I searched this forum and google before hand but couldn't find anything that helped me out. What I need to do is match a right bracket mark ( ] ) inside of brackets ( [] ). Actually I need to make that negative with ^ also.

I am trying to match
Where the text Hello World can be anything but the [say=] will always be the same. Using instinct I tried:
but can't get it to work, I know I'm doing something wrong but does anything know how to match a ] inside of []?

I might misunderstand the question, but I don't think you need the double [[]] construct
If you want to match this: [say=Hello Word] and Hello World can be anything, this will match what you want:

\[say=.*\]

The .* matches anything (including nothing) but it has to start with \[say= and end with \]. You need to escape them, otherwise everything in between is seen as possibilities for the regular expression, you need them literaly.

Like I said, I could have misunderstood the question. Hope this helps anyway

That would work except that if they use it twice or use another ] mark later it might match to more text that I want it to.

Example:
In the above example that regexp would match "[say=say test[testing more:]", is there a way to switch the PHP ereg function and the javascript string.replace function to ungreedy?

I guess I don't understand what it is you want.

"test [say=say test] testing more :]"
"[say=say test[testing more:]"

Both get valid hits by what you described earlier (Hello World can be anything, but the [say=] will always be the same). Both have the [say= and the ending ] and some text in between.

Using sed, this is what I get:

$ echo $TESTING
test [say=say test] testing more :]

Show all, except what's in between [say= and the ending ]
$ echo $TESTING | sed 's/\(\[say=\).*\(\]$\)/\1\2/'
test [say=]

Show all, except [say= and the ending ]
$ echo $TESTING | sed 's/\(\[say=\)\(.*\)\(\]$\)/\2/'
test say test] testing more :

And for the second example:

echo $TESTING
[say=say test[testing more:]

Show all, except what's in between [say= and the ending ]
$ echo $TESTING | sed 's/\(\[say=\).*\(\]$\)/\1\2/'
[say=]

Show all, except [say= and the ending ]
$ echo $TESTING | sed 's/\(\[say=\)\(.*\)\(\]$\)/\2/'
say test[testing more:

Both look correct to me, but we probably are talking about different things

I'm not familiar with javascript and have some experience with PHP, cannot tell you if there is a function in PHP that would do what you want.

Could/would you elaborate on what it is you are trying to do.

I am working on a chat room, when the message is sent I need to format certain tags that contain style information. One of these is the color tag that has the syntax: Black text or you can do black text. The problem is that the PHP eregi function (which finds a match based on a regular expression you give it) is greedy, so it doesn't stop when it has a match (ie. string = "[color=red]red text[/red] black text", regexp = \[color=.*\], it returns "[color=red]red text[/red]" since it starts with [color= and ends in ] but it's the biggest possible thing that it can match and I need it to return when it hits the first ] not the last ] so it only returns "[color=red]".

Anyone know how to do this? (It's most important in javascript, I can use preg_match in PHP to make it ungreedy)

I'm clear about your problem and it's a nice one.......

Have to think about it for a bit, but I don't think it can be done with just a regular expression (not knowing what javascript can/cannot do). This will take a few lines of code.

First that comes to mind is counting the amount of [say= that are in a line. If there's more then one, chop the line to chunks and work your magic on one chunck at the time. Will let you know if I find a(n elegant) solution.

How about something like: "\[say=.*[^\]].*\]"

This is just off the top of my head, but that should match
[say=<anything other than a ]>]

Ok, I tried again with :

\[size=[^\]]*\] and got it to work, the code above didn't because of hte .* after the [^\]] that matched everything to the end of hte string if there was a second ] mark.

I also got it to work with \[size=(.+?)?\] but I have no clue whta (.+?)? means, if anyone has a minute could they explain that part please.

Thanks for the help everyone RegExps are difficult for me

I can't say that I know anything at all about php... but I do know that perl has the same issue. Perl's matching by default is greedy and always try to find the largest match possile. You can use a ? to tell perl to be stingy in the match.

For example..

If you used the regex /a.+b/ to match "a xxx b funregex b" you find that perl matches the whole string.

If we wanted perl to just match the "a xxx b" we would use

/a.+?b/

This tells perl to stop being greedy directly after the .+ and to stop matching at the first "b".

I hope this helps.

-Shawn

Ah, thank you. That makes sense and that seems to be the way it is acting also. I believe regular expressions are pretty much the same in all programming languages and I have been able to use the same regular expressions for perl, php and javascript fine.