From 04d21c8a97da9463e6bedda620ecf1282722ea3e Mon Sep 17 00:00:00 2001
From: Dan Brown <ssddanbrown@googlemail.com>
Date: Wed, 22 Nov 2023 22:14:28 +0000
Subject: [PATCH] Includes: Started foundations for new include tag parser

---
 app/Entities/Tools/PageIncludeParser.php | 57 ++++++++++++++++++++++++
 tests/Unit/PageIncludeParserTest.php     | 46 +++++++++++++++++++
 2 files changed, 103 insertions(+)
 create mode 100644 app/Entities/Tools/PageIncludeParser.php
 create mode 100644 tests/Unit/PageIncludeParserTest.php

diff --git a/app/Entities/Tools/PageIncludeParser.php b/app/Entities/Tools/PageIncludeParser.php
new file mode 100644
index 000000000..63d3ea8d6
--- /dev/null
+++ b/app/Entities/Tools/PageIncludeParser.php
@@ -0,0 +1,57 @@
+<?php
+
+namespace BookStack\Entities\Tools;
+
+use BookStack\Util\HtmlDocument;
+use Closure;
+
+class PageIncludeParser
+{
+    protected static string $includeTagRegex = "/{{@\s?([0-9].*?)}}/";
+
+    public function __construct(
+        protected string $pageHtml,
+        protected Closure $pageContentForId,
+    ) {
+    }
+
+    public function parse(): string
+    {
+        $html = new HtmlDocument($this->pageHtml);
+
+        $includeHosts = $html->queryXPath("//body//*[contains(text(), '{{@')]");
+        $node = $includeHosts->item(0);
+
+        // One of the direct child textnodes of the "$includeHosts" should be
+        // the one with the include tag within.
+        $textNode = $node->childNodes->item(0);
+
+        // TODO:
+        // Hunt down the specific text nodes with matches
+        // Split out tag text node from rest of content
+        // Fetch tag content->
+          // If range or top-block: delete tag text node, [Promote to top-block], delete old top-block if empty
+          // If inline: Replace current text node with new text or elem
+        // !! "Range" or "inline" status should come from tag parser and content fetcher, not guessed direct from content
+        //     since we could have a range of inline elements
+
+        // [Promote to top-block]
+        // Tricky operation.
+        // Can throw in before or after current top-block depending on relative position
+        // Could [Split] top-block but complex past a single level depth.
+        // Maybe [Split] if one level depth, otherwise default to before/after block
+        // Should work for the vast majority of cases, and not for those which would
+        // technically be invalid in-editor anyway.
+
+        // [Split]
+        // Copy original top-block node type and attrs (apart from ID)
+        // Move nodes after promoted tag-node into copy
+        // Insert copy after original (after promoted top-block eventually)
+
+        // Notes: May want to eventually parse through backwards, which should avoid issues
+        // in changes affecting the next tag, where tags may be in the same/adjacent nodes.
+
+
+        return $html->getBodyInnerHtml();
+    }
+}
diff --git a/tests/Unit/PageIncludeParserTest.php b/tests/Unit/PageIncludeParserTest.php
new file mode 100644
index 000000000..de31504ff
--- /dev/null
+++ b/tests/Unit/PageIncludeParserTest.php
@@ -0,0 +1,46 @@
+<?php
+
+namespace Tests\Unit;
+
+use BookStack\Entities\Tools\PageIncludeParser;
+use Tests\TestCase;
+
+class PageIncludeParserTest extends TestCase
+{
+    public function test_include_simple_inline_text()
+    {
+        $this->runParserTest(
+            '<p>{{@45#content}}</p>',
+            ['45' => '<p id="content">Testing</p>'],
+            '<p>Testing</p>',
+        );
+    }
+
+    public function test_include_simple_inline_text_with_existing_siblings()
+    {
+        $this->runParserTest(
+            '<p>{{@45#content}} <strong>Hi</strong>there!</p>',
+            ['45' => '<p id="content">Testing</p>'],
+            '<p>Testing <strong>Hi</strong>there!</p>',
+        );
+    }
+
+    public function test_include_simple_inline_text_within_other_text()
+    {
+        $this->runParserTest(
+            '<p>Hello {{@45#content}}there!</p>',
+            ['45' => '<p id="content">Testing</p>'],
+            '<p>Hello Testingthere!</p>',
+        );
+    }
+
+    protected function runParserTest(string $html, array $contentById, string $expected)
+    {
+        $parser = new PageIncludeParser($html, function (int $id) use ($contentById) {
+            return $contentById[strval($id)] ?? null;
+        });
+
+        $result = $parser->parse();
+        $this->assertEquals($expected, $result);
+    }
+}