[RESOLVED] identifying same texts in 2 web pages

ajay600

i need to identify the similar texts in two web pages and remove themm...
i need to store the text that is between the and in an array...that is for example... hello everyone ..i need to store " hello everyone" in an array ...

i will do it for all tags in page 1 and then page 2 and then compare the text in both the pages to find similar matches and remove them ..

please help me code it in php

Weedpacket

Easiest would be to use [man]DOM[/man]. Then it would be $html_document->getElementsByTagName('b') to get a list of all the bold elements. Loop through that result and get the textContent property of each element. There's your text.

Once you've got the two arrays, some combination of [man]array_intersect[/man] and [man]array_difference[/man] will do the matching/removal.

ajay600

i found out all the b tags but how do i take the text content from the b tags..please help

<?php

$doc = new DOMDocument(); // An instance of DOMDocument
@$doc->loadHTMLFile('http://www.web-source.net/web_design_tips/');

$doc2 = new DOMDocument(); // An instance of DOMDocument
@$doc2->loadHTMLFile('http://www.web-source.net/html_codes_chart.htm');

$xpath = new DOMXPath($doc);
$xpath2 = new DOMXPath($doc2);

$List=array();

$List2=array();

$List[] = $doc->getElementsByTagName("b");
$List2[] = $doc2->getElementsByTagName("b");

$textBoth = array_intersect($List, $List2);

foreach ($textBoth as $text)
{ // Loops through the src strings that are common to both documents

$text->parentNode->removeChild($text);

}
echo $doc->saveHTML();

Weedpacket

Like I said; loop through the two lists and fetch each item's textContent into two arrays (I've provided a link this time).

ajay600

how do i extract the text contents from the tag and save it in an array...

please write the lines necessary to extract the text contents from the b tag and save it array...im new to php !

Weedpacket

And you were doing so well, too....

$bold_text = array();
$bold_elements = $doc->getElementsByTagName('b');
foreach($bold_elements as $element)
    $bold_text[] = $element->textContent;

ajay600

after comparing for the similar elements , when i try to remove the repeated elements , i get an error as follows "Fatal error: Call to a member function removeChild() on a non-object in c:\wamp\www\b.php on line 38 "..i have shown the line as bold in the code..how do i correct it

$doc = new DOMDocument(); // An instance of DOMDocument
@$doc->loadHTMLFile('http://www.web-source.net/web_design_tips/');

$doc2 = new DOMDocument(); // An instance of DOMDocument
@$doc2->loadHTMLFile('http://www.web-source.net/html_codes_chart.htm');

$bold_text = array();
$bold_text2 = array();

$bold_elements = $doc->getElementsByTagName('b');
foreach($bold_elements as $element)
$bold_text[] = $element->textContent; 

$bold_elements2 = $doc2->getElementsByTagName('b');
foreach($bold_elements2 as $element2)
$bold_text2[] = $element2->textContent; 


$domElemsToRemove=array();

$srcBoth = array_intersect($bold_text , $bold_text2);

foreach ($srcBoth as $img) 

{ 

$domElemsToRemove[] = $img;   

}

foreach( $domElemsToRemove as $img )
{ 

[B][U]$img->removeChild($img);[/U][/B]

} 
echo $doc->saveHTML();

ajay600

found the problem but i dont know how to correct it

right now i am only deleting only the variable ...but i should be removing the node instead ..
i should create a DOMNodeList with matching text cotent using xpath so that the node can be removed using remove child.. how do i do it ..

i tried using but this did'nt work

$imgs = $xpath->query('//b="$img");

..please help

Weedpacket

Well, predicates in XPath need to be in "[]" (in this case //b[string(.)='$img']).

Since you're wanting to remove from document2 elements that appear in document1 (a difference, rather than an intersection), you might be able to use a single XPath query to find all such nodes (by using [man]join[/man] on the array of textContents to make the predicate "[string(.)='$img1' or string(.)='$img2' or ....]". I think the string() function may not be strictly necessary, but I don't see it hurting, either.