regular expression for dummies

[sem_db]

We are trying to match a string beginning with

<?xml

and ending with

</OAI-PMH>

We don't care about anything between these start and end points.

[Weedpacket]

Regular expressions? Overkill

if([man]substr/man=='<?xml' && substr($string,-10)=='</OAI-PMH>')

[sem_db]

I am sorry,

We are in fact, trying to extract the string which is bounded by the beginning (<?xml) and end (</OAI-PMH) marks. We have been looking for a regular expression to perform this task.

Thanks for the post!

[jblack]

Try:

if(substr($string,0,5)=='<?xml' && substr($string,-10)=='</OAI-PMH>')
	{
	$stringbetweentags = substr($string, 5, strlen($string) - 15);
	}

I haven't tested this, but I think you can get the drift on what I'm doing and make change accordingly. Regular expressions still aren't need.

[sem_db]

We explained this a bit wrong, again🙁 What we really have is a really long string like:

xxxxxxxxxxxxxx<?xmlyyyyyyyyyyyyyyy</OAI-PMH>xxxxxxxxxxxxxxxxxxxxxxxx

*where x and y can be any character, including newline.

we want what's in blue.

I don't know if we can use substr because we don't know how many x's come before and after or how many y's are in between.

Thanks a lot everyone!

[sem_db]

The regular expression I found that worked is:

preg_match("/<\?xml[\d|\D|\w|\W|\s|\S]+<\/OAI-PMH>/", $this->response, $reg_array)

[xblue]

wow, were did you find that? [\d|\D|\w|\W|\s|\S] seem one of the most redundant ways to say 'just any character' 😉

a much simpler way is replacing this with a dot and setting the s modifier to make the dot match newlines as well.