I am working with some text that is full of stuff between brackets [] that I don't want. Since I can delete the brackets myself, I don't need the one-liner to do that for me, but I do need a one-liner that will remove everything between them.
What about parentheses () instead of brackets?
Replace [some text] by the empty string. Assuming you don't want to parse nested brackets, the some text can't contain any brackets.
sed -e 's/\[[^][]*\]//g'
Note that in the bracket expression [^][] to match anything but [ or ], the ] must come first. Normally a ] would end the character set, but if it's the first character in the set (here, after the ^ complementation character), the ] stands for itself.
If you do want to parse nested brackets, or if the bracketed text can span multiple lines, sed isn't the right tool.
s/\[[^\[\]]*\]//g(i.e. use backslashes to escape the members of the character set). – Gilles 'SO- stop being evil' Jun 12 '11 at 11:58[…]= character set.^as the first character means to complement the set.]would normally be mark the end of the set, but if it's the first character other than^, it's an ordinary character — an empty set or the complement of the empty set isn't permitted. Then[is ordinary and]ends the set. – Gilles 'SO- stop being evil' May 17 '13 at 18:35