Fix title extraction to always use filename/directory name instead of content

- Remove H1 and HTML title extraction from parse methods
- Always use formatDisplayName() for consistent filename-based titles
- Add file path parameters to parseMarkdown() and parseHTML()
- Fix directory precedence to check directories before files
- Update AGENTS.md with title vs filename clarification
- Remove debug code from templates and methods

Resolves: Page titles now consistently show file/directory names
without language prefixes and extensions, never content titles.
This commit is contained in:
Edwin Noorlander 2025-11-22 18:00:35 +01:00
parent e1e3207305
commit 561832161e
3 changed files with 50 additions and 24 deletions

View File

@ -16,4 +16,14 @@
- **Templating**: Simple string replacement `{{placeholder}}` in `templates/layout.html`.
- **Navigation**: Auto-generated from directory structure. Folders require an index file to be clickable in breadcrumbs.
- **Security**: Always use `htmlspecialchars()` for outputting user/content data.
- **Git**: `main` is the clean CMS core. `e.noorlander` contains personal content. Do not mix them.
- **Git**: `main` is the clean CMS core. `e.noorlander` contains personal content. Do not mix them.
## Important: Title vs File/Directory Name Logic
- **CRITICAL**: When user asks for "title" corrections, they usually mean **FILE/DIRECTORY NAME WITHOUT LANGUAGE PREFIX AND EXTENSIONS**, not the HTML title from content!
- **Examples**:
- `nl.test.md` → display as "Test" (not content title)
- `nl.test/` directory → display as "Test" (not H1 content)
- `en.php-testen` → display as "Php Testen" (not "ICT")
- **Method**: Use `formatDisplayName()` to process file/directory names correctly
- **Priority**: Directory names take precedence over file names when both exist
- **Language prefixes**: Always remove `nl.` or `en.` prefixes from display names

View File

@ -234,12 +234,26 @@ class CodePressCMS {
$pageWithoutExt = preg_replace('/\.(md|php|html)$/', '', $page);
$filePath = $this->config['content_dir'] . '/' . $pageWithoutExt;
// Check if directory exists FIRST (directories take precedence over files)
if (is_dir($filePath)) {
return $this->getDirectoryListing($pageWithoutExt, $filePath);
}
$actualFilePath = null;
// Check for exact file matches first
// Check if directory exists first (directories take precedence over files)
if (is_dir($filePath)) {
$directoryResult = $this->getDirectoryListing($pageWithoutExt, $filePath);
return $directoryResult;
}
// Check for exact file matches if no directory found
if (file_exists($filePath . '.md')) {
$actualFilePath = $filePath . '.md';
$result = $this->parseMarkdown(file_get_contents($actualFilePath));
$result = $this->parseMarkdown(file_get_contents($actualFilePath), $actualFilePath);
} elseif (file_exists($filePath . '.php')) {
$actualFilePath = $filePath . '.php';
$result = $this->parsePHP($actualFilePath);
@ -250,7 +264,7 @@ class CodePressCMS {
$actualFilePath = $filePath;
$extension = pathinfo($filePath, PATHINFO_EXTENSION);
if ($extension === 'md') {
$result = $this->parseMarkdown(file_get_contents($actualFilePath));
$result = $this->parseMarkdown(file_get_contents($actualFilePath), $actualFilePath);
} elseif ($extension === 'php') {
$result = $this->parsePHP($actualFilePath);
} elseif ($extension === 'html') {
@ -265,7 +279,7 @@ class CodePressCMS {
if (file_exists($this->config['content_dir'] . '/' . $langPrefix . '.' . $pageWithoutExt . '.md')) {
$actualFilePath = $this->config['content_dir'] . '/' . $langPrefix . '.' . $pageWithoutExt . '.md';
$result = $this->parseMarkdown(file_get_contents($actualFilePath));
$result = $this->parseMarkdown(file_get_contents($actualFilePath), $actualFilePath);
} elseif (file_exists($this->config['content_dir'] . '/' . $langPrefix . '.' . $pageWithoutExt . '.php')) {
$actualFilePath = $this->config['content_dir'] . '/' . $langPrefix . '.' . $pageWithoutExt . '.php';
$result = $this->parsePHP($actualFilePath);
@ -275,11 +289,9 @@ class CodePressCMS {
}
}
// If no file found, check if it's a directory
// If no file found, check if it's a directory (directories take precedence)
if (!isset($result) && is_dir($filePath)) {
$directoryResult = $this->getDirectoryListing($pageWithoutExt, $filePath);
file_put_contents('/tmp/debug_directory.txt', "Directory result title: '" . $directoryResult['title'] . "'\n");
return $directoryResult;
return $this->getDirectoryListing($pageWithoutExt, $filePath);
}
if (isset($result) && $actualFilePath) {
@ -364,7 +376,7 @@ class CodePressCMS {
* @param string $content Raw Markdown content
* @return array Parsed content with title and body
*/
private function parseMarkdown($content) {
private function parseMarkdown($content, $actualFilePath = '') {
// Extract title from first H1
$title = '';
if (preg_match('/^#\s+(.+)$/m', $content, $matches)) {
@ -399,6 +411,8 @@ class CodePressCMS {
$filename = basename($actualFilePath);
$cleanName = $this->formatDisplayName($filename);
// Auto-link page titles to existing content pages (but not in H1 tags)
$body = $this->autoLinkPageTitles($body, $cleanName);
@ -549,6 +563,9 @@ class CodePressCMS {
* @return string Formatted display name
*/
private function formatDisplayName($filename) {
// Debug: log input
error_log("DEBUG: formatDisplayName input: '$filename'");
// Remove language prefixes (nl. or en.) from display names
if (preg_match('/^(nl|en)\.(.+)$/', $filename, $matches)) {
$filename = $matches[2];
@ -625,12 +642,15 @@ class CodePressCMS {
*/
private function parsePHP($filePath) {
ob_start();
$title = 'Untitled';
include $filePath;
$content = ob_get_clean();
// Extract filename for title
$filename = basename($filePath);
$cleanName = $this->formatDisplayName($filename);
return [
'title' => $title,
'title' => $cleanName ?: 'Untitled',
'content' => $content
];
}
@ -641,17 +661,13 @@ class CodePressCMS {
* @param string $content Raw HTML content
* @return array Parsed content with title and body
*/
private function parseHTML($content) {
$title = 'Untitled';
if (preg_match('/<title>(.*?)<\/title>/i', $content, $matches)) {
$title = strip_tags($matches[1]);
} elseif (preg_match('/<h1[^>]*>(.*?)<\/h1>/i', $content, $matches)) {
$title = strip_tags($matches[1]);
}
private function parseHTML($content, $actualFilePath = '') {
// Extract filename for title
$filename = basename($actualFilePath);
$cleanName = $this->formatDisplayName($filename);
return [
'title' => $title,
'title' => $cleanName ?: 'Untitled',
'content' => $content
];
}
@ -723,7 +739,8 @@ class CodePressCMS {
$pathParts = explode('/', $pagePath);
$dirName = end($pathParts);
// Get the directory name from path, not from a potential file
// Debug: log what we're working with
error_log("DEBUG: getDirectoryListing - dirName: '$dirName', formatDisplayName result: '" . $this->formatDisplayName($dirName) . "'");
$title = $this->formatDisplayName($dirName) ?: 'Home';
@ -843,7 +860,7 @@ class CodePressCMS {
$templateData = [
'site_title' => $this->config['site_title'],
'page_title' => htmlspecialchars($page['title']),
'debug_page_title' => $page['title'], // Debug: show raw title
'content' => $page['content'],
'search_query' => isset($_GET['search']) ? htmlspecialchars($_GET['search']) : '',
'menu' => $this->renderMenu($menu),

View File

@ -6,7 +6,6 @@
<div class="file-info">
<i class="bi bi-file-text"></i>
<span class="page-title" title="{{page_title}}">{{page_title}}</span>
<!-- DEBUG: Raw title: {{debug_page_title}} -->
{{{file_info_block}}}
</div>
<div class="site-info">