CODEWITHSUNDEEP
bashawksed
spywill
spywill

Reputation: 3

copy and paste between two txt file

Hi I am new to bash and using sed need a little help

I have two txt files i need to copy and paste between them the first file I know what the text is and placed of the text but the second txt file I don't know the text but I do know the placed of the text is.

In file1 put the two text words or numbers from file2 and place them like I show below.

When I create file2 all I am going to know about it will have two words or numbers on the same line4

I have been trying with this

sed $'10{e sed "4!d" /home/Desktop/file1.txt\n;d}'  /home/Desktop/file2.txt

and

awk 'NR==4{a=$0}NR==FNR{next}FNR==10{print a}4' /home/Desktop/file2.txt /home/Desktop/file1.txt

This is what my files would look like

file1.txt

cat
hat
sat
fat
mat
rat

file2.txt

line1
line2
line3
text1 text2
line5

I need it to look like this

file1.txt

cat
hat
sat text1
fat text2
mat
rat

thanks for any help

Upvotes: 0

Views: 214

Answers (2)

potong
potong

Reputation: 58371

This might work for you (GNU sed):

sed -E '1{x;s#^#sed -n 4p file2#e;x};3{G;s/\n(\S+).*/ \1/};4{G;s/\n\S+//}' file1

Stuff the line from file2 into the hold space when processing file1 and append and manipulate that line when needed.

A more explicit explanation:

By default, sed reads each line of a file. For each cycle, it removes the newline, places the result in the pattern space, goes through a sequence of commands, re-appends the newline and prints the result e.g. sed '' file replicates the cat command. The sed commands are usually placed between '...' and represent a cycle, thus:

1{x;s#^#sed -n 4p file2#e;x}

  • 1{..} executes the commands between the ellipses on the first line of file1. Commands are separated by ;'s
  • x sed provides two buffers. After removing the newline that delimits each line of a file, the result is placed in the pattern space. Another buffer is provided empty, at the start of each invocation, called the hold space. The x swaps the pattern space for the hold space.
  • s#^#sed -n 4p file2#e this inserts another sed invocation into the empty hold space and evaluates it by the use of the e flag. The second invocation turns off implicit printing (-n option) and then prints line 4 of file2 only.
  • x the hold space is now swapped with the pattern space.Thus, line 4 of file2 is placed in the hold space.

3{G;s/\n(\S+).*/ \1/}

  • 3{..} executes the commands between the ellipses on the third line of file1.
  • G append the contents of hold space to the pattern space using a newline as a separator.
  • s/\n(\S+).*/ \1/ match on the appended hold space and replace it by a space and the first column.

4{G;s/\n\S+//}

  • 4{..} executes the commands between the ellipses on the fourth line of file1.
  • G append the contents of hold space to the pattern space using a newline as a separator.
  • s/\n\S+// match on the appended hold space and remove the newline and the first column, thus leaving a space and the second column. m

Upvotes: 1

tshiono
tshiono

Reputation: 22012

Assuming you want to append the fields of the 4th line of file2.txt to the 3rd and the following lines of file1.txt, how about:

awk 'FNR==NR {if (FNR==4) split($0, ary, " "); next} {print $0 " " ary[FNR - 3 + 1]}' /home/Desktop/file2.txt /home/Desktop/file1.txt

Result:

cat
hat
sat text1
fat text2
mat
rat

Upvotes: 1

Related Questions