preg_replace + regular expression

EasyExpat · #1 08-Jul-2003, 17:58

Hi,
I've got a flat file such as:

Quote:

<TITLE>Thursday 3rd March</TITLE><USER>Paul Smith</USER><ADDRESS>3, battersea square - SW12 3GE London - UK</ADDRESS>

I would like to retrieve the value of the parameters, i.e. have
$title='Thursday 3rd March'
$user='Paul Smith'
$address='3, battersea square - SW12 3GE London - UK'

So I tried the regular expression such as:
$title=preg_replace("'<TITLE[^>]*?>.*?</TITLE>'", "\\1", $file);

But it gives me a result such as:<USER>Paul Smith</USER><ADDRESS>3, battersea square - SW12 3GE London - UK</ADDRESS>

Anybody can help?
Thanks in advance

JdS · #2 08-Jul-2003, 23:46

The question is: why are you using preg_replace() when it isn't obvious that you want to replace anything?

Use preg_match() or preg_match_all() instead:

e.g.

PHP Code:


<?php
preg_match( '_<title>(.*)</title>_is', $file, $matches );
$title = $matches[1];
echo '$title is: '.$title;
?>

On the other hand, I wouldn't store my data like that in a text file.

EasyExpat · #3 09-Jul-2003, 13:42

Ok ok

apologize!
You are right I should use preg_match. I've never used it before so I was not familiar with.
Thanks for the solution again.

Cyril

PS: Actually, as I'm thinking about that. Can you explain me the difference between

PHP Code:


preg_match( '_<title>(.*)</title>_is', $file, $matches );

and

PHP Code:


preg_match("#<TITLE>(.{0,})</TITLE>#", $file, $matches);

?

It gives exactly the same result (second one is an example I found from another forum)

JdS · #4 09-Jul-2003, 21:40

Actually Cyril, there's no difference...

this bit:

Code:

(.{0,}) is the same as (.*)

it just ends up looking more complicated for someone starting out with REGEXs. Why he gave you that I cannot explain... perhaps he was just trying to mess with your mind?

Another is that the pattern I supplied earlier should work even if you had inadvertently stored some data like this (note the lowercase tags for 'title'):

Code:

<title>Thursday 3rd March</title><USER>Paul Smith</USER><ADDRESS>3, battersea square - SW12 3GE London - UK</ADDRESS>

that's done by the i modifier towards the end of the pattern.
i (PCRE_CASELESS)
s (PCRE_DOTALL)

You already know that the '_' and '#' characters were just pattern delimiters right?

EasyExpat · #5 10-Jul-2003, 05:36

Thanks for that. I know already for the regexp (.{0,}) being the same as (.*).

I was not aware of # or _ being pattern delimiters (I thought it was only /).

The fact is I've got sometime a text with end of line caracters (\n) in it. Your regexp can handle them nicely with no problems, the other one cannot.

But it's all good now. Thanks again for the help.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Rate This Thread
Rate This Thread:

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Microsoft Visual C++6.0 - Illegal use of this type as an expression	RealmMaster	C++ Forum	5	24-Aug-2006 14:51
Regular Expressions question.	JdS	MySQL / PHP Forum	3	04-Nov-2002 15:04

preg_replace + regular expression

Use preg_match() instead?

PCRE pattern explanation