Update export scripts for Just-the-docs theme compatibility

Agent-Logs-Url: https://github.com/bndtools/bnd/sessions/f984c310-80ca-4eb0-ab37-9e7d2186d4ed

Co-authored-by: peterkir <250545+peterkir@users.noreply.github.com>

Author: Copilot

docs/EXPORT_README.md

@@ -14,6 +14,11 @@ Both formats are ideal for:
 - Distributing documentation without requiring a web server
 - Creating searchable documentation archives
 
+**Compatibility:** The export scripts support multiple documentation structures:
+- Just-the-docs Jekyll theme (current)
+- Legacy custom Jekyll templates
+- Archived release documentation
+
 ## Quick Start
 
 ### Prerequisites

docs/IMPLEMENTATION_SUMMARY.md

@@ -109,11 +109,19 @@ Added section linking to export functionality and providing quick examples.
 
 The most complex part was extracting main content from different HTML structures:
 
-1. **_site structure**: Uses `<main data-pagefind-body>` tags
-2. **Release structure**: Uses `<div class="notes-margin">` for content
-3. **Navigation/menus**: Different structures need to be filtered out
-
-Solution: Simplified parser that looks for main content markers and captures everything within, using depth tracking to properly close tags.
+1. **Just-the-docs theme** (current): Uses `<main id="main-content">` tags
+2. **Legacy _site structure**: Uses `<main data-pagefind-body>` tags
+3. **Release structure**: Uses `<div class="notes-margin">` for content
+4. **Navigation/menus**: Different structures need to be filtered out
+
+Solution: Implemented a priority-based parser that looks for main content markers in order of specificity:
+- `<main id="main-content">` (highest priority - Just-the-docs specific)
+- `<main>` with `data-pagefind-body` attribute
+- Generic `<main>` tags
+- `<div class="notes-margin">` (legacy releases)
+- `<div id="main-content">` (fallback for Just-the-docs)
+
+The parser uses depth tracking to properly capture nested content and close tags correctly.
 
 ### Directory Structure Handling
 

docs/export-single-html.sh

@@ -71,12 +71,32 @@ class ContentExtractor(HTMLParser):
     def handle_starttag(self, tag, attrs):
         attrs_dict = dict(attrs)
         
-        # Track main content - handle both _site structure and release structure
-        # Once we find the main content area, we'll capture everything inside it
+        # Track main content - handle multiple documentation structures
+        # Priority order (most specific first):
+        # 1. <main id="main-content"> (Just-the-docs specific)
+        # 2. <main> with data-pagefind-body (legacy _site)
+        # 3. <div class="notes-margin"> (release structure)
+        # 4. <div id="main-content"> (alternative Just-the-docs)
         if not self.found_main:
-            if (tag == 'main' or 
-                attrs_dict.get('data-pagefind-body') is not None or
-                (tag == 'div' and 'notes-margin' in attrs_dict.get('class', ''))):
+            is_main_content = False
+            
+            # Check for main tag with id="main-content" (highest priority)
+            if tag == 'main' and attrs_dict.get('id') == 'main-content':
+                is_main_content = True
+            # Check for main tag with data-pagefind-body
+            elif tag == 'main' and attrs_dict.get('data-pagefind-body') is not None:
+                is_main_content = True
+            # Check for generic main tag
+            elif tag == 'main':
+                is_main_content = True
+            # Check for div with notes-margin class (legacy)
+            elif tag == 'div' and 'notes-margin' in attrs_dict.get('class', ''):
+                is_main_content = True
+            # Check for div with id="main-content" (fallback for Just-the-docs)
+            elif tag == 'div' and attrs_dict.get('id') == 'main-content':
+                is_main_content = True
+                
+            if is_main_content:
                 self.in_main = True
                 self.found_main = True
                 self.depth = 0