bash: lines of a file in reverse order?

[Gregory D Hough]

Scott Elcomb wrote:
> On 8/24/06, Gregory D Hough <mr6re9-mI4xJ4qlgtBiLUuM0BA3LQ at public.gmane.org> wrote:
> 
>> >    Use tac:
>> >
>> > tac filename
>> >
>> >> I want to find the "last matching line" in a large log file.  My hope
>> >> was:
>> >>
>> >> tail -50 -r filename | grep -m 1 "matching"
>> >> would grab it for me if it exists in the last 50 lines of the log.
>> >
>> >
>> > grep "PATTERN" filename | tail -n1
>> >
>> A
>> #!/bin/bash
>> grep "INVALID" /var/log/messages | tail -n1
>>
>>   time sh A
>> Aug 24 15:52:22 localhost kernel: INVALID: IN=eth1 OUT= SRC=69.63.65.165
>>
>> real    0m0.044s
>> user    0m0.001s
>> sys     0m0.015s
>>
>>
>> B
>> #!/bin/bash
>> tail -50 /var/log/messages | tac | grep -m1 "INVALID"
>>
>>   time sh B
>> Aug 24 15:52:22 localhost kernel: INVALID: IN=eth1 OUT= SRC=69.63.65.165
>>
>> real    0m0.048s
>> user    0m0.004s
>> sys     0m0.016s
>>
>> Yep, method A is 4 to 5 thou faster than B most of the time
> 
> 
> That's kinda cool.  Might get more accuracy on the last statment if
> others' results were averaged in.
> 
> Still, that's a really interesting test, thanks for the results.  Is
> there a way to track memory usage during the same periods?
> 
Yes but my time is wasted and the env is a tally whacker.
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml