fix missing images in paras

tests/io_tests/test_html_extraction.py

@@ -7,7 +7,7 @@ including styled content within paragraphs and block-level elements.
 
 import unittest
 from pyWebLayout.io.readers.html_extraction import parse_html_string
-from pyWebLayout.abstract.block import Paragraph, Heading, HeadingLevel, Quote, CodeBlock, HList, ListStyle, Table
+from pyWebLayout.abstract.block import Paragraph, Heading, HeadingLevel, Quote, CodeBlock, HList, ListStyle, Table, Image
 from pyWebLayout.abstract.document import Document
 from pyWebLayout.style import Font, FontWeight, FontStyle, TextDecoration
 
@@ -585,5 +585,103 @@ class TestHTMLFontRegistryIntegration(unittest.TestCase):
                            "Should create separate styles for style combinations")
 
 
+class TestHTMLImagesInParagraphs(unittest.TestCase):
+    """Test cases for handling images inside paragraph tags."""
+
+    def setUp(self):
+        """Set up test fixtures."""
+        self.base_font = Font(font_size=14)
+
+    def test_image_only_paragraph(self):
+        """Test paragraph containing only an image (common in EPUBs)."""
+        html = '<p><img src="cover.jpg" alt="Book Cover"/></p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should parse as an Image block, not a Paragraph
+        self.assertGreater(len(blocks), 0, "Should parse at least one block")
+
+        # Check that we have an Image block
+        image_blocks = [b for b in blocks if isinstance(b, Image)]
+        self.assertGreater(len(image_blocks), 0, "Should have at least one Image block")
+
+        # Verify image properties
+        img = image_blocks[0]
+        self.assertEqual(img.source, "cover.jpg")
+        self.assertEqual(img.alt_text, "Book Cover")
+
+    def test_paragraph_with_multiple_images(self):
+        """Test paragraph with multiple images."""
+        html = '<p><img src="img1.jpg" alt="First"/><img src="img2.jpg" alt="Second"/></p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should have multiple Image blocks
+        image_blocks = [b for b in blocks if isinstance(b, Image)]
+        self.assertEqual(len(image_blocks), 2, "Should have two Image blocks")
+
+        # Verify both images were parsed
+        sources = [img.source for img in image_blocks]
+        self.assertIn("img1.jpg", sources)
+        self.assertIn("img2.jpg", sources)
+
+    def test_paragraph_with_text_and_image(self):
+        """Test paragraph with mixed text and image content."""
+        html = '<p>Some text before <img src="inline.jpg" alt="Inline"/> and after</p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should have both paragraph and image blocks
+        paragraphs = [b for b in blocks if isinstance(b, Paragraph)]
+        images = [b for b in blocks if isinstance(b, Image)]
+
+        self.assertGreater(len(paragraphs), 0, "Should have a Paragraph block for text")
+        self.assertGreater(len(images), 0, "Should have an Image block")
+
+        # Verify image was parsed
+        self.assertEqual(images[0].source, "inline.jpg")
+
+        # Verify text was extracted (should have words like "Some", "text", etc.)
+        if paragraphs:
+            words = list(paragraphs[0].words_iter())
+            self.assertGreater(len(words), 0, "Paragraph should have words")
+
+    def test_regular_paragraph_still_works(self):
+        """Test that regular paragraphs without images still work correctly."""
+        html = '<p>Just regular text without any images.</p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should be exactly one Paragraph block
+        self.assertEqual(len(blocks), 1, "Should have exactly one block")
+        self.assertIsInstance(blocks[0], Paragraph, "Should be a Paragraph block")
+
+        # Should not have any Image blocks
+        image_blocks = [b for b in blocks if isinstance(b, Image)]
+        self.assertEqual(len(image_blocks), 0, "Should have no Image blocks")
+
+    def test_image_with_width_and_height(self):
+        """Test image parsing with width and height attributes."""
+        html = '<p><img src="sized.jpg" alt="Sized Image" width="400" height="300"/></p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should have an Image block
+        image_blocks = [b for b in blocks if isinstance(b, Image)]
+        self.assertEqual(len(image_blocks), 1, "Should have one Image block")
+
+        # Verify dimensions were parsed
+        img = image_blocks[0]
+        self.assertEqual(img.width, 400)
+        self.assertEqual(img.height, 300)
+
+    def test_nested_paragraph_with_image_in_span(self):
+        """Test image inside nested inline elements."""
+        html = '<p><span><img src="nested.jpg" alt="Nested"/></span></p>'
+        blocks = parse_html_string(html, base_font=self.base_font)
+
+        # Should still extract the image
+        image_blocks = [b for b in blocks if isinstance(b, Image)]
+        self.assertGreater(len(image_blocks), 0, "Should find image even when nested")
+
+        # Verify image was parsed correctly
+        self.assertEqual(image_blocks[0].source, "nested.jpg")
+
+
 if __name__ == '__main__':
     unittest.main()