Re: More about ksh_glob and completion

Re: More about ksh_glob and completion

[Peter Stephenson]

Sven wrote:
> Actually, I once thought about pre-compiling them, then had a look at
> the pattern code, was reminded of the options controlling pattern
> matching and...

Yes, the option kshglob is the one at the point of compilation.  This is
pretty much unavoidable in this case --- things like @(foo|bar) behave
completely differently in the two cases and it's just not worth compiling
it both ways on the off chance.  With globsubst, the only question is
tokenisation, and it's easy to make sure the pattern code is independent of
that (there were some glitches in that but I think they've gone).

But I still think precompiling patterns is OK as long as you do it at the
point of first use.  The difficult case

% unsetopt kshglob
% testpat() { setopt kshglob; [[ $1 = @(aardvark|bison) ]]; }

will always work so long as the pattern is compiled at that point.

I believe syntactic structures no longer get copied to be executed, so that
ought to be able to work.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Unit 300, Science Park, Milton Road,
Cambridge, CB4 0XL, UK                          Tel: +44 (0)1223 392070



**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

Re: More about ksh_glob and completion

[Sven Wischnowsky]

Peter Stephenson wrote:

> ...
> 
> But I still think precompiling patterns is OK as long as you do it at the
> point of first use.  The difficult case
> 
> % unsetopt kshglob
> % testpat() { setopt kshglob; [[ $1 = @(aardvark|bison) ]]; }
> 
> will always work so long as the pattern is compiled at that point.

Yes, that's what we do: see, for example, the reference to the `pats'
array in the Eprog in cond.c:199 (execcase() contains similar code).

Bye
  Sven


-- 
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de