Clean text in php.

Question

Hi, This is a string:

--0-1946616131-1282798399=:21360 Content-Type: text/plain; charset=us-ascii --------------
------ do not change ---------------------------- Ticket ID : #987336 --------------------
------------------------------------------- Hello, This is my problem try to solve this 
thank u --0-1946616131-1282798399=:21360 Content-Type: text/html; charset=us-ascii"

Now I want to remove -

--0-1946616131-1282798399=:21360 Content-Type: text/plain; charset=us-ascii

and

--0-1946616131-1282798399=:21360 Content-Type: text/html; charset=us-ascii

section from it. I mean clean the text.

How can I do that?

Answer 1

You could do two a regular expression, or you could try a couple splits. Here's the second option:

//the original string
$string = "--0-1946616131-1282798399=:21360 Content-Type: text/plain; charset=us-ascii -------------------- do not change ---------------------------- Ticket ID : #987336 --------------------------------------------------------------- Hello, This is my problem try to solve this thank u --0-1946616131-1282798399=:21360 Content-Type: text/html; charset=us-ascii";
//split the string into lines separated by --0-
$splitstring = explode("--0-",$string);
print "<pre>";
print_r($splitstring);
print "</pre>";
//create an array that will be our final clean strings
$cleanstrings = array();
//go through each of our lines
foreach($splitstring as $line){
    //if it has content
    if (strlen($line)>0) {
        //then split it again to get rid of the junk sections
        $splitline = explode("charset=us-ascii",$line);
        //if the second part of the string has content
        if (strlen($splitline[1])>0) {
            //then add it to our list of clean strings
            $cleanstrings[] = $splitline[1];
        }
    }
}
print "<pre>";
print_r($cleanstrings);
print "</pre>";

Answer 2

Use this simple one line code (where $text is the input text):

$newtext = str_replace('--0-1946616131-1282798399=:21360 Content-Type: text/plain; charset=us-ascii', '', $text);

Answer 3

Please clarify whether this string variates and how or is always the same?

Also it seems that you are doing something wrong in the first place to get this string. Or do you have no control over the incoming string?

Functions to look at: str_replace, preg_replace and explode

Answer 4

This seems like part of a MIME multipart message. If that's the case, the parts you are looking to remove are unpredictable.

The break between different parts should be specified in the message header like so:

Content-Type: multipart/mixed; boundary="frontier"

boundary="frontier" means that every new part of the message will be introduced by something like this:

--frontier
Content-Type: text/plain

Since the sender of the message is completely free to choose any text he likes for the boundaries, they're unpredictable without looking at the message header. Unless you have a really specific case of very specific boundaries, it's almost impossible to reliably remove the boundary text after the fact. It needs to be "cleaned up" while the message is being parsed.

If you are dealing with a very limited, predictable set of boundaries, you should specify their format and try to remove them with a regular expression.