Compile artifact entry HTML by default

php-transformer/README.md

@@ -126,7 +126,7 @@ if ('failed' === $result['status']) {
 
 ## Artifact Compiler Fallbacks
 
-The artifact compiler accepts loose generated-site bundles and normalizes them into an explicit result envelope. HTML entries are preserved as `core/html` serialized block markup, Markdown falls back to `core/html` when a Markdown adapter is not loaded, and MDX support is partial: source documents are preserved while imports and JSX component references are exposed as inspectable metadata and warnings.
+The artifact compiler accepts loose generated-site bundles and normalizes them into an explicit result envelope. Safe HTML entries are compiled through `HtmlTransformer` into native serialized block markup, Markdown falls back to `core/html` when a Markdown adapter is not loaded, and MDX support is partial: source documents are preserved while imports and JSX component references are exposed as inspectable metadata and warnings.
 
 Unsupported or unsafe artifact inputs are reported through diagnostics instead of hidden best-effort behavior. Empty, absolute, or root-escaping paths are rejected; oversized files are ignored according to the source report limits; and a bundle with neither an HTML entry nor source documents fails with `missing_entry_html`.
 

php-transformer/src/ArtifactCompiler/ArtifactCompiler.php

@@ -137,7 +137,7 @@ private function compileEntryBlocks(string $html, string $entryPath, array $file
             );
         }
 
-        if ( '' === trim($html) || ! $this->entryHtmlReferencesImageAsset($html, $entryPath, $files) ) {
+        if ( '' === trim($html) ) {
             return array(
                 'blocks'            => array(),
                 'serialized_blocks' => '',
@@ -159,25 +159,6 @@ private function compileEntryBlocks(string $html, string $entryPath, array $file
         );
     }
 
-    /**
-     * @param array<int, array<string, mixed>> $files
-     */
-    private function entryHtmlReferencesImageAsset(string $html, string $entryPath, array $files): bool
-    {
-        if ( ! preg_match_all('/<img\s+[^>]*src\s*=\s*(["\'])([^"\']+)\1/i', $html, $matches) ) {
-            return false;
-        }
-
-        foreach ( $matches[2] as $src ) {
-            $asset = $this->findAssetByHtmlReference((string) $src, $entryPath, $files);
-            if ( is_array($asset) && str_starts_with((string) ($asset['mime_type'] ?? ''), 'image/') && $this->isSafeImageAsset($asset) ) {
-                return true;
-            }
-        }
-
-        return false;
-    }
-
     /**
      * @param array<int, array<string, mixed>> $files
      */

php-transformer/tests/contract/run.php

@@ -154,12 +154,13 @@ function serialize_blocks(array $blocks): string
 $assert(ArtifactCompiler::INPUT_SCHEMA === ($simple['source_reports']['artifact']['schema'] ?? ''), 'artifact report exposes canonical site artifact schema');
 $assert(ArtifactCompiler::INPUT_SCHEMA === ($simple['source_reports']['artifact']['original_schema'] ?? ''), 'canonical site artifact input schema is accepted and preserved');
 $assert('index.html' === ($simple['source_reports']['artifact']['entry_path'] ?? ''), 'generated HTML becomes an index entry');
-$assert(str_contains((string) $simple['serialized_blocks'], '<!-- wp:html -->'), 'HTML is preserved as serialized block markup');
+$assert(str_contains((string) $simple['serialized_blocks'], '<!-- wp:heading'), 'artifact HTML is transformed into native serialized block markup');
+$assert(! str_contains((string) $simple['serialized_blocks'], '<!-- wp:html -->'), 'artifact HTML does not fall back to raw HTML when transformer-safe');
 $assert('hero' === ($simple['components'][0]['name'] ?? ''), 'component candidates are exposed');
 $assert(! array_key_exists('legacy_mapping', $simple), 'artifact result omits compatibility-only legacy mapping');
 $assert(strlen('<main><article data-component="Hero"><h1>Hello artifact</h1></article></main>') === ($simple['metrics']['input_bytes'] ?? null), 'artifact metrics expose input bytes');
 $assert(strlen((string) $simple['serialized_blocks']) === ($simple['metrics']['output_bytes'] ?? null), 'artifact metrics expose output bytes');
-$assert(0 === ($simple['metrics']['block_count'] ?? null), 'artifact metrics expose block count');
+$assert(2 === ($simple['metrics']['block_count'] ?? null), 'artifact metrics expose nested block count');
 $assert(0 === ($simple['metrics']['fallback_count'] ?? null), 'artifact metrics expose fallback count');
 $assert(0 === ($simple['metrics']['diagnostic_count'] ?? null), 'artifact metrics expose diagnostic count');
 $assert(is_float($simple['metrics']['transform_duration_ms'] ?? null), 'artifact metrics expose transform duration');

php-transformer/tests/fixtures/parity/artifact-generated-html.json

@@ -1,25 +1,32 @@
 {
   "schema": "blocks-engine/php-transformer/parity-fixture/v1",
   "name": "artifact-generated-html-entry",
-  "description": "Compiles generated HTML artifact input into the shared transformer result shape.",
+  "description": "Compiles generated static HTML artifact input without images into native block markup through the HTML transformer.",
   "source_reference": {
     "repo": "php-transformer",
     "path": "tests/fixtures/parity/artifact-generated-html.json",
-    "notes": "Covers the artifact compiler contract smoke path."
+    "notes": "Covers the artifact compiler default HTML transform path when entry markup has no local image references."
   },
   "operation": "artifact_compiler.compile",
   "input": {
     "artifact": {
       "generated_html": "<main><article data-component=\"Hero\"><h1>Hello artifact</h1></article></main>"
     }
   },
+  "expected_blocks": [
+    { "path": "blocks.0", "name": "core/group" },
+    { "path": "blocks.0.innerBlocks.0", "name": "core/heading", "attrs": { "content": "Hello artifact", "level": 1 } }
+  ],
+  "expected_fallbacks": [],
   "expect": [
     { "path": "schema", "assert": "equals", "value": "blocks-engine/php-transformer/result/v1" },
     { "path": "status", "assert": "equals", "value": "success" },
     { "path": "source_reports.artifact.entry_path", "assert": "equals", "value": "index.html" },
     { "path": "components.0.name", "assert": "equals", "value": "hero" },
-    { "path": "serialized_blocks", "assert": "contains", "value": "<!-- wp:html -->" },
+    { "path": "serialized_blocks", "assert": "contains", "value": "<!-- wp:group -->" },
+    { "path": "serialized_blocks", "assert": "contains", "value": "<!-- wp:heading {\"content\":\"Hello artifact\",\"level\":1} -->" },
     { "path": "serialized_blocks", "assert": "contains", "value": "<h1>Hello artifact</h1>" },
+    { "path": "serialized_blocks", "assert": "not_contains", "value": "<!-- wp:html -->" },
     { "path": "source_reports.artifact.html.element_count", "assert": "equals", "value": 3 },
     { "path": "source_reports.conversion_report.metrics.fallback_count", "assert": "equals", "value": 0 },
     { "path": "provenance.0.source_format", "assert": "equals", "value": "artifact" }