Globbing is relatively simple. More powerful string matching can be done with regular expressions. In Unix, modern regex style is derived from ed
(a predecessor to vi
).
Syntax
ed | shell | description |
---|---|---|
. |
? |
single character |
[ ] |
[ ] |
single from this set or range |
[^ ] |
`[^ ] | single character not in set/range |
* |
>=0 occurances of previous character | |
.* |
** |
>=0 occurances of any letter |
^ |
start of string | |
$ |
end of string | |
\ |
\ |
escape character |
\( \) |
capture match |
Capturing Matches
sed
and grep
provide a way capturing a match and using it again later.
Using \( \)
(up to 9 times), you can recall the captured text with \1
, \2
...