string manipulation -get all text between ""

php_noob2

If I have a bunch of text like so and want to get all of what is inside the quotes, how do I do that... example please?

{i:0;s:45:"http://localhost/~test/test.php?page=9";i:1;s:46:"http://localhost/~test/fun.php?page=11";i:2;s:45:"http://localhost/~test/grass.php?page=8";i:3;s:45:"http://localhost/~test/tree.php?page=7";i:4;s:45:"http://localhost/~test/log.php?page=8";}

to get only http://localhost/~test/test.php?page=9

It would also be helpful to have ; as a separator but maybe I should be leaning towards unserialize than string manipulation... 😕

thanks!!

nrg_alpha

Here is a quick and dirty way I quickly came up with:

$x = '{i:0;s:45:"http://localhost/~test/test.php?page=9";i:1;s:46:"http://localhost/~test/fun.php?page=11";i:2;s:45:"http://localhost/~test/grass.php?page=8";i:3;s:45:"http://localhost/~test/tree.php?page=7";i:4;s:45:"http://localhost/~test/log.php?page=8";}';

$stringArray = explode('"', $x);
for($i=0; $i < count($stringArray); ++$i){
	if(ereg('http', $stringArray[$i])){
        $stringArray[$i] .= ';';
		echo $stringArray[$i] . '<br />';
	}
}

But something tells me there is probably some sleeker preg expression that can handle this sort of thing better.. but in ancy case, it extracts everything in the double quotes, and due to the additional ereg expression inside, only the array keys with 'http' in its values will show up.

Hope this helps.

Cheers,

NRG

NogDog

<?php
$testText = '{i:0;s:45:"http://localhost/~test/test.php?page=9";i:1;s:46:"http://localhost/~test/fun.php?page=11";i:2;s:45:"http://localhost/~test/grass.php?page=8";i:3;s:45:"http://localhost/~test/tree.php?page=7";i:4;s:45:"http://localhost/~test/log.php?page=8";}';
preg_match_all('/(?<=:")[^"]+(?=")/', $testText, $matches);
echo "<pre>" . print_r($matches[0], 1) . "</pre>\n";

Output:

Array
(
    [0] => http://localhost/~test/test.php?page=9
    [1] => http://localhost/~test/fun.php?page=11
    [2] => http://localhost/~test/grass.php?page=8
    [3] => http://localhost/~test/tree.php?page=7
    [4] => http://localhost/~test/log.php?page=8
)

bradgrafelman

This is only a partial copy of a serialized array. What happened to the rest of the serialized string? It'd make sense to just keep it all intact, then you wouldn't have to mess around with funky regexp's and whatnot.
If, for whatever reason, you can't fix this error, I suppose you could do a simple preg_match_all() here:

$x = '{i:0;s:45:"http://localhost/~test/test.php?page=9";i:1;s:46:"http://localhost/~test/fun.php?page=11";i:2;s:45:"http://localhost/~test/grass.php?page=8";i:3;s:45:"http://localhost/~test/tree.php?page=7";i:4;s:45:"http://localhost/~test/log.php?page=8";}';

preg_match_all('/\"(.*?)\"/', $x, $matches);

foreach($matches[1] as $string)
	echo $string . "<br>";

EDIT: And of course NogDog beats me to it with a more robust regexp pattern. Still, I argue that you should figure out why this serialized data wasn't kept intact so that you could simply unserialize it. :p

nrg_alpha

Nog.. one of my bigger weaknesses (aside from not knowing how to code my way out of a paper bag) is regular / perl expressions..

I look at your snippet of code:

preg_match_all('/(?<=:")[^"]+(?=")/', $testText, $matches);

And I find some difficulty dicyphering what the perl expression is saying..
I suppose the brackets denote groups?

what exactly does ?<=:" imply?

then we have [^"]+ which I assume simply implys anything that is NOT a double quote one or more times?

then finally a third group? (?=") meaning equal sign is optional before the closing double quote?

Sorry..I'm slightly lost.. any help in explaining this would be greatly appreciated.
cheers,

NRG

P.S I knew there was a more elegant solution than mine involving preg! :rolleyes:

NogDog

(?=abc) is a look-ahead assertion. (?<=def) is a look-behind assertion. See http://www.php.net/manual/en/regexp.reference.php#regexp.reference.assertions for info on what the heck that means.

nrg_alpha

QUOTE=NogDog is a look-ahead assertion. (?<=def) is a look-behind assertion. See http://www.php.net/manual/en/regexp.reference.php#regexp.reference.assertions for info on what the heck that means.[/QUOTE]

Thanks for the link, Nog. As well as reading that, I have found a 'cheat sheet':
http://php-regex.blogspot.com/2008/01/php-regex-cheat-sheet.html

This will take a while to digest.. but I'm slowly understanding it a little more. I find that while the 'syntax' for expressions is not necessarily hard, the 'rules of engagement' are rather tricky. I'll keep looking this stuff over till I really become comfortable understanding it all.

Cheers,

NRG

NogDog

It is, essentially, it's own little programming language (and a very terse one).

php_noob2

Wow, thanks so much everyone...

bradgrafelman, yes that was just a portion of the entire thing. Basically it's just the full session stored in a second table apart from the sessions table in the database -one with no garbage collector, etc.

So from the 'session_data' field in table 'users' I have:

userID|s:2:"31";level|s:1:"1";advisor|s:1:"1";active|i:1;page_history|a:5:{i:0;s:45:"http://localhost/~test/test.php?page=9";i:1;s:46:"http://localhost/~test/fun.php?page=11";i:2;s:45:"http://localhost/~test/grass.php?page=8";i:3;s:45:"http://localhost/~test/tree.php?page=7";i:4;s:45:"http://localhost/~test/log.php?page=8";}

My end goal is to have nice and pretty url's from this so maybe there's a better way than my primordial, caveman php skills are leaning towards. I've been trying to recreate the array and then unserialize but no luck -can't get it unserialized at all. Then I found how to get all the string past "page_history" with strstr and then tried and just get the values between the quotes.

So I am hobbling along ...anyone have a better suggestion for what I'm tryin' to do?

Thanks so much!!! 😃

bradgrafelman

Well if you're storing the entire string, why not just [man]unserialize/man it so you can work with normal variables?

php_noob2

My last post: "I've been trying to recreate the array and then unserialize but no luck -can't get it unserialized at all. "

Weedpacket

The session contains more than just one serialised variable (it contains all of the session variables and their names).

userID The userID session variable
| a separator
s:2:"31" the serialised value of $_SESSION['userID']
; another separator
level the level session variable
...

There's no doubt some clever parsing method using regular expressions in combination with explode or maybe strtok(), but for now, you might want to look at session_decode and the user notes in particular.