Knah-Tsaeb/MyShaarli
Knah-Tsaeb/MyShaarli
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

namespacing: \Shaarli\Http\Url

Browse source 
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
This commit is contained in:
VirtualTam 2018-12-03 00:23:35 +01:00
parent 00af48d9d2
commit fb1b182fbf
4 changed files with 234 additions and 227 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
application/HttpUtils.php
View file

@ -1,4 +1,7 @@
<?php
use Shaarli\Http\Url;
/**
* GET an HTTP URL to retrieve its content
* Uses the cURL library or a fallback method

218
application/Url.php
View file

@ -34,7 +34,7 @@ function unparse_url($parsedUrl)
*/
function cleanup_url($url)
{
$obj_url = new Url($url);
$obj_url = new \Shaarli\Http\Url($url);
return $obj_url->cleanup();
}
@ -47,7 +47,7 @@ function cleanup_url($url)
*/
function get_url_scheme($url)
{
$obj_url = new Url($url);
$obj_url = new \Shaarli\Http\Url($url);
return $obj_url->getScheme();
}
@ -86,217 +86,3 @@ function whitelist_protocols($url, $protocols)
}
return $url;
}
/**
* URL representation and cleanup utilities
*
* Form
* scheme://[username:password@]host[:port][/path][?query][#fragment]
*
* Examples
* http://username:password@hostname:9090/path?arg1=value1&arg2=value2#anchor
* https://host.name.tld
* https://h2.g2/faq/?vendor=hitchhiker&item=guide&dest=galaxy#answer
*
* @see http://www.faqs.org/rfcs/rfc3986.html
*/
class Url
{
private static $annoyingQueryParams = array(
// Facebook
'action_object_map=',
'action_ref_map=',
'action_type_map=',
'fb_',
'fb=',
'PHPSESSID=',
// Scoop.it
'__scoop',
// Google Analytics & FeedProxy
'utm_',
// ATInternet
'xtor=',
// Other
'campaign_'
);
private static $annoyingFragments = array(
// ATInternet
'xtor=RSS-',
// Misc.
'tk.rss_all'
);
/*
* URL parts represented as an array
*
* @see http://php.net/parse_url
*/
protected $parts;
/**
* Parses a string containing a URL
*
* @param string $url a string containing a URL
*/
public function __construct($url)
{
$url = self::cleanupUnparsedUrl(trim($url));
$this->parts = parse_url($url);
if (!empty($url) && empty($this->parts['scheme'])) {
$this->parts['scheme'] = 'http';
}
}
/**
* Clean up URL before it's parsed.
* ie. handle urlencode, url prefixes, etc.
*
* @param string $url URL to clean.
*
* @return string cleaned URL.
*/
protected static function cleanupUnparsedUrl($url)
{
return self::removeFirefoxAboutReader($url);
}
/**
* Remove Firefox Reader prefix if it's present.
*
* @param string $input url
*
* @return string cleaned url
*/
protected static function removeFirefoxAboutReader($input)
{
$firefoxPrefix = 'about://reader?url=';
if (startsWith($input, $firefoxPrefix)) {
return urldecode(ltrim($input, $firefoxPrefix));
}
return $input;
}
/**
* Returns a string representation of this URL
*/
public function toString()
{
return unparse_url($this->parts);
}
/**
* Removes undesired query parameters
*/
protected function cleanupQuery()
{
if (! isset($this->parts['query'])) {
return;
}
$queryParams = explode('&', $this->parts['query']);
foreach (self::$annoyingQueryParams as $annoying) {
foreach ($queryParams as $param) {
if (startsWith($param, $annoying)) {
$queryParams = array_diff($queryParams, array($param));
continue;
}
}
}
if (count($queryParams) == 0) {
unset($this->parts['query']);
return;
}
$this->parts['query'] = implode('&', $queryParams);
}
/**
* Removes undesired fragments
*/
protected function cleanupFragment()
{
if (! isset($this->parts['fragment'])) {
return;
}
foreach (self::$annoyingFragments as $annoying) {
if (startsWith($this->parts['fragment'], $annoying)) {
unset($this->parts['fragment']);
break;
}
}
}
/**
* Removes undesired query parameters and fragments
*
* @return string the string representation of this URL after cleanup
*/
public function cleanup()
{
$this->cleanupQuery();
$this->cleanupFragment();
return $this->toString();
}
/**
* Converts an URL with an International Domain Name host to a ASCII one.
* This requires PHP-intl. If it's not available, just returns this->cleanup().
*
* @return string converted cleaned up URL.
*/
public function idnToAscii()
{
$out = $this->cleanup();
if (! function_exists('idn_to_ascii') || ! isset($this->parts['host'])) {
return $out;
}
$asciiHost = idn_to_ascii($this->parts['host'], 0, INTL_IDNA_VARIANT_UTS46);
return str_replace($this->parts['host'], $asciiHost, $out);
}
/**
* Get URL scheme.
*
* @return string the URL scheme or false if none is provided.
*/
public function getScheme()
{
if (!isset($this->parts['scheme'])) {
return false;
}
return $this->parts['scheme'];
}
/**
* Get URL host.
*
* @return string the URL host or false if none is provided.
*/
public function getHost()
{
if (empty($this->parts['host'])) {
return false;
}
return $this->parts['host'];
}
/**
* Test if the Url is an HTTP one.
*
* @return true is HTTP, false otherwise.
*/
public function isHttp()
{
return strpos(strtolower($this->parts['scheme']), 'http') !== false;
}
}

217
application/http/Url.php Normal file
View file

@ -0,0 +1,217 @@
<?php
namespace Shaarli\Http;
/**
* URL representation and cleanup utilities
*
* Form
* scheme://[username:password@]host[:port][/path][?query][#fragment]
*
* Examples
* http://username:password@hostname:9090/path?arg1=value1&arg2=value2#anchor
* https://host.name.tld
* https://h2.g2/faq/?vendor=hitchhiker&item=guide&dest=galaxy#answer
*
* @see http://www.faqs.org/rfcs/rfc3986.html
*/
class Url
{
private static $annoyingQueryParams = array(
// Facebook
'action_object_map=',
'action_ref_map=',
'action_type_map=',
'fb_',
'fb=',
'PHPSESSID=',
// Scoop.it
'__scoop',
// Google Analytics & FeedProxy
'utm_',
// ATInternet
'xtor=',
// Other
'campaign_'
);
private static $annoyingFragments = array(
// ATInternet
'xtor=RSS-',
// Misc.
'tk.rss_all'
);
/*
* URL parts represented as an array
*
* @see http://php.net/parse_url
*/
protected $parts;
/**
* Parses a string containing a URL
*
* @param string $url a string containing a URL
*/
public function __construct($url)
{
$url = self::cleanupUnparsedUrl(trim($url));
$this->parts = parse_url($url);
if (!empty($url) && empty($this->parts['scheme'])) {
$this->parts['scheme'] = 'http';
}
}
/**
* Clean up URL before it's parsed.
* ie. handle urlencode, url prefixes, etc.
*
* @param string $url URL to clean.
*
* @return string cleaned URL.
*/
protected static function cleanupUnparsedUrl($url)
{
return self::removeFirefoxAboutReader($url);
}
/**
* Remove Firefox Reader prefix if it's present.
*
* @param string $input url
*
* @return string cleaned url
*/
protected static function removeFirefoxAboutReader($input)
{
$firefoxPrefix = 'about://reader?url=';
if (startsWith($input, $firefoxPrefix)) {
return urldecode(ltrim($input, $firefoxPrefix));
}
return $input;
}
/**
* Returns a string representation of this URL
*/
public function toString()
{
return unparse_url($this->parts);
}
/**
* Removes undesired query parameters
*/
protected function cleanupQuery()
{
if (!isset($this->parts['query'])) {
return;
}
$queryParams = explode('&', $this->parts['query']);
foreach (self::$annoyingQueryParams as $annoying) {
foreach ($queryParams as $param) {
if (startsWith($param, $annoying)) {
$queryParams = array_diff($queryParams, array($param));
continue;
}
}
}
if (count($queryParams) == 0) {
unset($this->parts['query']);
return;
}
$this->parts['query'] = implode('&', $queryParams);
}
/**
* Removes undesired fragments
*/
protected function cleanupFragment()
{
if (!isset($this->parts['fragment'])) {
return;
}
foreach (self::$annoyingFragments as $annoying) {
if (startsWith($this->parts['fragment'], $annoying)) {
unset($this->parts['fragment']);
break;
}
}
}
/**
* Removes undesired query parameters and fragments
*
* @return string the string representation of this URL after cleanup
*/
public function cleanup()
{
$this->cleanupQuery();
$this->cleanupFragment();
return $this->toString();
}
/**
* Converts an URL with an International Domain Name host to a ASCII one.
* This requires PHP-intl. If it's not available, just returns this->cleanup().
*
* @return string converted cleaned up URL.
*/
public function idnToAscii()
{
$out = $this->cleanup();
if (!function_exists('idn_to_ascii') || !isset($this->parts['host'])) {
return $out;
}
$asciiHost = idn_to_ascii($this->parts['host'], 0, INTL_IDNA_VARIANT_UTS46);
return str_replace($this->parts['host'], $asciiHost, $out);
}
/**
* Get URL scheme.
*
* @return string the URL scheme or false if none is provided.
*/
public function getScheme()
{
if (!isset($this->parts['scheme'])) {
return false;
}
return $this->parts['scheme'];
}
/**
* Get URL host.
*
* @return string the URL host or false if none is provided.
*/
public function getHost()
{
if (empty($this->parts['host'])) {
return false;
}
return $this->parts['host'];
}
/**
* Test if the Url is an HTTP one.
*
* @return true is HTTP, false otherwise.
*/
public function isHttp()
{
return strpos(strtolower($this->parts['scheme']), 'http') !== false;
}
}

23
tests/Url/UrlTest.php → tests/http/UrlTest.php
View file

@ -3,12 +3,13 @@
* Url's tests
*/
require_once 'application/Url.php';
namespace Shaarli\Http;
/**
* Unitary tests for URL utilities
*/
class UrlTest extends PHPUnit_Framework_TestCase
class UrlTest extends \PHPUnit\Framework\TestCase
{
// base URL for tests
protected static $baseUrl = 'http://domain.tld:3000';
@ -18,7 +19,7 @@ class UrlTest extends PHPUnit_Framework_TestCase
*/
private function assertUrlIsCleaned($query = '', $fragment = '')
{
$url = new Url(self::$baseUrl.$query.$fragment);
$url = new Url(self::$baseUrl . $query . $fragment);
$url->cleanup();
$this->assertEquals(self::$baseUrl, $url->toString());
}
@ -38,7 +39,7 @@ public function testEmptyConstruct()
public function testConstruct()
{
$ref = 'http://username:password@hostname:9090/path'
.'?arg1=value1&arg2=value2#anchor';
. '?arg1=value1&arg2=value2#anchor';
$url = new Url($ref);
$this->assertEquals($ref, $url->toString());
}
@ -52,7 +53,7 @@ public function testNoCleanup()
$this->assertUrlIsCleaned();
// URL with no annoying elements
$ref = self::$baseUrl.'?p1=val1&p2=1234#edit';
$ref = self::$baseUrl . '?p1=val1&p2=1234#edit';
$url = new Url($ref);
$this->assertEquals($ref, $url->cleanup());
}
@ -115,26 +116,26 @@ public function testCleanupMixedContent()
// ditch annoying query params and fragment, keep useful params
$url = new Url(
self::$baseUrl
.'?fb=zomg&my=stuff&utm_medium=numnum&is=kept#tk.rss_all'
. '?fb=zomg&my=stuff&utm_medium=numnum&is=kept#tk.rss_all'
);
$this->assertEquals(self::$baseUrl.'?my=stuff&is=kept', $url->cleanup());
$this->assertEquals(self::$baseUrl . '?my=stuff&is=kept', $url->cleanup());
// ditch annoying query params, keep useful params and fragment
$url = new Url(
self::$baseUrl
.'?fb=zomg&my=stuff&utm_medium=numnum&is=kept#again'
. '?fb=zomg&my=stuff&utm_medium=numnum&is=kept#again'
);
$this->assertEquals(
self::$baseUrl.'?my=stuff&is=kept#again',
self::$baseUrl . '?my=stuff&is=kept#again',
$url->cleanup()
);
// test firefox reader url
$url = new Url(
'about://reader?url=' . urlencode(self::$baseUrl .'?my=stuff&is=kept')
'about://reader?url=' . urlencode(self::$baseUrl . '?my=stuff&is=kept')
);
$this->assertEquals(self::$baseUrl.'?my=stuff&is=kept', $url->cleanup());
$this->assertEquals(self::$baseUrl . '?my=stuff&is=kept', $url->cleanup());
}
/**