I'm trying to create the program to read each lines and calculate each line's average and store into array...For example, program will read first line, add all the numbers and divide by 24 to calculate the average which will be stored into Avg_list[1].
When I try to run the program, I encounter following error, I have no idea why it doesn't work... Can someone identify the problem?
Code: in Ksh
c=0
while read -r line ; do
v=$line
set -- $v ((g=($2+$3+$4+$5+$6+$7+$8+$9+$10+$11+$12+$13+$14+$15+$16+$17+$18+$19+$20+$21+$22+$23+$24+$25+$26)/24))
echo $g
Avg_list[${c}]=$g
((c=c+1))
done < daily.txt
daily.txt
CPU 55 54 54 54 54 54 54 54 54 54 54 54 54 54 54 55 54 54 55 56 57 54 57 54
CPEAK 56 56 57 55 58 56 56 56 57 55 60 56 55 56 55 56 58 55 57 56 63 56 72 57
RAM 97 97 97 97 97 96 96 96 96 96 96 93 91 89 86 84 90 90 95 97 97 97 97 97
RPEAK 97 97 97 97 97 97 96 96 96 96 96 96 92 90 91 81 94 89 97 97 97 97 97 97
Error note:
while read -r line ; do
v=$line
set -- $v
((g=($2+$3+$4+$5+$6+$7+$8+$9+$10+$11+$12+$13+$14+$15+$16+$17+$18+$19+$20+$21+$22+$23+$24+$25+$26)/24))
echo $g
Avg_list[${c}]=$g
((c=c+1))
done < daily.txt
+ 0< daily.txt
+ read -r line
+ v=CPU 54 54 54 54 54 54 54 54 54 54 54 54 54 54 55 54 54 55 56 57 54 57 54 54
+ set -- CPU 54 54 54 54 54 54 54 54 54 54 54 54 54 54 55 54 54 55 56 57 54 57 54 54
+ (( g=(54+54+54+54+54+54+54+54+CPU0+CPU1+CPU2+CPU3+CPU4+CPU5+CPU6+CPU7+CPU8+CPU9+540+541+542+543+544+545+546)/24 ))
PerformanceAM.sh[21]: g=(54+54+54+54+54+54+54+54+CPU0+CPU1+CPU2+CPU3+CPU4+CPU5+CPU6+CPU7+CPU8+CPU9+540+541+542+543+544+545+546)/24: 0403-009 The specified number is not valid for this command.
EDIT
while read -r line ; do
v=$line
set -- $v
((g=${2}+${3}+${4}+${5}+${6}+${7}+${8}+${9}+${10}+${11}+${12}+${13}+${14}+${15}+${16}+${17}+${18}+${19}+${20}+${21}+${22}+${23}+${24}+${25}+${26})/24))
echo $g
Avg_list[${c}]=$g
((c=c+1))
done < daily.txt
New error:
while read -r line ; do
v=$line
set -- $v
((g=${2}+${3}+${4}+${5}+${6}+${7}+${8}+${9}+${10}+${11}+${12}+${13}+${14}+${15}+${16}+${17}+${18}+${19}+${20}+${21}+${22}+${23}+${24}+${25}+${26})/24)PerformanceAM.sh[18]: 0403-057 Syntax error at line 21 : `/24' is not expected.
Thanks for your suggestions! when I tried used bracket i get this error...I'm now even more confused....it seems like it's not collecting numbers at all...
#!/bin/ksh
while read -r line ; do
v=$line
set -- $v
((g=(${2}+${3}+${4}+${5}+${6}+${7}+${8}+${9}+${10}+${11}+${12}+${13}+${14}+${15}+${16}+${17}+${18}+${19}+${20}+${21}+${22}+${23}+${24}+${25})/24))
echo $g
Avg_list[${c}]=$g
((c=c+1))
done < daily.txt
You were missing one ( at g=${2} and the arguments only go until ${25} not ${26}.
This should do it:
while read -r line; do
sum=0
total=0
for x in $line; do
# if x is numeric
if echo "$x" | grep -E '^[0-9]*$' > /dev/null ; then
((sum=sum+x))
((total=total+1))
else
echo -n "$x "
fi
done
if [ $total = 0 ]; then
echo
else
echo $((sum/total))
fi
done < daily.txt
This follows my general principle of never making long lists of $1 $2... This solution works for any number of integers per line, and it also prints out the line label (a feature easy to remove if you want).
For reference purposes, here's how awk can be used to solve this
array=( $(awk '{sum=0; for (i=2;i<=25; i++) sum=sum+$i; printf "%.0f ",sum/24 }' daily.txt ) )
Given a list of numbers, it's a little tidier to use reverse-polish notation for the calculations:
c=0
while read line; do
set -- $line
shift
script="3 k $* + + + + + + + + + + + + + + + + + + + + + + + 24 / p"
Avg_list[c++]=$( dc -e "$script" )
done < daily.txt
Then
printf "%s\n" "${Avg_list[#]}"
produces
54.458
57.250
94.333
94.875
Related
I encounter a strange behaviour with bash string substitution.
I expected the same substitution on $r1 and $var to yield the exact same results.
both strings seem to have the same value.
But It is not the case and I can't understand what I am missing....
maybe is because of the glob? I just don't know... I am not pure IT guys and maybe it's something that will be evident for you.
(bottom a Repl.it link)
mkdir -p T21805
touch T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
r1=T21805/*R1*
echo $r1;
echo ${r1%%_S1*z}
var=T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
echo ${var%%_S1*z}
echo $r1| hexdump -C
echo $var | hexdump -C
output :
echo $r1
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
echo ${r1%%_S1*z}
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
echo ${var%%_S1*z}
T21805/T21805_SI-GA-D8-BH25N7DSXY
echo $r1| hexdump -C
00000000 54 32 31 38 30 35 2f 54 32 31 38 30 35 5f 53 49
|T21805/T21805_SI|
00000010 2d 47 41 2d 44 38 2d 42 48 32 35 4e 37 44 53 58
|-GA-D8-BH25N7DSX|
00000020 59 5f 53 31 5f 4c 30 30 31 5f 52 31 5f 30 30 31
|Y_S1_L001_R1_001|
00000030 2e 66 61 73 74 71 2e 67 7a 0a
|.fastq.gz.| 0000003a
echo $var | hexdump -C
00000000 54 32 31 38 30 35 2f 54 32 31 38 30 35 5f 53 49
|T21805/T21805_SI|
00000010 2d 47 41 2d 44 38 2d 42 48 32 35 4e 37 44 53 58
|-GA-D8-BH25N7DSX|
00000020 59 5f 53 31 5f 4c 30 30 31 5f 52 31 5f 30 30 31
|Y_S1_L001_R1_001|
00000030 2e 66 61 73 74 71 2e 67 7a 0a
|.fastq.gz.| 0000003a
Repl.it
I am interested on understanding why this is not working, I can achieve my desire output using sed for example.
Glob expansion doesn't happen at assignment time.
$ mkdir -p T21805
$ touch T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
$ touch T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_002.fastq.gz
$ r1=T21805/*R1*
$ printf '%s\n' "$r1"
T21805/*R1*
$ printf '%s\n' $r1
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_002.fastq.gz
It happens after the unquoted r1 has been expanded. When you write ${r1%%_S1*z}, the value of r1 doesn't contain the string S1; only after ${r1} expands is there an S1 you could match against.
If you set an array, the assignment rules are different. The glob expands before the assignment, and so you can do your filtering on each element of the array.
$ r1=( T21805/*R1* )
$ printf '%2\n' "${r1[#]}"
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_002.fastq.gz
$ printf '%s\n' "${r1[#]%%_S1*z}"
T21805/T21805_SI-GA-D8-BH25N7DSXY
T21805/T21805_SI-GA-D8-BH25N7DSXY
I ran it after set -xv to see the contents of r1.
$ r1=T21805/*R1*
+ r1='T21805/*R1*'
$ var=T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
+ var=T21805/T21805_SI-GA-D8-BH25N7DSXY_S1_L001_R1_001.fastq.gz
The r1 of$ {r1 %% _ S1 * z}isT21805 / * R1 *.
r1 does not include_S1 * z.
i am trying to transpose a table (10k rows X 10K cols) using the following script.
A simple data example
$ cat rm1
t1 t2 t3
n1 1 2 3
n2 2 3 44
n3 1 1 1
$ sh transpose.sh rm1
n1 n2 n3
t1 1 2 1
t2 2 3 1
t3 3 44 1
However, I am getting memory error. Any help would be appreciated.
awk -F "\t" '{
for (f = 1; f <= NF; f++)
a[NR, f] = $f
}
NF > nf { nf = NF }
END {
for (f = 1; f <= nf; f++)
for (r = 1; r <= NR; r++)
printf a[r, f] (r==NR ? RS : FS)
}'
Error
awk: cmd. line:2: (FILENAME=input FNR=12658) fatal: dupnode: r->stptr: can't allocate 10 bytes of memory (Cannot allocate memory)
Here's one way to do it, as I mentioned in my comments, in chunks. Here I show the mechanics on a tiny 12r x 10c file, but I also ran a chunk of 1000 rows on a 10K x 10K file in not much more than a minute (Mac Powerbook).6
EDIT The following was updated to consider an M x N matrix with unequal number of rows and columns. The previous version only worked for an 'N x N' matrix.
$ cat et.awk
BEGIN {
start = chunk_start
limit = chunk_start + chunk_size - 1
}
{
n = (limit > NF) ? NF : limit
for (f = start; f <= n; f++) {
a[NR, f] = $f
}
}
END {
n = (limit > NF) ? NF : limit
for (f = start; f <= n; f++)
for (r = 1; r <= NR; r++)
printf a[r, f] (r==NR ? RS : FS)
}
$ cat t.txt
10 11 12 13 14 15 16 17 18 19
20 21 22 23 24 25 26 27 28 29
30 31 32 33 34 35 36 37 38 39
40 41 42 43 44 45 46 47 48 49
50 51 52 53 54 55 56 57 58 59
60 61 62 63 64 65 66 67 68 69
70 71 72 73 74 75 76 77 78 79
80 81 82 83 84 85 86 87 88 89
90 91 92 93 94 95 96 97 98 99
A0 A1 A2 A3 A4 A5 A6 A7 A8 A9
B0 B1 B2 B3 B4 B5 B6 B7 B8 B9
C0 C1 C2 C3 C4 C5 C6 C7 C8 C9
$ cat et.sh
inf=$1
outf=$2
rm -f $outf
for i in $(seq 1 2 12); do
echo chunk for rows $i $(expr $i + 1)
awk -v chunk_start=$i -v chunk_size=2 -f et.awk $inf >> $outf
done
$ sh et.sh t.txt t-transpose.txt
chunk for rows 1 2
chunk for rows 3 4
chunk for rows 5 6
chunk for rows 7 8
chunk for rows 9 10
chunk for rows 11 12
$ cat t-transpose.txt
10 20 30 40 50 60 70 80 90 A0 B0 C0
11 21 31 41 51 61 71 81 91 A1 B1 C1
12 22 32 42 52 62 72 82 92 A2 B2 C2
13 23 33 43 53 63 73 83 93 A3 B3 C3
14 24 34 44 54 64 74 84 94 A4 B4 C4
15 25 35 45 55 65 75 85 95 A5 B5 C5
16 26 36 46 56 66 76 86 96 A6 B6 C6
17 27 37 47 57 67 77 87 97 A7 B7 C7
18 28 38 48 58 68 78 88 98 A8 B8 C8
19 29 39 49 59 69 79 89 99 A9 B9 C9
And then running the first chunk on the huge file looks like:
$ time awk -v chunk_start=1 -v chunk_size=1000 -f et.awk tenk.txt > tenk-transpose.txt
real 1m7.899s
user 1m5.173s
sys 0m2.552s
Doing that ten times with the next chunk_start set to 1001, etc. (and appending with >> to the output, of course) should finally give you the full transposed result.
There is a simple and quick algorithm based on sorting:
1) Make a pass through the input, prepending the row number and column number to each field. Output is a three-tuple of row, column, value for each cell in the matrix. Write the output to a temporary file.
2) Sort the temporary file by column, then row.
3) Make a pass through the sorted temporary file, reconstructing the transposed matrix.
The two outer passes are done by awk. The sort is done by the system sort. Here's the code:
$ echo '1 2 3
2 3 44
1 1 1' |
awk '{ for (i=1; i<=NF; i++) print i, NR, $i}' |
sort -n |
awk ' NR>1 && $2==1 { print "" }; { printf "%s ", $3 }; END { print "" }'
1 2 1
2 3 1
3 44 1
So i'm issuing a query to mysql and it's returning say 1,000 rows,but each iteration of the program could return a different number of rows. I need to break up (without using a mysql limit) this result set into chunks of 100 rows that i can then programatically iterate through in these 100 row chunks.
So
MySQLOutPut='1 2 3 4 ... 10,000"
I need to turn that into an array that looks like
array[1]="1 2 3 ... 100"
array[2]="101 102 103 ... 200"
etc.
I have no clue how to accomplish this elegantly
Using Charles' data generation:
MySQLOutput=$(seq 1 10000 | tr '\n' ' ')
# the sed command will add a newline after every 100 words
# and the mapfile command will read the lines into an array
mapfile -t MySQLOutSplit < <(
sed -r 's/([^[:blank:]]+ ){100}/&\n/g; $s/\n$//' <<< "$MySQLOutput"
)
echo "${#MySQLOutSplit[#]}"
# 100
echo "${MySQLOutSplit[0]}"
# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
echo "${MySQLOutSplit[99]}"
# 9901 9902 9903 9904 9905 9906 9907 9908 9909 9910 9911 9912 9913 9914 9915 9916 9917 9918 9919 9920 9921 9922 9923 9924 9925 9926 9927 9928 9929 9930 9931 9932 9933 9934 9935 9936 9937 9938 9939 9940 9941 9942 9943 9944 9945 9946 9947 9948 9949 9950 9951 9952 9953 9954 9955 9956 9957 9958 9959 9960 9961 9962 9963 9964 9965 9966 9967 9968 9969 9970 9971 9972 9973 9974 9975 9976 9977 9978 9979 9980 9981 9982 9983 9984 9985 9986 9987 9988 9989 9990 9991 9992 9993 9994 9995 9996 9997 9998 9999 10000
Something like this:
# generate content
MySQLOutput=$(seq 1 10000 | tr '\n' ' ') # seq is awful, don't use in real life
# split into a large array, each item stored individually
read -r -a MySQLoutArr <<<"$MySQLOutput"
# add each batch of 100 items into a new array entry
batchSize=100
MySQLoutSplit=( )
for ((i=0; i<${#MySQLoutArr[#]}; i+=batchSize)); do
MySQLoutSplit+=( "${MySQLoutArr[*]:i:batchSize}" )
done
To explain some of the finer points:
read -r -a foo reads contents into an array named foo, split on IFS, up to the next character specified by read -d (none given here, thus reading only a single line). If you wanted each line to be a new array entry, consider IFS=$'\n' read -r -d '' -a foo, which will read each line into an array, terminated at the first NUL in the input stream.
"${foo[*]:i:batchSize}" expands to a list of items in array foo, starting at index i, and taking the next batchSize items, concatenated into a single string with the first character in $IFS used as a separator.
I want to print the numbers 1 to 100 numbers, but in columns: first 1 to 10 in first column, 11 to 20 in 2nd column, then 21 to 30 in 3rd column, ..., 91 to 100 in 10th column.
How can I achieve this in Bash? I have tried:
#!/bin/bash
for ((i=1; i <= 100 ; i++)) do
echo " $i"
done
Its a bit heavy since it spawns many subprocesses .. but posting it as a 1-liner
paste <(seq 1 10) <(seq 11 20) <(seq 21 30) <(seq 31 40) <(seq 41 50) <(seq 51 60) <(seq 61 70) <(seq 71 80) <(seq 81 90) <(seq 91 100)
for 1-10 in 1st column, 11-20 in 2nd column and so on..
and
seq 1 100 | paste - - - - - - - - - -
for 1-10 in 1st row, 11-20 in 2nd row and so on..
Note: There are 10 hypens in 2nd command and in the 1st one, <(command) means process substitution i.e substitutes the the output of the process
Edit: Approach purely using for loop
for ((i=1;i<=10;i++)); do
for ((j=i;j<=(i+90);j+=10)); do
printf "%2d " $j
done
echo
done
for 1-10 in 1st column, 11-20 in 2nd column and so on..
and
for ((i=0;i<10;i++)); do
for ((j=1;j<=10;j++)); do
printf "%2d " $[$i*10+$j]
done
echo
done
for 1-10 in 1st row, 11-20 in 2nd row and so on..
There is an easier way!
$ seq 100 | pr -10t
1 11 21 31 41 51 61 71 81 91
2 12 22 32 42 52 62 72 82 92
3 13 23 33 43 53 63 73 83 93
4 14 24 34 44 54 64 74 84 94
5 15 25 35 45 55 65 75 85 95
6 16 26 36 46 56 66 76 86 96
7 17 27 37 47 57 67 77 87 97
8 18 28 38 48 58 68 78 88 98
9 19 29 39 49 59 69 79 89 99
10 20 30 40 50 60 70 80 90 100
I have a some file with the following content
$ cat somefile
28 46 5d a2 26 7a 192 168 2 2
0 15 e c8 a8 a3 192 168 100 3
54 4 2b 8 c 26 192 168 20 3
As you can see the values in first six columns are represented in hex, the values in last four columns in decimal formats. I just want to add 0 to every single symbol hexidecimal value.
Thanks beforehand.
This one should work out for you:
while read -a line
do
hex=(${line[#]:0:6})
printf "%02x " ${hex[#]/#/0x}
echo ${line[#]:6:4}
done < somefile
Example:
$ cat somefile
28 46 5d a2 26 7a 192 168 2 2
0 15 e c8 a8 a3 192 168 100 3
54 4 2b 8 c 26 192 168 20 3
$ while read -a line
> do
> hex=(${line[#]:0:6})
> printf "%02x " ${hex[#]/#/0x}
> echo ${line[#]:6:4}
> done < somefile
28 46 5d a2 26 7a 192 168 2 2
00 15 0e c8 a8 a3 192 168 100 3
54 04 2b 08 0c 26 192 168 20 3
Here is a way with awk if that is an option:
awk '{for(i=1;i<=6;i++) if(length($i)<2) $i=0$i}1' file
Test:
$ cat file
28 46 5d a2 26 7a 192 168 2 2
0 15 e c8 a8 a3 192 168 100 3
54 4 2b 8 c 26 192 168 20 3
$ awk '{for(i=1;i<=6;i++) if(length($i)<2) $i=0$i}1' file
28 46 5d a2 26 7a 192 168 2 2
00 15 0e c8 a8 a3 192 168 100 3
54 04 2b 08 0c 26 192 168 20 3
Please try this too, if it helps (bash version 4.1.7(1)-release)
#!/bin/bash
while read line;do
arr=($line)
i=0
for num in "${arr[#]}";do
if [ $i -lt 6 ];then
if [ ${#num} -eq 1 ];then
arr[i]='0'${arr[i]};
fi
fi
i=$((i+1))
done
echo "${arr[*]}"
done<your_file
This might work for you (GNU sed):
sed 's/\b\S\s/0&/g' file
Finds a single non-space character and prepends a 0.