Very good.
Thank you.


On Jun 22, 2010, at 8:14 PM, Ashley Sheridan wrote:

On Tue, 2010-06-22 at 20:03 -0400, Rick Dwyer wrote:

Hello again list.

My code for stripping characters is below. I'm hoping to get feedback
as to how rock solid it will provide the desired output under any

My output must look like this (no quotes):


The code with string looks like this:

$old_string = 'ééééThis is my  & $string     -- with ƒ
lots˙˙˙of junk characters in it¡™£¢∞§¶•ªºœ∑´®†
¥¨ˆøπ“‘ååååååß∂ƒ©˙∆˚¬… æ`````````__________';

$find = '/[^a-z0-9]/i';
$replace = ' ';

$new_string = preg_replace($find, $replace, $old_string);
$new_string = preg_replace("/ {2,}/", "-", $new_string);
$new_string = preg_replace("/ {1,}/", "-", $new_string);

$new_string = rtrim($new_string, "-");
$new_string = ltrim($new_string, "-");

echo $new_string;

Will the logic above capture and remove every non alpha numeric
character and place a SINGLE hyphen between the non contiguous alpha
numeric characters?

Thanks for the help on this.


On Jun 22, 2010, at 4:52 PM, Rick Dwyer wrote:

> On Jun 22, 2010, at 1:41 PM, Ashley Sheridan wrote:
>> It is clean, but as Richard mentioned, it won't handle strings
>> outside of the traditional 128 ASCII range, so accented characters
>> and the like will be converted to an underscore. Also, spaces might
>> become an issue.
>> However, if you are happy that your input won't go beyond the a-
>> z0-9 range, then it should do what you need.
> No, actually I'm fairly confident characters outside the 128 range
> are what are causing me problems now.
> So I will try Richard's method.
> Thanks to all.
> --Rick

You can remove the second line of code, as the third one is replacing what line 2 does anyway.

Also, instead of a rtrim and ltrim, you can merge the two with a single call to trim, which will work on both ends of the string at once.


Reply via email to