From b46cb9598d70bba893f0c6710be681e55f976066 Mon Sep 17 00:00:00 2001
From: Tim Lim <tim@thewheatfield.org>
Date: Mon, 16 Oct 2023 20:49:38 +0800
Subject: [PATCH] Allow Python scraper to keep empty spans with ids

---
 lib/docs/filters/sphinx/clean_html.rb | 8 +++++---
 lib/docs/scrapers/python.rb           | 6 ++++++
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/lib/docs/filters/sphinx/clean_html.rb b/lib/docs/filters/sphinx/clean_html.rb
index 126ff47a..36bb96d9 100644
--- a/lib/docs/filters/sphinx/clean_html.rb
+++ b/lib/docs/filters/sphinx/clean_html.rb
@@ -36,9 +36,11 @@ module Docs
           node.replace(pre)
         end
 
-        css('span[id]:empty').each do |node|
-          (node.next_element || node.previous_element)['id'] ||= node['id'] if node.next_element || node.previous_element
-          node.remove
+        unless context[:sphinx_keep_empty_ids]
+          css('span[id]:empty').each do |node|
+            (node.next_element || node.previous_element)['id'] ||= node['id'] if node.next_element || node.previous_element
+            node.remove
+          end
         end
 
         css('.section').each do |node|
diff --git a/lib/docs/scrapers/python.rb b/lib/docs/scrapers/python.rb
index 39448bd5..52b2505f 100644
--- a/lib/docs/scrapers/python.rb
+++ b/lib/docs/scrapers/python.rb
@@ -7,6 +7,12 @@ module Docs
       code: 'https://github.com/python/cpython'
     }
 
+    # bypass the clean_text filter as it removes empty span with ids
+    options[:clean_text] = false
+
+    # bypass sphinx modifying empty ids
+    options[:sphinx_keep_empty_ids] = true
+
     options[:skip_patterns] = [/whatsnew/]
     options[:skip] = %w(
       library/2to3.html