LinuxQuestions.org - how can I seprate normal text from html tags spell check it & then again place it ins

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - how can I seprate normal text from html tags spell check it & then again place it ins (https://www.linuxquestions.org/questions/programming-9/how-can-i-seprate-normal-text-from-html-tags-spell-check-it-and-then-again-place-it-ins-214556/)

how can I seprate normal text from html tags spell check it & then again place it ins

Hello
I am devloping a spell check utility using PHP4.2.2 on Redhat9, Apache & Netscape7.1.
I have devloped it successfully for normal text.
but problem comes when I try to check spellings in html formated text.

heres how it works
I have one php named compose_richtext.php. I have provided a toolbar which I got from http://www.kevinroth.com/rte/demo.htm. here user can type formated text. Their is in built feature given in that toolbar for spell check but I don't want to use that for some reasons. So I implemnted my own code.
when user clicks on spell check the whole data is passed to next page i.e Spellcheck.php their I am using pspell_check & pspell_suggest for spellchecking.
But the problem is how to distinguish actual text from html tags. I looked in php manual & found

htmlentities
nl2br
htmlspecialchars

but I don' think this will work for me.
I can easily use strip_tags but then I will loose all the tags so how will I place the actual text again in formated form.

I read something like tidy->ishtml is their but that don't works with php4.2.2 also I was unable to find proper documantation of how to use it.

So can u please help me & tell me that how can I seprate normal text from html tags spell check it & then again place it inside html tags.

Amit

i dont have pspell on any php installation i have access to but you should find it simple to adapt this code

PHP Code:




<?php 
 $ary=preg_split('((<[^>]*>)|( ))', $_GET['stp'], -1,PREG_SPLIT_DELIM_CAPTURE|PREG_SPLIT_NO_EMPTY); 
 
 $out=""; 
 
 for($i=0;$i<count($ary);$i++) { 
  $a=$ary[$i]; 
  if($a==" " || substr($a,0,1)=="<") { 
   $out.=$a; 
  } else { 
   //spellcheck here 
   $out.=$a; 
  } 
 } 
 
 print($out); 
?>

Thanks for ur support kev.
but with this code what will happen if users text message contains < > signs for eg user text could be
Arrow looks like this >------->
& the second thing which I mentioned in my Question was how will I place all the text in html format again after checking spellings because in this case we will loose all the info about previous formating. Because String $out is just plain text.

Here is the code which am using

Regards
Amit

Code:

<?php 

                    



$formatedtext=base64_decode($rtedata);

$normaltext=strip_tags($formatedtext);

$word=' ';

$updated_msg='';

if($_REQUEST['Button']=="spellcheck_rich"){

        $pspell_link = pspell_new("en");

        //$text=base64_decode($Compose);

        $str_len=strlen($normaltext);

        for($i=0; $i<=$str_len; $i++){

                if(ord($normaltext[$i])==10){

                        echo "<br>";

                        $updated_msg=$updated_msg."\n";

                }else if( (((ord($normaltext[$i])<=64) || ( (ord($normaltext[$i]))>=91 && (ord($normaltext[$i]))<=96 ) ||  ( (ord($normaltext[$i])) >=123 )) &&  ( (ord($normaltext[$i])) != 10 ))) {

                        echo $normaltext[$i];

                        $updated_msg=$updated_msg.$normaltext[$i];

                }else{

                        

                        if($word==' ')

                        {

                                $word=$normaltext[$i];

                                $loc=$i;

                        }

                        else

                                $word=$word.$normaltext[$i];

                        if($normaltext[$i+1]==' '  || $normaltext[$i+1]=='\n' || $normaltext[$i+1]=='.' || $normaltext[$i+1]==',' || ( (((ord($normaltext[$i+1])<=64) || ( (ord($normaltext[$i+1]))>=91 && (ord($normaltext[$i+1]))<=96 ) ||  ( (ord($normaltext[$i+1])) >=123 )) &&  ( (ord($normaltext[$i+1])) != 10 )))){

                                if (!pspell_check($pspell_link,$word)) {

                                        $size=strlen($word);

?>

                <input type=text value="<?php echo $word  ?>" size="<? echo strlen($word); ?>" name=elements<? echo "_".$loc."_".$size ?>>

                <input type=button value="?" onclick="Suggest('<?php echo base64_encode($word) ?>','<?php echo $loc ?>','<?php echo $size ?>')">

                

                

<?php

                        $updated_msg=$updated_msg.$word;

                                }else{

                                        echo $word;

                                        $updated_msg=$updated_msg.$word;

                                }

                                $word=' ';

                        }

                }

        }//For closed

}



$updated_msg=base64_encode($updated_msg);

?>

<edit>
Re: your edit

by amit_28oct
but with this code what will happen if users text message contains < > signs for eg user text could be
Arrow looks like this >------->

if there not tags then you should have already converted them to &lt and &gt

the second thing which I mentioned in my Question was how will I place all the text in html format again after checking spellings because in this case we will loose all the info about previous formating. Because String $out is just plain text.

as i say $out is html, it is not just the plain text, run it and see.
</edit>

if the users message contains < > signs then how are you differentiating them from tags? The code i gave above does place stuff back into its original form with tags after youve spellchecked it so im not sure what you mean. below is what i believe to be correct code but as i dont have php compiled with the spelling libraries i cant check it. if it doesnt work i'll need to see the output.

PHP Code:




<?php 
$ary=preg_split('((<[^>]*> )|( ))', $_GET['stp'], -1,PREG_SPLIT_DELIM_CAPTURE|PREG_SPLIT_NO_EMPTY); 
 
$out=""; 
$psl=pspell_new("en"); 
 
for($i=0;$i<count($ary);$i++) { 
  $a=$ary[$i]; 
  if($a==" " || substr($a,0,1)=="<") { 
   $out.=$a; 
  } else { 
   if(pspell_check($psl,$a)) {    
    $out.=$a; 
   } else { 
    $out.=sprintf("<font color=\"#ff0000\">%s</font>", $a); 
   } 
  } 
} 
 
print($out); 
?>

this should take a GET variable stp and spellcheck each word and output it with tags preserved but each incorrectly spelt word should be in red.

hello kave
hear are 2 sample outputs for different values of stp
stp="say";
style="font-weight: bold;">say
stp="say sayy say";
say sayy say

Try yahoo's spell check function in Netscape & then try yahoo's facility of typing formated text (In IE) if possible. observe how spell checking works in boths cases that is what I am trying to do.

Amit

well one of us is doing something very wrong cos when i run your examples through my code it comes out fine

check out http://khn.homelinux.net/kev/spell.php?stp=say

it is definatly spellchecking the right stuff, and not the others.