Yeah, but it has some negative things too. The replacement of word 1 has to have the id 1.Vox Populi wrote:Ahhh, I see. Looks great. So a person can add to this easily. The array takes away the load... Good thought again.
Brocolli[c] becomesBroccoli[/c] without having Broccoli or Brocolli defined in the array. You'll only need to add the uncapitalized words.
Code: Select all
//Anti spam start
function message_validate_spam($message)
{
if(isset($message) && $message != '')
{
$corr = array(
1 => 'accidentally',
2 => 'accordion',
3 => 'acquaintance',
4 => 'acquaintance',
5 => 'aquire',
6 => 'acquit',
7 => 'a lot',
8 => 'a lot',
9 => 'argument',
10 => 'a while',
11 => 'axle',
12 => 'barbeque',
13 => 'believable',
14 => 'broccoli',
15 => 'cemetary',
16 => 'chauvinism',
17 => 'chocolaty',
18 => 'chauvinism',
19 => 'colosseum',
20 => 'collectible',
21 => 'definite',
22 => 'development',
23 => 'drunkenness',
24 => 'dumbbell',
25 => 'exercise',
);
$wrong = array(
1 => 'accidently',
2 => 'accordian',
3 => 'acquaintence',
4 => 'aquaintance',
5 => 'acquire',
6 => 'aquit',
7 => 'alot',
8 => 'allot',
9 => 'arguement',
10 => 'awhile',
11 => 'axel',
12 => 'barbeque',
13 => 'believeable',
14 => 'brocolli',
15 => 'cemetery',
16 => 'chauvanism',
17 => 'chocolatey',
18 => 'chauvanism',
19 => 'colliseum',
20 => 'collectable',
21 => 'definate',
22 => 'developement',
23 => 'drunkeness',
24 => 'dumbell',
25 => 'excercise',
);
$l['corr'] = count($wrong);
$l['wrong'] = count($corr);
if($l['corr'] == $l['wrong'])
{
while($l['corr'] > 0)
{
while(strstr($message,$wrong[$l['corr']]) !== false)
{
$message = str_replace($wrong[$l['corr']], $corr[$l['corr']], $message);
}
while(strstr($message, ucwords($wrong[$l['corr']])) !== false)
{
$message = str_replace(ucwords($wrong[$l['corr']]), ucwords($corr[$l['corr']]), $message);
}
$l['corr'] = $l['corr'] - 1;
}
return $message;
}
else
{
return $message;
}
}
else
{
return $message;
}
} Code: Select all
Removed because a newer version is available. Code: Select all
//Spelling corrector START
function message_validate_spam($message)
{
if(isset($message) && $message != '')
{
$corr = array(
'accidentally', // 0
'accordion',
'acquaintance',
'acquaintance',
'aquire',
'acquit', // 5
'a lot',
'a lot',
'argument',
'a while',
'axle', // 10
'barbeque',
'believable',
'broccoli',
'cemetary',
'chauvinism', // 15
'chocolaty',
'colosseum',
'collectible',
'definite', // 20
'development',
'drunkenness',
'dumbbell',
'exercise',
'existence', // 25
'fiery',
'flabbergast',
'flotation',
'frustum',
'genius', //30
'grammar',
'handkerchief',
'harass',
'inadvertent',
'incidentally', //35
'indispensable',
);
$wrong = array(
'accidently', // 1
'accordian',
'acquaintence',
'aquaintance',
'acquire',
'aquit', // 5
'alot',
'allot',
'arguement',
'awhile',
'axel', // 10
'barbeque',
'believeable',
'brocolli',
'cemetery',
'chauvanism', // 15
'chocolatey',
'colliseum',
'collectable',
'definate', // 20
'developement',
'drunkeness',
'dumbell',
'excercise',
'existance', // 25
'firey',
'flabberghast',
'floatation',
'frustrum',
'genious', // 30
'grammer',
'hankerchief',
'harrass',
'inadvertant',
'incidently', // 35
'indispensible',
);
$l['corr'] = count($wrong) - 1;
$l['wrong'] = count($corr) - 1;
$typos = 0;
if($l['corr'] == $l['wrong'])
{
while($l['corr'] >= 0&& $typos <= 20)
{
while(strstr($message,$wrong[$l['corr']]) !== false)
{
$message = str_replace($wrong[$l['corr']], $corr[$l['corr']], $message);
}
while(strstr($message, ucwords($wrong[$l['corr']])) !== false)
{
$message = str_replace(ucwords($wrong[$l['corr']]), ucwords($corr[$l['corr']]), $message);
}
$l['corr'] = $l['corr'] - 1;
$typos = $typos + 1;
}
return $message;
}
else
{
return $message;
}
}
else
{
return $message;
}
}
//Spelling corrector END How more spelling errors, how more server load. I've to say, 20 words isn't a big server load time yet, so you can change this if you want.Vox Populi wrote:I have not tested it out as of yet, but what is the server load? Is it directly related to how many items in the array or in the misspelled words when submitting.
How about on submit "You have misspelled words. Would you like me to correct them?" Yes / No...
Naw, that won't work either. It still has to read before asking the question.
It's not reading that take serverload, but it's the replacing.How about on submit "You have misspelled words. Would you like me to correct them?" Yes / No...