Re^3: vec overflow?

[LanX]

Ok, I had errors testing it out in the debugger, but turned out that just the automatic Data::Dump couldn't handle lvalues.

Asa suggestion:

I think using a foreach alias like shown in the docs for substr is a clearer alternative to lvalue-refs.

[BrowserUk]

I think using a foreach alias like shown in the docs for substr is a clearer alternative to lvalue-refs.

Maybe, though I've been aware of LVALUE refs, and using them regularly, for a long time, but I've never seen a use of a for alias used that way until I followed your link to a newer set of docs than I have locally.

Comparing:

my $ref = \ substr( $astring, $start, $len );
vec( $$ref, $offset, $size ) = 1;

To:

vec( $_, $offset, $size ) = 1 for substr( $astring, $start, $len );

The first one is very familiar (to me), whereas the latter just looks weird.

The presence of \ and $$ref seem stronger clues than the $_ and for.

[LanX]

A matter of taste ...

for me

vec( $_, $offset, $size ) = 1 for substr( $astring, $start, $len );

has no big advantage over

vec( substr( $astring, $start, $len ), $offset, $size ) = 1;

but

for my $chunk ( substr( $astring, $start, $len ) ) {
  ...
  vec( $chunk, $offset, $size ) = 1;
  ...
}

looks better for me than de/referencing and has the advantage to be documented in the docs.

And your use of \ substr( $astring, $start, $len ); is less frequent and makes me wondering about precedences.

(The practice to reference lvalues is still somehow new for me...)

OTOH your approach has the (dis?)advantage that the ref can be passed around w/o being restricted to the body of a for loop...

Well ... I'll better continue meditating about it after seeing more use cases! :)

[BrowserUk]

vec( $_, $offset, $size ) = 1 for substr( $astring, $start, $len ); has no big advantage over vec( substr( $astring, $start, $len ), $offset, $size ) = 1;

I agree. But I've only very recently 'discovered' nested lvalue subs -- this thread is the first time I've ever used the construct publicly -- and I've never seen it used anywhere else.

I can well imagine that there are lot of silent readers looking at

• vec( vec( ... ), ... ) = 1;
• vec( substr(... ), ... ) = 1;
• substr( vec( ... ), ... ) = $stuff;
• substr( substr( ...), ... ) = $stuff;
• substr( vec( substr( vec( substr( ...), ... ), ... ), ... ), ... ) = $something;

and thinking:

"I'd never use such an unholy constructs even if it meant my code running a 1000 times more slowly."