sed命令在多行搜索后插入多行字符串

Question

I would like to insert lines of text into another text file after two specific lines.

Inserting after something similar to

some text...
  example text
    (
    );
some text...

I have a text file(containing two lines of text) that I would like sed to insert between the parentheses.

If the text file I am trying to insert contains something similar to the following

need this;
in between the parentheses;

Then I would like the result to look like this

some text...
  example text
    (
     need this;
     in between the parentheses;
    );
some text...

What ever the best solution would be can work(doesn't have to be sed).

EDIT for clarification

There are other open parentheses before the section where the text needs to be inserted eg

sometext...
sometext (sometext)....
sometext
  (
  );
  exampletext
    (
    );
sometext...

So, I would think that the "exampletext" would need to be reference to then look for the parentheses. Also, it would would probably need to search for "exampletext" exactly since there are other lines within the document with "exampletextsometext..."

Once this has been completed the addition to the file would need to be saved over the original file.

Answer 1

If the open ( is on a line by itself you could just do

sed -e '/^(/r fileToInsert' firstFile

since the /^(/ finds the line where you want to insert ("a line starting with an open parenthesis"), and r means "read contents of file and insert at this point.

If the expression needed to determine the insertion point has to be more complex, please elaborate in comments. For example, "exactly an open parenthesis and nothing else" would be /^($/

EDIT thank you for clarifying the requirement. If you need to insert this text right after example text followed by an ( , the following script ought to work. Put it in a file of its own, and make it executable ( chmod 755 myScript ), then run with ./myScript .

#!/bin/bash
sed '
/exampletext/ {
  N
  /(/ r multi2.txt
}' multi1.txt

Explanation:

/exampletext/ {     find a match of this text, then…
N                   go to the next line
/(/                 match open parenthesis
r multi2.txt        insert file 'multi2.txt' here
}'                  end of script
multi1.txt          name of input file

Note that this produces output to stdout . You could direct this to a new file name - for example

./myScript > newFile.txt

I tested this with the following input file ( multi1.txt ):

some text...
sometext...
sometext (sometext)....
  exampletext
  not the right string
    (
    );
sometext
  (
  );
  exampletext
    (
    );
sometext...

And it gave the output

some text...
sometext...
sometext (sometext)....
  exampletext
  not the right string
    (
    );
sometext
  (
  );
  exampletext
    (
insert this
and that
    );
sometext...

Which I think is what you wanted? The text is inserted where the example text is followed by an open parenthesis - but not when there's another line in between...

Answer 2

Here is another way of doing it with awk :

awk '/^\(/{print $0; while((getline line <"filetoInsert") > 0) print line; next}1' firstFile

Explanation:

• We use a simple regex /^\\(/ to state that look at the line that starts with an opening parenthesis.
• Once we encounter that line, we print it.
• We then use getline function to read through the second file.
• Once it has been read and printed completely, we use next to avoid default printing of the line of original file (our paren) again as we have turned on default printing using the 1 you see at the end.

Answer 3

这可能适合你（GNU sed）：

sed -e '$!N;/^(\n);$/{r insert_file' -e '};P;D' first_file

Answer 4

also in python:

import sys

with open(sys.argv[2]) as secondFile:
    insertStrList=secondFile.readlines()

firstStrList=[]
with open(sys.argv[1]) as firstFile:
    while True:
        i_str=firstFile.readline()
        if not i_str: 
            break
        firstStrList.append(i_str)
        if i_str == '(\n':
            firstStrList+=insertStrList

print ''.join(firstStrList)

using it with python mergeParenthesis.py a.txt b.txt

But I agree Floris answer is easier ;)

cheers