jasm1nii
/
xml-feed-generator
mirror of https://github.com/jasm1nii/xml-feed-generator.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add variables, fallback for timestamps

This commit is contained in:
jasmine 2023-10-18 11:28:45 +07:00 committed by GitHub
parent 0382f8d4e4
commit e35a40bd47
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 80 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

109
feed_generator.php
View File

@ -1,52 +1,91 @@
<?php
// work in progress!!
// GENERAL SETTINGS -------------------------------------------------------------------
// the timezone referenced by the system for automatic timestamping.
// suported timezones: https://www.php.net/manual/en/timezones.php
$timezone = 'Asia/Jakarta';
// FEED METADATA //////////////////////////////////////////////////////////////////////
// certain characters must be escaped as HTML entities - note that XML only accepts five of them.
// reference: https://en.wikipedia.org/wiki/List_of_XML_and_HTML_character_entity_references
$feed_title = 'jasmine&apos;s b(rain)log | jasm1nii.xyz';
$feed_subtitle = 'blog articles by jasmine';
$blog_url = 'https://jasm1nii.xyz/blog/articles';
$feed_url = 'https://jasm1nii.xyz/blog/articles/articles.xml';
$author_name = 'jasmine';
$author_email = 'contact@jasm1nii.xyz';
$author_homepage = 'https://jasm1nii.xyz/';
$feed_icon = 'https://jasm1nii.xyz/assets/media/itchio-textless-white.svg';
$feed_logo = 'https://jasm1nii.xyz/assets/media/main/07042023-me_compressed.webp';
// 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);
// once i'm there, i specify the parent directory where i keep all of my blog pages.
$blog_root = $site_root.'/blog/articles';
// then i specify a pattern that matches the path of each individual page.
// my setup is /YYYY/MM/DD/entry.html
$blog_entries = $blog_root.'/*/*/*/*.html';
// ------------------------------------------------------------------------------------
// create beginning of feed template.
// reference for required elements: https://validator.w3.org/feed/docs/atom.html
ob_start();
date_default_timezone_set("Asia/Jakarta");
date_default_timezone_set($timezone);
echo '<?xml version="1.0" encoding="utf-8"?>'
.'<feed xmlns="http://www.w3.org/2005/Atom">'
// optionally specify feed generator for debugging purposes.
.'<generator version="1.0">jasmine&apos;s PHP feed generator!! | last modified by system at ' . strtoupper(date("h:i:sa")) . ' (GMT+7)</generator>'
.'<title>jasmine&apos;s b(rain)log | jasm1nii.xyz</title>'
.'<subtitle>blog articles by jasmine</subtitle>'
.'<id>https://jasm1nii.xyz/blog/articles</id>';
echo '<link rel="self" href="https://jasm1nii.xyz/blog/articles/articles.xml" type="application/atom+xml"/>';
echo '<link rel="alternate" href="https://jasm1nii.xyz/blog/articles" type="text/html"/>';
// define path to fetch pages from.
$root = dirname(__DIR__, 2);
$blog = $root.'/blog/articles';
.'<generator uri="https://github.com/jasm1nii/xml-feed-generator" version="1.1">PHP feed generator by jasm1nii.xyz | last modified by the system at ' . strtoupper(date("h:i:sa")) . ' (GMT' . date('P') . ')</generator>'
.'<title>' . $feed_title . '</title>'
.'<subtitle>' . $feed_subtitle . '</subtitle>'
.'<id>' . $blog_url . '</id>'
.'<link rel="self" href="'. $feed_url .'" type="application/atom+xml"/>'
.'<link rel="alternate" href="' . $blog_url .'" type="text/html"/>';
// 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);
// match feed update time with the newest entry.
// i'm personally not using the system modification time to avoid sending uneccessary notifications to feed readers.
$article_list = glob($blog.'/*/*/*/*.html');
$article_list = glob($blog_entries);
$first_article = array_pop($article_list);
$first_article_content = file_get_contents($first_article);
$first_article_dom = new DOMDocument;
$first_article_dom->loadHTML($first_article_content);
$feed_updated = $first_article_dom->getElementsByTagName('time');
$f = 0;
foreach ($feed_updated as $feed_updated_text) {
echo '<updated>' . $feed_updated_text->getAttribute('datetime') . 'T00:00:00+07:00</updated>';
$feed_datetime = $feed_updated_text->getAttribute('datetime');
if (strlen($feed_datetime) == 10) {
echo '<updated>' . $feed_datetime . 'T00:00:00' . date('P') .'</updated>';
}
elseif (strlen($feed_datetime) == 25 || strlen($feed_datetime) == 20) {
echo '<updated>' . $feed_datetime .'</updated>';
}
if(++$f > 0) break;
}
// if no RFC 3339 timestamp is found, use the file creation date.
if (empty($feed_updated)) {
$first_article_created = filectime($first_article);
echo '<updated>' . date(DATE_ATOM, $first_article_created) . '</updated>';
}
// rest of the template.
echo '<author>'
.'<name>jasmine</name>'
.'<email>contact@jasm1nii.xyz</email>'
.'<uri>https://jasm1nii.xyz/</uri>'
.'<name>' . $author_name . '</name>'
.'<email>' . $author_email . '</email>'
.'<uri>' . $author_homepage . '</uri>'
.'</author>'
.'<icon>https://jasm1nii.xyz/assets/media/itchio-textless-white.svg</icon>'
.'<logo>https://jasm1nii.xyz/assets/media/main/07042023-me_compressed.webp</logo>';
.'<icon>' . $feed_icon . '</icon>'
.'<logo>' . $feed_logo . '</logo>';
// output entries.
$i = 0;
foreach (array_reverse(glob($blog.'/*/*/*/*.html')) as $article) {
foreach (array_reverse(glob($blog_entries)) as $article) {
$article_content = file_get_contents($article);
$article_dom = new DOMDocument;
$article_dom->loadHTML($article_content);
@ -60,16 +99,27 @@
}
// id
echo '<id>https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog) . '</id>';
echo '<id>https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog_root) . '</id>';
// alternate link
echo '<link rel="alternate" type="text/html" href="https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog) . '"/>';
echo '<link rel="alternate" type="text/html" href="https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog_root) . '"/>';
$updated = $article_dom->getElementsByTagName('time');
$t = 0;
$a = 0;
foreach ($updated as $updated_text) {
echo '<updated>' . $updated_text->getAttribute('datetime') . 'T00:00:00+07:00</updated>';
if(++$t > 0) break;
$timestamp = $updated_text->getAttribute('datetime');
if (strlen($timestamp) == 10) {
echo '<updated>' . $timestamp . 'T00:00:00' . date('P'). '</updated>';
}
elseif (strlen($timestamp) == 25 || strlen($timestamp) == 20) {
echo '<updated>' . $timestamp .'</updated>';
}
if(++$a > 0) break;
}
// if no RFC 3339 timestamp is found, use the file creation date.
if (empty($updated)) {
$article_created = filectime($article);
echo '<updated>' . date(DATE_ATOM, $article_created) . '</updated>';
}
// summary
@ -83,12 +133,11 @@
}
// content
// HTML entities must be escaped - note that XML only defines five of them.
// reference: https://en.wikipedia.org/wiki/List_of_XML_and_HTML_character_entity_references
$content_class = 'e-content';
$content = $x->query("//*[@class='" . $content_class . "']");
if ($content->length > 0) {
echo '<content type="html">' . preg_replace('/\s\s+/', ' ',(substr($content->item(0)->nodeValue,0,500))) . '... (&lt;a href="https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog) . '"&gt;read more&lt;/a&gt;)' . '</content>';
// strip line breaks and output a maximum of 500 characters.
echo '<content type="html">' . preg_replace('/\s\s+/', ' ',(substr($content->item(0)->nodeValue,0,500))) . '... (&lt;a href="https://jasm1nii.xyz/blog/articles/' . ltrim($article, $blog_root) . '"&gt;read more&lt;/a&gt;)' . '</content>';
} else {
// fallback for older markup
$content_class = 'entry';
@ -99,11 +148,13 @@
}
echo '</entry>';
// add no more than 10 entries.
if(++$i > 9) break;
}
echo '</feed>';
$xml_str = ob_get_contents();
ob_end_clean();
file_put_contents($blog.'/articles.xml', $xml_str);
file_put_contents($blog_root.'/articles.xml', $xml_str);
?>