Regular expressions

Tinkster · 04-21-2010, 09:22 PM

Quote:

Originally Posted by pixellany

Postulate:
Whenever "[" is encountered, the following characters are taken as literal --until the first "]". After that, "]" is literal, but "[" is not

Nice, but:
That wouldn't explain why [][] gives the desired result. :/

grail · 04-21-2010, 10:06 PM

Quote:

Whenever "[" is encountered, the following characters are taken as literal --until the first "]". After that, "]" is literal, but "[" is not

Not sure this holds true for the previous example of:

Code:

[][]

Based on your postulation we meet the first "]" immediately so should this not then be deemed to be two empty character sets?

Kenhelm · 04-21-2010, 10:11 PM

From the regular expression section of the grep man page

Quote:

To include a literal ] place it first in the list.

Or, from the Grymoire tutorial
http://www.grymoire.com/Unix/Regular.html#uh-6

Quote:

Like the anchors in places that can't be considered an anchor, the characters "]" and "-" do not have a special meaning if they directly follow "[".

pixellany · 04-21-2010, 10:13 PM

Touche!!!

Add to the postulate: It continues to the first "]" assuming that there is something in the character set. We have to assume that there is a simple and deterministic rule in effect.....

"[][]" in fact matches "]" OR "["

grail · 04-21-2010, 11:09 PM

OK, so based on the following file:

Quote:

from
a
n
[]
[[
]
s
9
-
][
plus
]]
[

All of the following return the expected results:

Code:

grep -E "[][]" file
grep -E "[][]+" file
grep -E "[][]{2}" file
grep -E "[][9a-]" file //minus had to be last for me here??

All worked equally well with single or double quotes.