Using sed to match regex

I don't know much about sed, nor regex. I want to replace every line that contains only tabs by the string '0'. There are also lines in my file that contain only '\n'.

Basically I want to use the regular expression ^\h+$ and replace the matches with 0.

I tried:

sed -i 's/^\h+$/0/' file.txt

But it doesn't work

Answer

You can use:

sed -i.bak -E 's/^[[:blank:]]+$/0/' file

POSIX character class [[:blank:]] matches a space or tab which is same as \h in PCRE.

-i.bak is to keep original file in file.bak, in case you want to restore.