Add data validation to prevent rogue entries about pi-hole

[korhadris]

Here's a sed command to remove duplicated '.' characters:

sed 's/\.\+/./g'

This is probably faster as it will only make the replacement if there are two or more '.' characters ...

sed 's/\.\.\+/./g'

[kurumushi]

The real challenge is to see if these can be combined into one, I'm not very good at simplification.
Here's a few more useful ones:
sed -e 's/ \.//g': remove periods at the start and the middle of domains.
sed -e 's/\[. ]+$/$/g' remove domains ending in a period, and trailing white space.

[korhadris]

Yeah, simplifying regexp statements to be optimized is rough. As long as I really don't require the speed I usually combine statements like this:

sed -e 's/\.\.\+/./g' -e 's/^[. ]*//' -e 's/\.* \+\.*/ /g' -e 's/[. ]\+$//'

This way it's easier for a human to look and see what's going on.
This removes duplicated . characters and replaces with a single .; removes spaces and . characters at the front of the line; removes extra spaces and . characters in the middle (leaving a single space); removes spaces and . characters at the end of the line.

I tested it with the following example:

echo "  .192.168.23.515.. ...foo...bar..com.  " | sed -e 's/\.\.\+/./g' -e 's/^[. ]*//' -e 's/\.* \+\.*/ /g' -e 's/[. ]\+$//'
192.168.23.515 foo.bar.com

I also fixed our regexp examples in this post (we both had some errors). I had an extra / at the end of my commands above (now editted); sed needs to have + escaped, or it will try to match a + character; $ in the second substitution section will put in an actual $ character; there was an extra \ before the [, which would try to match a [ character instead of starting a list of possible characters to match. I also removed the g from the end of the that can only match at the end of the line, since the global part isn't needed (doesn't break having it there though).

[jacobsalmela]

It looks pretty good, but it's a bit out of my skill level. Maybe someone could make a pull request and I can test it out.