On Wednesday, April 24, 2002, at 02:45 , Kevin Cornmell wrote:
[..]
> Or try anchoring that regex.
>
> if ($radentry =~/\n\n$/){
> foo;
> }
this rules.
> -----Original Message-----
> From: John W. Krahn [mailto:[EMAIL PROTECTED]]
[..]
>> Quick question. How can I check if the last two characters of a string
>> are
>> both new line's (i,e, \n)
>>
>> This is what I have come up with so far:
>>
>> if ($radentry =~ /"*\n\n/")
>
> if ( substr( $radentry, -2 ) eq "\n\n" )
I tried to get these to work - and was not able to
get them to work....
So I figured, why not test the benchmark of Kevin's
idea over and against the two basic 'qr' strategies
if ($radentry =~ $regEx){ # Reg
or
if ($radentry =~ /$regEx/){ # Reg2
or
if ($radentry =~ /$regEx/o){ # Reg3 - note the compile once 'o'
where
my $regEx = qr/\n\n$/; # defined globally in the benchmark
and the survey says:
[jeeves:~/perl] drieux% perl SillyDoubleNewLine.pl
Benchmark: timing 1000000 iterations of Reg, Reg2, Reg3, Regular, Regular1.
...
Reg: 3 wallclock secs ( 3.43 usr + 0.00 sys = 3.43 CPU) @ 291545.
19/s (n=1000000)
Reg2: 3 wallclock secs ( 3.42 usr + 0.00 sys = 3.42 CPU) @ 292397.
66/s (n=1000000)
Reg3: 2 wallclock secs ( 2.98 usr + 0.00 sys = 2.98 CPU) @ 335570.
47/s (n=1000000)
Regular: 3 wallclock secs ( 3.04 usr + 0.00 sys = 3.04 CPU) @ 328947.
37/s (n=1000000)
Regular1: 2 wallclock secs ( 3.18 usr + 0.00 sys = 3.18 CPU) @ 314465.
41/s (n=1000000)
[jeeves:~/perl] drieux%
regular1 there is Regular with the '/o' compile once option....
so we are down into the 'scale problem' of what benchmark can handle?
Would using Time::HiRes really help us here????
ciao
drieux
---
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]