small ereg() help needed

Anon

Hi
here's the code

<?
$keywords="these, are, my, keywords";

if (ereg("(^[a-zA-Z]+[, ])+[a-zA-Z]+$", $keywords)) {
print("success");
} else print("failure");
?>

What I'm trying to say is (starts with any word char followed by a comma and space) and repeated one or many times but ending with a word char.

Help please?

Nick

Anon

Ok. I don't know what basics of regular expressions you know, but lets try explaining this. However, I'm going to use perl compatible regular expressions (PCRE), which you would you preg_match as the function.

/^{(([A-Za-z]),\s?)?([A-Za-z]*)$/}

The first / just means we are starting. The ^ means start at the beginning of the string. Then, we have any character and number of times, the is a greedy operator that just matches anything before it any number of times. Then we have a comma, and a space (maybe) the question mark means we can have a space or we can't, it doesn't matter. The next means match the previous any number of times. Then, match the last work and the final $ means match the end of the string. So if you do:

if (preg_match("/^{(([A-Za-z]),\s?)?([A-Za-z]*)$/",} $string)) {
echo "OK";
} else {
echo "Bad";
}

You should be ok. Hope that helps!

Chris King

Anon

Thanks
I'll check it out in the a.m. (late now), I've had some success matching a word followed by a comma and a space but much trouble letting it repeat that whole sequence however many times. I figuered I could use () to encapsulate the whole thing but it wouldn't work!
I'll check out your stuff tommorow aswell as the preg_match() function

Cheers

Nick

Anon

This is a bit more simple as it makes use of the eregi function(case insensitive):

eregi("^(([a-z], )+(([a-z])+))$", $string);

It just checks that the first part (string, ) matches at least once and is followed by the other part (string).

knickerlas

Hi
Thanks Stephen, that works just great!
Would you mind 'dissecting' the process for me so that I fully understand it? I know the basic ereg() stuff so I don't need an explanation of how the operators work but I did try this with () brackets because I figured it was the eay forward and failed horribly.
Could you explain how it's acomplished?

Cheers

Nick

Anon

No problem!
The first char ^ marks the start of the string. It is followed by the first 'Atom' definition (an Atom defines one part of a string):
([a-z], )+
The first part of the Atom ([a-z] states that the first part of the string must be any letter between a-z and can occur any number of times. It is followed by the comma and space to say that the character(s) must be followed by this. The + symbol after the Atom means that this Atom must occur at least once.
This is followed by the second Atom:
(([a-z])+)
which does the same thing but does not need to be followed by a comma and a space. I don't now why I used double braces this:
([a-z])+
should work just fine.
This is followed by the end-of-string marker $ and there is your definition.
I hope this makes sense.

Anon

After looking at your message again I realised that I should also tell you that the 'trick' to designing ereg() definitions is to create the definition Atom by Atom. In this case create an Atom to define the (string, ) part first and so on. It is important to remember that the (, +, ?) operators can not be placed in sucsession (+?) but can be used on ANY part of the ereg() definition. So you could difine a string such as this (string,,,,, ) like so: ([a-z], ).
I you have any more questions don't hesitate to contact me.

knickerlas

Great,
THanks for that, it was the double braces that threw me.
I think I understand it all much better now I know about the 'atom' bit.

Thanks again

Nick