mjhjmtu

I have a text file with this ending:

line1^I$^M$
line2^I$^M$
line3^I$^M$

I have trying many options I found in the web by using sed command and even 'dos2unix' but they only work to remove the ^M$ but not ^I. So my file is now like this:

line1^I$
line2^I$
line3^I$

I have scrutinized in the web but there is nothing about ^I. What exactly does it mean?

So your file will look like this using od or cat -A

$ od -c foo
0000000 l i n e 1 t n l i n e 2 t n l i
0000020 n e 3 t n
0000025
$ cat -A foo
line1^I$
line2^I$
line3^I$
$

To strip all ^I (Tab) characters, you could use sed -i 's/t//g' foo. To only strip tabs at the end of a line, used sed -i 's/t$//' foo

$ sed -i 's/t//g' foo
$ od -c foo
0000000 l i n e 1 n l i n e 2 n l i n e
0000020 3 n
0000022
$ cat -A foo
line1$
line2$
line3$
$

^I is probably not indicating a caret followed by an I, but rather is a representation of the character composed by pressing Ctrl+I. This character actually has a shortcut on most keyboards, to wit Tab. If you were to cat the file, you would probably see something like:

line1 $
line2 $
line3 $

Unless, that is, whatever you're using that is displaying literal ^Is is also showing the end of the line with a visible $.

You can use tr to eliminate them if you like, though:

$ cat file | tr -d 't'

t in this case is a magic sequence that is an easy way to represent a Tab character in several common tools (including, helpfully, tr).

_{dos2unix did nothing to the Tabs because they are identical between all the platforms that tool is used to "translate" to and from.}