Beefy Boxes and Bandwidth Generously Provided by pair Networks
"be consistent"
 
PerlMonks  

Re^4: Hard syntax error or disambiguable parsing?

by pobocks (Chaplain)
on Jan 29, 2009 at 05:38 UTC ( #739773=note: print w/replies, xml ) Need Help??


in reply to Re^3: Hard syntax error or disambiguable parsing?
in thread Hard syntax error or disambiguable parsing?

It doesn't reuse the one created in the above line my $i;? Or do you mean uninitialized, rather than previously undeclared. Either way, why does a scalar member of an array not qualify?

for(split(" ","tsuJ rehtonA lreP rekcaH")){print reverse . " "}print "\b.\n";

Replies are listed 'Best First'.
Re^5: Hard syntax error or disambiguable parsing?
by merlyn (Sage) on Jan 29, 2009 at 05:41 UTC
    my $i = 5; for $i (10..15) { } print "$i\n";
    prints 5. So clearly, the $i being iterated is not the $i from the my immediately above it.

    A scalar member of an array is not a simple scalar. By Larry's rule, it has to be a simple scalar.

      Ahhhh... and he was enlightened. Basically, $i here is operated on as if it was:

      my $i; foreach local $i (0 .. 5){ ... }

      Do you have any ideas/pointers to the rationale for "no array members?" I understand that it is so, but fail to understand why it is so, when (for instance) one can local-ize individual array members.

      for(split(" ","tsuJ rehtonA lreP rekcaH")){print reverse . " "}print "\b.\n";

        Nope, not that either. A Perl for loop variable is magical - there is no simple "equivalent" syntax that conveys what is going on under the hood. See my reply to your earlier node for (I hope) some real enlightenment.


        Perl's payment curve coincides with its learning curve.

      I just tried the following code...which is a variation on what you show.

      #!/user/bin/perl use warnings; use strict; my $i = 10; print "$i\n"; foreach $i (0..5){ print "$i\n"; } # end foreach $i loop print "$i\n"; exit(0);

      It prints the following:

      10 0 1 2 3 4 5 10

      This is exactly what I believe the Camel says it should do since the loop variable is, if I recall correctly, always created anew as a lexical whose scope is the subsequent loop block (even though it appears to be created before the loop block). The output above seems to confirm that.

      I, like BrowserUk, however am a bit perplexed by the inability to use the $i[0] construct as the loop variable.

      Is it because the $i[0] implies a list structure which does not get created (via, for example, autovivification...spelling?)?

      ack Albuquerque, NM
        Actually, the loop variable will be lexical if there's a current lexical variable with that name, otherwise it will be local variable (with a local scope).
        my $i = 10; our $j = 10; sub print_it {say "[$i, $j]"} print_it; foreach $i (0 .. 5) {print_it;} print_it; foreach $j (0 .. 5) {print_it;} print_it; __END__ [10, 10] [10, 10] [10, 10] [10, 10] [10, 10] [10, 10] [10, 10] [10, 10] [10, 0] [10, 1] [10, 2] [10, 3] [10, 4] [10, 5] [10, 10]
        As you can see, $i is lexical, and its value isn't visible outside the loop. But $j is a package variable, who gets a localized value inside the loop. And then its value is visible outside the loop.
      How could I get the for loop to use the predeclared $i instead of using a local one?
        How could I get the for loop to use the predeclared $i instead of using a local one?
        You don't, and that's a good thing. (You mean "foreach loop" there, I presume.)
Re^5: Hard syntax error or disambiguable parsing?
by GrandFather (Saint) on Jan 29, 2009 at 06:02 UTC

    The "variable" in a Perl for loop is aliased to each element that is being iterated over. It is effectively a symbolic place holder that is only valid within the scope of the loop body. However strict requires that variables are declared. Declaring the variable before the loop header makes it look like a normal variable - it ain't.


    Perl's payment curve coincides with its learning curve.
      ... a symbolic place holder that is only valid within the scope of the loop body.

      Right, but it is subject to the same scoping rules as any other perl variable, i.e. to the variable it aliased:

      use strict; our $i; sub bar { print "<$i>" } for $i (1..3) { bar } # localized for my $i (1..3) { bar } print "'$i'" __END__ <1><2><3><><><>''
      use strict; my $i; sub bar { print "<$i>" } for $i (1..3) { bar } # already 'my' for my $i (1..3) { bar } print "'$i'" __END__ <><><><><><>''

      AhHa! In my earlier comment on one of the other threads, I had forgotten that the loop variable is not only lexical and in-scope in the loop's block, but that it is also aliased to the iterator.

      Thanks GrandFather. I learn so much (sometimes mulitiple times since I am frequently not a very memory-rich Pilgrim) from the Monestary.

      I am curious, however. Since the loop variable is aliased to the iterator (or to the iterator's values?); how does that work when the iterator doesn't appear to have variables, but rather constants (as in the case this thread is addressing (i.e., the (0..5) iterator)? And what would happen if you tried to change the iterator loop variable within the loop block>

      By way of an example of that last question:

      foreach $i (0..5){ $i = 9; print "$i\n"; }

      Actually I can answer that myself by trying it. No need to answer that part of the question; I don't want to abuse the Monks with things I can easily do my self.

      But the question before that last one is still nagging me.

      ack Albuquerque, NM

      Danke schonn for this and your other reply.

      I think I have too much of a tendency to try and explain away magic as a result of regular syntax objects; sometimes, there's just a wizard behind that curtain ;-)

      for(split(" ","tsuJ rehtonA lreP rekcaH")){print reverse . " "}print "\b.\n";

        Well, a lot of the magic does just come down to pulling the right levers. ;)


        Perl's payment curve coincides with its learning curve.

Log In?
Username:
Password:

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: note [id://739773]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others studying the Monastery: (5)
As of 2022-12-10 00:01 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    No recent polls found

    Notices?