- Notifications
You must be signed in to change notification settings - Fork76
A universal web-util for PHP.
License
spekulatius/PHPScraper
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
For full documentation, visitphpscraper.de.
PHPScraper is a versatile web-utility for PHP. Its primary objective is to streamline the process of extracting information from websites, allowing you to focus on accomplishing tasks without getting caught up in the complexities of selectors, data structure preparation, and conversion.
Under the hood, it uses
- BrowserKit (formerlyGoutte) to access the web
- League/URI to process URLs
- donatello-za/rake-php-plus to extract and analyze keywords
Seecomposer.json for more details.
Here are a few impressions of the way the library works. More examples are on theproject website.
All scraping functionality can be accessed either as a function call or a property call. For example, the title can be accessed in two ways:
// Prep$web =new \Spekulatius\PHPScraper\PHPScraper;$web->go('https://google.com');// Returns "Google"echo$web->title;// Also returns "Google"echo$web->title();
Many common use cases are covered already. You can find prepared extractors for various HTML tags, including interesting attributes. You can filter and combine these to your needs. In some cases there is an option to get a simple or detailed version, here in the case oflinksWithDetails:
$web =new \Spekulatius\PHPScraper\PHPScraper;// Contains:// <a href="https://placekitten.com/456/500" rel="ugc">// <img src="https://placekitten.com/456/400">// <img src="https://placekitten.com/456/300">// </a>$web->go('https://test-pages.phpscraper.de/links/image-urls.html');// Get the first link on the page and print the resultprint_r($web->linksWithDetails[0]);// [// 'url' => 'https://placekitten.com/456/500',// 'protocol' => 'https',// 'text' => '',// 'title' => null,// 'target' => null,// 'rel' => 'ugc',// 'image' => [// 'https://placekitten.com/456/400',// 'https://placekitten.com/456/300'// ],// 'isNofollow' => false,// 'isUGC' => true,// 'isSponsored' => false,// 'isMe' => false,// 'isNoopener' => false,// 'isNoreferrer' => false,// ]
If there aren't any matching elements (here links) on the page, an empty array will be returned. If a method normally returns a string it might returnnull. Details such asfollow_redirects, etc. are optional configuration parameters (see below).
Most of the DOM should be covered using these methods:
- severalmeta-tags and other
<head>-information - Social-Media information like Twitter Card and Facebook Open Graph
- Content:Headings,Outline,Texts andLists
- Images
- Links
- Keywords
A full list of methods with example code can be found onphpscraper.de. Further examples are in thetests.
Besides processing the content on the page itself, you can download files usingfetchAsset:
// Absolute URL$csvString =$web->fetchAsset('https://test-pages.phpscraper.de/test.csv');// Relative URL after navigation$csvString =$web ->go('https://test-pages.phpscraper.de/meta/lorem-ipsum.html') ->fetchAsset('/test.csv');
You will only need to write the content into a file or cloud storage.
PHPScraper can assist in collecting feeds such asRSS feeds,sitemap.xml-entries and static search indexes. This can be useful when deciding on the next page to crawl or building up a list of pages on a website.
Here we are processing the sitemap into a set ofFeedEntry-DTOs:
(new \Spekulatius\PHPScraper\PHPScraper) ->go('https://phpscraper.de') ->sitemap// array(131) {// [0]=>// object(Spekulatius\PHPScraper\DataTransferObjects\FeedEntry)#165 (3) {// ["title"]=>// string(0) ""// ["description"]=>// string(0) ""// ["link"]=>// string(22) "https://phpscraper.de/"// }// [1]=>// ...
Whenever post-processing is applied, you can fall back to the underlying*Raw-methods.
PHPScraper comes out of the box with file / URL processing methods for CSV-, XML- and JSON:
parseJsonparseXmlparseCsvparseCsvWithHeader(generates an asso. array using the first row)
Each method can process both strings as well as URLs:
// Parse JSON into array:$json =$web->parseJson('[{"title": "PHP Scraper: a web utility for PHP", "url": "https://phpscraper.de"}]');// [// 'title' => 'PHP Scraper: a web utility for PHP',// 'url' => 'https://phpscraper.de'// ]// Fetch and parse CSV into a simple array:$csv =$web->parseCsv('https://test-pages.phpscraper.de/test.csv');// [// ['date', 'value'],// ['1945-02-06', 4.20],// ['1952-03-11', 42],// ]// Fetch and parse CSV with first row as header into an asso. array structure:$csv =$web->parseCsvWithHeader('https://test-pages.phpscraper.de/test.csv');// [// ['date' => '1945-02-06', 'value' => 4.20],// ['date' => '1952-03-11', 'value' => 42],// ]
Additional CSV parsing parameters such as separator, enclosure and escape are possible.
There are plenty of examples on thePHPScraper website and in thetests.
Check theplayground.php if you prefer learning by doing. You get it up and running with:
$ git clone git@github.com:spekulatius/PHPScraper.git&& composer updateThe future development is organized intomilestones. Releases followsemver.
- Improve documentation and examples.
- Organize code better (move websites into separate repos, etc.)
- Add support for feeds and some typical file types.
- Switch from Goutte toSymfony BrowserKit. Goutte has been archived.
- Expand to parse a wider range of types, elements, embeds, etc.
- Improve performance with caching and concurrent fetching of assets
- Minor improvements for parsing methods
TBC.
PHPScraper is sponsored by:
With your support, PHPScraper can became thePHP swiss army knife for the web. If you find PHPScraper useful to your work, please consider asponsorship ordonation. Thank you 💪
If needed, you can use the following configuration options:
You can set the browser agent usingsetConfig:
$web->setConfig(['agent' =>'Mozilla/5.0 (X11; Linux x86_64; rv:107.0) Gecko/20100101 Firefox/107.0']);
It defaults toMozilla/5.0 (compatible; PHP Scraper/1.x; +https://phpscraper.de).
You can configure proxy support withsetConfig:
$web->setConfig(['proxy' =>'http://user:password@127.0.0.1:3128']);
You can set thetimeout usingsetConfig:
$web->setConfig(['timeout' =>15]);
Setting the timeout to zero will disable it.
While unrecommended, it might be required to disable SSL checks. You can do so using:
$web->setConfig(['disable_ssl' =>true]);
You can callsetConfig multiple times. It stores the config and merges it with previous settings. This should be kept in mind in the unlikely use-case when unsetting values.
composer require spekulatius/phpscraper
After the installation, the package will be picked up by the Composer autoloader. If you are using a common PHP application or framework such as Laravel or Symfony you can start scraping now 🚀
If not or you are building a standalone-scraper, please include the autoloader invendor/ at the top of your file:
<?phprequire__DIR__ .'/vendor/autoload.php';// ...
Now you can now use any of the examples on the documentation website or from thetests/-folder.
Please consider supporting PHPScraper with a star orsponsorship:
composer thanks
Thank you 💪
The library comes with a PHPUnit test suite. To run the tests, run the following command from the project folder:
composertestYou can find the testshere. The test pages arepublicly available.
About
A universal web-util for PHP.