Created
January 13, 2016 05:43
-
-
Save jkopelioff/15191cf5b9fd0ad5736a to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| <?php | |
| include('../predis/lib/Predis/Autoloader.php'); | |
| $m = new MongoClient(); // connect | |
| $collection = $m->opencorkage->selectCollection("restaurants.losangeles"); | |
| $base_url = 'http://www.opentable.com/'; | |
| $simple_info = 'httphandlers/RestaurantInfoLiteNew.ashx?rid='; | |
| $restaurant_listings = 'los-angeles-restaurant-listings'; | |
| Predis\Autoloader::register(); | |
| $redis = new Predis\Client(); | |
| $page = $base_url.$restaurant_listings; | |
| echo "Retrieving... ".$page; | |
| $html_text = file_get_contents($page); | |
| $dom = @DOMDocument::loadHTML($html_text); | |
| $xpath = new DomXPath($dom); | |
| $nodes = $xpath->query("//div[contains(normalize-space(@class),'rinfo')]"); | |
| $count_total = count($nodes); | |
| $count = 1; | |
| foreach( $nodes as $node) | |
| { | |
| $dom_name_url = $xpath->query("a[contains(normalize-space(@class),'r')]", $node); | |
| $name = $dom_name_url->item(0)->nodeValue; | |
| $rid = $node->getAttribute('rid'); | |
| $href = $dom_name_url->item(0)->getAttribute('href'); | |
| $simple_page = $base_url.$href; | |
| $html_text = $redis->get($simple_page); | |
| $throttle = true; | |
| if (empty($html_text)) | |
| { | |
| echo "Retrieving ".$count." of ".$count_total."... ".$simple_page."\n"; | |
| $html_text = file_get_contents($simple_page); | |
| $redis->set($simple_page, $html_text); | |
| } else { | |
| echo "From Cache ".$count." of ".$count_total."... ".$simple_page."\n"; | |
| $throttle = false; | |
| } | |
| $dom_s = @DOMDocument::loadHTML($html_text); | |
| $element = $dom_s->getElementById('ProfileOverview_linkRestaurantImage'); | |
| $element = ($element)?$element->getElementsByTagName('img'):null; | |
| $image = ($element)?$element->item(0)->getAttribute('src'):""; | |
| $element = $dom_s->getElementById('ProfileOverview_DiningStyle'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $dining_style = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_Website'); | |
| $element = ($element)?$element->getElementsByTagName('a'):null; | |
| $website = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_Phone'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $phone = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_HoursOfOperation'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $hours = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_Payment'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $payment = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_AllCuisines'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $cuisine = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_DressCode'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $dress_code = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_AcceptsWalkins'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $walkins = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_Offers'); | |
| $element = ($element)?$element->getElementsByTagName('span'):null; | |
| $offers = ($element)?$element->item(0)->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_lblPriceText'); | |
| $price = ($element)?$element->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_lblAddressText'); | |
| $address = ($element)?$element->nodeValue:""; | |
| $element = $dom_s->getElementById('ProfileOverview_lblNeighborhoodText'); | |
| $neighborhood = ($element)?$element->nodeValue:""; | |
| $element = $dom_s->getElementById('DescriptionBlock'); | |
| $description = ($element)?$element->nodeValue:""; | |
| $restaurant = array("name" => $name, | |
| "description" => preg_replace('/<[^>]*>/', "", $description), | |
| "rid" => $rid, | |
| "ot_url" => $simple_page, | |
| "image" => $base_url.$image, | |
| "dining_style" => $dining_style, | |
| "website" => $website, | |
| "phone" => $phone, | |
| "hours" => $hours, | |
| "payment" => $payment, | |
| "cuisine" => $cuisine, | |
| "dress_code" => $dress_code, | |
| "walkins" => $walkins, | |
| "offers" => $offers, | |
| "price" => $price, | |
| "address" => $address, | |
| "neighborhood" => $neighborhood); | |
| $collection->insert($restaurant); | |
| if ($throttle) { | |
| $rand_num = rand(1,10); | |
| echo "Sleeping for ".$rand_num." sec\n"; | |
| //Sleep random interval so opentable doesn't trip | |
| sleep($rand_num); | |
| } | |
| $count++; | |
| } | |
| ?> |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment