot be parsed (e.g. for ``x - y`` the ``-`` could be unary if ``x`` is callable or binary otherwise).

Python, on the other hand, can be parsed entirely unambiguously.

Re:Perl Cannot Be Parsed: A Formal Proof (2008) by happy_dino on Sunday May 26, 2013 @10:14AM (#5771116)
> How do we define a language that's not parseable?

You can read up on that.

> Does it only mean that you can't know what the program will do until you run it?

No.