how to use AWK to merge two columns from each of 10 files - awk
I have 10 files that have the same tab-delimited column structure. I need to merge the columns 8 and 9 from each of the files. I came up with the following AWK code, but it only merge two files at a time. I am looking for help with merging all 10 files same time, I am not sure if this is feasible.
All my file names following the same pattern (s1s2.txt, s3s4.txt, s5s6.txt, .... s19s20.txt)
#!/bin/bash
awk '
BEGIN {
#load array with contents of the first file
while ( getline < "s1s2.txt" > 0)
{
s1s2_counter++
f1_8[s1s2_counter] = $8
f1_9[s1s2_counter] = $9
}
}
{OFS="\t"}
{ #output the columns 8 and 9 from the first file before the second file
print f1_8[NR],f1_9[NR], $8, $9
} ' s3s4.txt
awk -F'\t' '{a[FNR] = a[FNR] (NR==FNR?"":FS) $8 FS $9} END{for (i=1;i<=FNR;i++) print a[i]}' s1s2.txt .... s19s20.txt
Using getline is usually the wrong approach, see http://awk.info/?tip/getline.
Related
Edit multiple files using awk
I am trying to edit multiple files using awk in the following way:
awk '{F = FILENAME ".inp"; print $0 > F ; if(NR==3) print "%moinp usr/speciale/br/brhgooh/scan/newscan/freqBZ/cas/prov2/",FILENAME > F }' *.xyz
It works well with the first file, but in the rest of the files the change does not appear.
Any suggestions?
You need to use FNR not NR
NR is the record number of all records seen thus far
FNR is the record number of the current file
Adding some whitespace helps readability:
awk '
BEGINFILE {
close(F)
F = FILENAME ".inp"
}
{print > F}
FNR == 3 {print "%moinp usr/speciale/br/brhgooh/scan/newscan/freqBZ/cas/prov2/" FILENAME > F }
' *.xyz
If your awk does not have BEGINFILE, you can use FNR == 1 instead.
Other changes:
print instead of print $0 ($0 is the default)
print "%mo .../" FILENAME without a comma -- a comma will insert a space after the slash.
close(F) to prevent "too many open files" errors
selecting columns in awk discarding corresponding header
How to properly select columns in awk after some processing. My file here:
cat foo
A;B;C
9;6;7
8;5;4
1;2;3
I want to add a first column with line numbers and then extract some columns of the result. For the example let's get the new first (line numbers) and third columns. This way:
awk -F';' 'FNR==1{print "linenumber;"$0;next} {print FNR-1,$1,$3}' foo
gives me this unexpected output:
linenumber;A;B;C
1 9 7
2 8 4
3 1 3
but expected is (note B is now the third column as we added linenumber as first):
linenumber;B
1;6
2;5
3;2
[fixed and revised]
To get your expected output, use:
$ awk 'BEGIN {
FS=OFS=";"
}
{
print (FNR==1?"linenumber":FNR-1),$(FNR==1?3:1)
}' file
Output:
linenumber;C
1;9
2;8
3;1
To add a column with line number and extract first and last columns, use:
$ awk 'BEGIN {
FS=OFS=";"
}
{
print (FNR==1?"linenumber":FNR-1),$1,$NF
}' file
Output this time:
linenumber;A;C
1;9;7
2;8;4
3;1;3
Why do you print $0 (the complete record) in your header? And, if you want only two columns in your output, why to you print 3 (FNR-1, $1 and $3)? Finally, the reason why your output field separators are spaces instead of the expected ; is simply that... you did not specify the output field separator (OFS). You can do this with a command line variable assignment (OFS=\;), as shown in the second and third versions below, but also using the -v option (-v OFS=\;) or in a BEGIN block (BEGIN {OFS=";"}) as you wish (there are differences between these 3 methods but they don't matter here).
[EDIT]: see a generic solution at the end.
If the field you want to keep is the second of the input file (the B column), try:
$ awk -F\; 'FNR==1 {print "linenumber;" $2; next} {print FNR-1 ";" $2}' foo
linenumber;B
1;6
2;5
3;2
or
$ awk -F\; 'FNR==1 {print "linenumber",$2; next} {print FNR-1,$2}' OFS=\; foo
linenumber;B
1;6
2;5
3;2
Note that, as long as you don't want to keep the first field of the input file ($1), you could as well overwrite it with the line number:
$ awk -F\; '{$1=FNR==1?"linenumber":FNR-1; print $1,$2}' OFS=\; foo
linenumber;B
1;6
2;5
3;2
Finally, here is a more generic solution to which you can pass the list of indexes of the columns of the input file you want to print (1 and 3 in this example):
$ awk -F\; -v cols='1;3' '
BEGIN { OFS = ";"; n = split(cols, c); }
{ printf("%s", FNR == 1 ? "linenumber" : FNR - 1);
for(i = 1; i <= n; i++) printf("%s", OFS $(c[i]));
printf("\n");
}' foo
linenumber;A;C
1;9;7
2;8;4
3;1;3
linux csv file concatenate columns into one column
I've been looking to do this with sed, awk, or cut. I am willing to use any other command-line program that I can pipe data through.
I have a large set of data that is comma delimited. The rows have between 14 and 20 columns. I need to recursively concatenate column 10 with column 11 per row such that every row has exactly 14 columns. In other words, this:
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p
will become:
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
I can get the first 10 columns. I can get the last N columns. I can concatenate columns. I cannot think of how to do it in one line so I can pass a stream of endless data through it and end up with exactly 14 columns per row.
Examples (by request):
How many columns are in the row?
sed 's/[^,]//g' | wc -c
Get the first 10 columns:
cut -d, -f1-10
Get the last 4 columns:
rev | cut -d, -f1-4 | rev
Concatenate columns 10 and 11, showing columns 1-10 after that:
awk -F',' ' NF { print $1","$2","$3","$4","$5","$6","$7","$8","$9","$10$11}'
Awk solution:
awk 'BEGIN{ FS=OFS="," }
{
diff = NF - 14;
for (i=1; i <= NF; i++)
printf "%s%s", $i, (diff > 1 && i >= 10 && i < (10+diff)?
"": (i == NF? ORS : ","))
}' file
The output:
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
With GNU awk for the 3rd arg to match() and gensub():
$ cat tst.awk
BEGIN{ FS="," }
match($0,"(([^,]+,){9})(([^,]+,){"NF-14"})(.*)",a) {
$0 = a[1] gensub(/,/,"","g",a[3]) a[5]
}
{ print }
$ awk -f tst.awk file
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
If perl is okay - can be used just like awk for stream processing
$ cat ip.txt
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p
1,2,3,4,5,6,3,4,2,4,3,4,3,2,5,2,3,4
1,2,3,4,5,6,3,4,2,4,a,s,f,e,3,4,3,2,5,2,3,4
$ awk -F, '{print NF}' ip.txt
16
18
22
$ perl -F, -lane '$n = $#F - 4;
print join ",", (#F[0..8], join("", #F[9..$n]), #F[$n+1..$#F])
' ip.txt
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
1,2,3,4,5,6,3,4,2,43432,5,2,3,4
1,2,3,4,5,6,3,4,2,4asfe3432,5,2,3,4
-F, -lane split on , results saved in #F array
$n = $#F - 4 magic number, to ensure output ends with 14 columns. $#F gives the index of last element of array (won't work if input line has less than 14 columns)
join helps to stitch array elements together with specified string
#F[0..8] array slice with first 9 elements
#F[9..$n] and #F[$n+1..$#F] the other slices as needed
Borrowing from Ed Morton's regex based solution
$ perl -F, -lape '$n=$#F-13; s/^([^,]*,){9}\K([^,]*,){$n}/$&=~tr|,||dr/e' ip.txt
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
1,2,3,4,5,6,3,4,2,43432,5,2,3,4
1,2,3,4,5,6,3,4,2,4asfe3432,5,2,3,4
$n=$#F-13 magic number
^([^,]*,){9}\K first 9 fields
([^,]*,){$n} fields to change
$&=~tr|,||dr use tr to delete the commas
e this modifier allows use of Perl code in replacement section
this solution also has the added advantage of working even if input field is less than 14
You can try this gnu sed
sed -E '
s/,/\n/9g
:A
s/([^\n]*\n)(.*)(\n)(([^\n]*\n){4})/\1\2\4/
tA
s/\n/,/g
' infile
First variant - with awk
awk -F, '
{
for(i = 1; i <= NF; i++) {
OFS = (i > 9 && i < NF - 4) ? "" : ","
if(i == NF) OFS = "\n"
printf "%s%s", $i, OFS
}
}' input.txt
Second variant - with sed
sed -r 's/,/#/10g; :l; s/#(.*)((#[^#]){4})/\1\2/; tl; s/#/,/g' input.txt
or, more straightforwardly (without loop) and probably faster.
sed -r 's/,(.),(.),(.),(.)$/#\1#\2#\3#\4/; s/,//10g; s/#/,/g' input.txt
Testing
Input
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u
Output
a,b,c,d,e,f,g,h,i,jkl,m,n,o,p
a,b,c,d,e,f,g,h,i,jklmn,o,p,q,r
a,b,c,d,e,f,g,h,i,jklmnopq,r,s,t,u
Solved a similar problem using csvtool. Source file, copied from one of the other answers: $ cat input.txt a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p 1,2,3,4,5,6,3,4,2,4,3,4,3,2,5,2,3,4 1,2,3,4,5,6,3,4,2,4,a,s,f,e,3,4,3,2,5,2,3,4 Concatenating columns: $ cat input.txt | csvtool format '%1,%2,%3,%4,%5,%6,%7,%8,%9,%10%11%12,%13,%14,%15,%16,%17,%18,%19,%20,%21,%22\n' - a,b,c,d,e,f,g,h,i,jkl,m,n,o,p,,,,,, 1,2,3,4,5,6,3,4,2,434,3,2,5,2,3,4,,,, 1,2,3,4,5,6,3,4,2,4as,f,e,3,4,3,2,5,2,3,4 anatoly#anatoly-workstation:cbs$ cat input.txt
How to print two lines of several files to a new file with speicific order?
I have a task to do with awk. I am doing sequence analysis for some genes.
I have several files with sequences in order. I would like to extract first sequence of each file into new file and like till the last sequence. I know only how to do with first or any specific line with awk.
awk 'FNR == 2 {print; nextfile}' *.txt > newfile
Here I have input like this
File 1
Saureus081.1
ATCGGCCCTTAA
Saureus081.2
ATGCCTTAAGCTATA
Saureus081.3
ATCCTAAAGGTAAGG
File 2
SaureusRF1.1
ATCGGCCCTTAC
SauruesRF1.2
ATGCCTTAAGCTAGG
SaureusRF1.3
ATCCTAAAGGTAAGC
File 3
SaureusN305.1
ATCGGCCCTTACT
SauruesN305.2
ATGCCTTAAGCTAGA
SaureusN305.3
ATCCTAAAGGTAATG
And similar files 12 are there
File 4
.
.
.
.File 12
Required Output
Newfile
Saureus081.1
ATCGGCCCTTAA
SaureusRF1.1
ATCGGCCCTTAC
SaureusN305.1
ATCGGCCCTTACT
Saureus081.2
ATGCCTTAAGCTATA
SaureusRF1.2
ATGCCTTAAGCTAGG
SauruesN305.2
ATGCCTTAAGCTAGA
Saureus081.3
ATCCTAAAGGTAAGG
SaureusRF1.3
ATCCTAAAGGTAAGC
SaureusN305.3
ATCCTAAAGGTAATG
I guess this task can be done easily with awk but not getting any idea how to do for multiple lines
Based on the modified question, the answer shall be done with some changes.
$ awk -F'.' 'NR%2{k=$2;v=$0;getline;a[k]=a[k]?a[k] RS v RS $0:v RS $0} END{for(i in a)print a[i]}' file1 file2 file3
Saureus081.1
ATCGGCCCTTAA
SaureusRF1.1
ATCGGCCCTTAC
SaureusN305.1
ATCGGCCCTTACT
Saureus081.2
ATGCCTTAAGCTATA
SauruesRF1.2
ATGCCTTAAGCTAGG
SauruesN305.2
ATGCCTTAAGCTAGA
Saureus081.3
ATCCTAAAGGTAAGG
SaureusRF1.3
ATCCTAAAGGTAAGC
SaureusN305.3
ATCCTAAAGGTAATG
Brief explanation,
Set '.' as the delimeter
For every odd record, distinguish k=$2 as the key of array a
Invoke getline to set $0 of next record as the value corresponds to the key k
Print the whole array for the last step
If your data is very large, I would suggest creating temporary files:
awk 'FNR%2==1 { filename = $1 }
{ print $0 >> filename }' file1 ... filen
Afterwards, you can cat them together:
cat Seq1 ... Seqn > result
This has the additional advantage that it will work if not all sequences are present in all files.
paste + awk solution:
paste File1 File2 | awk '{ p=$2;$2="" }NR%2{ k=p; print }!(NR%2){ v=p; print $1 RS k RS v }'
paste File1 File2 - merge corresponding lines of files
p=$2;$2="" - capture the value of the 2nd field which is the respective key/value from File2
The output:
Seq1
ATCGGCCCTTAA
Seq1
ATCGGCCCTTAC
Seq2
ATGCCTTAAGCTATA
Seq2
ATGCCTTAAGCTAGG
Seq3
ATCCTAAAGGTAAGG
Seq3
ATCCTAAAGGTAAGC
Additional approach for multiple files:
paste Files[0-9]* | awk 'NR%2{ k=$1; n=NF; print k }
!(NR%2){ print $1; for(i=2;i<=n;i++) print k RS $i }'
How to do calculations over lines of a file in awk
I've got a file that looks like this:
88.3055
45.1482
37.7202
37.4035
53.777
What I have to do is isolate the value from the first line and divide it by the values of the other lines (it's a speedup calculation). I thought of maybe storing the first line in a variable (using NR) and then iterate over the other lines to obtain the values from the divisions. Desired output is:
1,9559
2,3410
2,3608
1,6420
UPDATE
Sorry Ed, my mistake, the desired decimal point is .
I made some small changes to Ed's answer so that awk prints the division of 88.3055 by itself and outputs it to a file speedup.dat:
awk 'NR==1{n=$0} {print n/$0}' tavg.dat > speedup.dat
Is it possible to combine the contents of speedup.dat and the results from another awk command without using intermediate files and in one single awk command?
First command:
awk 'BEGIN { FS = \"[ \\t]*=[ \\t]*\" } /Total processes/ { if (! CP) CP = $2 } END {print CP}' cg.B.".n.".log ".(n == 1 ? ">" : ">>")." processes.dat
This first command outputs:
1
2
4
8
16
Paste of the two files:
paste processes.dat speedup.dat > prsp.dat
which gives the now desired output:
1 1
2 1.9559
4 2.34107
8 2.36089
16 1.64207
$ awk 'NR==1{n=$0;next} {print n/$0}' file
1.9559
2.34107
2.36089
1.64207
$ awk 'NR==1{n=$0;next} {printf "%.4f\n", n/$0}' file
1.9559
2.3411
2.3609
1.6421
$ awk 'NR==1{n=$0;next} {printf "%.4f\n", int(n*10000/$0)/10000}' file
1.9559
2.3410
2.3608
1.6420
$ awk 'NR==1{n=$0;next} {x=sprintf("%.4f",int(n*10000/$0)/10000); sub(/\./,",",x); print x}' file
1,9559
2,3410
2,3608
1,6420
Normally you'd just use the correct locale to have . or , as your decimal point but your input uses . while your output uses , so I don't think that's an option.
awk '{if(n=="") n=$1; else print n/$1}' inputFile