librarian.php

<?php
    /*
     * RSS-Librarian - A read-it-later service for RSS purists
     *
     * https://github.com/thefranke/rss-librarian
     *
     */

    use fivefilters\Readability\Readability;
    use fivefilters\Readability\Configuration;
    use fivefilters\Readability\ParseException;

    // Global configuration
    $g_max_items = 100;
    $g_url_base = 'http' . (isset($_SERVER['HTTPS']) ? 's' : '') . '://' . $_SERVER['HTTP_HOST'];
    $g_url_librarian = $g_url_base . $_SERVER["PHP_SELF"];
    $g_dir_feeds = "feeds";


// echo "<pre>";
// echo $g_url_base."<br>";
// echo $g_url_librarian."<br>";
// echo $g_dir_feeds."<br>";
// echo "</pre>";


    // Fetch parameters given to librarian
    function fetch_param($param)
    {
        $params = $_GET;

        foreach($params as $k => $v)
        {
            if ($k == $param)
                return $v;
        }

        return "";
    }

    // Turn user ID into feed ID
    function get_feed_id($user_id)
    {
        return $user_id;

        // Later feature: Disconnect feed from user
        //return hash('sha256', $user_id);
    }

    // Produce path for local feed file
    function get_local_feed_file($param_id)
    {
        global $g_dir_feeds;
        $file_hash = get_feed_id($param_id);
        return $g_dir_feeds . "/" . $file_hash . ".xml";
    }

    // Update feed files with new header
    function update_feed_file($user_id)
    {
        global $g_dir_feeds;
        global $g_url_librarian;

        // check for subs dir
        if (!is_dir($g_dir_feeds))
            mkdir($g_dir_feeds);

        $personal_url = $g_url_librarian . '?id=' . $user_id;

        // recreate base file so changes in the header are put in with every new release
        $new_rss_base_text = '<?xml version="1.0" encoding="utf-8"?>
        <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
            <channel>
                <title>RSS-Librarian (' . substr($user_id, 0, 4) . ')</title>
                <description>A read-it-later service for RSS purists</description>
                <link>' . $personal_url . '</link>
                <atom:link href="' . $personal_url . '" rel="self" type="application/rss+xml" />
            </channel>
        </rss>
        ';

        $rss_xml = simplexml_load_string($new_rss_base_text);
        $local_feed_file = get_local_feed_file($user_id);

        // try to open local subscriptions and copy items over
        $local_feed_text = @file_get_contents($local_feed_file);
        if ($local_feed_text != "")
        {
            $old_rss_xml = simplexml_load_string($local_feed_text);

            foreach($old_rss_xml->channel->item as $item)
                sxml_append($rss_xml->channel, $item);
        }

        return $rss_xml;
    }

    // Helper to attach new XML element to existing one
    function sxml_append(SimpleXMLElement $to, SimpleXMLElement $from)
    {
        $toDom = dom_import_simplexml($to);
        $fromDom = dom_import_simplexml($from);
        $new_node = $toDom->ownerDocument->importNode($fromDom, true);
        $firstSibling = $toDom->getElementsByTagName('item')->item(0);
        $toDom->insertBefore($new_node, $firstSibling);
    }

    // Create XML element for an RSS item
    function make_feed_item($url, $title, $author, $content)
    {
        $pub_date = date("D, d M Y H:i:s T");

        $xmlstr = '
        <item>
            <link>' . $url . '</link>
            <title>' . $title . '</title>

            <guid isPermaLink="true">' . $url .'</guid>
            <description>'
                . htmlspecialchars($content) .
            '</description>
            <author>' . $author . '</author>
            <pubDate>' . $pub_date . '</pubDate>
        </item>
        ';

        return new SimpleXMLElement($xmlstr);
    }

    // Extract content by piping through Readability.php
    function extract_readability($url)
    {
        $autoload = __DIR__ . '/vendor/autoload.php';

        $html = "";

        if (file_exists($autoload))
        {
            require $autoload;
            // Pretend to be a browser to have an increased success rate of
            // downloading the contents compared to a simple `file_get_contents`.
            // See https://stackoverflow.com/a/11680776.
            $ch = curl_init();
            curl_setopt($ch, CURLOPT_URL, $url);
            curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
            curl_setopt($ch,CURLOPT_USERAGENT,'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.13) Gecko/20080311 Firefox/2.0.0.13');
            $html = curl_exec($ch);
            curl_close($ch);
        }

        // No local Readability.php installed, use FiveFilters
        if ($html == "")
        {
            //$feed_url = "https://darch.dk/read/Readability.php?url=" . urlencode($url);

            //echo "using ftr";
            $feed_url = "https://ftr.fivefilters.net/makefulltextfeed.php?url=" . urlencode($url);

            $feed_item = file_get_contents($feed_url);

            // error handling remove everything until first <
            $start = strpos($feed_item, "<");
            $feed_item = substr($feed_item, $start);
            $xml = simplexml_load_string($feed_item);
            $ff_item = $xml->channel->item[0];

            $title = $ff_item->title;
            $content = $ff_item->description;
            $author = "";
            return make_feed_item($url, $title, $author, $content);
        }

        if (function_exists('tidy_parse_string'))
        {
            $tidy = tidy_parse_string($html, array(), 'UTF8');
            $tidy->cleanRepair();
            $html = $tidy->value;
        }

        $readability = new Readability(new Configuration([
            'fixRelativeURLs' => true,
            'originalURL' => $url,
        ]));

        $item = "";
        try
        {
            $readability->parse($html);

            $title = $readability->getTitle();
            $content = $readability->getContent();
            $author = $readability->getAuthor();
            $item = make_feed_item($url, $title, $author, $content);
        }
        catch (ParseException $e)
        {
            $item = make_feed_item($url, $url, $url, "Could not extract content: " . $e->getMessage());
        }

        return $item;
    }

    // Add URL to personal feed
    function add_url($user_id, $param_url)
    {
        global $g_max_items;

        $local_feed_file = get_local_feed_file($user_id);

        $xml = update_feed_file($user_id);

        if ($xml->channel->item)
        {
            // check max item count, remove anything beyond
            $c = $xml->channel->item->count();
            while ($c > $g_max_items)
            {
                unset($xml->channel->item[$c - 1]);
                $c--;
            }

            // TODO: Shoud these two checks not be swapped?
            // If a URL already on the list then there is no need to check if it list is maxed out?

            // check if item already exists
            foreach($xml->channel->item as $item)
            {
                if ($item->link == $param_url)
                    return '<p class="notice">' . $param_url . ' already added</p>';
            }
        }

        // fetch rss content and add
        $item = extract_readability($param_url);
        sxml_append($xml->channel, $item);

        // write to rss file
        file_put_contents($local_feed_file, $xml->asXml());
        return '<p class="notice"><a href="' . $param_url . '">' . $param_url . '</a> added</p>';
    }

    // Remove URL from personal feed
    function delete_url($user_id, $param_remove)
    {
        $local_feed_file = get_local_feed_file($user_id);

        $xml = update_feed_file($user_id);

        if ($xml->channel->item)
        {
            // Find and remove item
            $c = $xml->channel->item->count();
            for( $i = 0; $i < $c; $i++)
            {
                $item = $xml->channel->item[$i];

                if ($item->link == $param_remove)
                {
                    //echo "<p>Remove [" . $i . "]" . $param_remove ."</p>";
                    unset($xml->channel->item[$i]);
                } /*else
                    return '<p class="notice">Cannot delete link to <code>' . $param_remove .'</code>, since it was not fond in the list</p>';
                    */
            }
        }

        // write to rss file
        file_put_contents($local_feed_file, $xml->asXml());
        return '<p class="notice">Link to <a href="' . $param_remove . '">' . $param_remove . '</a> was removed</p>';
    }

    // Count number of feeds in feed directory
    function count_feeds()
    {
        global $g_dir_feeds;
        $filecount = count(glob($g_dir_feeds . "/*.xml"));
        return $filecount;
    }
?>

<!DOCTYPE html>
<html lang="en">
<head>
  <title>RSS-Librarian</title>
  <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  <link rel="shortcut icon" href="favicon.png">
  <?php
    $param_url = fetch_param("url");
    $param_remove = fetch_param("remove");
    $param_id = fetch_param("id");
    $user_id = $param_id;

    // user exists?
    if ($param_id != "")
        print('<link rel="alternate" type="application/rss+xml" title="RSS Librarian (' . substr($param_id, 0, 4) . ')" href="' . get_local_feed_file($param_id) . '">');
    else
        $user_id = hash('sha256', random_bytes(18));
  ?>

  <style>

    /*
    html {
      font-family: monospace;
      font-size: 12pt;
      color: #66397C;
      text-align: center;
    }
    #main {
      text-align: center;
      margin: auto;
      max-width: 75%;
      display: inline-block;
    }
    hr {
      border: 1px dashed;
    }
    a:link, a:visited {
      color: #66397C;
    }
    ul {
      margin: 10px;
    }
    #header {
        text-align: center;
        margin: 24pt;
    }
    h1, h2, h3, h4 {
        margin: 0;
    }
    */

    html {
        font-family: sans-serif;
        line-height: 1.5;
        margin: 1rem;
    }

    body {
        max-width: 900px;
        margin: 1rem auto;
    }

    header {
        text-align: center;
    }

    header h1 {
        /*border-bottom: thin dashed;*/
        margin-bottom: 0.75rem;
    }

    header img {
        height: 1.5rem;
    }

    header .nav {
        border-top: thin dashed;
        padding: 0.75rem 1.5rem;
    }

    .notice {
        background-color: #ffdd55;
        border-radius: 0.5rem;
        padding: 0.5rem 1rem;
    }

    form {
        margin-top: 1rem;
        margin-bottom: 1.5rem;
        margin: 1.5rem 0;
        text-align: center;
    }

    fieldset {
        border-radius: 0.5rem;
        margin: 0;
    }

    legend {
        font-weight: bold;
        padding: 0 0.5rem;
    }

    input[type=text] {
        width: 100%;
        moz-box-sizing:border-box;
        -webkit-box-sizing:border-box;
        box-sizing:border-box;
        margin-bottom: 0.5rem;
    }

    details {
        margin: 1rem 0;
        border: thin solid lightgray;
        border-radius: 0.5rem;        
        padding: 0.5rem 1rem;
    }

    details img {
        max-width: 100%;
        margin: auto;
    }

    .title {
        font-weight: bold;
    }

    a.remove {
        color: tomato;
        text-decoration: none;
        float: right;

    }

    .content {
        margin-top: 0.5rem;
        /*
        border: thin solid lightgray;
        border-radius: 0.5rem;
        padding: 0.5rem 1rem;
        */
    }

    .date {
        font-size: small;
        color: gray;
        border-bottom: thin dashed;
        display: block;
        padding-bottom: 0.25rem;
        margin-bottom:  0.25rem;
    }

    footer {
        font-size: small;    
        border-top: thin dashed;
        text-align: center;
        padding-top: 0.5rem;
    }

    @media (prefers-color-scheme: dark) {
        html {
            background-color: #000;
            color: #99c683;
        }
        a:link, a:visited {
            color: #99c683;
        }
        img {
            filter:invert(1);
        }
    }
  </style>
</head>
<body>
    <main>
        <header>

            <h1><img alt="logo" src="https://raw.githubusercontent.com/Warhammer40kGroup/wh40k-icon/master/src/svgs/librarius-02.svg"> RSS-Librarian</h1>

<?php
    // Check if parameter was supplied to distinguish new users from existing ones
    if ($param_id != "")
    {
        $personal_url = $g_url_librarian . '?id=' . $user_id;
        $feed_id = get_feed_id($user_id);
        $local_feed_file = get_local_feed_file($user_id);
        //print_r('<h4>Managing articles of ' . substr($user_id, 0, 4) . '...' . substr($user_id, -4) .':</h4>');
        print_r('<span class="nav">Your <a href="'. $personal_url .'">personal URL</a> and <a href="' . $local_feed_file . '">personal RSS feed</a></span>');
    }
?>      </header>


        <form action="<?= $g_url_librarian ?>">
            <fieldset>
                <legend>Add URL to reading list</legend>
                <input type="text" id="url" name="url">
                <input type="hidden" id="id" name="id" value="<?= $user_id ?>">
                <input type="submit" value="Add to feed">
            </fieldset>
        </form>

<?php
        if ($param_url != "")
        {
            $result = add_url($user_id, $param_url);
            print_r($result);
        }

        if ($param_remove != "")
        {
            $result = delete_url($user_id, $param_remove);
            print_r($result);
        }

    // Build-in Reader
    // based on: https://gist.github.com/mburst/5230448

    if ($param_id) {

        $feed_url = $g_url_base  . '/read/' .  $local_feed_file; // TODO: Make not hardcoded by detecting propper path
        
        //echo '<p>'.$feed_url;

        $entries = array();
        $xml = simplexml_load_file($feed_url);
        $entries = array_merge($entries, $xml->xpath("//item"));

        //Sort feed entries by pubDate
        usort($entries, function ($feed1, $feed2) {
            return strtotime($feed2->pubDate) - strtotime($feed1->pubDate);
        });

    }

    //echo "<ol>";

    //Print all the entries
    foreach($entries as $entry)
    {
        ?>
        <!-- <li> -->
            <details>
                <summary>
                    <span class="title"><?= $entry->title ?></span>
                    (<a href="<?= $entry->link ?>" target="_blank"><?= parse_url($entry->link)['host'] ?></a>)
                    <a href="<?= $personal_url . "&remove=" .$entry->link ?>" class="remove">[x]</a>
                </summary>
                            
                <div class="content">
                    <span class="date"><?= strftime('%d/%m/%Y %H:%M', strtotime($entry->pubDate)) ?></span>
                    <?= $entry->description ?>
                </div>

            </details>
        <!-- </li> -->
    <?php
    } 

    // echo "</ol>";
?>
    
    <footer>
        <a href="https://github.com/thefranke/rss-librarian">RSS-Librarian on Github</a>
        | Instance managing <?php echo count_feeds() ?> feeds
<?php

    if ($param_id)
    {
        print_r(' | ');
        print_r('More tools: <a href="javascript:window.location.href=\'' . $personal_url . '&url=\' + window.location.href">Feed boomarklet</a>, <a href="https://feedreader.xyz/?url=' . urlencode($g_url_base . '/' . $local_feed_file) . '">Feed preview</a>');
        // BUG: Not working if rss-lib is in a folder like: domain.tld/reader/
    }
?>
    </footer>

    </main>
</body>

</html>