简单数据统计和处理（Data processing and statistic）

kim_wei

  实际工作中经常需要对一些数据列表进行相应的处理，比如，求和，最大／最小值，平均值等。 bash脚本中的expr只能处理简单整数的四则运算，不能处理浮点数据。在这种情况下，可以考虑bash脚本中嵌入另一种计算语言bc来增强数据统计功能。
  下例中，假设，数据源的格式为简单的，
....
234.02
3345.05
12.4039
....
#!/bin/bash
##################################
##
##Desc: noramlly, in bash script by using
##arithmetical command "expr", the integer  
##values can be operated. However, if other
##data types(float, double) are in data source, there would
##be problem by calling "expr", therefore, another  
##script language bc has to be used to perform
##such operation.
##
##Argument: 1.)-sum(summary)/-min(minimum)/-max(maximum)/-ave(average)
##        2.)file name (data source)
##
##
###################
##Sum()
##Calculate sum
##of all data entries in
##data source
##
Sum()
{
    echo "sum entries..."
    tmp=0            
    for i in `cat $DATA`; do
        tmp=`echo $tmp '+' $i | bc -l `   
        LINECNT=`expr $LINECNT + 1`
    done
   
    echo "----------------------"
    echo "Sum: " $tmp "Nr of Entry:" $LINECNT
        
    SUMVAL=$tmp
}
####################
##Max()
##Evaluate all data entry
##and find maximum one
##
Max()
{
    echo "maximum..."
    tmp=0
    cmp=0   
    for i in `cat $DATA`; do
        #compare,result saved in cmp
        cmp=`echo $tmp '>' $i | bc -l`
        
        #if necessary, exchange the maximum value        
        if [ $cmp -ne 1 ];then
            tmp=$i
        fi
    done   
   
    echo "-----------------------"
    echo "Max: " $tmp
}
#####################
##Min()
##Evaluate all data and
##find the minimum value
##
Min()
{
   
    echo "minimum..."
    tmp=0
    cmp=0   
    cnt=0   
    for i in `cat $DATA`; do
   
        #keep first data entry   
        cnt=`expr $cnt + 1`
        if [ $cnt -eq 1 ];then
            tmp=$i
            continue
        fi
   
        #compare,result saved in cmp
        cmp=`echo $tmp '>' $i | bc -l`
        
        #if necessary, exchange the maximum value        
        if [ $cmp -ne 0 ];then
            tmp=$i
        fi
    done   
    echo "-----------------------"
    echo "Min: " $tmp
}
#####################
##Avg()
##calculate the average
##value of all data entries
##
Avg()
{
    LINECNT=0
    Sum
    echo "Total Entries: "$LINECNT
   
    tmp=0
    tmp=`echo $SUMVAL '/' $LINECNT | bc -l`
   
    echo "------------------------"
    echo "Average: " $tmp
}
#####################
##Help()
##dispaly help information
##
##
Help()
{
    echo "--------------------"
    echo "Options: "
    echo "1.) -sum/-max/-min/ave: to calculate summary/maximum/minimun/average values"
    echo "2.)file name if necessary with path. "
    echo "--------------------"
}
#####################
## Programm Entry
## from here
##
echo "Staring evaluating..."
##check argument
if [ $# -ne 2 ]; then
   
    echo ""
    echo "Missing argument ..."
    echo ""
    Help
    exit
fi
##check file
if ! test -f $2; then
   
    #echo "Given file is existing !"
    #exit
#else
    echo "Given file is not existing !"
    exit
fi
DATA=$2
LINECNT=0
LINECNT=0
##do calculation
case "$1" in
   
    "-sum")
        Sum
    ;;
    "-min")
        Min
    ;;
    "-max")
        Max
    ;;
   
    "-ave")
        Avg
    ;;
    *)   
        echo "Unkonw argument"
        Help
    ;;
esac
  
               
               
               

本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u1/45602/showart_409583.html