Extract title text also from img alt attribute

mkdocs/structure/pages.py

@@ -2,6 +2,7 @@
 
 import copy
 import enum
+import io
 import logging
 import posixpath
 import warnings
@@ -548,6 +549,18 @@ def handle_starttag(self, tag: str, attrs: Sequence[tuple[str, str]]) -> None:
         return super().handle_starttag(tag, attrs)
 
 
+def _extract_text(el: etree.Element, buf: io.StringIO) -> None:
+    if text := el.get('alt'):
+        buf.write(text)
+        return
+    if text := el.text:
+        buf.write(text)
+    for child in el:
+        _extract_text(child, buf)
+        if text := child.tail:
+            buf.write(text)
+
+
 class _ExtractTitleTreeprocessor(markdown.treeprocessors.Treeprocessor):
     title: str | None = None
     postprocessors: Sequence[markdown.postprocessors.Postprocessor] = ()
@@ -559,8 +572,10 @@ def run(self, root: etree.Element) -> etree.Element:
                 if len(el) > 0 and el[-1].tag == 'a' and not (el[-1].tail or '').strip():
                     el = copy.copy(el)
                     del el[-1]
-                # Extract the text only, recursively.
-                title = ''.join(el.itertext())
+                # Extract the text only.
+                buf = io.StringIO()
+                _extract_text(el, buf)
+                title = buf.getvalue()
                 # Unescape per Markdown implementation details.
                 for pp in self.postprocessors:
                     title = pp.run(title)

mkdocs/tests/structure/page_tests.py

@@ -315,7 +315,7 @@ def test_page_title_from_markdown(self):
         self.assertEqual(pg.parent, None)
         self.assertEqual(pg.previous_page, None)
         self.assertEqual(pg.title, 'Welcome to MkDocs')
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Welcome to MkDocs')
 
     _SETEXT_CONTENT = dedent(
@@ -335,7 +335,7 @@ def test_page_title_from_setext_markdown(self, docs_dir):
         self.assertIsNone(pg.title)
         pg.read_source(cfg)
         self.assertEqual(pg.title, 'Testing setext title')
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Welcome to MkDocs Setext')
 
     @tempdir(files={'testing_setext_title.md': _SETEXT_CONTENT})
@@ -347,12 +347,12 @@ def test_page_title_from_markdown_stripped_anchorlinks(self, docs_dir):
         fl = File('testing_setext_title.md', docs_dir, docs_dir, use_directory_urls=True)
         pg = Page(None, fl, cfg)
         pg.read_source(cfg)
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Welcome to MkDocs Setext')
 
     _FORMATTING_CONTENT = dedent(
         '''
-        # \\*Hello --- *beautiful* `world`
+        # \\*Hello --- *beautiful* `wor<dl>`
 
         Hi.
         '''
@@ -365,8 +365,8 @@ def test_page_title_from_markdown_strip_formatting(self, docs_dir):
         fl = File('testing_formatting.md', docs_dir, docs_dir, use_directory_urls=True)
         pg = Page(None, fl, cfg)
         pg.read_source(cfg)
-        pg.render(cfg, fl)
-        self.assertEqual(pg.title, '*Hello — beautiful world')
+        pg.render(cfg, Files([fl]))
+        self.assertEqual(pg.title, '*Hello — beautiful wor<dl>')
 
     _RAW_CONTENT = dedent(
         '''
@@ -382,9 +382,21 @@ def test_page_title_from_markdown_strip_raw_tags(self, docs_dir):
         fl = File('testing_raw_content.md', docs_dir, docs_dir, use_directory_urls=True)
         pg = Page(None, fl, cfg)
         pg.read_source(cfg)
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Hello world')
 
+    _IMAGE_CONTENT = '''# ![😄](hah.png)\n'''
+
+    @tempdir(files={'testing_image.md': _IMAGE_CONTENT, 'hah.png': ''})
+    def test_page_title_from_markdown_strip_image(self, docs_dir):
+        cfg = load_config()
+        fl = File('testing_image.md', docs_dir, docs_dir, use_directory_urls=True)
+        fl2 = File('hah.png', docs_dir, docs_dir, use_directory_urls=True)
+        pg = Page(None, fl, cfg)
+        pg.read_source(cfg)
+        pg.render(cfg, Files([fl, fl2]))
+        self.assertEqual(pg.title, '😄')
+
     _ATTRLIST_CONTENT = dedent(
         '''
         # Welcome to MkDocs Attr { #welcome }
@@ -409,7 +421,7 @@ def test_page_title_from_markdown_preserved_attr_list(self, docs_dir):
         fl = File('testing_attr_list.md', docs_dir, docs_dir, use_directory_urls=True)
         pg = Page(None, fl, cfg)
         pg.read_source(cfg)
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Welcome to MkDocs Attr { #welcome }')
 
     def test_page_title_from_meta(self):
@@ -435,7 +447,7 @@ def test_page_title_from_meta(self):
         self.assertEqual(pg.previous_page, None)
         self.assertEqual(pg.title, 'A Page Title')
         self.assertEqual(pg.toc, [])
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'A Page Title')
 
     def test_page_title_from_filename(self):
@@ -460,7 +472,7 @@ def test_page_title_from_filename(self):
         self.assertEqual(pg.parent, None)
         self.assertEqual(pg.previous_page, None)
         self.assertEqual(pg.title, 'Page title')
-        pg.render(cfg, fl)
+        pg.render(cfg, Files([fl]))
         self.assertEqual(pg.title, 'Page title')
 
     def test_page_title_from_capitalized_filename(self):
@@ -721,7 +733,7 @@ def test_page_render(self):
         pg.read_source(cfg)
         self.assertEqual(pg.content, None)
         self.assertEqual(pg.toc, [])
-        pg.render(cfg, [fl])
+        pg.render(cfg, Files([fl]))
         self.assertTrue(
             pg.content.startswith('<h1 id="welcome-to-mkdocs">Welcome to MkDocs</h1>\n')
         )