I've been struggling to write a code for extracting every N columns from an input file and write them into output files according to their extracting order.
(My real world case is to extract every 800 columns from a total 24005 columns file starting at column 6, so I need a loop)
In a simpler case below, extracting every 3 columns(fields) from an input file with a start point of the 2nd column.
for example, if the input file looks like:
aa 1 2 3 4 5 6 7 8 9
bb 1 2 3 4 5 6 7 8 9
cc 1 2 3 4 5 6 7 8 9
dd 1 2 3 4 5 6 7 8 9
and I want the output to look like this:
output_file_1:
1 2 3
1 2 3
1 2 3
1 2 3
output_file_2:
4 5 6
4 5 6
4 5 6
4 5 6
output_file_3:
7 8 9
7 8 9
7 8 9
7 8 9
I tried this, but it doesn't work:
awk 'for(i=2;i<=10;i+a) {{printf "%s ",$i};a=3}' <inputfile>
It gave me syntax error and the more I fix the more problems coming out.
I also tried the linux command cut but while I was dealing with large files this seems effortless. And I wonder if cut would do a loop cut of every 3 fields just like the awk.
Can someone please help me with this and give a quick explanation? Thanks in advance.
Actions to be performed by awk on the input data must be included in curled braces, so the reason the awk one-liner you tried results in a syntax error is that the for cycle does not respect this rule. A syntactically correct version will be:
awk '{for(i=2;i<=10;i+a) {printf "%s ",$i};a=3}' <inputfile>
This is syntactically correct (almost, see end of this post.), but does not do what you think.
To separate the output by columns on different files, the best thing is to use awk redirection operator >. This will give you the desired output, given that your input files always has 10 columns:
awk '{ print $2,$3,$4 > "file_1"; print $5,$6,$7 > "file_2"; print $8,$9,$10 > "file_3"}' <inputfile>
mind the " " to specify the filenames.
EDITED: REAL WORLD CASE
If you have to loop along the columns because you have too many of them, you can still use awk (gawk), with two loops: one on the output files and one on the columns per file. This is a possible way:
#!/usr/bin/gawk -f
BEGIN{
CTOT = 24005 # total number of columns, you can use NF as well
DELTA = 800 # columns per file
START = 6 # first useful column
d = CTOT/DELTA # number of output files.
}
{
for ( i = 0 ; i < d ; i++)
{
for ( j = 0 ; j < DELTA ; j++)
{
printf("%f\t",$(START+j+i*DELTA)) > "file_out_"i
}
printf("\n") > "file_out_"i
}
}
I have tried this on the simple input files in your example. It works if CTOT can be divided by DELTA. I assumed you had floats (%f) just change that with what you need.
Let me know.
P.s. going back to your original one-liner, note that the loop is an infinite one, as i is not incremented: i+a must be substituted by i+=a, and a=3 must be inside the inner braces:
awk '{for(i=2;i<=10;i+=a) {printf "%s ",$i;a=3}}' <inputfile>
this evaluates a=3 at every cycle, which is a bit pointless. A better version would thus be:
awk '{for(i=2;i<=10;i+=3) {printf "%s ",$i}}' <inputfile>
Still, this will just print the 2nd, 5th and 8th column of your file, which is not what you wanted.
awk '{ print $2, $3, $4 >"output_file_1";
print $5, $6, $7 >"output_file_2";
print $8, $9, $10 >"output_file_3";
}' input_file
This makes one pass through the input file, which is preferable to multiple passes. Clearly, the code shown only deals with the fixed number of columns (and therefore a fixed number of output files). It can be modified, if necessary, to deal with variable numbers of columns and generating variable file names, etc.
(My real world case is to extract every 800 columns from a total 24005 columns file starting at column 6, so I need a loop)
In that case, you're correct; you need a loop. In fact, you need two loops:
awk 'BEGIN { gap = 800; start = 6; filebase = "output_file_"; }
{
for (i = start; i < start + gap; i++)
{
file = sprintf("%s%d", filebase, i);
for (j = i; j <= NF; j += gap)
printf("%s ", $j) > file;
printf "\n" > file;
}
}' input_file
I demonstrated this to my satisfaction with an input file with 25 columns (numbers 1-25 in the corresponding columns) and gap set to 8 and start set to 2. The output below is the resulting 8 files pasted horizontally.
2 10 18 3 11 19 4 12 20 5 13 21 6 14 22 7 15 23 8 16 24 9 17 25
2 10 18 3 11 19 4 12 20 5 13 21 6 14 22 7 15 23 8 16 24 9 17 25
2 10 18 3 11 19 4 12 20 5 13 21 6 14 22 7 15 23 8 16 24 9 17 25
2 10 18 3 11 19 4 12 20 5 13 21 6 14 22 7 15 23 8 16 24 9 17 25
With GNU awk:
$ awk -v d=3 '{for(i=2;i<NF;i+=d) print gensub("(([^ ]+ +){" i-1 "})(([^ ]+( +|$)){" d "}).*","\\3",""); print "----"}' file
1 2 3
4 5 6
7 8 9
----
1 2 3
4 5 6
7 8 9
----
1 2 3
4 5 6
7 8 9
----
1 2 3
4 5 6
7 8 9
----
Just redirect the output to files if desired:
$ awk -v d=3 '{sfx=0; for(i=2;i<NF;i+=d) print gensub("(([^ ]+ +){" i-1 "})(([^ ]+( +|$)){" d "}).*","\\3","") > ("output_file_" ++sfx)}' file
The idea is just to tell gensub() to skip the first few (i-1) fields then print the number of fields you want (d = 3) and ignore the rest (.*). If you're not printing exact multiples of the number of fields you'll need to massage how many fields get printed on the last loop iteration. Do the math...
Here's a version that'd work in any awk. It requires 2 loops and modifies the spaces between fields but it's probably easier to understand:
$ awk -v d=3 '{sfx=0; for(i=2;i<=NF;i+=d) {str=fs=""; for(j=i;j<i+d;j++) {str = str fs $j; fs=" "}; print str > ("output_file_" ++sfx)} }' file
I was successful using the following command line. :) It uses a for loop and pipes the awk program into it's stdin using -f -. The awk program itself is created using bash variable math.
for i in 0 1 2; do
echo "{print \$$((i*3+2)) \" \" \$$((i*3+3)) \" \" \$$((i*3+4))}" \
| awk -f - t.file > "file$((i+1))"
done
Update: After the question has updated I tried to hack a script that creates the requested 800-cols-awk script dynamically ( a version according to Jonathan Lefflers answer) and pipe that to awk. Although the scripts looks good (for me ) it produces an awk syntax error. The question is, is this too much for awk or am I missing something? Would really appreciate feedback!
Update: Investigated this and found documentation that says awk has a lot af restrictions. They told to use gawk in this situations. (GNU's awk implementation). I've done that. But still I'll get an syntax error. Still feedback appreciated!
#!/bin/bash
# Note! Although the script's output looks ok (for me)
# it produces an awk syntax error. is this just too much for awk?
# open pipe to stdin of awk
exec 3> >(gawk -f - test.file)
# verify output using cat
#exec 3> >(cat)
echo '{' >&3
# write dynamic script to awk
for i in {0..24005..800} ; do
echo -n " print " >&3
for (( j=$i; j <= $((i+800)); j++ )) ; do
echo -n "\$$j " >&3
if [ $j = 24005 ] ; then
break
fi
done
echo "> \"file$((i/800+1))\";" >&3
done
echo "}"
Related
As the title suggests I'm trying to find all rows in an large tsv file, where at least 50% of the columns have a value bigger than a value x using awk.
E.g for x=5:
9 6 7 2 3
0 1 2 7 6
1 3 8 9 10
should return
9 6 7 2 3
1 3 8 9 10
awk to the rescue!
$ awk -v t=5 '{c=0; for(i=1;i<=NF;i++) c+=($i>t)} c/NF>0.5' file
9 6 7 2 3
1 3 8 9 10
Using Perl:
perl -ane '$x = 5; print if #F / 2 <= grep $_ > $x, #F' -- file.tsv
Using an input .tsv file which looks like this:
Num1 Num2 Num3 Num4 Num5
9 6 7 2 3
0 1 2 7 6
1 3 8 9 10
This code will do it in a awk script. I've left comments to see
the form of a script so you can adjust accordingly.
#!/usr/bin/awk -f
# reads from stdin.
# Usage: $ ./bigcols.awk < input1.tsv
# Run at start.
BEGIN {
# print "Start"
# print "TSV setting. Field seperator set to tab."
FS = "\t"
# He wants to find lines with avg greater than var x
x=5
}
# main. Run for each record. This code uses newlines to denote records.
{
# Find lines which are of this form: (skip header)
# #+,
# ie. start with one or more numbers in column 1.
if ($1 ~ /^[0-9]+/) {
the_avg = ($1 + $2 + $3 + $4 + $5)/5
if (the_avg > x) {
print $1, $2, $3, $4, $5
}
}
}
# run at end
#END { print "Stop" }
I have a .txt file with 25,000 lines. Each line there is a number from 1 to 20. I want to compute the total occurrence of each number in the file. I don't know should I use grep or awk and how to use it. And I'm worried about I got confused with 1 and 11, which both contain 1's. Thank you very much for helping!
I was trying but this would double count my numbers.
grep -o '1' degreeDistirbution.txt | wc -l
With grep you can match the beginning and end of a line with '^' and '$' respectively. For the whole thing I'll use an array, but to illustrate this point I'll just use one variable:
one="$(grep -c "^1$" ./$inputfile)"
then we put that together with the magic of bash loops and loop through all the numbers with a while like so:
i=1
while [[ $i -le 20 ]]
do
arr[i]="$(grep -c "^$i$" ./$inputfile)"
i=$[$i+1]
done
if you like you can of course use a for as well
An easier method is:
sort -n file | uniq -c
Which will count the occurrences of each number in the sorted file and display the results like:
$ sort -n dat/twenty.txt | uniq -c
3 1
3 2
3 3
4 4
4 5
4 6
4 7
4 8
4 9
4 10
4 11
3 12
2 13
2 14
4 15
4 16
4 17
2 18
2 19
2 20
Showing I have 3 ones, 3 twos, etc.. in the sample file.
Is it possible to format seq in a way that it will display the range desired but with N numbers per line?
Let say that I want seq 20 but with the following output:
1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
My next guess would be a nested loop but I'm not sure how...
Any help would be appreciated :)
Use can use awk to format it as per your needs.
$ seq 20 | awk '{ORS=NR%5?FS:RS}1'
1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
ORS is awk's built-in variable which stands for Output Record Separator and has a default value of \n. NR is awk's built-in variable which holds the line number. FS is built-in variable that stands for Field Separator and has the default value of space. RS is built-in variable that stands for Record Separator and has the default value of \n.
Our action which is a ternary operator, to check if NR%5 is true. When it NR%5 is not 0 (hence true) it uses FS as Output Record Separator. When it is false we use RS which is newline as Output Record Separator.
1 at the end triggers awk default action that is to print the line.
You can use xargs to limit the sequence displayed per line.
$ seq 20 | xargs -n 5
1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
The parameter -n 5 tells xargs to only display 5 sequence numbers.
If you have bash you can use the builtin sequence.
echo {1..20} | xargs -n 5
Using Bash:
while read num; do
((num % 5)) && printf "$line " || echo "$line"
done < <(seq 20)
Or:
for i in {1..20}; do
s+="$i "
if ! ((i % 5)); then
echo $s
s=""
fi
done
I have two text files A.txt and B.txt. Each line of A.txt
A.txt
100
222
398
B.txt
1 2 103 2
4 5 1026 74
7 8 209 55
10 11 122 78
What I am looking for is something like this:
for each line of A
search B;
if (the value of third column in a line of B - the value of the variable in A > 10)
print that line of B;
Any awk for doing that??
How about something like this,
I had some troubles understanding your question, but maybe this will give you some pointers,
#!/bin/bash
# Read intresting values from file2 into an array,
for line in $(cat 2.txt | awk '{print $3}')
do
arr+=($line)
done
# Linecounter,
linenr=0
# Loop through every line in file 1,
for val in $(cat 1.txt)
do
# Increment linecounter,
((linenr++))
# Loop through every element in the array (containing values from 3 colum from file2)
for el in "${!arr[#]}";
do
# If that value - the value from file 1 is bigger than 10, print values
if [[ $((${arr[$el]} - $val )) -gt 10 ]]
then
sed -n "$(($el+1))p" 2.txt
# echo "Value ${arr[$el]} (on line $(($el+1)) from 2.txt) - $val (on line $linenr from 1.txt) equals $((${arr[$el]} - $val )) and is hence bigger than 10"
fi
done
done
Note,
This is a quick and dirty thing, there is room for improvements. But I think it'll do the job.
Use awk like this:
cat f1
1
4
9
16
cat f2
2 4 10 8
3 9 20 8
5 1 15 8
7 0 30 8
awk 'FNR==NR{a[NR]=$1;next} $3-a[FNR] < 10' f1 f2
2 4 10 8
5 1 15 8
UPDATE: Based on OP's edited question:
awk 'FNR==NR{a[NR]=$1;next} {for (i in a) if ($3-a[i] > 10) print}'
and see how simple awk based solution is as compared to nested for loops.
I want to extract the 5th column from multiple files, named in a numerical order, and paste those columns in sequence, side by side, into one output file.
The file names look like:
sample_problem1_part1.txt
sample_problem1_part2.txt
sample_problem2_part1.txt
sample_problem2_part2.txt
sample_problem3_part1.txt
sample_problem3_part2.txt
......
Each problem file (1,2,3...) has two parts (part1, part2). Each file has the same number of lines.
The content looks like:
sample_problem1_part1.txt
1 1 20 20 1
1 7 21 21 2
3 1 22 22 3
1 5 23 23 4
6 1 24 24 5
2 9 25 25 6
1 0 26 26 7
sample_problem1_part2.txt
1 1 88 88 8
1 1 89 89 9
2 1 90 90 10
1 3 91 91 11
1 1 92 92 12
7 1 93 93 13
1 5 94 94 14
sample_problem2_part1.txt
1 4 330 30 a
3 4 331 31 b
1 4 332 32 c
2 4 333 33 d
1 4 334 34 e
1 4 335 35 f
9 4 336 36 g
The output should look like: (in a sequence of problem1_part1, problem1_part2, problem2_part1, problem2_part2, problem3_part1, problem3_part2,etc.,)
1 8 a ...
2 9 b ...
3 10 c ...
4 11 d ...
5 12 e ...
6 13 f ...
7 14 g ...
I was using:
paste sample_problem1_part1.txt sample_problem1_part2.txt > \
sample_problem1_partall.txt
paste sample_problem2_part1.txt sample_problem2_part2.txt > \
sample_problem2_partall.txt
paste sample_problem3_part1.txt sample_problem3_part2.txt > \
sample_problem3_partall.txt
And then:
for i in `find . -name "sample_problem*_partall.txt"`
do
l=`echo $i | sed 's/sample/extracted_col_/'`
`awk '{print $5, $10}' $i > $l`
done
And:
paste extracted_col_problem1_partall.txt \
extracted_col_problem2_partall.txt \
extracted_col_problem3_partall.txt > \
extracted_col_problemall_partall.txt
It works fine with a few files, but it's a crazy method when the number of files is large (over 4000).
Could anyone help me with simpler solutions that are capable of dealing with multiple files, please?
Thanks!
Here's one way using awk and a sorted glob of files:
awk '{ a[FNR] = (a[FNR] ? a[FNR] FS : "") $5 } END { for(i=1;i<=FNR;i++) print a[i] }' $(ls -1v *)
Results:
1 8 a
2 9 b
3 10 c
4 11 d
5 12 e
6 13 f
7 14 g
Explanation:
For each line of input of each input file:
Add the files line number to an array with a value of column 5.
(a[FNR] ? a[FNR] FS : "") is a ternary operation, which is set up to build up the arrays value as a record. It simply asks if the files line number is already in the array. If so, add the arrays value followed by the default file separator before adding the fifth column. Else, if the line number is not in the array, don't prepend anything, just let it equal the fifth column.
At the end of the script:
Use a C-style loop to iterate through the array, printing each of the arrays values.
For only ~4000 files, you should be able to do:
find . -name sample_problem*_part*.txt | xargs paste
If find is giving names in the wrong order, pipe it to sort:
find . -name sample_problem*_part*.txt | sort ... | xargs paste
# print filenames in sorted order
find -name sample\*.txt | sort |
# extract 5-th column from each file and print it on a single line
xargs -n1 -I{} sh -c '{ cut -s -d " " -f 5 $0 | tr "\n" " "; echo; }' {} |
# transpose
python transpose.py ?
where transpose.py:
#!/usr/bin/env python
"""Write lines from stdin as columns to stdout."""
import sys
from itertools import izip_longest
missing_value = sys.argv[1] if len(sys.argv) > 1 else '-'
for row in izip_longest(*[column.split() for column in sys.stdin],
fillvalue=missing_value):
print " ".join(row)
Output
1 8 a
2 9 b
3 10 c
4 11 d
5 ? e
6 ? f
? ? g
Assuming the first and second files have less lines than the third one (missing values are replaced by '?').
Try this one. My script assumes that every file has the same number of lines.
# get number of lines
lines=$(wc -l sample_problem1_part1.txt | cut -d' ' -f1)
for ((i=1; i<=$lines; i++)); do
for file in sample_problem*; do
# get line number $i and delete everything except the last column
# and then print it
# echo -n means that no newline is appended
echo -n $(sed -n ${i}'s%.*\ %%p' $file)" "
done
echo
done
This works. For 4800 files, each 7 lines long it took 2 minutes 57.865 seconds on a AMD Athlon(tm) X2 Dual Core Processor BE-2400.
PS: The time for my script increases linearly with the number of lines. It would take very long time to merge files with 1000 lines. You should consider learning awk and use the script from steve. I tested it: For 4800 files, each with 1000 lines it took only 65 seconds!
You can pass awk output to paste and redirect it to a new file as follows:
paste <(awk '{print $3}' file1) <(awk '{print $3}' file2) <(awk '{print $3}' file3) > file.txt