XCVI. Cha�ne de caract�res

I had the problem that in some of the guestbooks I programmed, people
entererd trash like 400 characters without spaces and such.  Without any
spaces there was no way to break the lines and the whole guestbooked
looked screwed.  I wrote this little function that makes sure no words are
more than a specified number of characters long.  If a word is too long,
spaces are inserted to get the word to the maximum length allowed. Maybe
it's useful to some of you:

function strmaxwordlen($input,$len) {
 $l = 0;
 $output = "";
 for ($i = 0; $i < strlen($input); $i++) {
  $char = substr($input,$i,1);
  if ($char != " ") { $l++; } else { $l = 0; }
  if ($l == $len) { $l = 0; $output .= " "; }
  $output .= $char;
 }
 return($output);
}

Regarding MS-Dos style paths, and the escaping problems thereof:  Don't use
backslashes.

Train yourself to use forward slashes.  They actually work just as well on
Win32, and your code will be quickly portable to UNIX.

"c:/WinNT/System32" is a valid path. 
 Mind you, I'm not sure
what the various web servers will do to these strings...

For a MySQL problem (where a user might put quotes in a text field from a
form) use addslashes() to escape all quotes (single and double).  For
example:

// Assuming 'textstuff' is from a form
$mytextstuff = addslashes($textstuff);
$sql = "insert into MyTable set MyField='$mytextstuff'";

Also if you are working with html forms and/or text fields from databases
a lot you'll probably want to read up on commands like htmlspecialchars()
as well (a field like "foo<bar>" might not throw MySQL off
but it won't display correctly in an html page).

// Makes all words starting with  clickable
function urls_clickable($string) {
	for($n=0; $n < strlen($string); $n++)
	{
		if(strtolower($string[$n]) == 'h') {
			if(!strcmp("
strtolower($string[$n]) . strtolower($string[$n+1]) .
strtolower($string[$n+2]) . strtolower($string[$n+3]) . $string[$n+4] .
$string[$n+5] . $string[$n+6])) {
				$startpos = $n;
				while($n < strlen($string) &&
eregi("[a-z0-9\.\:\?\/\~\-\_\&\=\%\+\'\"]",
$string[$n])) $n++;
				if(!eregi("[a-z0-9]", $string[$n-1])) $n--;
				$link = substr($string, $startpos, ($n-$startpos));
				$link = $link;
				$string_tmp = $string;
				$string = substr($string_tmp, 0, $startpos);
				$string .= "<a
href=\"$link\">$link</a>";
				$string .= substr($string_tmp, $n, strlen($string_tmp));
				$n = $n + 15;
			}
		}
	}
	return $string;
}

In order to properly display text from a form into an html page you need
code
like this:

<? echo htmlentities(stripslashes($form_text)); ?>

Here is a much better more thought out ucsetence() function. I thought
about it more and it know understands .,?! and all other punctuations and
allows for making any word or character uppercase. Just put words to be
upper case in array $uppercase. Fixes common user email problems like
CAPS, !!!, and i.

Ex:
$paragraph="I Am AN ItiOT!!! I cAN Not TYPe. WHat IS WRONG with ME???
ME (i) do NOT knOW....";

$newParagraph=ucsentence($paragraph);

$newParagraph now is:
I am an itiot! I can not type. What is wrong with me? Me (I) do not know.

function ucsentence($paragraph,$uppercase=0) {
  //default capitol words
  $uppercase[]="I";

  //make sentences from $paragraph
  $paragraph=strtolower(trim($paragraph));

  //replace duplicate chars
  $paragraph=ereg_replace('[ ]+'," ",$paragraph);
  $paragraph=ereg_replace('[.]+',".",$paragraph);
  $paragraph=ereg_replace('[?]+',"?",$paragraph);
  $paragraph=ereg_replace('[!]+',"!",$paragraph);
 
  $words=explode(" ",$paragraph);
  
  FOR ($i=0;$i<count($words);$i++) {
    //fix sentence capitolization
    IF (strstr($words[$i],".")) {
      $words[$i+1]=ucfirst($words[$i+1]);
      }

    IF (strstr($words[$i],"!")) {
      $words[$i+1]=ucfirst($words[$i+1]);
      }

    IF (strstr($words[$i],"?")) {
      $words[$i+1]=ucfirst($words[$i+1]);
      }

    //make array $uppercase uppercase
    IF ($uppercase) {
      IF
(in_array(strtoupper(ereg_replace('[!@#\$%^&*()_+-=,.?\'\";:]',"",$words[$i])),$uppercase))
{
        $words[$i]=strtoupper($words[$i]);
        }
      }
    }

    $new_paragraph=ucfirst(trim(implode(" ",$words)));

  return $new_paragraph;
} //--end ucsentence--

I wanted to convert numbers to an adjective form (so from "1" to
"1st", from "352" to "352nd", etc.), but
couldn't find it in PHP, so I wrote this little code to do that for me.
Hope you find it useful:

function nth($in) {
  if (!is_long($in))
    return(-1);
  switch ($in) {
    case 0 : return("0th");
    case 1 : return("1st");
    case 2 : return("2nd");
    case 3 : return("3rd");
    case 4 :
    case 5 :
    case 6 :
    case 7 :
    case 8 :
    case 9 :
    case 10 :
    case 11 :
    case 12 :
    case 13 :
    case 14 :
    case 15 :
    case 16 :
    case 17 :
    case 18 :
    case 19 : return($in."th");
  }
  switch (substr($in, -1, 1)) {
    case 0 : return($in."th");
    case 1 : return($in."st");
    case 2 : return($in."nd");
    case 3 : return($in."rd");
    case 4 :
    case 5 :
    case 6 :
    case 7 :
    case 8 :
    case 9 : return($in."th");
    default : return(0);
  }
}

It will return -1 when the input is not an integer and 0 when it couldn't
find a match (although that shouldn't happen).

Here's some of my code for inserting spaces so people can't screw the page,
I have not tested it, but it's cleaner than other's.

	list ($words) = array (split (" ", $comment));
	$comment = '';
	foreach ($words as $c => $word) {
		if (strlen ($word) > 60 and !ereg ("[\[|\]|\/\/]", $word))
			$word = chunk_split ($word, 60);
		if ($c)
			$comment .= ' ';
		$comment .= $word;
	}

Here is a very slight fix to this function to make URLs in strings
clickable.  The change allows URLs that have a dash character
"-" to not stop the href tag just before the "-".

The change un-literals the "-" in the eregi function since the
eregi function appears to not want this when you are trying to match an
actual "-".  Seems that different languages have different
requirements about what should be literal.

This:
eregi("[a-z0-9\.\:\?\/\~\-\_\&\=\%\+\'\"]",
$string[$n])

becomes this:
eregi("[a-z0-9\.\:\?\/\~\_\&\=\%\+\'\"-]",
$string[$n])

// Makes all words starting with  clickable 
function urls_clickable($string) { 
for($n=0; $n < strlen($string); $n++) 
{ 
if(strtolower($string[$n]) == 'h') { 
if(!strcmp("
strtolower($string[$n]) . strtolower($string[$n+1]) .
strtolower($string[$n+2]) . strtolower($string[$n+3]) . $string[$n+4] .
$string[$n+5] . $string[$n+6])) { 
$startpos = $n; 
while($n < strlen($string) &&
eregi("[a-z0-9\.\:\?\/\~\_\&\=\%\+\'\"-]",
$string[$n])) $n++; 
if(!eregi("[a-z0-9]", $string[$n-1])) $n--; 
$link = substr($string, $startpos, ($n-$startpos)); 
$link = $link; 
$string_tmp = $string; 
$string = substr($string_tmp, 0, $startpos); 
$string .= "<a href=\"$link\">$link</a>";

$string .= substr($string_tmp, $n, strlen($string_tmp)); 
$n = $n + 15; 
} 
} 
} 
return $string; 
}

Minor correction to my HTMLEncode function.

function HTMLEncode($text)
{
  $searcharray =  array(
   "'([-_\w\d.]+@[-_\w\d.]+)'",
   "'((?:(?!://).{3}|^.{0,2}))(www\.[-\d\w\.\/]+)'", 
   "'(http[s]?:\/\/[-_~\w\d\.\/]+)'");  
  $replacearray = array(
   "<a href=\">\\1</a>",
   "\\1
   "<a href=\"\\1\">\\1</a>");
  return nl2br(preg_replace($searcharray, $replacearray,
stripslashes($text) ));
}

To shorten a string to a certain number of characters, replacing surplus
chars with a shortening symbol such as "[...]", use the
following simple function:

========= CUT ==========

function shorten_string($text, $length, $symbol = "[...]")
{
  $length_text = strlen($text);
  $length_symbol = strlen($symbol);

  if($length_text <= $length || $length_text <= $length_symbol ||
$length <= $length_symbol)
    return($text);
  else
    return(substr($text, 0, $length - $length_symbol) . $symbol);
}

========= CUT ==========

The parameter $text ist the string to be shortened. $length is the maximum
number of characters in the new string. $symbol is the shortening symbol
(default is "[...]"). If shortening would make no sense, the
original string is returned.

urls_clickable seems to miss the last character from any hyperlink which
ends with a /. This can be a problem for a link to certain web servers
which need a terminating slash to indicate that the last section refers to
a directory.

e.g. 

Here's the altered function:

// Makes all words starting with  clickable 
function urls_clickable($string) {
	for($n=0; $n < strlen($string); $n++) { 
		if(strtolower($string[$n]) == 'h') { 
			if(!strcmp("
strtolower($string[$n]) . strtolower($string[$n+1]) .
strtolower($string[$n+2]) . strtolower($string[$n+3]) . $string[$n+4] .
$string[$n+5] . $string[$n+6])) { 
				$startpos = $n; 
				while($n < strlen($string) &&
eregi("[a-z0-9\.\:\?\/\~\_\&\=\%\+\'\"-]",
$string[$n])) $n++; 
					if(!eregi("[a-z0-9]", $string[$n-1])) $n--; 
					$link = substr($string, $startpos, ($n-$startpos+1)); 
					$link = $link; 
					$string_tmp = $string; 
					$string = substr($string_tmp, 0, $startpos); 
					$string .= "<a
href=\"$link\">$link</a>"; 
					$string .= substr($string_tmp, $n+1, strlen($string_tmp)); 
					$n = $n + 15; 
					} 
			} 
		} 
	return $string; 
}

// lines with a additional +1 are:
// the $link = substr(...
// and
// the $string .= substr(...
// lines

Browsers (Internet Explorer at least) won't wrap text immediately before
puntuation . ? ! ( ) [ ] { } / even if the preceeding character is
whitespace.  I believe this is intentional widow/orphan prevention
behavior.  This complicates the processing of user-inputed text (e.g. the
message board junk problem).  Even if you use <textarea wrap=hard>
to force newlines into the input text (which is imperfect since a clever
user can just modify the HTML before posting), people posting long strings
of punctuation characters, or just punctuation characters at the beginning
of each line in the <textarea> will mess up the format of the page
for users viewing their post.

You obviously can't solve this by just inserting a spaces - the browser
still won't wrap.  My solution is to use str_replace to put a &nbsp;
(the least visually impacting non-whitespace character I can think of)
between the \n and any pucntuation so the wrap can happen in the
whitespace before the &nbsp;.  Kindof ironic to be using &nbsp; to
*cause* a break:)

I have written an *evil* little function that will cut a string on a
character given. Features:
1) Minimum length.
2) If a string doesn't contain the character, it cuts your string to
minimum length.
3) If the next character beyond the minimumlength equals the given
character, it will cut the string to minimulength as well.
Code:

<?php
function cuttext($tring,$cuton) {

/* Define the character to cut on */
/* One could make an array with all characters to cut on */
/* But I wouldn't know how to :) */

$space=" ";

if (!strstr($tring,$space)) {

	/* No space is found in the whole string: cut at 20th character */
	
	$tring=substr($tring,0,$cuton);
}
if (substr($tring,$cuton,1)==$space) {

	/* 21st character is a space: cut at 20th character */
	
	$tring=substr($tring,0,$cuton);
} else {

	/* 21st Character is NOT a space: return to last space and cut there */
	
	while ($teller <= $cuton) {
		if (substr($tring,$cuton-$teller,1)==$space) {
			$tring=substr($tring,0,$cuton-$teller);
			break;
		}
		$teller++;
	}
}
return $tring;
}
echo(cuttext("waf groink muha zz gruu wrrreea iii gargghhh
>:)>:)",20));
?>

there's a problem when i used the addslashes function for my odbc
connection... in order to fix it, I used the htmlspecialchars($str_var,
ENT_QUOTES); instead and it worked fine.  example is as follow:

// $str - a string w/ single quotes
$str = htmlspecialchars($str, ENT_QUOTES);
.
.
.
.
$sql = "<something> $str <something>";
<etc.>

while we're waiting that ctype functions get out from experimental phase,
look at this function.
It check if a string contain only alphanumeric characters (a-zA-Z0-9)

function isalphanumstr($string) {
          for($i = 0; $i < strlen($string); $i++)
              if (!in_array($string[$i],
array("q","w","e","r","t","y","u","i","o","p","l","k","j","h",
"g","f","d","s","a","z","x","c","v","b","n","m","Q","W",
"E","R","T","Y","U","I","O","P","L","K","J","H","G","F",
"D","S","A","Z","X","C","V","B","N","M","0","1","2","3",
"4","5","6","7","8","9")))
{
                  return false;
              } else {
                  return true;
              }
      }

I've ported a function to PHP that will find the longest common substring
in two different strings.  Here it is:

function LongestCommonSubstring($string1, $string2)
{
	// find length of longest common substring first
	$L = array();
	$length1 = strlen($string1);
	$length2 = strlen($string2);
		
	for ($i = $length1; $i >= 0; $i--)
	{
		for ($j = $length2; $j >= 0; $j--)
		{
			if ($string1[$i] == '' || $string2[$j] == '') $L[$i][$j] = 0;
			elseif ($string1[$i] == $string2[$j]) $L[$i][$j] =
			       1 + $L[$i + 1][$j + 1];
			else $L[$i][$j] = max($L[$i + 1][$j], $L[$i][$j + 1]);
		}
	}

	// then find the actual substring
	$substring = '';
	$i = 0;
	$j = 0;
    
	while ($i < $length1 && $j < $length2)
	{
		if ($string1[$i] == $string2[$j])
		{
			$substring .= $string1[$i];
			$i++; $j++;
		}
		elseif ($L[$i + 1][$j] >= $L[$i][$j + 1]) $i++;
		else $j++;
	}
	
	return $substring;
}

Here is where I found the original:

I found a nice replacement vor the "swap" command, which used to
swap two variables...

it even works with more than one element and I used it to change the
bgcolor of a <table> each row (makes reading more comfortable)

first you define $color:
$color= array('#0066CC', '#0099CC');
this piece of code just gives out another color, each time you call it
$tmp = $color[0];
$col = array_shift ($color);
$color[] = $tmp;

Much faster and much better way to check does string consists
from alphanumeric chars 'only' ...

By the way, author of previous isalphanumstr()
forgot to add whitespace to his array.

It's not an alphanumeric char,
but who would need this function without it ? :)

function isalphanumstr($string) {
    $res=true;
    for($i = 0; $i < strlen($string); $i++)
    if((ord($string[$i])<48 && ord($string[$i])!=32) ||
    (ord($string[$i])>57 && ord($string[$i])<65) ||
    (ord($string[$i])>90 && ord($string[$i])<97) ||
    ord($string[$i])>122) {$res=false; break;}
return $res;
}

echo isalphanumstr(" 1234567890qwertyuiopasdfghjklzxcvbnm
QWERTYUIOPASDFGHJKLZXCVBNM");

//... returns "1" //;)

This is a more compact function to get a numeric adjective.  No idea if
it's faster than the other "nth" function here (though probably
is, since "th" is more likely to occur), but the code is
definitely shorter.

Assumes integer input.  Returns a string.

function getNth ($n) {
  $qn = (int) (($n%100) / 10);
  $rn = $n % 10;

  $suffix = "th";

  if ($qn != 1) {
    switch ($rn) {
    case 1: $suffix = "st"; break;
    case 2: $suffix = "nd"; break;
    case 3: $suffix = "rd"; break;
    }
  }

  return $n . $suffix;
}

One could make an ever quicker isalphanumeric() function by using regular
expressions:

function isalphanumeric($test) {
    return !(preg_match("/[^a-z,A-Z,0-9 ]/", $test));
}

Note: that is a space after the 9, in order to inclue spaces in the
comparison.  One could also add \t and \n so that it did not fail out if
it found a tab or newline.  This is much cleaner and more efficient than
the array approach described earlier, and nicer than the ordinal
characters approach as well.

Earlier I posted a LongestCommonSubstring function.  The function does not
find the long common CONTIGUOUS substring, so I've modified it to do that,
in case anyone needs it:

function LongestCommonContiguousSubstring($string1, $string2)
{
	// find length of longest common substring first
	$L = array();
	$length1 = strlen($string1);
	$length2 = strlen($string2);
		
	for ($i = $length1; $i >= 0; $i--)
	{
		for ($j = $length2; $j >= 0; $j--)
		{
			if ($string1[$i] == '' || $string2[$j] == '') $L[$i][$j] = 0;
			elseif ($string1[$i] == $string2[$j]) $L[$i][$j] = 1 + $L[$i+1][$j+1];
			else $L[$i][$j] = max($L[$i + 1][$j], $L[$i][$j + 1]);
		}
	}

	// then find the actual substring
	$substrings = array();
	$i = 0;
	$j = 0;
	$s = 0;
    
	while ($i < $length1 && $j < $length2)
	{
		if ($string1[$i] == $string2[$j])
		{
			$substrings[$s] .= $string1[$i];
			$i++; $j++;
			continue;
		}
		elseif ($L[$i + 1][$j] >= $L[$i][$j + 1]) $i++;
		else $j++;
		$s++;
	}
	
	$maxlength = 0;
	while (list($key, $val) = each($substrings))
	{
		$length = strlen($val);
		if ($length > $maxlength)
		{
			$maxlength = $length;
			$winner = $key;
		}
	}
	
	return $substrings[$winner];
}