How to replace a line that starts with a word (and use that old line on the new line)

Question

I have a file with many lines and I would like to replace specific lines that start with a new line but include the old line in it. See below.

for example, if a line starts with (xyz is different for every line)

"#EXT-1,xyz"

I would like to have a line like this

!group=12, "#EXT-1,xyz", name="#EXT-1,xyz"

Is this possible to do with sed, and if so, how?

I would like to replace specific lines that start with a new line but include the old line in it So , you mean if line is blank, insert the line before it in place of blank one ? — Sergiy Kolodyazhnyy, Sep 03 '16 at 18:42

steeldriver · Accepted Answer · 2016-09-03T18:36:57.290

6

There are a number of ways of formulating it - one would be

sed '/^\"#EXT-1,.*\"/ s//!group=12, &, name=&/' file

If you want to modify the file in place, then add the -i or --in-place switch

sed -i '/^\"#EXT-1,.*\"/ s//!group=12, &, name=&/' file

edited Sep 03 '16 at 18:36

answered Sep 03 '16 at 17:01

steeldriver

it did't write anything to the file, it just listed the lines in the file – LewHam Sep 03 '16 at 17:17
Now I get an message saying "extra characters at the end of d command" – LewHam Sep 03 '16 at 18:59
found the error, my bad.. :-) – LewHam Sep 03 '16 at 19:27

1 Answers1