I liked this one because I've always wanted "max" and "min" programs.
What I usually do is sort and use head or tail. That is very inefficient.
So based on your work I made this:
awk 'BEGIN {max=$1} {if ($1>max) max=$1} END {print max}'
It seems to do what I want for whole lines if I do this:
awk 'BEGIN {max=$0} {if ($0>max) max=$0} END {print max}'
It looks like these two lines give the same output for most files:
awk 'BEGIN {max=$0} {if ($0>max) max=$0} END {print max}' filename
sort filename | tail -n1
Of course the awk code is much faster and I assume it uses much less
memory, especially for larger input files.
If input is a single column of numbers, awk will provide the numerical
maximum number, but sort will not do that unless the -g option is used
(sometimes -n would be enough, but not always).
Any ideas about this?
I'm having a problem getting min to work. This gives me a blank line as
output:
echo -e "1\n2" | awk 'BEGIN {min=$1} {if ($1<min) min=$1} END {print min}'
I'm obviously missing something!
Mike
On Fri, 30 Oct 2015, Saul Alanis wrote:
> Thanks Gerry.
>
> With your suggestion a solution was found:
>
> awk -F"|" 'BEGIN {max = 0} $2 ~ /foo/ {if ($1>max) max=$1} END{print max}'
>
> Sincerely,
> SDA
>
>
> On Fri, Oct 30, 2015 at 7:52 PM, gerry <gsker at skerbitz.org> wrote:
>
>> Saul,
>> It sounds like you just want the max for the first field?
>> gawk '/foo/{if (max < $1) max=$1} END {print $1}' file
>>
>> But that doesn't seem likely since it's too simple - no offence intended.
>>
>> Did you want the 2nd field of the row with the max of the first field?
>> gawk '/foo/{if (max < $1) {max=$1;save3=$3} } END {print save3;}'
>>
>> I'm ignoring the vertical bars because your example had spaces which awk
>> recognizes by default.
>> If the spaces are not consistent then you might want to use -F.
>>
>> You can probably find a good example of what you want
>> on http://commandlinefu.com
>>
>> --
>> gsker at skerbitz.org
>>
>>
>> On Fri, 30 Oct 2015, Saul Alanis wrote:
>>
>>
>>> I have a file with multiple fields;
>>>
>>> 2 | foo | bar
>>> 4 | bar | foo
>>> 1 | bar | foo
>>> 3 | foo | bar
>>>
>>> My goal is to sort the first field numerically and print the first field
>>> of the last result.
>>>
>>> awk -F"|" '/foo/ {print $1 | "sort"}'
>>>
>>> awk -F"|" '/foo/ {number=$1} END {print version}'
>>>
>>> Help is greatly appreciated :)
>>>
>>> SDA
>>>
>>>
>>>
>>> _______________________________________________
>> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota
>> tclug-list at mn-linux.org
>> http://mailman.mn-linux.org/mailman/listinfo/tclug-list
>>
>