I am trying to compare two files and output a file which consists of common names for both.
File1
1990.A.BHT.s_fil 4.70
1991.H.BHT.s_fil 2.34
1992.O.BHT.s_fil 3.67
1993.C.BHT.s_fil -1.50
1994.I.BHT.s_fil -3.29
1995.K.BHT.s_fil -4.01
File2
1990.A.BHT_ScS.dat 1537 -2.21
1993.C.BHT_ScS.dat 1494 1.13
1994.I.BHT_ScS.dat 1545 0.15
1995.K.BHT_ScS.dat 1624 1.15
I want to compare the first parts of the names ** (ex:1990.A.BHT ) ** on both files and output a file which has common names with the values on 2nd column in file1 to file3
ex: file3 (output)
1990.A.BHT.s_fil 4.70
1993.C.BHT.s_fil -1.50
1994.I.BHT.s_fil -3.29
1995.K.BHT.s_fil -4.01
I used following codes which uses grep command
while read line
do
grep $line file1 >> file3
done < file2
and
grep -wf file1 file2 > file3
I sort the files before using this script.
But I get an empty file3. Can someone help me with this please?
You need to remove everything starting from _SCS.dat from the lines in file2. Then you can use that as a pattern to match lines in file1.
grep -F -f <(sed 's/_SCS\.dat.*//' file2) file1 > file3
The -F option matches fixed strings rather than treating them as regular expressions.
In your example data, the lines appear to be in sorted order. If you can guarantee that they always are, comm -1 -2 file1 file2 would do the job. If they can be unsorted, do a
comm -1 -2 <(sort file1) <(sort file2)
Related
Given two dotenv files,
# file1
FOO="X"
BAR="B"
and
# file2
FOO="A"
BAZ="C"
I want to run
$ ./merge.sh file1.env file2.env > file3.env
to get the following output:
# file3
FOO="A"
BAR="B"
BAZ="C"
So far, I used the python-dotenv module to parse the files into dictionaries, merge them and write them back. However, I feel like there should be a simple solution in shell that rids myself of a third-party module for such a basic task.
Answer
Alright, so I ended up using
$ sort -u -t '=' -k 1,1 file1 file2 | grep -v '^$\|^\s*\#' > file3
which omits blank lines and comments. Nevertheless, the proposed awk solution works just as fine.
Another quite simple approach is to use sort:
sort -u -t '=' -k 1,1 file1 file2 > file3
results in a file where the keys from file1 take precedence over the keys from file2.
Using a simple awk script:
awk -F= '{a[$1]=$2}END{for(i in a) print i "=" a[i]}' file1 file2
This stores all key values in the array a and prints the array content when both files are parsed.
The keys that are in file2 override the ones in file1.
To add new values only from file2 and NOT overwrite initial values from file1. Omit spaces from file 2.
grep "\S" file2 >> file1
awk -F "=" '!a[$1]++' file1 > file3
I have two files file1 and file2. I want to print the new line added to file2 using diff.
file1
/root/a
/root/b
/root/c
/root/d
file2
/root/new
/root/new_new
/root/a
/root/b
/root/c
/root/d
Expected output
/root/new
/root/new_new
I looked into man page but there was no any info on this
If you don't need to preserve the order, you could use the comm command like:
comm -13 <(sort file1) <(sort file2)
comm compares 2 sorted files and will print 3 columns of output. First is the lines unique to file1, then lines unique to file2 then lines common to both. You can supress any columns, so we turn of 1 and 3 in this example with -13 so we will see only lines unique to the second file.
or you could use grep:
grep -wvFf file1 file2
Here we use -f to have grep get its patterns from file1. We then tell it to treat them as fixed strings with -F instead of as patterns, match whole words with -w, and print only lines with no matches with -v
Following awk may help you on same. This will tell you all those lines which are present in Input_file2 and not in Input_file1.
awk 'FNR==NR{a[$0];next} !($0 in a)' Input_file1 Input_file2
Try using a combination of diff and sed.
The raw diff output is:
$ diff file1 file2
0a1,2
> /root/new
> /root/new_new
Add sed to strip out everything but the lines beginning with ">":
$ diff file1 file2 | sed -n -e 's/^> //p'
/root/new
/root/new_new
This preserves the order. Note that it also assumes you are only adding lines to the second file.
I have the files file1 and file2, where file2 is a subset of file1. That means, if I iterate over file1, there are some lines that are in file2, and some that aren't, but there is no line in file2 that is not in file1. There may be several lines with the same content in a file. Now I want to get the difference between them, that is, all lines of file1 that aren't in file2.
According to this well received answer
diff(1) isn't the answer, comm(1) is.
(For whatever reason)
But as I understand, for comm the files need to be sorted first. The problem: Both files are ordered (not sorted!), and this order needs to be kept. So what I really want is to iterate over file1, and check for every line, if it is also in file2. If not, write it to file3. If the same content occurs more than once, it should be kept more than once!
Is there any way to do this with the command line?
Try this with GNU grep:
grep -vFf file2 file1 > file3
Update:
grep -vxFf file2 file1 > file3
I think you do not want to sort for avoiding temp files. This is possible with process substitution:
diff <(sort file1) <(sort file2)
# or
comm <(sort file1) <(sort file2)
Edit: Using https://stackoverflow.com/a/4544925/3220113 I found another alternative (for text files with short lines):
diff -a --suppress-common-lines -y file2 file1 | sed 's/\s*>.//'
file1 contains:
someword0
someword2
someword4
someword6
someword8
someword9
someword1
file2 contains:
someword2
someword3
someword4
someword5
someword7
someword11
someword1
So I wan't to have only lines from file1 which file2 doesn't contains. How can I do this in bash ?
That's the answer:
grep -v -x -f file2 file1
-v for select non-matching lines
-x for matching whole lines only
-f f2 to get patterns from f2.
You can use grep -vf:
grep -vwFf file2 file1
someword0
someword6
someword8
someword9
Check man grep for detailed info on all the grep options used here.
You could use the comm command as well:
comm -23 file1 file2
Explanation:
comm compares two files and prints, in 3 columns, lines unique to file1, file2, and lines in both.
Using the options -2 and -3 (or simply -23) suppresses printing of these columns, so you just get the lines unique to file1.
If your lines are unique, do a left join and filter out lines that exist in both tables.
join <(sort file1) <(sort file2) -o0,2.1 -a1 | awk '!$2'
I am having time with my syntax here:
I have 3 files with various content file1 file2 file3 (100+ lines). I am trying to merge them together, but only the first line of each file should be merged. The point is to do it using one line of bash code:
sed -n 1p file1 file2 file3 returns only the first line of file1
You might want to try
head -n1 -q file1 file2 file3.
It's not clear if by merge you mean concatenate or join?
In awk by joining (each first line in the files printed side by side):
$ awk 'FNR==1{printf "%s ",$0}' file1 file2 file3
1 2 3
In awk by concatenating (each first line in the files printed one after another):
$ awk 'FNR==1' file1 file2 file3
1
2
3
I suggest you use head as explained by themel's answer. However, if you insist in using sed you cannot simply pass all files to it, since they are implicitly concatenated and you lose information about what the first line is in each file respectively. So, if you really want to do it in sed, you need bash to help you out:
for f in file1 file2 file3; do sed -n 1p "$f"; done
You can avoid calling external processes by using the read built-in command:
for f in file1 file2 file3; do read l < $f; echo "$l"; done > merged.txt