initial commit

.gitignore

@@ -0,0 +1,4 @@
+/composer.lock
+/config.json
+/data/
+/vendor/

README.md

@@ -1,2 +1,33 @@
 # Pulsar
-RSS Aggregator for Gemini Protocol
+
+RSS Aggregator for [Gemini Protocol](https://geminiprotocol.net)
+
+Simple RSS feed converter to static Gemtext format, useful for news portals or localhost usage.
+
+## Usage
+
+1. `git clone https://github.com/YGGverse/Pulsar.git`
+2. `cp example/config.json config.json` - setup your feeds there!
+3. `php src/crawler.php` - crontab schedule
+
+## Config
+
+Configuration file supports multiple feed channels with custom configurations:
+
+* `source` - string, filepath or URL to the valid RSS feed
+* `target` - string, relative or absolute path to Gemtext dumps
+* `item`
+  * `limit` - integer, how many items to display
+  * `template` - string, custom pattern for feed item, that supports following macros
+    * `{nl}` - new line separator
+    * `{link}` - item link
+    * `{guid}` - item guid
+    * `{pubDate}` - item pubDate, soon with custom time format e.g. `{pubDate:Y-m-d H:s}`
+    * `{title}` - item title
+    * `{description}` - item description
+
+Resulting files could be generated to the any folder for personal reading on localhost, or shared with others using [gmid](https://github.com/omar-polo/gmid), [twins](https://code.rocket9labs.com/tslocum/twins) or any other [Gemini server](https://github.com/kr1sp1n/awesome-gemini#servers).
+
+## Instances
+
+Coming soon!

composer.json

@@ -0,0 +1,19 @@
+{
+    "name": "yggverse/pulsar",
+    "description": "RSS Aggregator for Gemini Protocol",
+    "keywords": [ "yggverse", "gemini", "gemini-protocol", "gemtext", "gmi", "rss", "feed", "converter", "aggregator" ],
+    "homepage": "https://github.com/yggverse/pulsar",
+    "type": "project",
+    "license": "MIT",
+    "autoload": {
+        "psr-4": {
+            "Yggverse\\Pulsar\\": "src/"
+        }
+    },
+    "authors": [
+        {
+            "name": "YGGverse"
+        }
+    ],
+    "require": {}
+}

example/config.json

@@ -0,0 +1,23 @@
+{
+    "feed":
+    [
+        {
+            "source":"https://www.omglinux.com/feed",
+            "target":"data/omglinux.com/feed.gmi",
+            "item":
+            {
+                "template":"=> {link} {title}{nl}{nl}{description}",
+                "limit":20
+            }
+        },
+        {
+            "source":"https://omgubuntu.co.uk/feed",
+            "target":"data/omgubuntu.co.uk/feed.gmi",
+            "item":
+            {
+                "template":"=> {link} {title}{nl}{nl}{description}",
+                "limit":20
+            }
+        }
+    ]
+}

src/crawler.php

@@ -0,0 +1,132 @@
+<?php
+
+// Prevent multi-thread execution
+$semaphore = sem_get(
+    crc32(
+        __DIR__
+    ), 1
+);
+
+if (false === sem_acquire($semaphore, true))
+{
+    exit;
+}
+
+// Init config
+$config = json_decode(
+    file_get_contents(
+        __DIR__ . DIRECTORY_SEPARATOR . '..' . DIRECTORY_SEPARATOR . 'config.json'
+    )
+);
+
+// Update feeds
+foreach ($config->feed as $feed)
+{
+    // Init feed location
+    $filename = str_starts_with(
+        $feed->target,
+        DIRECTORY_SEPARATOR
+    ) ? $feed->target : __DIR__ . DIRECTORY_SEPARATOR . '..' . DIRECTORY_SEPARATOR . $feed->target;
+
+    // Init destination storage
+    @mkdir(
+        dirname(
+            $filename
+        ),
+        0755,
+        true
+    );
+
+    // Get feed data
+    if (!$channel = simplexml_load_file($feed->source)->channel)
+    {
+        continue;
+    }
+
+    // Update title
+    if (!empty($channel->title))
+    {
+        $title = trim(
+            strip_tags(
+                html_entity_decode(
+                    $channel->title
+                )
+            )
+        );
+    }
+
+    else
+    {
+        $title = parse_url(
+            $feed->source,
+            PHP_URL_HOST
+        );
+    }
+
+    file_put_contents(
+        $filename,
+        sprintf(
+            '# %s',
+            $title
+        ) . PHP_EOL
+    );
+
+    // Append description
+    if (!empty($channel->description))
+    {
+        file_put_contents(
+            $filename,
+            PHP_EOL . trim(
+                strip_tags(
+                    html_entity_decode(
+                        $channel->description
+                    )
+                )
+            ) . PHP_EOL,
+            FILE_APPEND | LOCK_EX
+        );
+    }
+
+    // Append items
+    $i = 1; foreach ($channel->item as $item)
+    {
+        // Apply items limit
+        if ($i > $feed->item->limit)
+        {
+            break;
+        }
+
+        // Format item
+        file_put_contents(
+            $filename,
+            PHP_EOL . trim(
+                preg_replace(
+                    '/[\s]{3,}/ui',
+                    PHP_EOL . PHP_EOL,
+                    str_replace(
+                        [
+                            '{nl}',
+                            '{link}',
+                            '{guid}',
+                            '{pubDate}',
+                            '{title}',
+                            '{description}'
+                        ],
+                        [
+                            PHP_EOL,
+                            !empty($item->link)        ? trim($item->link) : '',
+                            !empty($item->guid)        ? trim($item->guid) : '',
+                            !empty($item->pubDate)     ? trim($item->pubDate) : '',
+                            !empty($item->title)       ? trim(strip_tags(html_entity_decode($item->title))) : '',
+                            !empty($item->description) ? trim(strip_tags(html_entity_decode($item->description))) : ''
+                        ],
+                        $feed->item->template
+                    ) . PHP_EOL
+                )
+            ) . PHP_EOL,
+            FILE_APPEND | LOCK_EX
+        );
+
+        $i++;
+    }
+}