From eb61c1bae2ffad114e17393d91d86fb00d7ac8cd Mon Sep 17 00:00:00 2001
From: Hannah Isopod <me@bathynomus.xyz>
Date: Sat, 10 May 2025 03:03:58 -0600
Subject: Initialer commit Just remade the repo because I don't *need* need a
 comprehensive source history for my personal website and this is easier than
 getting filter-repo to work good

---
 stuff/rssfilter.php | 288 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 288 insertions(+)
 create mode 100644 stuff/rssfilter.php

(limited to 'stuff/rssfilter.php')

diff --git a/stuff/rssfilter.php b/stuff/rssfilter.php
new file mode 100644
index 0000000..ea749cd
--- /dev/null
+++ b/stuff/rssfilter.php
@@ -0,0 +1,288 @@
+<?php
+
+// Here's my embarrassing ass source code. It is public domain, do with it what you will
+
+/*ini_set('display_errors', 1);
+ini_set('display_startup_errors', 1);
+error_reporting(E_ALL);*/
+
+if($_GET['downloadsource'] == 'yes') { // just serve the source code as plain text and exit
+  header('Content-Type: text/plain');
+  echo file_get_contents('rssfilter.php');
+  die();
+}
+
+$feed = $_GET['feed'];
+$filter = $_GET['filter'];
+$filtertype = $_GET['type'];
+$useregex = false;
+if($_GET['regex'] == 'yes' or $_GET['regex'] == 'y') {
+  $useregex = true;
+}
+
+function rss_testforfilter($item) { // I wrote two different functions for filtering RSS and Atom feeds under the assumption that they would be radically different 
+                                    // in some way. The only real difference ended up being how the categories are stored within the category tags. I could probably 
+                                    // consolidate them into one function, but these boolean expressions are big and complicated and I'm too scared to touch it now.
+
+  global $useregex; // PHP is silly and makes you re-declare global variables if you want to use them inside functions.
+  global $filter;
+  foreach($item->category as $category) { // loops through all the <category> tags and returns true if any of them contain the filter string
+    if(
+      (!$useregex and strpos($category, $filter) !== false) or // Don't use regex and filter string is found
+      ($useregex and preg_match($filter, $category))           // DO use regex and filter expression is found
+      ) {
+      return true;
+    }
+  }
+  if(
+    (!$useregex and (strpos($item->description, $filter) !== false or strpos($item->title, $filter) !== false)) or // Don't use regex and filter string is found
+    ($useregex and (preg_match($filter, $item->description) or preg_match($filter, $item->title)))                 // DO use regex and filter expression is found
+    ) {
+    return true;
+  }
+  return false;
+}
+
+function atom_testforfilter($entry) {
+  global $useregex;
+  global $filter;
+  foreach($entry->category as $category) { // loops through all the <category> tags and returns true if any of them contain the filter string
+    if(
+      (!$useregex and strpos($category['term'], $filter) !== false) or // Don't use regex and filter string is found
+      ($useregex and preg_match($filter, $category['term']))           // DO use regex and filter expression is found
+      ) {
+      return true;
+    }
+  }
+  if(
+    (!$useregex and (strpos($entry->description, $filter) !== false or strpos($entry->title, $filter) !== false)) or // Don't use regex and filter string is found
+    ($useregex and (preg_match($filter, $entry->description) or preg_match($filter, $entry->title)))                 // DO use regex and filter expression is found
+    ) {
+    return true;
+  }
+  return false;
+}
+
+if($feed                                                   // A basic check to make sure the URL is formed properly. It doesn't actually check for a valid URL or valid regex. 
+  and $filter                                              // If this check fails, the HTML document below this gets served.
+  and ($filtertype === "white" or $filtertype === "black")
+){
+  header('Content-Type: text/xml'); // Not sure if I actually need to do this, but it seems like good practice and it makes Chromium display the XML tree all nice.
+
+  $content = simplexml_load_string(file_get_contents($feed)); // I've encountered a problem where something between the server this command pulls from and this script seems to
+                                                              // be caching the result. I'm guessing it's some kind of Nginx configuration error, but I'm not sure. Maybe I
+                                                              // should be using a different function to do this? rss-bridge uses the cURL library, which I know because I had
+                                                              // to install it myself.
+
+  if($content->getName() == 'rss') { // Here I just assume everything not explicitly an RSS feed is an Atom feed. I am 100% certain there is an edge case I'm missing here.
+
+    echo '<rss version="2.0"><channel>';                        // This is the most embarrassing part of this code by far. I wanted to do this by looping over all the entries
+    echo $content->channel->title->asXML();                     // in the XML object and specifically removing the ones that didn't pass the filter *from* the object, but
+    echo $content->channel->link->asXML();                      // I couldn't for the life of me get that to work, so I settled for the next best thing: just echoing all the
+    echo $content->channel->description->asXML();               // fields that are part of the spec out to a brand new XML document and then adding all the entries that did
+    echo $content->channel->language->asXML();                  // pass. I mean, if it works it works, but sorry to anyone that was using some nonstandard feature here.
+    echo $content->channel->pubDate->asXML();
+    echo $content->channel->lastBuildDate->asXML();
+    echo $content->channel->docs->asXML();
+    echo "<generator>will's RSS filter script</generator>";
+    echo $content->channel->managingEditor->asXML();
+    echo $content->channel->webMaster->asXML();
+
+    foreach($content->channel->item as $item) {
+
+      $found = rss_testforfilter($item, $filter);
+
+      if(($found and $filtertype == 'white') or (!$found and $filtertype == 'black')) {
+        echo $item->asXML();
+      }
+      
+    }
+
+    echo '</channel></rss>';
+    
+    die();
+  } else {
+
+    echo '<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom">';
+    echo $content->title->asXML();
+    echo $content->link->asXML();
+    echo $content->updated->asXML();
+    echo $content->author->asXML();
+    echo $content->id->asXML();
+    foreach($content->category as $category) {
+      echo $category->asXML();
+    }
+    foreach($content->contributor as $contributor) {
+      echo $contributor->asXML();
+    }
+    echo "<generator>will's Atom filter script</generator>";
+    echo $content->icon->asXML();
+    echo $content->logo->asXML();
+    echo $content->rights->asXML();
+    echo $content->subtitle->asXML();
+
+    foreach($content->entry as $entry) {
+
+      $found = atom_testforfilter($entry, $filter);
+
+      if(($found and $filtertype == 'white') or (!$found and $filtertype == 'black')) {
+        echo $entry->asXML();
+      }
+      
+    }
+
+    echo '</feed>';
+    
+    die();
+  }
+}
+
+?>
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8"/>
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>RSS Filter</title>
+    <style type="text/css">
+      html, body {
+        margin: 0;
+        padding: 0;
+        height: 100%;
+      }
+      body {
+        margin: auto;
+        width: fit-content;
+        max-width: 32rem;
+        background-color: #222222;
+        font-family: monospace;
+        color: white;
+        display: flex;
+        align-items: center;
+        font-size: 1rem;
+        flex-direction: column;
+        justify-content: center;
+        text-shadow: 1px 1px 3px black;
+      }
+      td:first-child {
+        padding-right: 1em;
+        font-weight: bold;
+      }
+      input[type="text"] {
+        width: 100%;
+      }
+      code {
+        font-size: initial;
+        background-color: #101010;
+        padding: 1em;
+        border-radius: 1em;
+        display: block;
+        text-align: center;
+        margin: auto;
+        box-shadow: 2px 2px 4px black;
+      }
+      code span {
+        color: #ccc;
+      }
+      table {
+        font-size: 0.9rem;
+      }
+      a {
+        color: #df1955;
+      }
+      a:hover {
+        color: #00ffd5;
+      }
+      @media only screen and (orientation: portrait) {
+        td {
+          display: block;
+          padding-top: 0.5rem;
+        }
+        body {
+          max-width: 20rem;
+        }
+      }
+    </style>
+  </head>
+  <body>
+    <h1 style="margin: 0;">RSS Filter</h1>    
+    <hr style="width: 30%; margin: 1.25rem;">
+    <table>
+      <noscript>
+        <tr>
+          <td colspan="2">You need to have Javascript enabled for this generator to work, sorry.</td>
+        </tr>
+      </noscript>
+      <tr>
+        <td>Feed URL</td>
+        <td><input type="text" placeholder="https://example.com/rss" id="feed" value=""></input></td>
+      </tr>
+      <tr>
+        <td>Filter String</td>
+        <td><input type="text" placeholder="foobar" id="filter" value=""></input></td>
+      </tr>
+      <tr>
+        <td>Filter Type</td>
+        <td>
+          <input type="radio" id="whitelist" name="filtertype" value="white" checked>
+          <label for="whitelist">Whitelist</label>
+          <input type="radio" id="blacklist" name="filtertype" value="black">
+          <label for="blacklist">Blacklist</label>
+        </td>
+      </tr>
+      <tr>
+        <td><label for="regex">Use Regex</label></td>
+        <td><input type="checkbox" id="regex"></td>
+      </tr>
+    </table>
+    <p>
+      <code><a id="generatedlink" href=""></a></code>
+    </p>
+    <p style="position: absolute; bottom: 1em; margin: 0; text-align: center;">
+      Made by <a href="https://isopod.cool/">an isopod</a><br>
+      <a href="?downloadsource=yes">Download source code</a>
+    </p>
+    <script>
+
+      let feedURL = "https://example.com/rss";
+      let filter = "foobar";
+      let type = "white";
+      let regex = "";
+
+      function updateLink() {
+        let link = `${window.location.href}?feed=${feedURL}&filter=${filter}&type=${type}${regex}`;
+        document.getElementById('generatedlink').href = link;
+        document.getElementById('generatedlink').innerHTML = link.replace(/&/g, "&amp;");
+      }
+
+      updateLink();
+
+      document.getElementById('feed').addEventListener('input', function(evt) {
+        feedURL = encodeURI(this.value);
+        updateLink();
+      });
+      document.getElementById('filter').addEventListener('input', function(evt) {
+        filter = encodeURI(this.value);
+        updateLink();
+      });
+      document.getElementById('whitelist').addEventListener('input', function(evt) {
+        type = "white";
+        updateLink();
+      });
+      document.getElementById('blacklist').addEventListener('input', function(evt) {
+        type = "black";
+        updateLink();
+      });
+      document.getElementById('regex').addEventListener('change', function(evt) {
+        if(this.checked) {
+          regex = "&regex=yes";
+          document.getElementById('filter').placeholder = "/foobar/";
+        } else {
+          regex = "";
+          document.getElementById('filter').placeholder = "foobar";
+        }
+        updateLink();
+      });
+    </script>
+  </body>
+</html>
\ No newline at end of file
-- 
cgit v1.2.3