PHP str_get_html函数代码示例

 function parseModelsPage($brandId, $brandName, $page)
 {
     $html_content = scraperwiki::scrape($page);
     $this->html = str_get_html($html_content);
     foreach ($this->html->find("div.makers a") as $el) {
         $img = $el->find('img', 0);
         $m['name'] = $brandName . ' ' . $el->find('strong', 0)->innertext;
         $m['img'] = $img->src;
         $m['link'] = 'http://www.gsmarena.com/' . $el->href;
         $m['desc'] = $img->title;
         $temp = explode('-', $el->href);
         $m['id'] = (int) substr($temp[1], 0, -4);
         $m['brand_id'] = $brandId;
         scraperwiki::save_sqlite(array("id" => $m['id']), $m, "cell_model");
         $this->models++;
     }
     $pagination = $this->html->find("div.nav-pages", 0);
     if ($pagination) {
         $nextPageLink = $pagination->lastChild();
         if ($nextPageLink && $nextPageLink->title == "Next page") {
             $this->parseModelsPage($brandId, $brandName, 'http://www.gsmarena.com/' . $nextPageLink->href);
         }
     }
     $this->html->__destruct();
 }

function getHypem($item)
{
    $ch = curl_init("http://hypem.com/item/{$item}?ax=1&ts=1295726809");
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
    curl_setopt($ch, CURLOPT_COOKIESESSION, true);
    curl_setopt($ch, CURLOPT_USERAGENT, "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.1) Gecko/20061204 Firefox/2.0.0.1");
    curl_setopt($ch, CURLOPT_FORBID_REUSE, true);
    curl_setopt($ch, CURLOPT_COOKIE, "AUTH=" . $_GET["hash"] . ";");
    $output = curl_exec($ch);
    $html = str_get_html($output);
    $scripts = $html->find('script');
    $script = $scripts[1];
    $informatie = array();
    $value = $script->innertext;
    for ($i = 0; $i < 15; $i++) {
        $pat = "/\\'([^\\']*?)\\'/";
        $value = str_replace("\\'", '&#39;', $value);
        $value = str_replace("&", "", $value);
        preg_match($pat, $value, $matches);
        $value = str_replace(@$matches[0], '', $value);
        $str = @$matches[1];
        $str = str_replace("&hellip;", "...", $str);
        $informatie[] = $str;
    }
    $result = array("id" => "http://hypem.com/item/" . $informatie[1], "blog" => $informatie[3], "secret" => $informatie[7], "artist" => $informatie[9], "song" => $informatie[10], "songurl" => 'http://hypem.com/serve/play/' . $informatie[1] . '/' . $informatie[7] . '.mp3', "duration" => $informatie[4]);
    return $result;
}

function get_product($url, $page = 0)
{
    global $save_folder;
    $html = curl_get($url);
    //Загружает страницу товара
    $dom = str_get_html($html);
    $article = $dom->find('article', 0);
    //Берем артикул
    $str = $article->attr['id'];
    sscanf($str, 'post-%d', $art);
    $scripts = $dom->find('script');
    foreach ($scripts as $script) {
        if (strpos($script->src, "script.js")) {
            $str = "script[src='" . $script->src . "']";
        }
    }
    $dom->find($str, 0)->outertext = '';
    //Ajax запрос
    $html = get_ajax($art);
    //Получили данные из ajax
    $dom2 = str_get_html($html);
    //Ищем в 1-й странице div куда будем вставлять данные из ajax
    $dom->find('div[id=order-variables]', 0)->innertext = $dom2;
    //Сохраняем HTML
    file_put_contents($save_folder . 'product--' . $page . '.html', $dom);
}

 public function dispatchLoopShutdown()
 {
     if (!Tool::isHtmlResponse($this->getResponse())) {
         return;
     }
     if (!Tool::useFrontendOutputFilters($this->getRequest()) && !$this->getRequest()->getParam("pimcore_preview")) {
         return;
     }
     if (\Pimcore::inDebugMode()) {
         return;
     }
     if ($this->enabled) {
         include_once "simple_html_dom.php";
         $body = $this->getResponse()->getBody();
         $html = str_get_html($body);
         if ($html) {
             $html = $this->searchForScriptSrcAndReplace($html);
             $html = $this->searchForInlineScriptAndReplace($html);
             $body = $html->save();
             $html->clear();
             unset($html);
         }
         $this->getResponse()->setBody($body);
     }
 }

 public function getDom($url, $post = false)
 {
     $f = fopen(CURL_LOG_FILE, 'a+');
     // curl session log file
     if ($this->lastUrl) {
         $header[] = "Referer: {$this->lastUrl}";
     }
     $curlOptions = array(CURLOPT_ENCODING => 'gzip,deflate', CURLOPT_AUTOREFERER => 1, CURLOPT_CONNECTTIMEOUT => 120, CURLOPT_TIMEOUT => 120, CURLOPT_URL => $url, CURLOPT_SSL_VERIFYPEER => false, CURLOPT_SSL_VERIFYHOST => false, CURLOPT_FOLLOWLOCATION => true, CURLOPT_MAXREDIRS => 9, CURLOPT_RETURNTRANSFER => 1, CURLOPT_HEADER => 0, CURLOPT_USERAGENT => "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36", CURLOPT_COOKIEFILE => COOKIE_FILE, CURLOPT_COOKIEJAR => COOKIE_FILE, CURLOPT_STDERR => $f, CURLOPT_VERBOSE => true);
     if ($post) {
         // add post options
         $curlOptions[CURLOPT_POSTFIELDS] = $post;
         $curlOptions[CURLOPT_POST] = true;
     }
     $curl = curl_init();
     curl_setopt_array($curl, $curlOptions);
     $data = curl_exec($curl);
     $this->lastUrl = curl_getinfo($curl, CURLINFO_EFFECTIVE_URL);
     // get url we've been redirected to
     curl_close($curl);
     if ($this->dom) {
         $this->dom->clear();
         $this->dom = false;
     }
     $dom = $this->dom = str_get_html($data);
     fwrite($f, "{$post}\n\n");
     fwrite($f, "-----------------------------------------------------------\n\n");
     fclose($f);
     return $dom;
 }

 protected function parsePageDynamicContent($content)
 {
     $html = str_get_html($content);
     $count = 0;
     // There are no dynamic widgets...
     foreach ($widgets = $html->find('.dm_widget') as $widget) {
         $widgetId = intval(str_replace('dm_widget_', '', $widget->id));
         if ($widgetId) {
             $cache = $widget->find('.dm_widget_cacheable', 0);
             if ($cache) {
                 $count++;
                 // We have found a dynamic widget...
                 $cache->innertext = sprintf('{#page#%s#page#}{#widget#%s#widget#}', $this->getPage()->getId(), $widgetId);
             }
         }
     }
     if ($count == 0) {
         // The page did not had any dynamic content...
         return null;
     }
     $code = $html->innertext;
     $code = str_replace('{#page#', '<?php echo $helper->renderWidgetInner(array(\'page_id\'=>', $code);
     $code = str_replace('#page#}{#widget#', ', \'widget_id\'=>', $code);
     $code = str_replace('#widget#}', ')); ?>', $code);
     return $code;
 }

function scrape($url, $path, $parse)
{
    $config = HTMLPurifier_Config::createDefault();
    $config->set('Core.Encoding', 'UTF-8');
    //encoding of output
    $config->set('HTML.Doctype', 'XHTML 1.1');
    //doctype of output
    $purifier = new HTMLPurifier($config);
    $dirty_html = file_get_contents($url);
    $clean_html = $purifier->purify($dirty_html);
    $html = str_get_html($clean_html);
    switch ($parse) {
        case 'tag':
            $ret = $html->find($path)->tag;
            break;
        case 'outertext':
            $ret = $html->find($path)->outertext;
            break;
        case 'innertext':
            $ret = $html->find($path)->innertext;
            break;
        case 'plaintext':
            $ret = $html->find($path)->plaintext;
            break;
        default:
            $ret = $html->find($path);
            break;
    }
    // clean up memory
    $html->clear();
    unset($dirty_html);
    unset($clean_html);
    unset($html);
    return $ret;
}

 public function testRadioInput()
 {
     // Create a Salutation
     $new_field = magic_form_field_radio::factory($this->input_default_name, $this->input_default_label);
     // Add some options to it.
     $new_field->add_options(array(1 => 'Test Value 1', 2 => 'Test Value 2'));
     // Add them to the form
     $this->magic_form->add_fields($new_field);
     $html = $this->magic_form->__toString();
     //Get HTML Dom
     $dom = str_get_html($html);
     $form = $dom->find("//form")[0];
     //Find Radio Inputs
     $test_radio_field1 = $form->find("input[id=" . $this->input_default_name . "-test-value-1]")[0];
     $test_radio_field2 = $form->find("input[id=" . $this->input_default_name . "-test-value-2]")[0];
     //Check Radio Inputs
     $this->assertEquals($this->input_default_name, $test_radio_field1->attr['name'], "Radio 1 Name Equals " . $this->input_default_name);
     $this->assertEquals($this->input_default_name, $test_radio_field2->attr['name'], "Radio 2 Name Equals " . $this->input_default_name);
     //Check Labels
     $test_radio_1_field_label = $form->find("label[for=" . $this->input_default_name . "-test-value-1]")[0];
     $this->assertEquals($this->input_default_name . "-test-value-1", $test_radio_1_field_label->attr['for'], "Check Label Radio 1");
     $this->assertEquals("Test Value 1", $test_radio_1_field_label->innertext(), "Check Label Text Radio 1");
     $test_radio_2_field_label = $form->find("label[for=" . $this->input_default_name . "-test-value-2]")[0];
     $this->assertEquals($this->input_default_name . "-test-value-2", $test_radio_2_field_label->attr['for'], "Check Label Radio 2");
     $this->assertEquals("Test Value 2", $test_radio_2_field_label->innertext(), "Check Label Text Radio 2");
 }

 private function clean_discourse($discourse)
 {
     App::import('Vendor', 'simple_html_dom');
     $html = str_get_html($discourse['Discourse']['content']);
     foreach ($html->find('div.pagebreak') as $pagebreak) {
         $anchor_name = $pagebreak->find('a[name]', 0)->name;
         $pdf_text = $pagebreak->find('a[target="_pdfwin"]', 0)->href;
         $pdf_image = $pagebreak->find('a[target="_pdfwin2"]', 0)->href;
         $pagebreak->innertext = '<a name="' . $anchor_name . '"></a><a class="pdf-image" href="' . $pdf_image . '"></a>';
         $pagebreak->tag = 'span';
     }
     foreach ($html->find('div.columnbreak') as $columnbreak) {
         $anchor_name = $columnbreak->find('a[name]', 0)->name;
         $pdf_text = $columnbreak->find('a[target="_pdfwin"]', 0)->href;
         $pdf_image = $columnbreak->find('a[target="_pdfwin2"]', 0)->href;
         $columnbreak->innertext = '<a name="' . $anchor_name . '"></a><a class="pdf-image" href="' . $pdf_image . '"></a>';
         $columnbreak->tag = 'span';
     }
     foreach ($html->find('div.paragraph') as $paragraph) {
         $paragraph->tag = "p";
         $paragraph->class = null;
     }
     foreach ($html->find('div.hyphen') as $hyphen) {
         $hyphen->tag = "span";
     }
     $this->Discourse->id = $discourse['Discourse']['id'];
     $this->Discourse->saveField('content', $html);
 }

function dados_telefone($num)
{
    $num = preg_replace("/[^\\d]/", "", $num);
    if (strlen($num) < 10) {
        return false;
    } else {
        include_once 'src/simple_html_dom.php';
        $query = http_build_query(array('tel' => $num));
        $options = array('http' => array('header' => "Content-Type: application/x-www-form-urlencoded\r\n" . "Content-Length: " . strlen($query) . "\r\n" . "User-Agent:Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/42.0.2311.90 Safari/537.36\r\n", 'method' => "POST", 'content' => $query));
        $context = stream_context_create($options);
        $result = file_get_html("http://consultanumero.info/consulta", false, $context);
        $resultado = @$result->find('div[class=resultado]', 0)->children(1)->outertext;
        if (empty($resultado)) {
            return false;
        } else {
            $resultado = str_get_html($resultado);
            $img = @$result->find('div[class=a]', 0)->children(0)->outertext;
            preg_match('%<img.*?title=["\'](.*?)["\'].*?/>%i', $img, $operadora);
            $data['operadora'] = $operadora[1];
            $data['tipo'] = substr(strrchr(strip_tags($resultado->find('p', 0)->outertext), ' &raquo; '), 1);
            $data['portabilidade'] = strtolower(substr(strrchr(strip_tags($resultado->find('p', 1)->outertext), ' &raquo; '), 1)) == 'sim' ? "Sim" : "Não";
            $data['estado'] = str_replace(array('(', ')'), '', substr(strrchr(strip_tags($resultado->find('p', 2)->outertext), ' &raquo; '), 1));
            $cidade = explode(' &raquo; ', strip_tags($resultado->find('p', 3)->outertext));
            $data['cidade'] = $cidade[1];
            return $data;
        }
    }
}

 function processPage($pageContent)
 {
     try {
         $links = array();
         $html = str_get_html($pageContent);
         //reparam html stricat
         if (!$html->find('body', 0, true)) {
             $html = $this->fixHtml($html);
         }
         $body = $html->find('body', 0, true);
         $this->extractText($body);
         foreach ($body->find("a") as $link) {
             $links[] = $link->href;
         }
         //cata memorie consuma
         //si eliberare referinte pierdute
         $html->clear();
         MemoryManagement::showUsage('before cleaning', true, 'KB');
         MemoryManagement::clean(true);
         MemoryManagement::showUsage('after cleaning', true, 'KB');
         return $links;
     } catch (Exception $ex) {
         Applog::exceptionLog($ex);
     }
 }

 /**
  * HTML属性削除処理の実行
  *
  * Pickles2の状態を参照し、自動的に処理を振り分けます。
  *
  * - パブリッシュする場合、DECコメントを削除します。
  * - プレビューの場合、DECライブラリを埋め込み、URIパラメータからDECの表示・非表示を切り替えられるようにします。
  *
  * @param object $px Picklesオブジェクト
  * @param object $options オプション
  * @return boolean true
  */
 public static function exec($px, $options = null)
 {
     require_once __DIR__ . '/simple_html_dom.php';
     if (!$px->is_publish_tool()) {
         // パブリッシュ時にのみ働きます。
         return true;
     }
     if (!@is_array($options->attrs)) {
         @($options->attrs = array());
     }
     // var_dump($options);
     foreach ($px->bowl()->get_keys() as $key) {
         $src = $px->bowl()->pull($key);
         // HTML属性を削除
         $html = str_get_html($src, true, true, DEFAULT_TARGET_CHARSET, false, DEFAULT_BR_TEXT, DEFAULT_SPAN_TEXT);
         foreach ($options->attrs as $attr) {
             $ret = $html->find('*[' . $attr . ']');
             foreach ($ret as $retRow) {
                 // var_dump($retRow->$attr);
                 $retRow->{$attr} = null;
             }
         }
         $src = $html->outertext;
         $px->bowl()->replace($src, $key);
     }
     return true;
 }

 /**
  * 
  * @param string $html
  */
 public function exec($html)
 {
     mb_language('Japanese');
     // 1.プリプロセス
     // scriptテキスト削除
     // script内に文字列リテラルの閉じタグがあるとDomDocumentがscriptのソースを#text扱いしてしまうので
     // script内の文字を削除する
     // 正規表現で削除しようとするとSegmentation faultが発生する（StackOverFlow?）ので
     // simple_html_domでscript内文字列を削除
     // MAX_FILE_SIZEの制限にひっかかったので、ソースを編集してデフォルトの3倍に変更している
     $simpleHtml = str_get_html($html);
     foreach ($simpleHtml->find('script') as $script) {
         $script->innertext = '';
     }
     $html = $simpleHtml->outertext;
     // トリム
     //		$html = preg_replace('/(\s|　)+/mi', ' ', $html);
     // 2. dom生成
     $doc = new DomDocument("1.0", "utf-8");
     @$doc->loadHTML(mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8'));
     $node = $doc->getElementsByTagName('body')->item(0);
     $this->preProcessedInput = $node->textContent;
     // 3.プロパティを初期化
     $this->domXPath = new DomXPath($doc);
     $this->title = @$doc->getElementsByTagName('title')->item(0)->textContent;
     $text = $this->scan($node);
     $this->textAll = $text;
     $this->domCountAll = $this->domCount;
     $this->pancutuationCountAll = $this->calcKutenScore($text) + $this->calcTotenScore($text);
     $this->textLengthAll = mb_strlen($text);
     $this->highScore = -1000000;
     $this->extracedNode = null;
     // 4.実行
     $this->extract($node);
 }

 function parse($url)
 {
     $url = $this->readUrl($url);
     var_dump($url);
     die;
     if (!$url or $this->cacheurl[$url] or $this->cacheurl[preg_replace('#/$#', '', $url)]) {
         return false;
     }
     var_dump('111');
     die;
     $this->_allcount--;
     if ($this->_allcount <= 0) {
         return false;
     }
     $this->cacheurl[$url] = true;
     $item = array();
     $data = str_get_html(request($url));
     $item['url'] = $url;
     $item['title'] = count($data->find('title')) ? $data->find('title', 0)->plaintext : '';
     $item['text'] = $data->plaintext;
     $this->result[] = $item;
     if (count($data->find('a'))) {
         foreach ($data->find('a') as $a) {
             $this->parse($a->href);
         }
     }
     $data->clear();
     unset($data);
 }

 function translate($lang = 'ru')
 {
     $config = App;
     $content = ob_get_contents();
     ob_end_clean();
     if (self::$active) {
         self::$locale = self::getAppLocale($lang);
         preg_match_all(LOCALE_TEMPLATE, $content, $mathes);
         if (sizeof($mathes[1]) > 0) {
             foreach ($mathes[1] as $word) {
                 if (array_key_exists($word, self::$locale)) {
                     $content = str_replace(sprintf(LOCALE_TEMPLATE_CHANGE, $word), self::$locale[$word], $content);
                 }
             }
         }
     }
     if (($selector = Request::post('html_element')) || ($selector = Request::get('html_element'))) {
         Load::dependence('simple_html_dom');
         $html = str_get_html($content);
         $elements = $html->find($selector);
         if (sizeof($elements) > 0) {
             echo $elements[0]->innertext;
         } else {
             echo $content;
         }
     } else {
         echo $content;
     }
 }