Merge pull request #3 from imagewize/feature/unbalanced-html-tags-check

Feature/unbalanced Html Tags Check

Author: jasperf

.gitignore

@@ -43,6 +43,7 @@ create-pr.sh
 # Testing
 /coverage/
 .phpunit.result.cache
+.phpunit.cache/
 phpunit.xml
 
 # General

CHANGELOG.md

@@ -4,6 +4,20 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [2.1.0] - 2026-05-08
+
+### Added
+
+- **Unbalanced HTML tags check** in `BaseRules` (`src/Compliance/Rules/BaseRules.php`) — Detects unbalanced HTML tags (`<div>`, `<ul>`, `<ol>`, `<li>`, `<figure>`, `<figcaption>`, `<section>`, `<article>`, `<header>`, `<footer>`, `<nav>`, `<aside>`, `<main>`, `<p>`) in pattern files. Reports violations with opening/closing tag counts to help authors identify mismatches. Addresses issue #2 where missing `</div>` tags caused DOM nesting issues but went undetected by both pt-cli and `wp pattern validate`.
+- **PHPUnit test suite** (`tests/`) — First test suite for the project with comprehensive tests for the unbalanced HTML tags rule:
+  - `tests/bootstrap.php` — Test bootstrap file
+  - `tests/Compliance/Rules/BaseRulesTest.php` — Tests for BaseRules including unbalanced HTML tags detection
+
+### Changed
+
+- `.gitignore` — Added `.phpunit.cache/` directory
+- `README.md` — Added "Balanced HTML tags" rule to the Base rules table
+
 ## [2.0.0] - 2026-05-08
 
 ### Added

README.md

@@ -204,6 +204,7 @@ Config lookup order:
 | No hardcoded font sizes (CSS px/rem/em) | No | Error |
 | No spacer blocks | No | Error |
 | Margin reset on alignfull patterns | No | Error |
+| Balanced HTML tags (`<div>`, `<ul>`, `<ol>`, `<li>`, `<figure>`, etc.) | No | Error |
 | Responsive grid for 3+ columns (warn on `wp:columns`) | No | Warning |
 | No hardcoded media IDs | No | Error |
 | Translated strings (HTML tags + alt attributes) | No | Error |

src/Compliance/Rules/BaseRules.php

@@ -69,6 +69,7 @@ public function check(string $filePath): array
         $violations = array_merge($violations, $this->checkFontPresetClasses($content));
         $violations = array_merge($violations, $this->checkBlockGapInlineStyles($content));
         $violations = array_merge($violations, $this->checkUntranslatedStrings($content));
+        $violations = array_merge($violations, $this->checkUnbalancedHtmlTags($content));
 
         if ($rules['requirePatternName'] ?? false) {
             $isTemplate = str_starts_with($patternName, 'template-');
@@ -898,4 +899,45 @@ function ($m) use ($property) {
             1
         );
     }
+
+    /**
+     * Check for unbalanced HTML tags in pattern files.
+     * This catches missing closing tags like </div> that can cause DOM nesting issues.
+     * Only checks structural HTML outside of block comments and PHP code.
+     *
+     * Note: <p> tags are included but WordPress block patterns use explicit <p class="wp-block-...">
+     * with matching </p>, so false positives are unlikely. The regex [^>]* correctly avoids
+     * matching <?php since ? != p after <\s*.
+     */
+    private function checkUnbalancedHtmlTags(string $content): array
+    {
+        $violations = [];
+
+        // Structural tags that should be balanced in pattern HTML
+        // Note: <li> is included even though HTML5 allows implicit closing, because
+        // WP block patterns use wp:list-item which explicitly closes them
+        $tags = ['div', 'ul', 'ol', 'li', 'figure', 'figcaption', 'section', 'article', 'header', 'footer', 'nav', 'aside', 'main', 'p'];
+
+        foreach ($tags as $tag) {
+            // Count opening tags (match <div, <div>, <div class="foo">, etc.)
+            $openingCount = preg_match_all('/<\s*' . $tag . '\b[^>]*>/i', $content);
+
+            // Count closing tags (match </div>, </div >, etc.)
+            $closingCount = preg_match_all('/<\/\s*' . $tag . '\s*>/i', $content);
+
+            if ($openingCount !== $closingCount) {
+                $violations[] = $this->violation(
+                    'unbalanced-html-tags',
+                    sprintf(
+                        'Unbalanced <%s> tags: %d opening, %d closing',
+                        $tag,
+                        $openingCount,
+                        $closingCount
+                    )
+                );
+            }
+        }
+
+        return $violations;
+    }
 }