From 49da67cb33669c07c8a0ab5d4716e22be807a654 Mon Sep 17 00:00:00 2001
From: Roliga <roliga.here@gmail.com>
Date: Mon, 26 Nov 2018 18:05:41 +0100
Subject: [PATCH] core: Automatically select a bridge based on a URL (#928)

* core: Add bridge parameter auto detection

This adds a new 'detect' action which accepts a URL from which an
appropriate bridge is selected and relevant parameters are extracted.
The user is then automatically redirected to the selected bridge.

For example to get a feed from: https://twitter.com/search?q=%23rss-bridge
we could send a request to:
'/?action=detect&format=Atom&url=twitter.com/search%3Fq%3D%2523rss-bridge'
which would redirect to:
'/?action=display&q=%23rss-bridge&bridge=Twitter&format=Atom'.

This auto detection happens on a per-bridge basis, so a new function
'detectParameters' is added to BridgeInterface which bridges may implement.
It takes a URL for an argument and returns a list of parameters that were
extracted, or null if the URL isn't relevant for the bridge.

* [TwitterBridge] Add parameter auto detection

* [BridgeAbstract] Add generic parameter detection

This adds generic "paramater detection" for bridges that don't have any
parameters defined. If the queried URL matches the URI defined in the
bridge (ignoring https://, www. and trailing /) an emtpy list of parameters is
returned.
---
 bridges/TwitterBridge.php | 35 +++++++++++++++++++++++++++++++++++
 index.php                 | 36 ++++++++++++++++++++++++++++++++++++
 lib/BridgeAbstract.php    | 13 +++++++++++++
 lib/BridgeInterface.php   |  8 ++++++++
 4 files changed, 92 insertions(+)

diff --git a/bridges/TwitterBridge.php b/bridges/TwitterBridge.php
index fc7b78bd..20f90d74 100644
--- a/bridges/TwitterBridge.php
+++ b/bridges/TwitterBridge.php
@@ -66,6 +66,41 @@ class TwitterBridge extends BridgeAbstract {
 		)
 	);
 
+	public function detectParameters($url){
+		$params = array();
+
+		// By keyword or hashtag (search)
+		$regex = '/^(https?:\/\/)?(www\.)?twitter\.com\/search.*(\?|&)q=([^\/&?\n]+)/';
+		if(preg_match($regex, $url, $matches) > 0) {
+			$params['q'] = urldecode($matches[4]);
+			return $params;
+		}
+
+		// By hashtag
+		$regex = '/^(https?:\/\/)?(www\.)?twitter\.com\/hashtag\/([^\/?\n]+)/';
+		if(preg_match($regex, $url, $matches) > 0) {
+			$params['q'] = urldecode($matches[3]);
+			return $params;
+		}
+
+		// By list
+		$regex = '/^(https?:\/\/)?(www\.)?twitter\.com\/([^\/?\n]+)\/lists\/([^\/?\n]+)/';
+		if(preg_match($regex, $url, $matches) > 0) {
+			$params['user'] = urldecode($matches[3]);
+			$params['list'] = urldecode($matches[4]);
+			return $params;
+		}
+
+		// By username
+		$regex = '/^(https?:\/\/)?(www\.)?twitter\.com\/([^\/?\n]+)/';
+		if(preg_match($regex, $url, $matches) > 0) {
+			$params['u'] = urldecode($matches[3]);
+			return $params;
+		}
+
+		return null;
+	}
+
 	public function getName(){
 		switch($this->queriedContext) {
 		case 'By keyword or hashtag':
diff --git a/index.php b/index.php
index c093644a..772f4942 100644
--- a/index.php
+++ b/index.php
@@ -96,6 +96,42 @@ try {
 		header('Content-Type: application/json');
 		echo json_encode($list, JSON_PRETTY_PRINT);
 
+	} elseif($action === 'detect') {
+
+		$targetURL = $params['url']
+			or returnClientError('You must specify a url!');
+
+		$format = $params['format']
+			or returnClientError('You must specify a format!');
+
+		foreach(Bridge::listBridges() as $bridgeName) {
+
+			if(!Bridge::isWhitelisted($bridgeName)) {
+				continue;
+			}
+
+			$bridge = Bridge::create($bridgeName);
+
+			if($bridge === false) {
+				continue;
+			}
+
+			$bridgeParams = $bridge->detectParameters($targetURL);
+
+			if(is_null($bridgeParams)) {
+				continue;
+			}
+
+			$bridgeParams['bridge'] = $bridgeName;
+			$bridgeParams['format'] = $format;
+
+			header('Location: ?action=display&' . http_build_query($bridgeParams), true, 301);
+			die();
+
+		}
+
+		returnClientError('No bridge found for given URL: ' . $targetURL);
+
 	} elseif($action === 'display' && !empty($bridge)) {
 
 		$format = $params['format']
diff --git a/lib/BridgeAbstract.php b/lib/BridgeAbstract.php
index b02280e4..a2b15802 100644
--- a/lib/BridgeAbstract.php
+++ b/lib/BridgeAbstract.php
@@ -278,4 +278,17 @@ abstract class BridgeAbstract implements BridgeInterface {
 		return static::CACHE_TIMEOUT;
 	}
 
+	/** {@inheritdoc} */
+	public function detectParameters($url){
+		$regex = '/^(https?:\/\/)?(www\.)?(.+?)(\/)?$/';
+		if(empty(static::PARAMETERS)
+		&& preg_match($regex, $url, $urlMatches) > 0
+		&& preg_match($regex, static::URI, $bridgeUriMatches) > 0
+		&& $urlMatches[3] === $bridgeUriMatches[3]) {
+			return array();
+		} else {
+			return null;
+		}
+	}
+
 }
diff --git a/lib/BridgeInterface.php b/lib/BridgeInterface.php
index bc48fe04..803bb3cd 100644
--- a/lib/BridgeInterface.php
+++ b/lib/BridgeInterface.php
@@ -114,4 +114,12 @@ interface BridgeInterface {
 	 * @return int Cache timeout
 	 */
 	public function getCacheTimeout();
+
+	/**
+	 * Returns parameters from given URL or null if URL is not applicable
+	 *
+	 * @param string $url URL to extract parameters from
+	 * @return array|null List of bridge parameters or null if detection failed.
+	 */
+	public function detectParameters($url);
 }