[RESOLVED] Help interpreting this Function?

Rivet

Hi,

I know what the following code does, but I don't understand the code specifically. Would anyone care to give me a blow-by-blow English interpretation of just what it's doing, please?

function checkOK($field)
{
if (eregi("\r",$field) || eregi("\n",$field)){
die("Invalid Input!");
}
}

TIA

Rivet`

dalecosp

// this is a generic function for checking input.  Probably not a very good one, though.

function checkOK($field) {   

   // if a character-return *type* character appears ("\r" as carriage return, "\n" as newline)
   if (eregi("\r",$field) || eregi("\n",$field)) {  

  die("Invalid Input!");    //halt execution and print this string.
   } 
}

IIRC, the ereg() family of functions is deprecated in PHP now.

And die() isn't considered "graceful" for handling errors.

Hope this helps,

NogDog

A quick reworking of the if condition:

if (preg_match('#[\r\n]#',$field)) {

Rivet

dalecosp;11030491,

That helped, a LOT!

ereg is indeed deprecated as of 5.3; thanks for that.

I agree 'die() 'can be a little rude $ crude; I'm only playing with test files at the moment, so no one but me is seeing it. I'll go to a function or two eventually and am currently also using Filters, which I'm also working to learn more about. Making good progress now; thanks.

Rivet`

Rivet

Thanks; works like a charm of course!

traq

In addition to NogDog's suggestion of using [man]preg_match/man, I would suggest changing the function name also, to be more descriptive of what it actually does. (Just running this function does not necessarily make the input "okay.") Maybe something like [font=monospace]checkNewlines()[/font].

NogDog

Yeah, ultimately you probably want it to return a Boolean, then let the calling code decide what to do based on the result. You can then include it with other validation functions in an include file, or maybe even create a validation class (perhaps full of stand-alone static functions that).

function hasNewline($text)
{
    return preg_match('#[\r\n]#', $text);
}

app code:

if(hasNewline("Foo\nBar")) {
    echo "<p class='error'>Hey! Don't do that!</p>";
    exit;
}

Rivet

🆒

traq;11030515 wrote:
In addition to NogDog's suggestion of using [man]preg_match/man, I would suggest changing the function name also, to be more descriptive of what it actually does. (Just running this function does not necessarily make the input "okay.") Maybe something like [font=monospace]checkNewlines()[/font].

Agreed; I always try to use names that are meaningful to me so not a problem. I do appreciate the followup! clarification is always good to have.

Rivet`

Weedpacket

Another quick rework of the condition:

 strpos($field, "\n") || strpos($field, "\r")

Rivet

Thanks; looking into pregmatch, pregreplace, etc. has made a few things quite a bit easier. I have a couple instances where the regular Filters aren't quite what I want and this will fit the bill well, I believe. I have a couple instances where it just seems easier to code what I want to "keep" and just eliminate everything else; sort of a whitelist approach. Allow only what I want and nothing else.
At least that's the direction I'm thinking in right now 🙂. I might change my mind when I actually start to test such code, who knows? lol At least it's fun to check such things out.

Cheers,

Rivet`

Rivet

Weedpacket;11030549 wrote:
Another quick rework of the condition:
 strpos($field, "\n") || strpos($field, "\r")

At first the use wasn't clear to me, but apparently the point is to use to determine if they are NOT present, and then react accordingly, right?

Sounds like a case for:
if (strpos($str, $substr) === false) {
OR if (strpos($str, $substr) !== false)
, right?
It didn't take me long to find the "gotchas" with it; like "if (strpos($str, $substr) == false) {" lol!

If I'm right, that's a decent example of thinking "outside the box", eh? I'd have never thought of that approach.

Regards,

Rivet`