Combine two text files in bash script

Question

I have a text file (FILE_A.txt) with this content:

Content1
Content2
Content3

And other text file (FILE_B.txt) with this content:

A[#]
B[#]
C[#]

I would like to combine FILE_A.txt and FILE_B.txt in other file (FILE_C.txt) in this way:

A[Content1]
B[Content2]
C[Content3]

How could I make this using bash shell in linux (sed, cut, grep, etc)?

I've tried again with diferents contents for FILE_A and FILE_B, and I don't know why it doesn't works. File_B.txt now has "something(#,@);" and File_A.txt has "__OtherTexts". When I try to combine the two files with dogbane method, it copies the content of each line at the end of the lines contained in FILE_B.txt. Any ideas? — Cesar Ortiz
– Cesar Ortiz, Commented Dec 12, 2012 at 15:43

ghoti · Accepted Answer · 2012-12-11 15:34:47Z

3

Here we go.

# awk 'NR==FNR{a[NR]=$0;next;} sub(/#/,a[FNR])' FILE_A.txt FILE_B.txt
A[Content1]
B[Content2]
C[Content3]

How does this work?

NR==FNR - causes the following statements to be run if the record number matches the FILE record number - that is, we are currently reading only the firs tfile.
{a[NR]=$0;next;} - Store values from the first file in an array.
sub(/#/,a[FNR]) - Once we're in the second file, substitute # for the matching value stored from the first file. Note that this isn't inside curly brackets, so it's being evaluated as a condition. If the sub() statement succeeds, the current line is printed.

answered Dec 11, 2012 at 15:34

ghoti

47.2k8 gold badges70 silver badges108 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

ghoti Over a year ago

No problem, Cesar. The advantage of this also is that because we're using sub() to replace a substring, this solution basically makes the lines of FILE_A.txt into a template for the data in FILE_B.txt. If your requirements become more complex than your sample data, this may be preferable to handling the output from paste.

dogbane · Accepted Answer · 2012-12-11 15:32:11Z

2

Use paste and sed as follows:

$ paste File_B.txt File_A.txt | sed 's/#]\s*\(.*$\)/\1]/g'
A[Content1]
B[Content2]
C[Content3]

answered Dec 11, 2012 at 15:32

dogbane

276k77 gold badges407 silver badges415 bronze badges

2 Comments

ghoti Over a year ago

Works in Linux, but the \s makes it non-portable. Won't work in BSD, OSX, Solaris, etc, unless you've installed GNU sed.

dogbane Over a year ago

@ghoti That's easily fixed. You can change the paste command to use a specific delimiter of your choice (using the -d option) and then use the same delimiter in the sed command instead of the \s*.

Shawn Chin · Accepted Answer · 2012-12-11 15:54:08Z

2

The following reads both files concurrently, one line at a time, and store the lines in $value and $template. We then use bash's variable substring replacement to replace # within $template with the contents of $value.

exec 6<"FILE_B.txt"  # open file for reading and assign file descriptor 6

while read -r value; do  # loop through FILE_A.txt, storing each line as $value
  read -r template <&6   # read a line from FILE_B.txt, store as $template
  echo ${template/\#/$value}  # replace value into the template in place of `#`
done <"FILE_A.txt"

exec 6<&-  # close input file descriptor 6

edited Dec 11, 2012 at 15:54

answered Dec 11, 2012 at 15:42

Shawn Chin

87.4k20 gold badges168 silver badges193 bronze badges

1 Comment

Cesar Ortiz Over a year ago

Complex but very educational. Thanks for your answer.

Collectives™ on Stack Overflow

Combine two text files in bash script

3 Answers 3

1 Comment

2 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

1 Comment

2 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related