From 92b9566b0af2f8af60dfbe17268955d5a3a80b1d Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Mon, 19 Aug 2024 12:23:20 +0530
Subject: [PATCH] Store page list in book metadata when rendering

---
 src/calibre/ebooks/oeb/polish/toc.py | 21 ++++++++++++++++++---
 src/calibre/srv/render_book.py       |  5 ++++-
 2 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/src/calibre/ebooks/oeb/polish/toc.py b/src/calibre/ebooks/oeb/polish/toc.py
index 1147f26938..602b278940 100644
--- a/src/calibre/ebooks/oeb/polish/toc.py
+++ b/src/calibre/ebooks/oeb/polish/toc.py
@@ -219,18 +219,33 @@ def parse_nav(container, nav_name):
     root = container.parsed(nav_name)
     toc_root = TOC()
     toc_root.lang = toc_root.uid = None
+    seen_toc = seen_pagelist = False
     et = '{%s}type' % EPUB_NS
-    for nav in root.iterdescendants(XHTML('nav')):
-        if nav.get(et) == 'toc':
+    for nav in XPath('descendant::h:nav[@epub:type]')(root):
+        nt = nav.get(et)
+        if nt == 'toc' and not seen_toc:
             ol = first_child(nav, XHTML('ol'))
             if ol is not None:
+                seen_toc = True
                 process_nav_node(container, ol, toc_root, nav_name)
                 for h in nav.iterchildren(*map(XHTML, 'h1 h2 h3 h4 h5 h6'.split())):
                     text = etree.tostring(h, method='text', encoding='unicode', with_tail=False) or h.get('title')
                     if text:
                         toc_root.toc_title = text
                         break
-                break
+        elif nt == 'page-list' and not seen_pagelist:
+            ol = first_child(nav, XHTML('ol'))
+            if ol is not None and not seen_pagelist:
+                seen_pagelist = True
+                for li in ol.iterchildren(XHTML('li')):
+                    for a in li.iterchildren(XHTML('a')):
+                        href = a.get('href')
+                        if href:
+                            text = (etree.tostring(a, method='text', encoding='unicode', with_tail=False) or a.get('title')).strip()
+                            if text:
+                                dest = nav_name if href.startswith('#') else container.href_to_name(href, base=nav_name)
+                                frag = urlparse(href).fragment or None
+                                toc_root.page_list.append({'dest': dest, 'pagenum': text, 'frag': frag})
     return toc_root
 
 
diff --git a/src/calibre/srv/render_book.py b/src/calibre/srv/render_book.py
index bce7fb7ac4..8e5c52e289 100644
--- a/src/calibre/srv/render_book.py
+++ b/src/calibre/srv/render_book.py
@@ -713,7 +713,9 @@ def process_exploded_book(
         name == 'mimetype' or not container.has_name_and_is_not_empty(name)}
     raster_cover_name, titlepage_name = create_cover_page(container, input_fmt.lower(), is_comic, book_metadata)
 
-    toc = get_toc(container, verify_destinations=False).to_dict(count())
+    tocobj = get_toc(container, verify_destinations=False)
+    page_list = tocobj.page_list or []
+    toc = tocobj.to_dict(count())
     if not toc or not toc.get('children'):
         toc = from_xpaths(container, ['//h:h1', '//h:h2', '//h:h3']).to_dict(count())
     spine = [name for name, is_linear in container.spine_names]
@@ -746,6 +748,7 @@ def process_exploded_book(
         'landmarks': landmarks,
         'link_to_map': {},
         'page_progression_direction': page_progression_direction,
+        'page_list': page_list,
     }
 
     names = sorted(