@kirigami/php-prepros 3.0.0 → 3.0.1

package/README.md

@@ -34,6 +34,7 @@ Part of the **Kirigami** project ecosystem.
 
 - [@kirigami/php-prepros](#kirigamiphp-prepros)
 - [Overview](#overview)
+- [What's new in 3.0.1](#whats-new-in-301)
 - [What's new in 3.0.0](#whats-new-in-300)
 - [What's new in 2.0.0](#whats-new-in-200)
 - [What's new in 1.9.3](#whats-new-in-193)
@@ -105,6 +106,12 @@ Part of the **Kirigami** project ecosystem.
 
 ---
 
+## What's new in 3.0.1
+
+A registered tag written inside Markdown code is no longer processed. `<markdown>`, `<img asset="...">` or a plugin tag shown in an inline code span (`` `<markdown prose>` ``) or a fenced block (` ``` ` / `~~~`) stays example text: before, a `<markdown>` in a code span paired with the real block's closing tag and broke the rest of the page, and an `<img asset>` in one became an empty image. Indented (four-space) code blocks are not recognized, since `<markdown>` bodies are indented; use a fence there.
+
+---
+
 ## What's new in 3.0.0
 
 This release switches `YAML::` to the native YAML extension, `MD::` to native mdhtml, `SCHEMA` to native jsonk, and `Normalizer` to native norm. It also includes page types and request lifecycle hooks.
@@ -1602,7 +1609,7 @@ STR::normalize(string $str): string
 STR::slug(string $str, string $sep = ''): string
 ```
 
-`STR::replaceTags()` is the engine behind `PREPROS::registerTag()`. It finds all occurrences of `<tagname ...>...</tagname>` in an HTML string and replaces each with the return value of `$callback($fullMatch, $attrs, $body)`.
+`STR::replaceTags()` is the engine behind `PREPROS::registerTag()`. It finds all occurrences of `<tagname ...>...</tagname>` in an HTML string and replaces each with the return value of `$callback($fullMatch, $attrs, $body)`. Occurrences inside Markdown code (an inline code span or a fenced block) are left as written, so a tag shown as an example is not run.
 
 `STR::trimIndent()` strips the common leading whitespace from a multi-line string — handy when pulling content out of indented `<markdown>` blocks.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kirigami/php-prepros",
-  "version": "3.0.0",
+  "version": "3.0.1",
   "description": "PHP preprocessor for the Kirigami static site generator. Compile PHP page templates to clean, deployable HTML — with zero server dependency.",
   "keywords": [
     "kirigami",

package/src/libraries/str.class.php

@@ -18,7 +18,27 @@ class STR
 			. '|<' . $t . '([^>]*)/>'
 			. '|<' . $t . '([^>]*)>#msi';
 
-		return preg_replace_callback($pattern, function ($m) use ($clb) {
+		// A tag written inside Markdown code (a fenced block or an inline
+		// code span) is example text, not a tag: `<markdown>` shown in a
+		// code span must not close the real block around it, and
+		// `<img asset="...">` shown in one must not become an image. Mask
+		// that code before matching and put it back afterwards, both in the
+		// output and in what the callback receives.
+		$code = [];
+		if (!str_contains($contents, "\x1A")) $contents = preg_replace_callback(
+			'#^[ \t]*(`{3,}|~{3,})[^\n]*\n.*?^[ \t]*\1[ \t]*$'
+			. '|(?<!`)(`+)(?!`)[^\n]*?(?<!`)\2(?!`)#ms',
+			function ($m) use (&$code) {
+				$code[] = $m[0];
+				return "\x1A" . (count($code) - 1) . "\x1A";
+			},
+			$contents
+		);
+		$restore = fn(?string $s) => $s === null || !$code ? $s
+			: preg_replace_callback("#\x1A(\d+)\x1A#", fn($m) => $code[$m[1]], $s);
+
+		$result = preg_replace_callback($pattern, function ($m) use ($clb, $restore) {
+			$m = array_map($restore, $m);
 			if (isset($m[1]) || isset($m[2])) {
 				// paired form: <tag>content</tag>
 				$attrs = $m[1] ?? '';
@@ -34,6 +54,8 @@ class STR
 			}
 			return call_user_func($clb, $m[0], self::parseHtmlAttributes($attrs), $inner);
 		}, $contents, -1, $count, PREG_UNMATCHED_AS_NULL); // <-- the flag that fixes everything
+
+		return $restore($result);
 	}