Re: basic question about runtime query parsing

On Tue, 26 Jul 2005, Chris Dollin wrote:


> Laurence Finston wrote:


> I don't see as those are particularly difficult to handle; you have to
> be prepared for the dynamically linked code to fail, [...]


My point was that compile-time and link-time errors in the "inner"
code would have to be caught when the "outer" program was running.
It seems to me this would make the actual programming more complicated.
If the generated code could be fairly complex, containing
conditionals and nested function calls, there could easily be
a huge number of possible cases. On the other hand, if the generated
code falls into a few stereotypical patterns, this would be less of a
problem.




> It may be portable enough. If you're prepared to stick to POSIX
> systems.


That seems like a reasonable constraint.


> I only meant to imply "this does not require a modern technology".


Generally speaking, I think one of the advantages of modern systems
is that they permit simpler and more straightforward approaches.
I would expect to find trickier kinds of programming on more limited
systems.


> Run-time code generation eg JIT.


Sorry, I don't know what this is. Please correct me if I'm wrong,
but I would assume that "code generation" means that new code is
created, not that existing code in memory is overwritten while the
program is running. I'm not talking about the kind of modifications
performed by linkers or debuggers. To the best of my knowledge,
on Unix-like systems, the actual machine instructions of a running
program are in read-only memory, so I wouldn't know how to
overwrite them. I suppose there's a way of doing it, though.


Laurence

Return to the comp.compilers page.
Search the comp.compilers archives again.