add BW image conversion

pyWebLayout/io/readers/epub_reader.py

@@ -8,10 +8,11 @@ to pyWebLayout's abstract document model.
 import os
 import zipfile
 import tempfile
-from typing import Dict, List, Optional, Any, Tuple
+from typing import Dict, List, Optional, Any, Tuple, Callable
 import xml.etree.ElementTree as ET
 import re
 import urllib.parse
+from PIL import Image as PILImage, ImageOps
 
 from pyWebLayout.abstract.document import Document, Book, Chapter, MetadataType
 from pyWebLayout.io.readers.html_extraction import parse_html_string
@@ -27,6 +28,30 @@ NAMESPACES = {
 }
 
 
+def default_eink_processor(img: PILImage.Image) -> PILImage.Image:
+    """
+    Process image for 4-bit e-ink display using PIL only.
+    Applies histogram equalization and 4-bit quantization.
+    
+    Args:
+        img: PIL Image to process
+        
+    Returns:
+        Processed PIL Image in L mode (grayscale) with 4-bit quantization
+    """
+    # Convert to grayscale if needed
+    if img.mode != 'L':
+        img = img.convert('L')
+    
+    # Apply histogram equalization for contrast enhancement
+    img = ImageOps.equalize(img)
+    
+    # Quantize to 4-bit (16 grayscale levels: 0, 17, 34, ..., 255)
+    img = img.point(lambda x: (x // 16) * 17)
+    
+    return img
+
+
 class EPUBReader:
     """
     Reader for EPUB documents.
@@ -35,14 +60,19 @@ class EPUBReader:
     pyWebLayout's abstract document model.
     """
     
-    def __init__(self, epub_path: str):
+    def __init__(self, epub_path: str, image_processor: Optional[Callable[[PILImage.Image], PILImage.Image]] = default_eink_processor):
         """
         Initialize an EPUB reader.
         
         Args:
             epub_path: Path to the EPUB file
+            image_processor: Optional function to process images for display optimization.
+                           Defaults to default_eink_processor for 4-bit e-ink displays.
+                           Set to None to disable image processing.
+                           Custom processor should accept and return a PIL Image.
         """
         self.epub_path = epub_path
+        self.image_processor = image_processor
         self.book = Book()
         self.temp_dir = None
         self.content_dir = None
@@ -70,6 +100,9 @@ class EPUBReader:
             # Add chapters to the book
             self._add_chapters()
             
+            # Process images for e-ink display optimization
+            self._process_content_images()
+            
             return self.book
             
         finally:
@@ -365,6 +398,14 @@ class EPUBReader:
             # Create a copy to ensure all data is in memory
             pil_image = pil_image.copy()
             
+            # Apply image processing if enabled
+            if self.image_processor:
+                try:
+                    pil_image = self.image_processor(pil_image)
+                except Exception as e:
+                    print(f"Warning: Image processing failed for cover: {str(e)}")
+                    # Continue with unprocessed image
+            
             # Create an AbstractImage block with the cover image path
             cover_image = AbstractImage(source=cover_path, alt_text="Cover Image")
             
@@ -386,6 +427,33 @@ class EPUBReader:
             if hasattr(self.book, 'chapters') and cover_chapter in self.book.chapters:
                 self.book.chapters.remove(cover_chapter)
     
+    def _process_chapter_images(self, chapter: Chapter):
+        """
+        Process images in a single chapter.
+        
+        Args:
+            chapter: The chapter containing images to process
+        """
+        from pyWebLayout.abstract.block import Image as AbstractImage
+        
+        for block in chapter.blocks:
+            if isinstance(block, AbstractImage):
+                # Only process if image has been loaded and processor is enabled
+                if hasattr(block, '_loaded_image') and block._loaded_image:
+                    try:
+                        block._loaded_image = self.image_processor(block._loaded_image)
+                    except Exception as e:
+                        print(f"Warning: Image processing failed for image '{block.alt_text}': {str(e)}")
+                        # Continue with unprocessed image
+    
+    def _process_content_images(self):
+        """Apply image processing to all images in chapters."""
+        if not self.image_processor:
+            return
+        
+        for chapter in self.book.chapters:
+            self._process_chapter_images(chapter)
+    
     def _add_chapters(self):
         """Add chapters to the book based on the spine and TOC."""
         # Add cover chapter first if available