From 3f1287d1721124e3a0752678a8c928959442b333 Mon Sep 17 00:00:00 2001
From: Jasmine Amalia <67263692+jasm1nii@users.noreply.github.com>
Date: Sun, 22 Oct 2023 15:42:07 +0700
Subject: [PATCH] add PHP precompiling

Signed-off-by: Jasmine Amalia <67263692+jasm1nii@users.noreply.github.com>
---
 README.md                    |  7 ++++---
 feed_generator.php           |  8 ++++----
 feed_generator_functions.php | 25 ++++++++++++++++++++++---
 3 files changed, 30 insertions(+), 10 deletions(-)

diff --git a/README.md b/README.md
index 8431a15..833b39c 100644
--- a/README.md
+++ b/README.md
@@ -5,8 +5,9 @@ an RSS/atom feed generator for my personal site :cat: tested to work with PHP ve
 
 ## how it works
 1. match files in a specificied directory.
-2. load the DOM of each file.
-3. for each child of `<entry>` in the XML feed, output a string value from one of the following HTML elements/file properties, in order of priority:
+2. precompile any PHP files, if found.
+3. load the DOM of each file.
+4. for each child of `<entry>` in the XML feed, output a string value from one of the following HTML elements/file properties, in order of priority:
 
 | original HTML                 | XML output                      |
 |-------------------------------|---------------------------------|
@@ -32,7 +33,7 @@ an RSS/atom feed generator for my personal site :cat: tested to work with PHP ve
 
 
 
-4. save all of the above into a new file named **articles.xml** (default name, but can be changed).
+5. save all of the above into a new file named **articles.xml** (default name, but can be changed).
 
 ## ways to use
 - configure a cron job on your web server to run automatically every now and then.
diff --git a/feed_generator.php b/feed_generator.php
index db7ae38..fd35671 100644
--- a/feed_generator.php
+++ b/feed_generator.php
@@ -32,15 +32,15 @@
     /* -------------------- */
 
     // PATH TO FETCH PAGES FROM
-    ## __DIR__ is the directory where *this script* is located. in my case, i first need to go up two directories to get to the site root.
-    $site_root = dirname(__DIR__, 2);
+    ## __DIR__ is the directory where *this script* is located.
+    $site_root = dirname(__DIR__, 2).'/public_html';
 
     ## once i'm there, i specify the parent directory where i keep all of my blog pages.
     ## because the values of $blog_root and $blog_entries will be used for generating entry links, forward slashes are a *must*.
     $blog_root = $site_root.'/blog/articles';
 
-    ## then, specify a pattern that matches the path of each individual page. for example, this will match /YYYY/MM/DD/entry.html.
-    $blog_entries = $blog_root.'/*/*/*/*.html';
+    ## then, specify a pattern that matches the path of each individual page. for example, this will match /YYYY/MM/DD/entry (any file extension).
+    $blog_entries = $blog_root.'/*/*/*/[entry]*';
 
     /* -------------------- */
 
diff --git a/feed_generator_functions.php b/feed_generator_functions.php
index 5621fa9..ea8ba50 100644
--- a/feed_generator_functions.php
+++ b/feed_generator_functions.php
@@ -1,8 +1,18 @@
 <?php
     date_default_timezone_set($timezone);
+    
+    $i = 0;
+    foreach ((glob($blog_entries.'.php')) as $article_php) {
+        ob_start();
+        include $article_php;
+        $compiled = ob_get_contents();
+        ob_end_clean();
+        file_put_contents($article_php.'_generator-tmp.html', $compiled);
+        if(++$i > ($max_entries-1) ) break;
+    }
 
     $i = 0;
-    foreach ((glob($blog_entries)) as $article) {
+    foreach ((glob($blog_entries.'html')) as $article) {
         $article_content = file_get_contents($article);
         // force libxml to parse all HTML elements, including HTML 5. by default, the extension can only read valid HTML 4.
         libxml_use_internal_errors(true);
@@ -88,6 +98,7 @@
         }
 
         if(++$i > ($max_entries-1) ) break;
+
         $data[$i] = [
             'title'=>$title_data,
             'id'=>$id_data,
@@ -151,11 +162,19 @@
         $entry->addChild('published',$published);
 
         $content = $data[$i]['content'];
-        $entry->addChild('content', nl2br(preg_replace("/\n\s+/", "",(htmlspecialchars($content, ENT_XML1)))));
+        $content_child = $entry->addChild('content', nl2br(preg_replace("/\n\s+/", "",(htmlspecialchars($content, ENT_XML1)))));
+        $content_child->addAttribute('type','html');
     }
 
-    echo $sxe->saveXML($blog_root . DIRECTORY_SEPARATOR . $file);
+    $sxe->saveXML($blog_root . DIRECTORY_SEPARATOR . $file);
+
+    $del_tmp = unlink($article_php.'_generator-tmp.html');
 
     echo    nl2br(strtoupper(date("h:i:sa")) . ' - Feed successfully generated in ' . realpath($blog_root) . DIRECTORY_SEPARATOR . $file . "\n");
+    if ($del_tmp) {
+        echo 'Temporary generator files successfully deleted.';
+    } else {
+        echo 'Temporary generator files could not be automatically deleted - check your directories to delete them manually.';
+    }
     echo    'Validate your feed at https://validator.w3.org/feed/';
 ?>
\ No newline at end of file