On 25 February 2014 03:45, Paul Makepeace pa...@paulm.com wrote:
Finally. But don't believe the python/perl comparison troll, as
python, for once, actually outguns perl on a character chomping basis,
sub pairwise_sum ($arg1, $arg2) {
return map { $arg1-[$_] + $arg2-[$_] } 0 .. $#$arg1;
On 25 Feb 2014, at 03:45, Paul Makepeace pa...@paulm.com wrote:
On Mon, Feb 24, 2014 at 2:04 PM, Steve Mynott steve.myn...@gmail.com wrote:
http://perltricks.com/article/72/2014/2/24/Perl-levels-up-with-native-subroutine-signatures
Finally. But don't believe the python/perl comparison troll,
On 25 Feb 2014, at 11:18, James Laver james.la...@gmail.com wrote:
But I probably already have List::MoreUtils imported (because Perl).
sub pairwise_sum ($a1, $a2) {
zip @$a1, @$a2;
}
And lets ignore the fact that the perl version that was used in the article
was buggy. If $arg2 is
On 25/02/14 11:18, James Laver wrote:
On 25 Feb 2014, at 03:45, Paul Makepeace pa...@paulm.com wrote:
On Mon, Feb 24, 2014 at 2:04 PM, Steve Mynott steve.myn...@gmail.com wrote:
http://perltricks.com/article/72/2014/2/24/Perl-levels-up-with-native-subroutine-signatures
Finally. But don't
On 2014-02-25 04:45, Paul Makepeace wrote:
On Mon, Feb 24, 2014 at 2:04 PM, Steve Mynott steve.myn...@gmail.com wrote:
http://perltricks.com/article/72/2014/2/24/Perl-levels-up-with-native-subroutine-signatures
Finally. But don't believe the python/perl comparison troll, as
python, for
What about Regexp::Assemble ?
J.
On 25 February 2014 12:49, Dirk Koopman d...@tobit.co.uk wrote:
I am looking at an application where some incoming string needs to be
compared to many (certainly 10s, probably 100s but not 1000s) regexes. This
being a message passing application, speed is more
Dirk Koopman writes:
... some incoming string needs to be compared to many (certainly 10s,
probably 100s but not 1000s) regexes. ... speed is more important than
it might otherwise be.
Aaron Crane mentions a couple of modules which might help (in explaining
his module which does something
Hey Dirk,
The search time you're after is Trie.
http://en.wikipedia.org/wiki/Trie
https://metacpan.org/search?q=trie
The POD of Dan Kogai's Regexp::Trie lists some alternatives and the
description compares them a little.
https://metacpan.org/pod/Regexp::Trie#DESCRIPTION
The search time you're after is Trie.
I mean search term...
- Alex
If you're looking to find the matches of a large number of strings in
a body of text, there's also https://metacpan.org/pod/Text::Scan .
AFAIK, it's very fast, as it's a C implementation of
http://en.wikipedia.org/wiki/Aho%E2%80%93Corasick_string_matching_algorithm
It doesn't do regexes though,
On 25 Feb 2014, at 11:31, Matt Lawrence matt.lawre...@virgin.net wrote:
On 25/02/14 11:18, James Laver wrote:
But I probably already have List::MoreUtils imported (because Perl).
sub pairwise_sum ($a1, $a2) {
zip @$a1, @$a2;
}
And lets ignore the fact that the perl version that was
Alex Balhatchet ka...@slackwise.net wrote:
The POD of Dan Kogai's Regexp::Trie lists some alternatives and the
description compares them a little.
The regexes generated by Regexp::Trie will almost certainly be slower
to execute than the regex you'd get with the naive approach:
my $pattern =
On 25 February 2014 14:59, Aaron Crane p...@aaroncrane.co.uk wrote:
That's because, thanks to demerphq++, Perl 5.10 and above have a
built-in trie optimisation which is defeated by the cleverness of the
Regexp::Trie regexes. To take the example from the Regexp::Trie
documentation:
Oh my,
On Tue, Feb 25, 2014 at 03:25:54PM +, Alex Balhatchet wrote:
On 25 February 2014 14:59, Aaron Crane p...@aaroncrane.co.uk wrote:
That's because, thanks to demerphq++, Perl 5.10 and above have a
built-in trie optimisation which is defeated by the cleverness of the
Regexp::Trie regexes.
On Feb 25, 2014 7:16 AM, James Laver james.la...@gmail.com wrote:
On 25 Feb 2014, at 11:31, Matt Lawrence matt.lawre...@virgin.net wrote:
On 25/02/14 11:18, James Laver wrote:
But I probably already have List::MoreUtils imported (because Perl).
sub pairwise_sum ($a1, $a2) {
zip
On Tue, Feb 25, 2014 at 03:59:21PM +, Nicholas Clark wrote:
On Tue, Feb 25, 2014 at 03:25:54PM +, Alex Balhatchet wrote:
On 25 February 2014 14:59, Aaron Crane p...@aaroncrane.co.uk wrote:
That's because, thanks to demerphq++, Perl 5.10 and above have a
built-in trie optimisation
On 25 Feb 2014, at 16:45, Nicholas Clark n...@ccl4.org wrote:
Vendor supplied perl is going to be slower than one you compile yourself.
Or at least, it won't be faster.
Historically this wasn’t always the case*. When ubuntu first got going, their
packages were a little faster to load than
The blog post modified the original from returning an arrayref to
returning a list.
Restoring the arrayref, I'd:
use Algorithm::Loops 'MapCarE';
use List::Util 'sum';
sub pairwise_sum { [ MapCarE { sum @_ } @_ ] }
(note that the only improvement offered by 5.19.9 would be to
arbitrarily limit
On 26 February 2014 03:59, James Laver james.la...@gmail.com wrote:
Ah yes. I’d just woken up.
sub pairwise_sum($a2,$a2) {
map {shift($_)+shift($_)} (zip @$a1, @$a2);
}
... I think you need that book we're writing more than you need to write it
;)
--
Kent
19 matches
Mail list logo
