Re: Print Number With Implied Decimal Point

Replies are listed 'Best First'.
Re^2: Print Number With Implied Decimal Point by Eliya (Vicar) on Apr 18, 2012 at 15:22 UTC
`use bigint` applied globally severely distorts the results. As both of your test cases work on strings anyway, you'd get a more useful comparison when restricting the scope of `bigint` to where it's required. With your original version, I get on my machine: `Rate substr regex substr 481/s -- -14% regex 558/s 16% --` [download] while with the restricted scope of `bigint`, I get `#!/usr/bin/perl use warnings; use strict; use feature 'say'; use Benchmark qw/cmpthese/; my @list; { use bigint; my @chars = ('0' .. '9', 'a' .. 'f'); @list = map {join q(), map $chars[rand @chars], 1 .. 32} 1 .. 100; say for @list; $_ = (hex $_)."" for @list; say for @list; } cmpthese(0, { substr => sub { my @l = @list; substr $_, -2, 0, '.' for @l; }, regex => sub { my @l = @list; s/(..)$/.$1/ for @l; } }); __END__ Rate regex substr regex 5672/s -- -80% substr 28469/s 402% --` [download]	[reply] [d/l] [select]
Re^3: Print Number With Implied Decimal Point by kennethk (Abbot) on Apr 18, 2012 at 15:38 UTC
Note that with a properly crafted regex, the timing difference between it and substr is pretty small; limiting the scope of bigint and using the look-ahead regex from Re^2: Print Number With Implied Decimal Point: `Rate regex substr regex2 regex 2926/s -- -78% -80% substr 13087/s 347% -- -11% regex2 14782/s 405% 13% --` [download] Update: see Re^2: Print Number With Implied Decimal Point #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.	[reply] [d/l]
Re^4: Print Number With Implied Decimal Point by Eliya (Vicar) on Apr 18, 2012 at 15:46 UTC
the timing difference between it and substr is pretty small Yes, sure. My main point was the side effects of applying `use bigint` globally (which completely washes out any difference there might be), and not so much to support that substr is faster than regex.	[reply] [d/l]
Re^3: Print Number With Implied Decimal Point by j355ga (Initiate) on Apr 19, 2012 at 14:01 UTC
Limiting the bigint scope makes a HUGE difference in my tests. 2,100 rows per second vs 700 per sec. Thanks for the tip!	[reply]
Re^2: Print Number With Implied Decimal Point by kennethk (Abbot) on Apr 18, 2012 at 15:08 UTC
If you want to make the regex solution run much more quickly, use Look Around Assertions in place of the capture and reinsert. `cmpthese(0, { substr => sub { my @l = @list; substr $_, -2, 0, '.' for @l; }, regex => sub { my @l = @list; s/(..)$/.$1/ for @l; }, regex2 => sub { my @l = @list; s/(?=..)$/./ for @l; }, });` [download] yields `Rate substr regex regex2 substr 73.1/s -- -39% -71% regex 121/s 65% -- -52% regex2 250/s 242% 107% --` [download] Update: As per JavaFan's comment, I had a typo in my sub. Replaced `s/(?=..)$/./` with `s/(?=..$)/./` which returns the correct result, but at substantially poorer performance. #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.	[reply] [d/l] [select]
Re^3: Print Number With Implied Decimal Point by JavaFan (Canon) on Apr 18, 2012 at 17:11 UTC
These numbers look too good to be true: `$_ = "9876543"; s/(?=..)$/./; say; __END__ 9876543` [download] You cannot actual replace a zero-width assertion. And before benchmarking, you should always check whether you're producing the correct results. Bogus solutions noone cares for, even if they're fast.	[reply] [d/l]