2013-05-11 5 views
0

Привет, поэтому я пытаюсь разобрать веб-сайт ratemyprofessor для имени профессора и комментариев и преобразовать каждый div в открытый текст. Вот структура div, с которой я работаю.Простая проблема с парсером HTML

<div id="ratingTable"> 
<div class="ratingTableHeader"></div> 
<div class="entry odd"><a name="18947089"></a> 
<div class="date"> 
    8/24/11 // the date which I want to parse 
</div><div class="class"><p> 
    ENGL2323 // the class which I want to parse 
</p></div><div class="rating"></div><div class="comment" style="width:350px;"> 
    <!-- comment section --> 
<p class="commentText"> // this is what I want to parse as plaintext for each entry 
    I have had Altimont for 4 classes. He is absolutely one of my favorite professors at St. Ed's. He's generous with his time, extremely knowledgeable, and such an all around great guy to know. Having class with him he would always have insightful comments on what we were reading, and he speaks with a lot of passion about literature. Just the best! 
</p><div class="flagsIcons"></div></div> 
    <!-- closes comment --> 
</div> 
    <!-- closes even or odd --> 
<div class="entry even"></div> // these divs are the entries for each professor 
    <!-- closes even or odd --> 
<div class="entry odd"></div> 
    <!-- closes even or odd --> 
</div> 
    <!-- closes rating table --> 

Таким образом, каждый вход инкапсулируются под этим «ratingtable» DIV и каждая запись является либо «запись нечетными» или «запись даже» делами.

Вот моя попытка до сих пор, но она просто производит огромный искаженный массив с большим количеством мусора.

<?php 
header('Content-type: text/html; charset=utf-8'); // this just makes sure encoding is right 
include('simple_html_dom.php'); // the parser library 

$html = file_get_html('http://www.ratemyprofessors.com/SelectTeacher.jsp?sid=834'); // the url for the teacher rating profile 

//first attempt, rendered nothing though 

    foreach($html->find("div[class=commentText]") as $content){ 
     echo $content.'<hr />'; 
    } 

foreach($html->find("div[class=commentText]") as $content){ 
    $content = <div class="commentText"> // first_child() should be the <p> 
    echo $content->first_child().'<hr />'; 

//Get the <p>'s following the <div class="commentText"> 

    $next = $content->next_sibling(); 
    while ($next->tag == 'p') { 
     echo $next.'<hr />'; 
     $next = $next->next_sibling(); 
    } 
} 
?> 
+0

Я думаю, что вы пытаетесь разобрать неправильную страницу ... – Dinesh

ответ

0

Confused HTML ... Не могли бы вы попытаться посмотреть, работает ли это?

foreach (DOM($html, '//div[@class="commentText"]//div[contains(@class,"entry")]') as $comment) 
{ 
    echo strval($comment); 
} 

О, и да - я не люблю simple_html_dom, используйте вместо этого:

function DOM($html, $xpath = null, $key = null, $default = false) 
{ 
    if (is_string($html) === true) 
    { 
     $dom = new \DOMDocument(); 

     if (libxml_use_internal_errors(true) === true) 
     { 
      libxml_clear_errors(); 
     } 

     if (@$dom->loadHTML(mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8')) === true) 
     { 
      return DOM(simplexml_import_dom($dom), $xpath, $key, $default); 
     } 
    } 

    else if (is_object($html) === true) 
    { 
     if (isset($xpath) === true) 
     { 
      $html = $html->xpath($xpath); 
     } 

     if (isset($key) === true) 
     { 
      if (is_array($key) !== true) 
      { 
       $key = explode('.', $key); 
      } 

      foreach ((array) $key as $value) 
      { 
       $html = (is_object($html) === true) ? get_object_vars($html) : $html; 

       if ((is_array($html) !== true) || (array_key_exists($value, $html) !== true)) 
       { 
        return $default; 
       } 

       $html = $html[$value]; 
      } 
     } 

     return $html; 
    } 

    return false; 
} 
0

Если вы все еще хотите использовать simple_html_dom .. смотри ниже код для ошибок в вашей код:

<?php 
header('Content-type: text/html; charset=utf-8'); // this just makes sure encoding is right 
include('simple_html_dom.php'); // the parser library 

// you were trying to parse the wrong link.. your previous link did not have <div> tag with commentText class .. I chose a random link.. choose link for whichever professor you like or grab the links of professor from previous page store it in an array and loopr through them to get comments 

$html = file_get_html('http://www.ratemyprofessors.com/ShowRatings.jsp?tid=1398302'); // the url for the teacher rating profile 

//first attempt, rendered nothing though 

//your div tag has class "comment" not "commentText" 
    foreach($html->find("div[class=comment]") as $content){ 
     echo $content.'<hr />'; 
    } 


foreach($html->find("div[class=comment]") as $content){ 

// I am not sure what you are trying to do here but watever it is it's wrong 
//$content = <div class='commentText'>"; // first_child() should be the <p> 
//echo $content->first_child().'<hr />'; 

    //correct way to do it 
    echo $html->firstChild();// ->first_child().'<hr />'; 


//this whole code does not make any sense since you are already retrieving the comments from the above code.. but if you still want to use it .. I can figure out what to do 

//Get the <p>'s following the <div class="commentText"> 
//  $next = $html->firstChild()->next_sibling(); 
// while ($next->tag == 'p') { 
//  echo $next.'<hr />'; 
//  $next = $next->next_sibling(); 
//  } 
    } 
?> 

Выход

Comment 

Dr.Alexander was the best. I would recommend him for American Experience or any class he teaches really. He is an amazing professor and one of the nicest most kind hearted people i've ever met. 
Report this rating 

Professor Alexander is SO great. I would recommend him to everyone for american experience. He has a huge heart and he's really interested in getting to know his students as actual people. The class isn't difficult and is super interesting. He's amazing. 
Report this rating 

Dins

Смежные вопросы