NAME

stats - helps you to calculate data from lines

SYNOPSIS

$ stats [options] FILE ...

or

$ cat FILE | stats [option]

options:
         --file          file path
    -f,  --fields        select only these fields to take stats
    -d,  --delimiter     specify the delimiter to split the log
         --more          calculate more results(NOTE: require more memory)
    -t,  --through       STDIN is shown
         --tsv           show the result as TSV
         --csv           show the result as CSV
    -di, --digit         number of the digit after decimal point(default:2)
         --no-comma      not add the comma
         --crlf          put a line feed as CRLF
         --cr            put a line feed as CR
                         (default line feed is LF)
    -s,  --strict        treats only the strict numbers field
         --rc            rc file(default: .statsrc)
    -h,  --help          show this help
    -v,  --version       show the version

EXAMPLE

If you have the log file which include response time, and response time were logged at the 7th element, for example you write the command like below

$ stats -f7 access_log

then show the result like this

--------- -------- 
                7  
--------- -------- 
 count      3,096  
 sum       747.20  
--------- -------- 
 average     0.24  
--------- -------- 
 max         5.01  
 min         0.01  
 range          5  
--------- -------- 

Or no option was spesified, stats will calculate the first element.

$ stats access_log

 --------- ------ 
               1  
 --------- ------ 
  count       10  
  sum         55  
 --------- ------ 
  average   5.50  
 --------- ------ 
  max         10  
  min          1  
  range        9  
 --------- ------ 

This is a bit complex example.

$ cat example_log
1100,1200
2300,1300
3400,1400

$ cat example_log | stats -d',' -f1,2 --no-comma --digit 3 --more

 ---------- ------------- -------
                       1       2
 ---------- ------------- -------
  count                3       3
  sum               6800    3900
 ---------- ------------- -------
  average       2266.667    1300
  median            2300    1300
  mode          2266.667    1300
 ---------- ------------- -------
  max               3400    1400
  min               1100    1200
  range             2300     200
  variance   1323333.333   10000
  stddev        1150.362     100
 ---------- ------------- -------

OK, you can get the result as CSV

$ cat example_log | stats -d',' -f1,2 --no-comma --digit 3 --more --csv

,"1","2"
"count","3","3"
"sum","6800","3900"
"average","2266.667","1300"
"median","2300","1300"
"mode","2266.667","1300"
"max","3400","1400"
"min","1100","1200"
"range","2300","200"
"variance","1323333.333","10000"
"stddev","1150.362","100"

CONFIGURATION

stats command will look for a configuration file before reading its command line parameters.

The default configuration file is .statsrc but you can specify it by --rc option. And the location of a configuration file is $HOME but if the STATSRC_DIR environment variable is set, stats will look for config in that directory or current.

The configuration file is just a list of names and values, separated by colons. Values that take lists are just space separated. A sample configuration file might read:

more: 1
tsv: 1

AUTHOR

Dai Okabayashi <bayashi@cpan.org>

SEE ALSO

App::LogStats

LICENSE

This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. See perlartistic.