Shell Programming and Scripting

Star1 ,NetWork,09/02/2008
Star1 ,NetWork,10/02/2008
Star1 ,NetWork,11/02/2008
Star2 ,NetWork,08/03/2008
Star2 ,NetWork,09/04/2008
Star2 ,NetWork,10/05/2008

csadev:/home/jmcnama> cat newfile
Star1 ,NetWork,09/02/2008
Star1 ,NetWork,11/02/2008
Star2 ,NetWork,08/03/2008
Star2 ,NetWork,10/05/2008

#!/bin/ksh
epoch_secs()
{
        perl -e '
                use POSIX qw(strftime);
                $fmt = "%s";  # %s = epoch seconds
                # mm/dd/yyyy
                $mday = substr("$ARGV[0]",3 , 2);
                $mon =  substr("$ARGV[0]", 0 ,2);
                $year = substr("$ARGV[0]", 6 ,4);
                $epoch_sec =
                  strftime($fmt, 0, 0, 0, $mday , $mon - 1, $year - 1900, -1, -1, -1);
                print "$epoch_sec";
                ' "$1"
}
while IFS=, && read fld1 fld2 fld3
do
	echo "$fld1,$fld2,$fld3,$(epoch_secs $fld3)"      	
done < filename  | sort -t"," -k1.1,1.6 -k4.1,4.10 | \
    awk -F, '{ if(NR==1) {
                     print sprintf("%s,%s,%s",$1, $2, $3) 
                     keep=$1; 
                     old=sprintf("%s,%s,%s",$1, $2, $3) 
              }
              if($1!=keep) { 
                       print old;                        
                       print sprintf("%s,%s,%s",$1, $2, $3) 
                       keep=$1
                       }
              old=sprintf("%s,%s,%s",$1, $2, $3 )
            }  
              END { print sprintf("%s,%s,%s",$1, $2, $3)  }              
           '  > newfile

#!/usr/bin/env python

"""
   Read the log file, returning the highest and lowest dates for each star_num
"""



import time
import sys
import os

##########
#Make sure we have an input file.
##########
try:
    input_file = sys.argv[1]
except IndexError:
    sys.stderr.write("No input file specified.")
    sys.exit(1)

##########
#Make sure we can open the file
##########
try :
    input = open(input_file, 'r')
except IOError:
    sys.stderr.write("File is invalid.")
    sys.exit(1)
    
def compareDates(x, y, prefer="high"):

    """
        Return higher or lower of two dates
    """

    x_epoch = time.mktime(time.strptime(x, "%m/%d/%Y"))
    y_epoch = time.mktime(time.strptime(y, "%m/%d/%Y"))

    if prefer == "high":
        if x_epoch <= y_epoch:
            myPick = y
        else:
            myPick = x
    else:
        if x_epoch >= y_epoch:
            myPick = y
        else:
            myPick = x

    return myPick


tracker = {}

for line in input:

    ################
    #strip newline
    ################
    line = line.rstrip("\n")

    ################
    #get fields
    ################
    star_num, location, date = line.split(",")

    ################
    #dictionary holding current min and max values
    ################
    minMax = tracker.get(star_num, {})

    old_min = minMax.get("min", "")
    old_max = minMax.get("max", "")

    if old_min == "":
        minMax["min"] = date
    else:
        minMax["min"] = compareDates(date, minMax["min"], "low")

    if old_max == "":
        minMax["max"] = date
    else:
        minMax["max"] = compareDates(date, minMax["max"], "high")

    tracker[star_num] = minMax

input.close()


################
#sort star_nums
################
keys = tracker.keys()
keys.sort()

for key in keys:

    print key, tracker[key]["min"]
    print key, tracker[key]["max"]

awk 'BEGIN{FS=","}
{
 n=split($3,dm,"/")
 y=dm[3];m=dm[2];d=dm[1]
 da=mktime(y" "m" "d" 0 0 0") 
 if ( da > date[$1]  ){
   date[$1]=da" "date[$1]
 }else {
   date[$1]=date[$1]" "da
 } 
 odate[da]=$3
}
END{
 for ( i in date) {
  m=split(date[i],temp," ")
  printf "max: %s, min: %s->%s\n" , odate[temp[1]],odate[temp[3]],i  
 }
}' file

# ./test.sh
max: 11/02/2008, min: 09/02/2008->Star1
max: 10/05/2008, min: 08/03/2008->Star2

#!/usr/bin/env python
import time
d={};e={}
for n,l in enumerate(open("file")):    
    line=[i.strip() for i in l.strip().split(",")]
    thedate = [ int(k) for k in  line[2].split("/") ]
    t = time.mktime( (thedate[2],thedate[1],thedate[0],0,0,0,0,0,0) ) 
    d.setdefault(line[0],[])
    d[line[0]].append(t)
    e[t]=line[2]
for m,n in d.iteritems():
    s=sorted(n)
    print "Max: %s ,  Min: %s -> %s" %( e[s[2]],e[s[0]],m )

 # ./test.py
Max: 11/02/2008 ,  Min: 09/02/2008 -> Star1
Max: 10/05/2008 ,  Min: 08/03/2008 -> Star2

awk 'END { for (k in min) 
printf "%s %s\n%s %s\n\n", 
k, min[k], k, max[k] }
!x[$1]++ { min[$1] = $3 }
fmtd($3) > fmtd(max[$1]) { max[$1] = $3 }
fmtd($3) < fmtd(min[$1]) { min[$1] = $3 } 
func fmtd(dt) { split(dt, t, "/") 
  return t[3]t[1]t[2] 
}' FS="," file