sed wildcard substitution

regex sed grep wildcard string-substitution
Steve picture Steve · Jun 7, 2011 · Viewed 19.4k times · Source

I want to do a substitution based on a wildcard. For example, change all "tenure" to "disposition" only if the word "tenure" comes after an '=' sign. Basically a regex that would match this =.*tenure

The sed command that I have so for this is:

sed 's/=.*tenure/=.*disposition/g' file.txt

However, if I pass this to a file containing:

blah blah blah = change "tenure" to "disposition"

I get 

blah blah blah =.*disposition" to "disposition"

instead of:

blah blah blah = change "disposition" to "disposition"

How do I do the substitution such that the wildcard in the regex won't be part of the destination file?

Answer

MGwynne picture MGwynne · Jun 7, 2011

You need to use a capturing group to capture the text that appears between your equals sign and "tenure".

So

sed 's/=\(.*\)tenure/=\1disposition/g'

Note the use of \1 to reference and use the group you captured.

So in 

echo 'blah blah blah = change "tenure" to "disposition"' | sed 's/=\(.*\)tenure/=\1disposition/g'

we get

blah blah blah = change "disposition" to "disposition".

See Regex grouping.

Related questions

Using sed and grep/egrep to search and replace

I am using egrep -R followed by a regular expression containing about 10 unions, so like: .jpg | .png | .gif etc. This works well, now I would like to replace all strings found with .bmp I was thinking of something like egrep …

Read More
Counting regex pattern matches in one line using sed or grep?

I want to count the number of matches there is on one single line (or all lines as there always will be only one line). I want to count not just one match per line as in echo "123 123 123" | grep -c …

Read More
Does lookbehind work in sed?

I created a test using grep but it does not work in sed. grep -P '(?<=foo)bar' file.txt This works correctly by returning bar. sed 's/(?<=foo)bar/test/g' file.txt I was expecting footest …

Read More