Reputation: 19
How do i split the input into chunks of six entries each using bash?
This is the script which i run to output the raw data of data_tripwire.sh
#!/bin/sh
LOG=/var/log/syslog-ng/svrs/sec2tes1
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
CBS=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.22.41 |sort|uniq | wc -l`
echo $CBS >> /home/secmgr/attmrms1/data_tripwire1.sh
done
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
GFS=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.22.31 |sort|uniq | wc -l`
echo $GFS >> /home/secmgr/attmrms1/data_tripwire1.sh
done
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
HR1=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.10.1 |sort|uniq | wc -l `
echo $HR1 >> /home/secmgr/attmrms1/data_tripwire1.sh
done
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
HR2=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.21.12 |sort|uniq | wc -l`
echo $HR2 >> /home/secmgr/attmrms1/data_tripwire1.sh
done
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
PAYROLL=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.21.18 |sort|uniq | wc -l`
echo $PAYROLL >> /home/secmgr/attmrms1/data_tripwire1.sh
done
for count in 6 5 4 3 2 1 0
do
MONTH=`date -d"$count month ago" +"%Y-%m"`
INCV=`bzcat $LOG/$MONTH*.log.bz2|grep 10.55.22.71 |sort|uniq | wc -l`
echo $INCV >> /home/secmgr/attmrms1/data_tripwire1.sh
done
data_tripwire.sh
91
58
54
108
52
18
8
81
103
110
129
137
84
15
14
18
11
17
12
6
1
28
6
14
8
8
0
0
28
24
25
23
21
13
9
4
18
17
18
30
13
3
I want to do the first 6 entries(91,58,54,108,52,18) from the output above. Then it will break out of the loop.After that it will continue for the next 6 entries.Then it will break out of the loop again....
The problem now is that it reads all the 42 numbers without breaking out of the loop.
This is the output of the table
Tripwire
Month CBS GFS HR HR Payroll INCV
cb2db1 gfs2db1 hr2web1 hrm2db1 hrm2db1a incv2svr1
2013-07 85 76 12 28 26 4
2013-08 58 103 18 6 24 18
2013-09 54 110 11 14 25 17
2013-10 108 129 17 8 23 18
2013-11 52 137 12 8 21 30
2013-12 18 84 6 0 13 13
2014-01 8 16 1 0 9 3
The problem now is that it read the total 42 numbers from 85...3 I want to make a loop which run from july till jan for one server.Then it will do the average mean and standard deviation calculation which is already done below. After that done, it will continue the next cycle of 6 numbers for the next server and it will do the same like initial cycle.Assistance is required for the for loops which has break and continue in it or any simpler.
This is my standard deviation calculation
count=0 # Number of data points; global.
SC=3 # Scale to be used by bc. three decimal places.
E_DATAFILE=90 # Data file error
## ----------------- Set data file ---------------------
if [ ! -z "$1" ] # Specify filename as cmd-line arg?
then
datafile="$1" # ASCII text file,
else #+ one (numerical) data point per line!
datafile=/home/secmgr/attmrms1/data_tripwire1.sh
fi # See example data file, below.
if [ ! -e "$datafile" ]
then
echo "\""$datafile"\" does not exist!"
exit $E_DATAFILE
fi
Calculate the mean
arith_mean ()
{
local rt=0 # Running total.
local am=0 # Arithmetic mean.
local ct=0 # Number of data points.
while read value # Read one data point at a time.
do
rt=$(echo "scale=$SC; $rt + $value" | bc)
(( ct++ ))
done
am=$(echo "scale=$SC; $rt / $ct" | bc)
echo $am; return $ct # This function "returns" TWO values!
# Caution: This little trick will not work if $ct > 255!
# To handle a larger number of data points,
#+ simply comment out the "return $ct" above.
} <"$datafile" # Feed in data file.
sd ()
{
mean1=$1 # Arithmetic mean (passed to function).
n=$2 # How many data points.
sum2=0 # Sum of squared differences ("variance").
avg2=0 # Average of $sum2.
sdev=0 # Standard Deviation.
while read value # Read one line at a time.
do
diff=$(echo "scale=$SC; $mean1 - $value" | bc)
# Difference between arith. mean and data point.
dif2=$(echo "scale=$SC; $diff * $diff" | bc) # Squared.
sum2=$(echo "scale=$SC; $sum2 + $dif2" | bc) # Sum of squares.
done
avg2=$(echo "scale=$SC; $sum2 / $n" | bc) # Avg. of sum of squares.
sdev=$(echo "scale=$SC; sqrt($avg2)" | bc) # Square root =
echo $sdev # Standard Deviation.
} <"$datafile" # Rewinds data file.
Showing the output
mean=$(arith_mean); count=$? # Two returns from function!
std_dev=$(sd $mean $count)
echo
echo "<tr><th>Servers</th><th>"Number of data points in \"$datafile"\"</th> <th>Arithmetic mean (average)</th><th>Standard Deviation</th></tr>" >> $HTML
echo "<tr><td>cb2db1<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo "<tr><td>gfs2db1<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo "<tr><td>hr2web1<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo "<tr><td>hrm2db1<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo "<tr><td>hrm2db1a<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo "<tr><td>incv21svr1<td>$count<td>$mean<td>$std_dev</tr>" >> $HTML
echo
I want to split the input into chunks of six entries each with the arithmetic mean and the sd of the entries 1..6, then of the entries 7..12, then of 13..18 etc.
This is the output of the table i want.
Tripwire
Month CBS GFS HR HR Payroll INCV
cb2db1 gfs2db1 hr2web1 hrm2db1 hrm2db1a incv2svr1
2013-07 85 76 12 28 26 4
2013-08 58 103 18 6 24 18
2013-09 54 110 11 14 25 17
2013-10 108 129 17 8 23 18
2013-11 52 137 12 8 21 30
2013-12 18 84 6 0 13 13
2014-01 8 16 1 0 9 3
*Standard
deviation
(7mths) 31.172 35.559 5.248 8.935 5.799 8.580
* Mean
(7mths) 54.428 94.285 11.142 9.142 20.285 14.714
Upvotes: 0
Views: 571
Answers (3)
Reputation: 6758
The functions will now be able to only read 6 items in datafile.
arith_mean ()
{
local rt=0 # Running total.
local am=0 # Arithmetic mean.
local ct=0 # Number of data points.
while read value # Read one data point at a time.
do
rt=$(echo "scale=$SC; $rt + $value" | bc)
(( ct++ ))
done
am=$(echo "scale=$SC; $rt / $ct" | bc)
echo $am; return $ct # This function "returns" TWO values!
# Caution: This little trick will not work if $ct > 255!
# To handle a larger number of data points,
#+ simply comment out the "return $ct" above.
} <(awk -v block=$i 'NR > (6* (block - 1)) && NR < (6 * block + 1) {print}' "$datafile") # Feed in data file.
sd ()
{
mean1=$1 # Arithmetic mean (passed to function).
n=$2 # How many data points.
sum2=0 # Sum of squared differences ("variance").
avg2=0 # Average of $sum2.
sdev=0 # Standard Deviation.
while read value # Read one line at a time.
do
diff=$(echo "scale=$SC; $mean1 - $value" | bc)
# Difference between arith. mean and data point.
dif2=$(echo "scale=$SC; $diff * $diff" | bc) # Squared.
sum2=$(echo "scale=$SC; $sum2 + $dif2" | bc) # Sum of squares.
done
avg2=$(echo "scale=$SC; $sum2 / $n" | bc) # Avg. of sum of squares.
sdev=$(echo "scale=$SC; sqrt($avg2)" | bc) # Square root =
echo $sdev # Standard Deviation.
} <(awk -v block=$i 'NR > (6 * (block - 1)) && NR < (6 * block + 1) {print}' "$datafile") # Rewinds data file.
From main you will need to set your blocks to read.
for((i=1; i <= $(( $(wc -l $datafile | sed 's/[A-Za-z \/]*//g') / 6 )); i++))
do
mean=$(arith_mean); count=$? # Two returns from function!
std_dev=$(sd $mean $count)
done
Of course it is better to move the wc -l outside of the loop for faster execution. But you get the idea.
The syntax error occured between < and ( due to space. There shouldn't be a space between them. Sorry for the typo.
cat <(awk -F: '{print $1}' /etc/passwd) works.
cat < (awk -F: '{print $1}' /etc/passwd) syntax error near unexpected token `('
Upvotes: 0
Reputation: 246799
paste - - - - - - < data_tripwire.sh | while read -a values; do
# values is an array with 6 values
# ${values[0]} .. ${values[5]}
arith_mean "${values[@]}"
done
This means you have to rewrite your function so they don't use read: change
while read value
to
for value in "$@"
@Matt, yes change both functions to iterate over arguments instead of reading from stdin. Then, you will pass the data file (now called "data_tripwire1.sh" (terrible file extension for data, use .txt or .dat)) into paste to reformat the data so that the first 6 values now form the first row. Read the line into the array values (using read -a values) and invoke the functions :
arith_mean () {
local sum=$(IFS=+; echo "$*")
echo "scale=$SC; ($sum)/$#" | bc
}
sd () {
local mean=$1
shift
local sum2=0
for i in "$@"; do
sum2=$(echo "scale=$SC; $sum2 + ($mean-$i)^2" | bc)
done
echo "scale=$SC; sqrt($sum2/$#)"|bc
}
paste - - - - - - < data_tripwire1.sh | while read -a values; do
mean=$(arith_mean "${values[@]}")
sd=$(sd $mean "${values[@]}")
echo "${values[@]} $mean $sd"
done | column -t
91 58 54 108 52 18 63.500 29.038
8 81 103 110 129 137 94.666 42.765
84 15 14 18 11 17 26.500 25.811
12 6 1 28 6 14 11.166 8.648
8 8 0 0 28 24 11.333 10.934
25 23 21 13 9 4 15.833 7.711
18 17 18 30 13 3 16.500 7.973
Note you don't need to return a fancy value from the functions: you know how many points you pass in.
Upvotes: 2
Reputation: 59426
Based on Glenn's answer I propose this which needs very little changes to the original:
paste - - - - - - < data_tripwire.sh | while read -a values
do
for value in "${values[@]}"
do
echo "$value"
done | arith_mean
for value in "${values[@]}"
do
echo "$value"
done | sd
done
You can type (or copy & paste) this code directly in an interactive shell. It should work out of the box. Of course, this is not feasible if you intend to use this often, so you can put that code into a text file, make that executable and call that text file as a shell script. In this case you should add #!/bin/bash as first line in that file.
Credit to Glenn Jackman for the use of paste - - - - - - which is the real solution I'd say.
Upvotes: 0
Related Questions
- Is it possible to split a file content after every 6. row with the `split` function?
- Run a bash command and split the output by line not by space
- How to split a long string of text with bash
- How to split a string in Bash?
- How to split a string into columns in bash
- Split text into equal parts
- How can I split input using bash?
- Split file into unequal chunks in Linux
- piping split input into command
- Split input into multiple outputs based on content?