Diese Seite bei php.net
The PHP manual text and comments are covered by the Creative Commons Attribution 3.0 License © the PHP Documentation Group - Impressum - mail("TO:Reinhard Neidl",...)
(PHP 4 >= 4.0.2, PHP 5)
pspell_check — Überprüft ein Wort
pspell_check() überprüft die Schreibweise eines Wortes.
Das überprüfte Wort
Gibt TRUE zurück, wenn die Schreibweise korrekt ist oder FALSE, wenn nicht.
Beispiel #1 pspell_check()-Beispiel
<?php
$pspell_link = pspell_new("de");
if (pspell_check($pspell_link, "Testt")) {
echo "Das ist eine gültige Schreibweise";
} else {
echo "Sorry, falsche Schreibweise";
}
?>
<?php
//should be using explode instead of implode
//$word = implode(" ", $message);
$word = explode(" ", $message);
foreach($word as $k => $v) {
if (pspell_check($pspell_link, $v)) {
echo "spelled right";
} else {
echo "Sorry, wrong spelling";
};
};
?>
<?php
/*
I had to write these routines to highlight spellings in a WYSIWYG editor.
pspell() barfed at HTML tags and entities, so this code deals with them.
ClearSpell() allows you to clear up the spellchecker mark up afterwards.*/
?>
<html>
<head>
<style>
acronym.spell
{
text-decoration:underline;
color:red;
cursor:help;
}
</style>
</head>
<body>
<?php
$t = "<font color=blue>text herre & some more</font>";
echo "Before:$t";
$t = SpellCheck($t);
echo "<hr>After SpellCheck: $t";
$t = ClearSpell($t);
echo "<hr>After ClearSpell: $t";
?>
</body>
</html>
<?php
function SpellCheck($text)
{
//depends on fnSpell()
// Extracts text from HTML code. In addition to normal word separators, HTML tags
// and HTML entities also function as word delimiters
$pspell_link = pspell_new("en"); //0. Get the dictionary
$strings = explode(">", $text); //1. Split $text on '>' to give us $strings with 0 or 1 HTML tags at the end
$nStrings = count($strings);
for ($cStrings=0; $cStrings < $nStrings; $cStrings++)
{
$string = $strings[$cStrings]; //2. For each string from 1
if ($string =='')
continue;
$temp = explode('<', $string); //2.1 Split $string from $strings on '>' to give us a $tag and $cdata
$tag = $temp[1];
$cdata = $temp[0];
$subCdatas = explode(";", $cdata); //2.2 Split &cdata on ';' to give $subcdatas with 0 or 1 HTML entities on the end
$nSubCdatas = count($subCdatas); //2.3 For each $subCdata from $subcdatas in 2.2
for ($cSubCdatas = 0; $cSubCdatas < $nSubCdatas; $cSubCdatas++)
{
$subCdata = $subCdatas[$cSubCdatas];
if ($subCdata == '')
continue;
$temp = explode('&', $subCdata); //2.3.1 Split the $subCdata on '&' to give us a $subCdataEntity and a $subCdataWithNoEntities
$subCdataEntity = $temp[1];
$subCdataWithNoEntities = $temp[0];
$subCdataWithNoEntities = fnSpell($pspell_link, $subCdataWithNoEntities); //2.3.2 Spellcheck the $cdataWithNoEntities
if (!$subCdataEntity ) //2.3.3 Put the $subCdataEntity, a '&' and the $cdataWithNoEntities back into the $subCdata from 2.2
$subCdata = $subCdataWithNoEntities;
else
$subCdata = $subCdataWithNoEntities. '&' . $subCdataEntity . ';' ;
$subCdatas[$cSubCdatas] = $subCdata; //2.3.4 Put the $subCdata back into the array of $subCdatas
}
$cdata = implode("", $subCdatas); //2.4 Implode the array of $subCdatas back into the $cdata
if ($tag) //2.5 Put the $tag , '>' and $cdata back into $string
$string = $cdata . '<' . $tag . '>';
else
$string = $cdata;
$strings[$cStrings] = $string; //2.6 Put $string back in its place in $strings
}
$text = implode('', $strings); //3 Implode the $strings back into $text
return $text;
}
function fnSpell($pspell_link, $string)
{
preg_match_all("/[A-Z\']{1,16}/i", $string, $words);
for ($i = 0; $i < count($words[0]); $i++)
{
$currentword = $words[0][$i];
if (!pspell_check($pspell_link, $currentword))
{
$wordarray = pspell_suggest($pspell_link, $currentword);
$words = implode(', ', $wordarray);
$suggest = "<acronym class='spell' title='$words'>$currentword</acronym class='spell'>";
$string = str_replace($currentword, $suggest, $string);
}
}
return $string;
}
function ClearSpell($text)
{
$strings = explode(">", $text);
$nStrings = count($strings);
for ($cStrings=0; $cStrings < $nStrings; $cStrings++)
{
$string = $strings[$cStrings];
if ($string =='')
continue;
$temp = explode('<', $string);
$tag = $temp[1];
$cdata = $temp[0];
if ( strstr($tag, 'acronym') && strstr($tag, "class='spell'") )
$string = $cdata;
else
$string = $cdata . '<' . $tag . '>';
$strings[$cStrings] = $string;
}
$text = implode('', $strings);
return $text;
}
?>
I felt that it would help to expand on batch spell checking using this function. The previously posted example implodes using spaces as the separator for each word. There are however situations in which doing this will not return the desired result. For example, "Hello, I like coding." will return an array with two problems, "Hello," and "coding.", both these words are spelt correctly, but pspell_check() will deem them as spelled incorrectly because a comma or a period is being passed in to the function along with the word. The following example allows you to extract only the words (using regular expressions to ignore grammar such as periods or commas) in to an array and then add in html font tags to highlight all words spelled incorrectly red before returning the string.
<?
Function SpellCheck($string) {
$pspell_link = pspell_new("en");
preg_match_all("/[A-Z\']{1,16}/i", $string, $words);
for ($i = 0; $i < count($words[0]); $i++) {
if (!pspell_check($pspell_link, $words[0][$i])) {
$string = str_replace($words[0][$i], "<font color=\"#FF0000\">" . $words[0][$i] . "</font>", $string);
}
}
return $string;
}
?>
Diese Seite bei php.net