James Edward Gray II <[EMAIL PROTECTED]> wrote:
> On Dec 8, 2003, at 11:28 AM, Jeff Westman wrote:
>
> > [EMAIL PROTECTED] wrote:
> >
> >> Here is some syntax I found a little funny, and I was hoping somebody
> >> could explain this too me:
> >>
> >> opendir(DIR, $some_dir) || die "can't opendir $some_dir: $!";
> >> @dots = grep { /^\./ && -f "$some_dir/$_" } readdir(DIR);
> >> closedir DIR;
> >>
> >> The first and 3rd lines I have no problem with.
> >
> > The first line is better written as
> >
> > opendir(DIR, $some_dir) or die "can't opendir $some_dir: $!";
> >
> > Note the use of 'or' instead of '||'. Precedence is extremely
> > important,
> > especially in perl where so much of the code can be "short-circuited".
>
> Changing the || to or in the above line has no effect.
Right, but too often one uses '||' and meant to use 'or'. It bites me every
time <g>.
> However, as
> Jeff said, parens in Perl are often optional on subroutine calls and if
> we leave those out:
>
> opendir DIR, $some_dir or die "can't opendir $some_dir: $!";
>
> We get the traditional Perl open() call and here it matters if we use
> || or or. Why do we write it like that? We like the way it reads.
>
> >> Its that line with the grep in it.
> >> grep is obviously a built-in perl function.
> >>
> >> This is something I've seen before ...
> >> the output of the function readdir is "trickling down" to become the
> >> input of the function grep ?
> >
> > One minor problem, there is a comma missing before the 'readdir':
> >
> > @dots = grep { /^\./ && -f "$some_dir/$_" }, readdir(DIR);
>
> The line compiles fine. The block version of grep() does not require a
> comma.
Right again.... this is a case where using parens will get you :) ...... I
usually do:
@dots = grep /^\./ && -f "$some_dir/$_", readdir(DIR);
(no parens)
> perldoc -f grep
>
> > No, you would use a while or for loop and then search for files not
> > beginning with a dot.
>
> > So while-readdir, search for files (only) and test if they DONT begin
> > with a dot, then store them into an array. Now that we have the
>
> You say this twice, but that still doesn't make it true.
Woops ... my bad!
> I think you
> need to reread that grep() call. ;) We're searching for files that DO
> begin with a dot here and placing them in the fitting @dots array.
My bad :)
> >> If that is so, what is all that business with the curly braces ?
> >> I thought curly braces are supposed to denote code sections ?
>
> This is correct, it's a "chunk of code". It will be executed once for
> each item returned by readdir().
>
> >> p.s. It seems some of my confusion comes from the fact that I am used
> >> to
> >> putting () around funtion calls. That still doesn't help with the
> >> curly
> >> braces though ... %^)
> >
> > Parens are often optional. When in doubt, use them. I never use them
> > in
> > simple print statemtents, but some functions like 'split', it just
> > makes more
> > sense.
>
> I think the easiest rule to start learning with in use parens for your
> own subs and leave them off of built-ins, except when you need them to
> clarify/enforce precedence.
I might add that parens can make your code more readable, but can also make
it harder to read some times too. Parens often denote precedence, while
braces tend to be used for scoping. This isn't a hard and fast rule though.
Case-in-point, read: perldoc -f map
-Jeff
__________________________________
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing.
http://photos.yahoo.com/
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
