diff --git a/CHANGELOG.md b/CHANGELOG.md
index a987142..37a0963 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,7 @@
- Added profile system to voice mixer, allowing users to create and manage multiple voice profiles.
- Improvements in the voice mixer, mostly for organizing controls and enhancing user experience.
- Added icons for flags and genders in the GUI, making it easier to identify different options.
+- Improved the content and chapter extraction process for EPUB files, ensuring better handling of various structures.
- Switched to platformdirs for determining the correct desktop path, instead of using old methods.
- Fixed preview voices was not using GPU acceleration, which was causing performance issues.
- Improvements in code and documentation.
\ No newline at end of file
diff --git a/README.md b/README.md
index 55abb0c..4ee7998 100644
--- a/README.md
+++ b/README.md
@@ -101,7 +101,7 @@ Here’s Abogen in action: in this demo, it processes ∼3,000 characters of tex
## `Voice Mixer`
-With voice mixer, you can create custom voices by mixing different voice models. You can adjust the weight of each voice and save your custom voice as a profile for future use. The voice mixer allows you to create unique and personalized voices. (Huge thanks to @jborza for making this possible through his contributions in #5)
+With voice mixer, you can create custom voices by mixing different voice models. You can adjust the weight of each voice and save your custom voice as a profile for future use. The voice mixer allows you to create unique and personalized voices. (Huge thanks to [@jborza](https://github.com/jborza) for making this possible through his contributions in [#5](https://github.com/denizsafak/abogen/pull/5))
## `Supported Languages`
```
@@ -168,8 +168,9 @@ Feel free to explore the code and make any changes you like.
## `Credits`
- Abogen uses [Kokoro](https://github.com/hexgrad/kokoro) for its high-quality, natural-sounding text-to-speech synthesis. Huge thanks to the Kokoro team for making this possible.
- Thanks to [@wojiushixiaobai](https://github.com/wojiushixiaobai) for [Embedded Python](https://github.com/wojiushixiaobai/Python-Embed-Win64) packages. These modified packages include pip pre-installed, enabling Abogen to function as a standalone application without requiring users to separately install Python in Windows.
+- Thanks to creators of [EbookLib](https://github.com/aerkalov/ebooklib), a Python library for reading and writing ePub files, which is used for extracting text from ePub files.
- Special thanks to the [PyQt](https://www.riverbankcomputing.com/software/pyqt/) team for providing the cross-platform GUI toolkit that powers Abogen's interface.
-- Icons: [US](https://icons8.com/icon/aRiu1GGi6Aoe/usa), [Great Britain](https://icons8.com/icon/t3NE3BsOAQwq/great-britain), [Spain](https://icons8.com/icon/ly7tzANRt33n/spain), [France](https://icons8.com/icon/3muzEmi4dpD5/france), [India](https://icons8.com/icon/esGVrxg9VCJ1/india), [Italy](https://icons8.com/icon/PW8KZnP7qXzO/italy), [Japan](https://icons8.com/icon/McQbrq9qaQye/japan), [Brazil](https://icons8.com/icon/zHmH8HpOmM90/brazil), [China](https://icons8.com/icon/Ej50Oe3crXwF/china), [Female](https://icons8.com/icon/uI49hxbpxTkp/female), [Male](https://icons8.com/icon/12351/male) and [Voice Id](https://icons8.com/icon/GskSeVoroQ7u/voice-id), [Person](https://icons8.com/icon/34105/person) icons by [Icons8](https://icons8.com/).
+- Icons: [US](https://icons8.com/icon/aRiu1GGi6Aoe/usa), [Great Britain](https://icons8.com/icon/t3NE3BsOAQwq/great-britain), [Spain](https://icons8.com/icon/ly7tzANRt33n/spain), [France](https://icons8.com/icon/3muzEmi4dpD5/france), [India](https://icons8.com/icon/esGVrxg9VCJ1/india), [Italy](https://icons8.com/icon/PW8KZnP7qXzO/italy), [Japan](https://icons8.com/icon/McQbrq9qaQye/japan), [Brazil](https://icons8.com/icon/zHmH8HpOmM90/brazil), [China](https://icons8.com/icon/Ej50Oe3crXwF/china), [Female](https://icons8.com/icon/uI49hxbpxTkp/female), [Male](https://icons8.com/icon/12351/male) and [Voice Id](https://icons8.com/icon/GskSeVoroQ7u/voice-id) icons by [Icons8](https://icons8.com/).
## `License`
This project is available under the MIT License - see the [LICENSE](https://github.com/denizsafak/abogen/blob/main/LICENSE) file for details.
diff --git a/abogen/VERSION b/abogen/VERSION
index e6d5cb8..e4c0d46 100644
--- a/abogen/VERSION
+++ b/abogen/VERSION
@@ -1 +1 @@
-1.0.2
\ No newline at end of file
+1.0.3
\ No newline at end of file
diff --git a/abogen/book_handler.py b/abogen/book_handler.py
index 3248f3d..fb79b9d 100644
--- a/abogen/book_handler.py
+++ b/abogen/book_handler.py
@@ -4,7 +4,7 @@ import ebooklib
import base64
import fitz # PyMuPDF for PDF support
from ebooklib import epub
-from bs4 import BeautifulSoup
+from bs4 import BeautifulSoup, NavigableString
from PyQt5.QtWidgets import (
QDialog,
QTreeWidget,
@@ -24,6 +24,13 @@ from PyQt5.QtWidgets import (
from PyQt5.QtCore import Qt
from utils import clean_text, calculate_text_length
import os
+import logging # Add logging
+import urllib.parse
+
+# Setup logging
+logging.basicConfig(
+ level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
+)
class HandlerDialog(QDialog):
@@ -59,7 +66,37 @@ class HandlerDialog(QDialog):
self.merge_chapters_at_end = HandlerDialog._merge_chapters_at_end
# Load the book based on file type
- self.book = epub.read_epub(book_path) if self.file_type == "epub" else None
+ try:
+ self.book = epub.read_epub(book_path) if self.file_type == "epub" else None
+ except KeyError as e:
+ logging.error(
+ f"EPUB file is missing a referenced file: {e}. Skipping missing file."
+ )
+ # Try to patch ebooklib to skip missing files (monkey-patch read_file)
+ import types
+
+ orig_read_file = None
+ try:
+ from ebooklib import epub as _epub_module
+
+ reader_class = _epub_module.EpubReader
+ orig_read_file = reader_class.read_file
+
+ def safe_read_file(self, name):
+ try:
+ return orig_read_file(self, name)
+ except KeyError:
+ logging.warning(
+ f"Missing file in EPUB: {name}. Returning empty bytes."
+ )
+ return b""
+
+ reader_class.read_file = safe_read_file
+ self.book = epub.read_epub(book_path)
+ reader_class.read_file = orig_read_file # Restore
+ except Exception as patch_e:
+ logging.error(f"Failed to patch ebooklib for missing files: {patch_e}")
+ raise e
self.pdf_doc = fitz.open(book_path) if self.file_type == "pdf" else None
# Extract book metadata
@@ -129,250 +166,18 @@ class HandlerDialog(QDialog):
def _preprocess_content(self):
"""Pre-process content from the document"""
if self.file_type == "epub":
- # Always process EPUB content using the anchor-based approach
- self._process_epub_content()
+ try:
+ self._process_epub_content_nav() # Use the new navigation-based method
+ except Exception as e:
+ logging.error(
+ f"Error processing EPUB with navigation: {e}. Falling back to TOC/spine.",
+ exc_info=True,
+ )
+ # Fallback to a simpler spine-based processing if nav fails
+ self._process_epub_content_spine_fallback()
else:
self._preprocess_pdf_content()
- def _process_epub_content(self, split_anchors=True):
- """
- Process EPUB content by globally ordering TOC entries and slicing content between them.
- Ensures all content between defined TOC start points is captured.
- """
- # split_anchors parameter kept for compatibility but always treated as True
- book = self.book
-
- # 1. Cache all document HTML and determine spine order
- self.doc_content = {}
- # Correctly get hrefs from spine items
- spine_docs = []
- for spine_item_tuple in book.spine:
- item_id = spine_item_tuple[0]
- item = book.get_item_with_id(item_id)
- if item:
- spine_docs.append(item.get_name()) # Use get_name() for href
- else:
- print(
- f"Warning: Spine item with id '{item_id}' not found in book items."
- )
-
- doc_order = {href: i for i, href in enumerate(spine_docs)}
-
- for item in book.get_items_of_type(ebooklib.ITEM_DOCUMENT):
- href = item.get_name()
- if href in doc_order: # Only process docs in spine
- try:
- html_content = item.get_content().decode("utf-8", errors="ignore")
- self.doc_content[href] = html_content
- except Exception:
- self.doc_content[href] = "" # Handle decoding errors
-
- # 2. Get all TOC entries with hrefs and determine their positions
- toc_entries_with_pos = []
-
- def find_position(doc_href, fragment_id):
- if doc_href not in self.doc_content:
- return -1
- html_content = self.doc_content[doc_href]
- if not fragment_id: # No fragment, position is 0
- return 0
-
- # Find position of fragment identifier (id= or name=)
- id_match_str = f'id="{fragment_id}"'
- name_match_str = f'name="{fragment_id}"'
-
- id_pos = html_content.find(id_match_str)
- name_pos = html_content.find(name_match_str)
-
- pos = -1
- if id_pos != -1 and name_pos != -1:
- pos = min(id_pos, name_pos)
- elif id_pos != -1:
- pos = id_pos
- elif name_pos != -1:
- pos = name_pos
- else:
- return -1 # Anchor not found by simple string search
-
- # Backtrack to the start of the tag '<'
- tag_start_pos = html_content.rfind("<", 0, pos)
- return (
- tag_start_pos if tag_start_pos != -1 else 0
- ) # Default to 0 if '<' not found
-
- def collect_toc_entries(entries):
- collected = []
- for entry in entries:
- href, title = None, "Unknown"
- children = []
- entry_obj = None # Store the original entry object
-
- if isinstance(entry, ebooklib.epub.Link):
- href, title = entry.href, entry.title or entry.href
- entry_obj = entry
- elif isinstance(entry, tuple) and len(entry) >= 1:
- section_or_link = entry[0]
- entry_obj = section_or_link
- if isinstance(section_or_link, ebooklib.epub.Section):
- title = section_or_link.title
- href = getattr(section_or_link, "href", None)
- elif isinstance(section_or_link, ebooklib.epub.Link):
- href, title = (
- section_or_link.href,
- section_or_link.title or section_or_link.href,
- )
-
- if len(entry) > 1 and isinstance(entry[1], list):
- children = entry[1]
-
- if href:
- base_href, fragment = (
- href.split("#", 1) if "#" in href else (href, None)
- )
- if (
- base_href in doc_order
- ): # Only consider entries pointing to spine documents
- position = find_position(base_href, fragment)
- if position != -1: # Only add if position is valid
- collected.append(
- {
- "href": href, # Use the original href from TOC as the key
- "title": title,
- "doc_href": base_href,
- "position": position,
- "doc_order": doc_order[base_href],
- }
- )
-
- if children:
- collected.extend(collect_toc_entries(children))
- return collected
-
- all_toc_entries = collect_toc_entries(self.book.toc)
-
- # Handle case where book has no TOC or empty TOC
- if not all_toc_entries:
- # Create a synthetic TOC entry for the first spine document
- if spine_docs:
- # Process all content as a single chapter
- all_content_html = ""
- for doc_href in spine_docs:
- all_content_html += self.doc_content.get(doc_href, "")
-
- if all_content_html:
- soup = BeautifulSoup(all_content_html, "html.parser")
- text = clean_text(soup.get_text()).strip()
-
- # Use the first spine document as the identifier
- first_doc = spine_docs[0]
- self.content_texts[first_doc] = text
- self.content_lengths[first_doc] = len(text)
-
- # Create a synthetic TOC entry for tree building
- self.book.toc = [(epub.Link(first_doc, "Main Content", first_doc),)]
- return
-
- # 3. Sort TOC entries globally
- all_toc_entries.sort(key=lambda x: (x["doc_order"], x["position"]))
-
- # 4. Slice content between sorted entries
- self.content_texts = {}
- self.content_lengths = {}
- num_entries = len(all_toc_entries)
-
- for i in range(num_entries):
- current_entry = all_toc_entries[i]
- current_href = current_entry["href"]
- current_doc = current_entry["doc_href"]
- current_pos = current_entry["position"]
- current_doc_html = self.doc_content.get(current_doc, "")
-
- start_slice_pos = current_pos
- slice_html = ""
-
- # Find the start of the next TOC entry
- next_entry = all_toc_entries[i + 1] if (i + 1) < num_entries else None
-
- if next_entry:
- next_doc = next_entry["doc_href"]
- next_pos = next_entry["position"]
-
- if current_doc == next_doc:
- # Next entry is in the same document
- slice_html = current_doc_html[start_slice_pos:next_pos]
- else:
- # Next entry is in a different document
- # Take content from current position to end of current document
- slice_html = current_doc_html[start_slice_pos:]
- # Include content from intermediate documents in the spine
- current_doc_index = current_entry["doc_order"]
- next_doc_index = next_entry["doc_order"]
- for doc_idx in range(current_doc_index + 1, next_doc_index):
- intermediate_doc_href = spine_docs[doc_idx]
- slice_html += self.doc_content.get(intermediate_doc_href, "")
- # Add content from the beginning of the next document up to the next entry's position
- next_doc_html = self.doc_content.get(next_doc, "")
- slice_html += next_doc_html[:next_pos]
- else:
- # This is the last TOC entry
- # Take content from current position to end of current document
- slice_html = current_doc_html[start_slice_pos:]
- # Include content from all remaining documents in the spine
- current_doc_index = current_entry["doc_order"]
- for doc_idx in range(current_doc_index + 1, len(spine_docs)):
- intermediate_doc_href = spine_docs[doc_idx]
- slice_html += self.doc_content.get(intermediate_doc_href, "")
-
- # 5. Extract text and store
- slice_soup = BeautifulSoup(slice_html, "html.parser")
-
- # Remove sup and sub tags from the HTML before extracting text
- for tag in slice_soup.find_all(["sup", "sub"]):
- tag.decompose()
-
- text = clean_text(slice_soup.get_text()).strip()
- self.content_texts[current_href] = text # Store using the original TOC href
- self.content_lengths[current_href] = len(text)
-
- # 6. Handle content BEFORE the first TOC entry
- if all_toc_entries:
- first_entry = all_toc_entries[0]
- first_doc_href = first_entry["doc_href"]
- first_pos = first_entry["position"]
- first_doc_order = first_entry["doc_order"]
- prefix_html = ""
- # Include content from documents before the first entry's document
- for doc_idx in range(first_doc_order):
- intermediate_doc_href = spine_docs[doc_idx]
- prefix_html += self.doc_content.get(intermediate_doc_href, "")
- # Include content from the start of the first entry's document up to its position
- first_doc_html = self.doc_content.get(first_doc_href, "")
- prefix_html += first_doc_html[:first_pos]
-
- if prefix_html.strip():
- prefix_soup = BeautifulSoup(prefix_html, "html.parser")
- # Remove sup and sub tags
- for tag in prefix_soup.find_all(["sup", "sub"]):
- tag.decompose()
- prefix_text = clean_text(prefix_soup.get_text()).strip()
-
- if prefix_text:
- # Create a new chapter for content before the first TOC entry
- # Use a synthetic href to avoid collision with real TOC entries
- prefix_chapter_href = "prefix_content_chapter"
- self.content_texts[prefix_chapter_href] = prefix_text
- self.content_lengths[prefix_chapter_href] = len(prefix_text)
-
- # Add a new entry to the TOC for the prefix content
- prefix_link = epub.Link(
- prefix_chapter_href, "Introduction", prefix_chapter_href
- )
- # Insert at beginning of TOC
- if isinstance(self.book.toc, list):
- self.book.toc.insert(0, (prefix_link,))
- else:
- self.book.toc = [(prefix_link,)] + (self.book.toc or [])
-
def _preprocess_pdf_content(self):
"""Pre-process all page contents from PDF document"""
for page_num in range(len(self.pdf_doc)):
@@ -395,16 +200,624 @@ class HandlerDialog(QDialog):
self.content_texts[page_id] = text
self.content_lengths[page_id] = calculate_text_length(text)
- def _build_tree(self):
- """Build tree based on file type"""
- if self.file_type == "epub":
- self._build_epub_tree()
- else:
- self._build_pdf_tree()
+ def _process_epub_content_spine_fallback(self):
+ """Fallback EPUB processing based purely on spine order."""
+ logging.info("Using spine fallback for EPUB processing.")
+ self.doc_content = {}
+ spine_docs = []
+ for spine_item_tuple in self.book.spine:
+ item_id = spine_item_tuple[0]
+ item = self.book.get_item_with_id(item_id)
+ if item:
+ spine_docs.append(item.get_name())
+ else:
+ logging.warning(f"Spine item with id '{item_id}' not found.")
- def _build_epub_tree(self):
- """Build the tree for EPUB files from TOC"""
+ # Cache content
+ for item in self.book.get_items_of_type(ebooklib.ITEM_DOCUMENT):
+ href = item.get_name()
+ if href in spine_docs:
+ try:
+ html_content = item.get_content().decode("utf-8", errors="ignore")
+ self.doc_content[href] = html_content
+ except Exception as e:
+ logging.error(f"Error decoding content for {href}: {e}")
+ self.doc_content[href] = ""
+
+ # Create a simple TOC based on spine order
+ synthetic_toc = []
+ self.content_texts = {}
+ self.content_lengths = {}
+ for i, doc_href in enumerate(spine_docs):
+ html_content = self.doc_content.get(doc_href, "")
+ if html_content:
+ soup = BeautifulSoup(html_content, "html.parser")
+ # Remove sup and sub tags
+ for tag in soup.find_all(["sup", "sub"]):
+ tag.decompose()
+ text = clean_text(soup.get_text()).strip()
+ if text:
+ # Use doc_href as the identifier
+ self.content_texts[doc_href] = text
+ self.content_lengths[doc_href] = len(text)
+ # Create a synthetic TOC entry
+ title = f"Chapter {i+1}: {doc_href}"
+ # Try to get a better title from or
+ h1 = soup.find("h1")
+ if h1 and h1.get_text(strip=True):
+ title = h1.get_text(strip=True)
+ else:
+ title_tag = soup.find("title")
+ if title_tag and title_tag.get_text(strip=True):
+ title = title_tag.get_text(strip=True)
+
+ synthetic_toc.append(
+ (epub.Link(doc_href, title, doc_href), [])
+ ) # Wrap in tuple and empty list for compatibility
+
+ # Replace book.toc with the synthetic one if it was empty or fallback was triggered
+ if not self.book.toc or not hasattr(
+ self, "processed_nav_structure"
+ ): # Check if nav processing failed
+ self.book.toc = synthetic_toc
+ logging.info(f"Generated synthetic TOC with {len(synthetic_toc)} entries.")
+
+ def _process_epub_content_nav(self):
+ """
+ Process EPUB content using ITEM_NAVIGATION (NAV HTML) or ITEM_NCX.
+ Globally orders navigation entries and slices content between them.
+ """
+ logging.info(
+ "Attempting to process EPUB using navigation document (NAV/NCX)..."
+ )
+ nav_item = None
+ nav_type = None
+
+ # 1. Check ITEM_NAVIGATION for actual NAV HTML (.xhtml/.html)
+ nav_items = list(self.book.get_items_of_type(ebooklib.ITEM_NAVIGATION))
+ if nav_items:
+ # Prefer files explicitly named 'nav.xhtml' or similar
+ preferred_nav = next(
+ (
+ item
+ for item in nav_items
+ if "nav" in item.get_name().lower()
+ and item.get_name().lower().endswith((".xhtml", ".html"))
+ ),
+ None,
+ )
+ if preferred_nav:
+ nav_item = preferred_nav
+ nav_type = "html"
+ logging.info(f"Found preferred NAV HTML item: {nav_item.get_name()}")
+ else:
+ # Check if any ITEM_NAVIGATION is actually HTML
+ html_nav = next(
+ (
+ item
+ for item in nav_items
+ if item.get_name().lower().endswith((".xhtml", ".html"))
+ ),
+ None,
+ )
+ if html_nav:
+ nav_item = html_nav
+ nav_type = "html"
+ logging.info(
+ f"Found NAV HTML item in ITEM_NAVIGATION: {html_nav.get_name()}"
+ )
+
+ # 2. If no NAV HTML found via ITEM_NAVIGATION, check if ITEM_NAVIGATION points to NCX
+ if not nav_item and nav_items:
+ ncx_in_nav = next(
+ (
+ item
+ for item in nav_items
+ if item.get_name().lower().endswith(".ncx")
+ ),
+ None,
+ )
+ if ncx_in_nav:
+ nav_item = ncx_in_nav
+ nav_type = "ncx"
+ logging.info(
+ f"Found NCX item via ITEM_NAVIGATION: {ncx_in_nav.get_name()}"
+ )
+
+ # 3. If still no nav_item, check for ITEM_NCX directly
+ if not nav_item:
+ ncx_items = list(self.book.get_items_of_type(ebooklib.ITEM_NCX))
+ if ncx_items:
+ nav_item = ncx_items[0] # Take the first one
+ nav_type = "ncx"
+ logging.info(f"Found NCX item via ITEM_NCX: {ncx_items[0].get_name()}")
+
+ # 4. If no navigation item found by any method, trigger fallback
+ if not nav_item or not nav_type:
+ logging.warning(
+ "No suitable EPUB navigation document (NAV HTML or NCX) found. Falling back."
+ )
+ raise ValueError("No navigation document found") # Trigger fallback
+
+ # Determine parser based on the confirmed nav_type
+ parser_type = "html.parser" if nav_type == "html" else "xml"
+ logging.info(f"Using parser: '{parser_type}' for {nav_item.get_name()}")
+ try:
+ nav_content = nav_item.get_content().decode("utf-8", errors="ignore")
+ nav_soup = BeautifulSoup(nav_content, parser_type)
+ except Exception as e:
+ logging.error(
+ f"Failed to parse navigation content ({nav_item.get_name()}) using {parser_type}: {e}",
+ exc_info=True,
+ )
+ raise ValueError(
+ f"Failed to parse navigation content: {e}"
+ ) # Trigger fallback
+
+ # --- Rest of the processing logic ---
+ # 1. Cache all document HTML and determine spine order (no changes needed here)
+ self.doc_content = {}
+ spine_docs = []
+ for spine_item_tuple in self.book.spine:
+ item_id = spine_item_tuple[0]
+ item = self.book.get_item_with_id(item_id)
+ if item:
+ spine_docs.append(item.get_name())
+ else:
+ logging.warning(f"Spine item with id '{item_id}' not found.")
+ doc_order = {href: i for i, href in enumerate(spine_docs)}
+ # Add a mapping for unquoted (decoded) hrefs as well
+ doc_order_decoded = {
+ urllib.parse.unquote(href): i for href, i in doc_order.items()
+ }
+
+ # Clear previous content/lengths before processing
+ self.content_texts = {}
+ self.content_lengths = {}
+
+ for item in self.book.get_items_of_type(ebooklib.ITEM_DOCUMENT):
+ href = item.get_name()
+ if href in doc_order or any(
+ href in nav_point.get("src", "")
+ for nav_point in nav_soup.find_all(["content", "a"])
+ ):
+ try:
+ html_content = item.get_content().decode("utf-8", errors="ignore")
+ self.doc_content[href] = html_content
+ except Exception as e:
+ logging.error(f"Error decoding content for {href}: {e}")
+ self.doc_content[href] = ""
+
+ # 2. Extract and order navigation entries globally
+ ordered_nav_entries = []
+
+ # Define find_position locally or ensure self._find_position_robust is used correctly
+ # Using self._find_position_robust is preferred as it's a method of the class
+ find_position_func = self._find_position_robust
+
+ # Store the parsed structure for tree building later
+ self.processed_nav_structure = []
+
+ # Call the correct parsing function based on confirmed nav_type
+ parse_successful = False
+ if nav_type == "ncx":
+ nav_map = nav_soup.find("navMap")
+ if nav_map:
+ logging.info("Parsing NCX <navMap>...")
+ for nav_point in nav_map.find_all("navPoint", recursive=False):
+ self._parse_ncx_navpoint(
+ nav_point,
+ ordered_nav_entries,
+ doc_order,
+ doc_order_decoded,
+ self.processed_nav_structure,
+ find_position_func,
+ )
+ parse_successful = bool(
+ ordered_nav_entries
+ ) # Success if entries were added
+ else:
+ logging.warning("Could not find <navMap> in NCX file.")
+ elif nav_type == "html":
+ logging.info("Parsing NAV HTML...")
+ toc_nav = nav_soup.find("nav", attrs={"epub:type": "toc"})
+ if not toc_nav:
+ # Fallback: look for any <nav> element containing an <ol>
+ all_navs = nav_soup.find_all("nav")
+ for nav in all_navs:
+ if nav.find("ol"):
+ toc_nav = nav
+ logging.info("Found fallback TOC structure in <nav> with <ol>.")
+ break
+ if toc_nav:
+ top_ol = toc_nav.find("ol", recursive=False)
+ if top_ol:
+ for li in top_ol.find_all("li", recursive=False):
+ self._parse_html_nav_li(
+ li,
+ ordered_nav_entries,
+ doc_order,
+ doc_order_decoded,
+ self.processed_nav_structure,
+ find_position_func,
+ )
+ parse_successful = bool(
+ ordered_nav_entries
+ ) # Success if entries were added
+ else:
+ logging.warning("Found <nav> for TOC but no top-level <ol> inside.")
+ else:
+ logging.warning(
+ "Could not find TOC structure (<nav epub:type='toc'> or <nav><ol>) in NAV HTML."
+ )
+
+ # Handle case where parsing ran but found no valid entries OR parsing failed
+ if not parse_successful:
+ logging.warning(
+ "Navigation parsing completed but found no valid entries, or parsing failed. Falling back."
+ )
+ raise ValueError("No valid navigation entries found after parsing")
+
+ # Sort entries globally by document order and position within the document
+ ordered_nav_entries.sort(key=lambda x: (x["doc_order"], x["position"]))
+ logging.info(f"Sorted {len(ordered_nav_entries)} navigation entries.")
+
+ # 3. Slice content ONLY between sorted TOC entries
+ num_entries = len(ordered_nav_entries)
+ for i in range(num_entries):
+ current_entry = ordered_nav_entries[i]
+ current_src = current_entry["src"]
+ current_doc = current_entry["doc_href"]
+ current_pos = current_entry["position"]
+ current_doc_html = self.doc_content.get(current_doc, "")
+
+ start_slice_pos = current_pos
+ slice_html = ""
+
+ next_entry = ordered_nav_entries[i + 1] if (i + 1) < num_entries else None
+
+ if next_entry:
+ next_doc = next_entry["doc_href"]
+ next_pos = next_entry["position"]
+
+ # Always include all content from current position to next position, even if next_doc is before current_doc
+ if current_doc == next_doc:
+ slice_html = current_doc_html[start_slice_pos:next_pos]
+ else:
+ # Collect all content from current_doc (from start_slice_pos to end),
+ # then all intermediate docs (in spine order),
+ # then up to next_pos in next_doc (even if next_doc is before current_doc in spine)
+ slice_html = current_doc_html[start_slice_pos:]
+ docs_between = []
+ try:
+ idx_current = spine_docs.index(current_doc)
+ idx_next = spine_docs.index(next_doc)
+ if idx_current < idx_next:
+ for doc_idx in range(idx_current + 1, idx_next):
+ docs_between.append(spine_docs[doc_idx])
+ elif idx_current > idx_next:
+ for doc_idx in range(idx_current + 1, len(spine_docs)):
+ docs_between.append(spine_docs[doc_idx])
+ for doc_idx in range(0, idx_next):
+ docs_between.append(spine_docs[doc_idx])
+ except Exception:
+ pass
+ for doc_href in docs_between:
+ slice_html += self.doc_content.get(doc_href, "")
+ next_doc_html = self.doc_content.get(next_doc, "")
+ slice_html += next_doc_html[:next_pos]
+ else:
+ # Last TOC entry: include all content from current position to end of book
+ slice_html = current_doc_html[start_slice_pos:]
+ try:
+ idx_current = spine_docs.index(current_doc)
+ for doc_idx in range(idx_current + 1, len(spine_docs)):
+ intermediate_doc_href = spine_docs[doc_idx]
+ slice_html += self.doc_content.get(intermediate_doc_href, "")
+ except Exception:
+ pass
+
+ if slice_html.strip():
+ slice_soup = BeautifulSoup(slice_html, "html.parser")
+ # Add double newlines after <p> and <div> tags
+ for tag in slice_soup.find_all(["p", "div"]):
+ tag.append("\n\n")
+ for tag in slice_soup.find_all(["sup", "sub"]):
+ tag.decompose()
+ text = clean_text(slice_soup.get_text()).strip()
+ if text:
+ self.content_texts[current_src] = text
+ self.content_lengths[current_src] = len(text)
+ else:
+ self.content_texts[current_src] = ""
+ self.content_lengths[current_src] = 0
+ else:
+ self.content_texts[current_src] = ""
+ self.content_lengths[current_src] = 0
+
+ # 4. Extract text and store using the original TOC entry src as the key
+ if ordered_nav_entries:
+ first_entry = ordered_nav_entries[0]
+ first_doc_href = first_entry["doc_href"]
+ first_pos = first_entry["position"]
+ first_doc_order = first_entry["doc_order"]
+ prefix_html = ""
+
+ for doc_idx in range(first_doc_order):
+ if doc_idx < len(spine_docs):
+ intermediate_doc_href = spine_docs[doc_idx]
+ prefix_html += self.doc_content.get(intermediate_doc_href, "")
+ else:
+ logging.warning(
+ f"Document index {doc_idx} out of bounds for spine (length {len(spine_docs)})."
+ )
+
+ first_doc_html = self.doc_content.get(first_doc_href, "")
+ prefix_html += first_doc_html[:first_pos]
+
+ if prefix_html.strip():
+ prefix_soup = BeautifulSoup(prefix_html, "html.parser")
+ for tag in prefix_soup.find_all(["sup", "sub"]):
+ tag.decompose()
+ prefix_text = clean_text(prefix_soup.get_text()).strip()
+
+ if prefix_text:
+ prefix_chapter_src = "internal:prefix_content"
+ self.content_texts[prefix_chapter_src] = prefix_text
+ self.content_lengths[prefix_chapter_src] = len(prefix_text)
+ self.processed_nav_structure.insert(
+ 0,
+ {
+ "src": prefix_chapter_src,
+ "title": "Introduction",
+ "children": [],
+ },
+ )
+ logging.info(
+ f"Added prefix content chapter '{prefix_chapter_src}'."
+ )
+
+ logging.info(
+ f"Finished processing EPUB navigation. Found {len(self.content_texts)} content sections linked to TOC."
+ )
+
+ def _parse_ncx_navpoint(
+ self,
+ nav_point,
+ ordered_entries,
+ doc_order,
+ doc_order_decoded,
+ tree_structure_list,
+ find_position_func,
+ ):
+ nav_label = nav_point.find("navLabel")
+ content = nav_point.find("content")
+ title = (
+ nav_label.find("text").get_text(strip=True)
+ if nav_label and nav_label.find("text")
+ else "Untitled Section"
+ )
+ src = content["src"] if content and "src" in content.attrs else None
+
+ current_entry_node = {"title": title, "src": src, "children": []}
+
+ if src:
+ base_href, fragment = src.split("#", 1) if "#" in src else (src, None)
+ # Try both original and decoded hrefs
+ doc_key = None
+ if base_href in doc_order:
+ doc_key = base_href
+ doc_idx = doc_order[base_href]
+ elif urllib.parse.unquote(base_href) in doc_order:
+ doc_key = urllib.parse.unquote(base_href)
+ doc_idx = doc_order[doc_key]
+ elif base_href in doc_order_decoded:
+ doc_key = base_href
+ doc_idx = doc_order_decoded[base_href]
+ elif urllib.parse.unquote(base_href) in doc_order_decoded:
+ doc_key = urllib.parse.unquote(base_href)
+ doc_idx = doc_order_decoded[doc_key]
+ else:
+ logging.warning(
+ f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list."
+ )
+ current_entry_node["has_content"] = False
+ doc_key = None
+ if doc_key is not None:
+ position = find_position_func(doc_key, fragment)
+ entry_data = {
+ "src": src,
+ "title": title,
+ "doc_href": doc_key,
+ "position": position,
+ "doc_order": doc_idx,
+ }
+ ordered_entries.append(entry_data)
+ current_entry_node["has_content"] = True
+ else:
+ logging.warning(f"Navigation entry '{title}' has no 'src' attribute.")
+ current_entry_node["has_content"] = False
+
+ child_navpoints = nav_point.find_all("navPoint", recursive=False)
+ if child_navpoints:
+ for child_np in child_navpoints:
+ # Pass find_position_func down recursively
+ self._parse_ncx_navpoint(
+ child_np,
+ ordered_entries,
+ doc_order,
+ doc_order_decoded,
+ current_entry_node["children"],
+ find_position_func,
+ )
+
+ if title and (
+ current_entry_node.get("has_content", False)
+ or current_entry_node["children"]
+ ):
+ tree_structure_list.append(current_entry_node)
+
+ def _parse_html_nav_li(
+ self,
+ li_element,
+ ordered_entries,
+ doc_order,
+ doc_order_decoded,
+ tree_structure_list,
+ find_position_func,
+ ):
+ link = li_element.find("a", recursive=False)
+ span_text = li_element.find("span", recursive=False)
+ title = "Untitled Section"
+ src = None
+ current_entry_node = {"children": []}
+
+ if link and "href" in link.attrs:
+ src = link["href"]
+ title = link.get_text(strip=True) or title
+ if not title.strip() and span_text:
+ title = span_text.get_text(strip=True) or title
+ if not title.strip():
+ li_text = "".join(
+ t for t in li_element.contents if isinstance(t, NavigableString)
+ ).strip()
+ title = li_text or title
+ elif span_text:
+ title = span_text.get_text(strip=True) or title
+ if not title.strip():
+ li_text = "".join(
+ t for t in li_element.contents if isinstance(t, NavigableString)
+ ).strip()
+ title = li_text or title
+ else:
+ li_text = "".join(
+ t for t in li_element.contents if isinstance(t, NavigableString)
+ ).strip()
+ title = li_text or title
+
+ current_entry_node["title"] = title
+ current_entry_node["src"] = src
+
+ if src:
+ base_href, fragment = src.split("#", 1) if "#" in src else (src, None)
+ # Try both original and decoded hrefs
+ doc_key = None
+ if base_href in doc_order:
+ doc_key = base_href
+ doc_idx = doc_order[base_href]
+ elif urllib.parse.unquote(base_href) in doc_order:
+ doc_key = urllib.parse.unquote(base_href)
+ doc_idx = doc_order[doc_key]
+ elif base_href in doc_order_decoded:
+ doc_key = base_href
+ doc_idx = doc_order_decoded[base_href]
+ elif urllib.parse.unquote(base_href) in doc_order_decoded:
+ doc_key = urllib.parse.unquote(base_href)
+ doc_idx = doc_order_decoded[doc_key]
+ else:
+ logging.warning(
+ f"Navigation entry '{title}' points to '{base_href}', which is not in the spine or document list."
+ )
+ current_entry_node["has_content"] = False
+ doc_key = None
+ if doc_key is not None:
+ position = find_position_func(doc_key, fragment)
+ entry_data = {
+ "src": src,
+ "title": title,
+ "doc_href": doc_key,
+ "position": position,
+ "doc_order": doc_idx,
+ }
+ ordered_entries.append(entry_data)
+ current_entry_node["has_content"] = True
+ else:
+ current_entry_node["has_content"] = False
+
+ child_ol = li_element.find("ol", recursive=False)
+ if child_ol:
+ for child_li in child_ol.find_all("li", recursive=False):
+ # Pass find_position_func down recursively
+ self._parse_html_nav_li(
+ child_li,
+ ordered_entries,
+ doc_order,
+ doc_order_decoded,
+ current_entry_node["children"],
+ find_position_func,
+ )
+
+ if title and (
+ current_entry_node.get("has_content", False)
+ or current_entry_node["children"]
+ ):
+ tree_structure_list.append(current_entry_node)
+
+ def _find_position_robust(self, doc_href, fragment_id):
+ if doc_href not in self.doc_content:
+ logging.warning(f"Document '{doc_href}' not found in cached content.")
+ return 0
+ html_content = self.doc_content[doc_href]
+ if not fragment_id:
+ return 0
+
+ try:
+ temp_soup = BeautifulSoup(f"<div>{html_content}</div>", "html.parser")
+ target_element = temp_soup.find(id=fragment_id)
+ if target_element:
+ tag_str = str(target_element)
+ pos = html_content.find(tag_str[: min(len(tag_str), 200)])
+ if pos != -1:
+ logging.debug(
+ f"Found position for id='{fragment_id}' in {doc_href} using BeautifulSoup: {pos}"
+ )
+ return pos
+ except Exception as e:
+ logging.warning(
+ f"BeautifulSoup failed to find id='{fragment_id}' in {doc_href}: {e}"
+ )
+
+ safe_fragment_id = re.escape(fragment_id)
+ id_name_pattern = re.compile(
+ f"<[^>]+(?:id|name)\\s*=\\s*[\"']{safe_fragment_id}[\"']", re.IGNORECASE
+ )
+ match = id_name_pattern.search(html_content)
+ if match:
+ pos = match.start()
+ logging.debug(
+ f"Found position for id/name='{fragment_id}' in {doc_href} using regex: {pos}"
+ )
+ return pos
+
+ id_match_str = f'id="{fragment_id}"'
+ name_match_str = f'name="{fragment_id}"'
+ id_pos = html_content.find(id_match_str)
+ name_pos = html_content.find(name_match_str)
+
+ pos = -1
+ if id_pos != -1 and name_pos != -1:
+ pos = min(id_pos, name_pos)
+ elif id_pos != -1:
+ pos = id_pos
+ elif name_pos != -1:
+ pos = name_pos
+
+ if pos != -1:
+ tag_start_pos = html_content.rfind("<", 0, pos)
+ final_pos = tag_start_pos if tag_start_pos != -1 else 0
+ logging.debug(
+ f"Found position for id/name='{fragment_id}' in {doc_href} using string search: {final_pos}"
+ )
+ return final_pos
+
+ logging.warning(
+ f"Anchor '{fragment_id}' not found in {doc_href}. Defaulting to position 0."
+ )
+ return 0
+
+ def _build_tree(self):
self.treeWidget.clear()
+
info_item = QTreeWidgetItem(self.treeWidget, ["Information"])
info_item.setData(0, Qt.UserRole, "info:bookinfo")
info_item.setFlags(info_item.flags() & ~Qt.ItemIsUserCheckable)
@@ -412,76 +825,129 @@ class HandlerDialog(QDialog):
font.setBold(True)
info_item.setFont(0, font)
- # Regular tree building
- def build_tree(toc_entries, parent_item):
- for entry in toc_entries:
- href, title, children = None, "Unknown", []
- if isinstance(entry, ebooklib.epub.Link):
- href, title = entry.href, entry.title or entry.href
- elif isinstance(entry, tuple) and len(entry) >= 1:
- section_or_link = entry[0]
- if isinstance(section_or_link, ebooklib.epub.Section):
- title = section_or_link.title
- href = getattr(section_or_link, "href", None)
- elif isinstance(section_or_link, ebooklib.epub.Link):
- href, title = (
- section_or_link.href,
- section_or_link.title or section_or_link.href,
- )
- if len(entry) > 1 and isinstance(entry[1], list):
- children = entry[1]
- else:
- continue
-
- # Create tree item
- item = QTreeWidgetItem(parent_item, [title])
- item.setData(0, Qt.UserRole, href)
-
- # Make item checkable if it has content
- has_content = (
- href
- and href in self.content_texts
- and self.content_texts[href].strip()
+ if self.file_type == "epub":
+ if (
+ hasattr(self, "processed_nav_structure")
+ and self.processed_nav_structure
+ ):
+ self._build_epub_tree_from_nav(
+ self.processed_nav_structure, self.treeWidget
)
- if has_content or children:
- item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
- is_checked = href and href in self.checked_chapters
- item.setCheckState(0, Qt.Checked if is_checked else Qt.Unchecked)
+ else:
+ logging.warning("Building EPUB tree using fallback book.toc.")
+ self._build_epub_tree_fallback(self.book.toc, self.treeWidget)
+ else:
+ self._build_pdf_tree()
+
+ has_parents = False
+ iterator = QTreeWidgetItemIterator(
+ self.treeWidget, QTreeWidgetItemIterator.HasChildren
+ )
+ if iterator.value():
+ has_parents = True
+ self.treeWidget.setRootIsDecorated(has_parents)
+
+ def _build_epub_tree_from_nav(
+ self, nav_nodes, parent_item, seen_content_hashes=None
+ ):
+ if seen_content_hashes is None:
+ seen_content_hashes = set()
+ for node in nav_nodes:
+ title = node.get("title", "Unknown")
+ src = node.get("src")
+ children = node.get("children", [])
+
+ item = QTreeWidgetItem(parent_item, [title])
+ item.setData(0, Qt.UserRole, src)
+
+ is_empty = (
+ src
+ and (src in self.content_texts)
+ and (not self.content_texts[src].strip())
+ )
+ is_duplicate = False
+ if src and src in self.content_texts and self.content_texts[src].strip():
+ content_hash = hash(self.content_texts[src])
+ if content_hash in seen_content_hashes:
+ is_duplicate = True
else:
- item.setFlags(item.flags() & ~Qt.ItemIsUserCheckable)
+ seen_content_hashes.add(content_hash)
- # Process children
- if children:
- build_tree(children, item)
+ if src and not is_empty and not is_duplicate:
+ item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
+ is_checked = src in self.checked_chapters
+ item.setCheckState(0, Qt.Checked if is_checked else Qt.Unchecked)
+ elif is_duplicate:
+ # Mark as duplicate and remove checkbox
+ item.setText(0, f"{title} (Duplicate)")
+ item.setFlags(item.flags() & ~Qt.ItemIsUserCheckable)
+ elif children:
+ item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
+ item.setCheckState(0, Qt.Unchecked)
+ else:
+ item.setFlags(item.flags() & ~Qt.ItemIsUserCheckable)
- build_tree(self.book.toc, self.treeWidget)
+ if children:
+ self._build_epub_tree_from_nav(children, item, seen_content_hashes)
+
+ def _build_epub_tree_fallback(self, toc_entries, parent_item):
+ for entry in toc_entries:
+ href, title, children = None, "Unknown", []
+ entry_obj = None
+ if isinstance(entry, ebooklib.epub.Link):
+ href, title = entry.href, entry.title or entry.href
+ entry_obj = entry
+ elif isinstance(entry, tuple) and len(entry) >= 1:
+ section_or_link = entry[0]
+ entry_obj = section_or_link
+ if isinstance(section_or_link, ebooklib.epub.Section):
+ title = section_or_link.title
+ href = getattr(section_or_link, "href", None)
+ elif isinstance(section_or_link, ebooklib.epub.Link):
+ href, title = (
+ section_or_link.href,
+ section_or_link.title or section_or_link.href,
+ )
+
+ if len(entry) > 1 and isinstance(entry[1], list):
+ children = entry[1]
+ else:
+ continue
+
+ item = QTreeWidgetItem(parent_item, [title])
+ item.setData(0, Qt.UserRole, href)
+
+ has_content = (
+ href and href in self.content_texts and self.content_texts[href].strip()
+ )
+
+ if has_content or children:
+ item.setFlags(item.flags() | Qt.ItemIsUserCheckable)
+ is_checked = href and href in self.checked_chapters
+ item.setCheckState(0, Qt.Checked if is_checked else Qt.Unchecked)
+ else:
+ item.setFlags(item.flags() & ~Qt.ItemIsUserCheckable)
+
+ if children:
+ self._build_epub_tree_fallback(children, item)
def _build_pdf_tree(self):
- """Build the tree for PDF files combining outline/bookmarks with pages"""
- # Get outline and store if this PDF has bookmarks
outline = self.pdf_doc.get_toc()
self.has_pdf_bookmarks = bool(outline)
if not outline:
- # No bookmarks/outline available, create a simple page list
self._build_pdf_pages_tree()
return
- # Process the outline to determine page ranges
bookmark_pages = []
page_to_bookmark = {}
next_page_boundaries = {}
- # Track added pages to prevent duplicates
added_pages = set()
- # Extract page numbers from outline recursively
def extract_page_numbers(entries):
for entry in entries:
- if (
- len(entry) >= 3
- ): # Valid outline entry has at least level, title, page
+ if len(entry) >= 3:
_, title, page = entry[:3]
- # Convert page reference to actual page number (0-based)
page_num = (
page - 1
if isinstance(page, int)
@@ -489,27 +955,21 @@ class HandlerDialog(QDialog):
)
bookmark_pages.append((page_num, title))
- # Process children recursively
if len(entry) > 3 and isinstance(entry[3], list):
extract_page_numbers(entry[3])
extract_page_numbers(outline)
bookmark_pages.sort()
- # Determine page ranges for each bookmark
for i, (page_num, title) in enumerate(bookmark_pages):
if i < len(bookmark_pages) - 1:
next_page_boundaries[page_num] = bookmark_pages[i + 1][0]
page_to_bookmark[page_num] = title
- # Helper function to build the tree structure recursively
def build_outline_tree(entries, parent_item):
for entry in entries:
- if (
- len(entry) >= 3
- ): # Valid outline entry has at least level, title, page
+ if len(entry) >= 3:
entry_level, title, page = entry[:3]
- # Get actual page number (0-based)
page_num = (
page - 1
if isinstance(page, int)
@@ -517,7 +977,6 @@ class HandlerDialog(QDialog):
)
page_id = f"page_{page_num+1}"
- # Create bookmark item
bookmark_item = QTreeWidgetItem(parent_item, [title])
bookmark_item.setData(0, Qt.UserRole, page_id)
bookmark_item.setFlags(
@@ -532,15 +991,10 @@ class HandlerDialog(QDialog):
),
)
- # Mark this page as added
added_pages.add(page_num)
- # Add child pages that belong to this bookmark
next_page = next_page_boundaries.get(page_num, len(self.pdf_doc))
- for sub_page_num in range(
- page_num + 1, next_page
- ): # Skip the bookmark page itself
- # Skip if this page is a bookmark itself or already added as a child elsewhere
+ for sub_page_num in range(page_num + 1, next_page):
if (
sub_page_num in page_to_bookmark
or sub_page_num in added_pages
@@ -550,7 +1004,6 @@ class HandlerDialog(QDialog):
page_id = f"page_{sub_page_num+1}"
page_title = f"Page {sub_page_num+1}"
- # Try to get a better title from the first line of content
page_text = self.content_texts.get(page_id, "").strip()
if page_text:
first_line = page_text.split("\n", 1)[0].strip()
@@ -569,22 +1022,15 @@ class HandlerDialog(QDialog):
),
)
- # Mark this page as added
added_pages.add(sub_page_num)
- # Process child bookmarks if any
if len(entry) > 3 and isinstance(entry[3], list):
build_outline_tree(entry[3], bookmark_item)
- # Start building the tree from the outline
build_outline_tree(outline, self.treeWidget)
- # Add pages not covered by bookmarks
- covered_pages = set(
- added_pages
- ) # Use our tracked pages to find uncategorized ones
+ covered_pages = set(added_pages)
- # Add remaining pages as top-level items under "Other Pages"
uncategorized_pages = [
i for i in range(len(self.pdf_doc)) if i not in covered_pages
]
@@ -592,7 +1038,6 @@ class HandlerDialog(QDialog):
self._add_other_pages(uncategorized_pages)
def _build_pdf_pages_tree(self):
- """Build a simple page list for PDFs without bookmarks"""
pages_item = QTreeWidgetItem(self.treeWidget, ["Pages"])
pages_item.setFlags(pages_item.flags() & ~Qt.ItemIsUserCheckable)
font = pages_item.font(0)
@@ -603,7 +1048,6 @@ class HandlerDialog(QDialog):
page_id = f"page_{page_num+1}"
page_title = f"Page {page_num+1}"
- # Try to get a better title from the first line of content
page_text = self.content_texts.get(page_id, "").strip()
if page_text:
first_line = page_text.split("\n", 1)[0].strip()
@@ -618,7 +1062,6 @@ class HandlerDialog(QDialog):
)
def _add_other_pages(self, uncategorized_pages):
- """Add uncategorized pages to the tree"""
other_pages = QTreeWidgetItem(self.treeWidget, ["Other Pages"])
other_pages.setFlags(other_pages.flags() & ~Qt.ItemIsUserCheckable)
font = other_pages.font(0)
@@ -629,7 +1072,6 @@ class HandlerDialog(QDialog):
page_id = f"page_{page_num+1}"
page_title = f"Page {page_num+1}"
- # Try to get better title from first line
page_text = self.content_texts.get(page_id, "").strip()
if page_text:
first_line = page_text.split("\n", 1)[0].strip()
@@ -644,11 +1086,9 @@ class HandlerDialog(QDialog):
)
def _are_provided_checks_relevant(self):
- """Check if provided checks are relevant to this book"""
if not self.checked_chapters:
return False
- # Collect all identifiers present in tree
all_identifiers = set()
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -659,18 +1099,14 @@ class HandlerDialog(QDialog):
all_identifiers.add(identifier)
iterator += 1
- # Check for any intersection with provided chapters
return bool(self.checked_chapters.intersection(all_identifiers))
def _setup_ui(self):
- """Set up the user interface"""
- # Add preview panel
self.previewEdit = QTextEdit(self)
self.previewEdit.setReadOnly(True)
self.previewEdit.setMinimumWidth(300)
self.previewEdit.setStyleSheet("QTextEdit { border: none; }")
- # Create informative text label below preview
self.previewInfoLabel = QLabel(
'*Note: You can modify the content later using the "Edit" button in the input box or by accessing the temporary files directory through settings.',
self,
@@ -680,7 +1116,6 @@ class HandlerDialog(QDialog):
"QLabel { color: #666; font-style: italic; }"
)
- # Right panel layout (preview and info label)
previewLayout = QVBoxLayout()
previewLayout.setContentsMargins(0, 0, 0, 0)
previewLayout.addWidget(self.previewEdit, 1)
@@ -689,30 +1124,24 @@ class HandlerDialog(QDialog):
rightWidget = QWidget()
rightWidget.setLayout(previewLayout)
- # Dialog buttons
buttons = QDialogButtonBox(QDialogButtonBox.Ok | QDialogButtonBox.Cancel, self)
buttons.accepted.connect(self.accept)
buttons.rejected.connect(self.reject)
- # Selection buttons
item_type = "chapters" if self.file_type == "epub" else "pages"
- # Auto-select button
self.auto_select_btn = QPushButton(f"Auto-select {item_type}", self)
self.auto_select_btn.clicked.connect(self.auto_select_chapters)
self.auto_select_btn.setToolTip(f"Automatically select main {item_type}")
- # Selection buttons layout
buttons_layout = QVBoxLayout()
buttons_layout.setContentsMargins(0, 0, 0, 0)
buttons_layout.setSpacing(10)
- # Row 1: Auto-select
auto_select_layout = QHBoxLayout()
auto_select_layout.addWidget(self.auto_select_btn)
buttons_layout.addLayout(auto_select_layout)
- # Row 2: Select/Deselect All
select_layout = QHBoxLayout()
self.select_all_btn = QPushButton("Select all", self)
self.select_all_btn.clicked.connect(self.select_all_chapters)
@@ -722,7 +1151,6 @@ class HandlerDialog(QDialog):
select_layout.addWidget(self.deselect_all_btn)
buttons_layout.addLayout(select_layout)
- # Row 3: Parent selection
parent_layout = QHBoxLayout()
self.select_parents_btn = QPushButton("Select parents", self)
self.select_parents_btn.clicked.connect(self.select_parent_chapters)
@@ -732,7 +1160,6 @@ class HandlerDialog(QDialog):
parent_layout.addWidget(self.deselect_parents_btn)
buttons_layout.addLayout(parent_layout)
- # Row 4: Expand/Collapse
expand_layout = QHBoxLayout()
self.expand_all_btn = QPushButton("Expand All", self)
self.expand_all_btn.clicked.connect(self.treeWidget.expandAll)
@@ -742,13 +1169,11 @@ class HandlerDialog(QDialog):
expand_layout.addWidget(self.collapse_all_btn)
buttons_layout.addLayout(expand_layout)
- # Left panel layout
leftLayout = QVBoxLayout()
leftLayout.setContentsMargins(0, 0, 5, 0)
leftLayout.addLayout(buttons_layout)
leftLayout.addWidget(self.treeWidget)
- # Save options checkboxes
checkbox_text = (
"Save each chapter separately"
if self.file_type == "epub"
@@ -770,33 +1195,25 @@ class HandlerDialog(QDialog):
leftLayout.addWidget(buttons)
- # Create left panel widget
leftWidget = QWidget()
leftWidget.setLayout(leftLayout)
- # Create splitter for left panel and preview
self.splitter = QSplitter(Qt.Horizontal)
self.splitter.addWidget(leftWidget)
- self.splitter.addWidget(
- rightWidget
- ) # Now using rightWidget that includes preview and label
+ self.splitter.addWidget(rightWidget)
self.splitter.setSizes([280, 420])
- # Set main layout
mainLayout = QVBoxLayout(self)
mainLayout.addWidget(self.splitter)
self.setLayout(mainLayout)
def _update_checkbox_states(self):
- """Update checkboxes enabled states based on document type and selection"""
- # Make sure checkboxes exist before trying to modify them
if (
not hasattr(self, "save_chapters_checkbox")
or not self.save_chapters_checkbox
):
return
- # For PDFs without bookmarks, always disable separate chapters option
if (
self.file_type == "pdf"
and hasattr(self, "has_pdf_bookmarks")
@@ -806,11 +1223,9 @@ class HandlerDialog(QDialog):
self.merge_chapters_checkbox.setEnabled(False)
return
- # Count checked items differently based on file type
checked_count = 0
if self.file_type == "epub":
- # For EPUB: Count all checked items
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
@@ -823,9 +1238,7 @@ class HandlerDialog(QDialog):
break
iterator += 1
- else: # PDF
- # For PDF: Count distinct parent groups
- # We need content from at least 2 different parents to enable "save separately"
+ else:
parent_groups = set()
iterator = QTreeWidgetItemIterator(self.treeWidget)
@@ -835,30 +1248,24 @@ class HandlerDialog(QDialog):
item.flags() & Qt.ItemIsUserCheckable
and item.checkState(0) == Qt.Checked
):
- # Get the parent (or the item itself if it's a top-level item)
parent = item.parent()
if parent and parent != self.treeWidget.invisibleRootItem():
- # Use memory address as a unique identifier since QTreeWidgetItem is not hashable
parent_groups.add(id(parent))
else:
- # Top-level items count as their own parent group
parent_groups.add(id(item))
iterator += 1
checked_count = len(parent_groups)
- # Enable save separately only if enough distinct groups are checked
min_groups_required = 2
self.save_chapters_checkbox.setEnabled(checked_count >= min_groups_required)
- # Enable merge only if save separately is enabled and checked
self.merge_chapters_checkbox.setEnabled(
self.save_chapters_checkbox.isEnabled()
and self.save_chapters_checkbox.isChecked()
)
def select_all_chapters(self):
- """Select all chapters/pages"""
self._block_signals = True
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -870,7 +1277,6 @@ class HandlerDialog(QDialog):
self._update_checked_set_from_tree()
def deselect_all_chapters(self):
- """Deselect all chapters/pages"""
self._block_signals = True
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -882,7 +1288,6 @@ class HandlerDialog(QDialog):
self._update_checked_set_from_tree()
def select_parent_chapters(self):
- """Select only parent chapters/sections"""
self._block_signals = True
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -894,7 +1299,6 @@ class HandlerDialog(QDialog):
self._update_checked_set_from_tree()
def deselect_parent_chapters(self):
- """Deselect only parent chapters/sections"""
self._block_signals = True
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -906,23 +1310,20 @@ class HandlerDialog(QDialog):
self._update_checked_set_from_tree()
def auto_select_chapters(self):
- """Auto-select chapters/pages"""
self._run_auto_check()
def _run_auto_check(self):
- """Run automatic content selection based on file type"""
self._block_signals = True
if self.file_type == "epub":
self._run_epub_auto_check()
- else: # PDF
+ else:
self._run_pdf_auto_check()
self._block_signals = False
self._update_checked_set_from_tree()
def _run_epub_auto_check(self):
- """Auto-check logic for EPUB files"""
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
@@ -930,25 +1331,30 @@ class HandlerDialog(QDialog):
iterator += 1
continue
- href = item.data(0, Qt.UserRole)
- lookup_href = href.split("#")[0] if href else None
+ src = item.data(0, Qt.UserRole)
- # Check based on length (> 1000 chars) and parent status
- if (
- lookup_href and self.content_lengths.get(lookup_href, 0) > 1000
- ) or item.childCount() > 0:
+ has_significant_content = src and self.content_lengths.get(src, 0) > 1000
+ is_parent = item.childCount() > 0
+
+ if has_significant_content or is_parent:
item.setCheckState(0, Qt.Checked)
- # Check children of parents
- if item.childCount() > 0:
+ if is_parent:
for i in range(item.childCount()):
child = item.child(i)
if child.flags() & Qt.ItemIsUserCheckable:
- child.setCheckState(0, Qt.Checked)
+ child_src = child.data(0, Qt.UserRole)
+ child_has_content = (
+ child_src and self.content_lengths.get(child_src, 0) > 0
+ )
+ child_is_parent = child.childCount() > 0
+ if child_has_content or child_is_parent:
+ child.setCheckState(0, Qt.Checked)
+ else:
+ item.setCheckState(0, Qt.Unchecked)
+
iterator += 1
def _run_pdf_auto_check(self):
- """Auto-check logic for PDF files"""
- # If there are no bookmarks, just check all pages
if hasattr(self, "has_pdf_bookmarks") and not self.has_pdf_bookmarks:
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -958,7 +1364,6 @@ class HandlerDialog(QDialog):
iterator += 1
return
- # For PDFs with bookmarks, select all bookmark items and non-empty pages
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
@@ -968,7 +1373,6 @@ class HandlerDialog(QDialog):
identifier = item.data(0, Qt.UserRole)
- # Always select bookmark items or non-empty pages
if not identifier:
iterator += 1
continue
@@ -982,7 +1386,6 @@ class HandlerDialog(QDialog):
iterator += 1
def _update_checked_set_from_tree(self):
- """Update the internal set of checked items"""
self.checked_chapters.clear()
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
@@ -992,18 +1395,15 @@ class HandlerDialog(QDialog):
if identifier:
self.checked_chapters.add(identifier)
iterator += 1
- # Only update checkbox states if they exist
if hasattr(self, "save_chapters_checkbox") and self.save_chapters_checkbox:
self._update_checkbox_states()
def handle_item_check(self, item):
- """Handle item check/uncheck by updating children"""
if self._block_signals:
return
self._block_signals = True
- # Update children recursively
if item.flags() & Qt.ItemIsUserCheckable:
for i in range(item.childCount()):
child = item.child(i)
@@ -1014,49 +1414,37 @@ class HandlerDialog(QDialog):
self._update_checked_set_from_tree()
def handle_item_double_click(self, item, column=0):
- """Toggle check state when a non-parent item is double-clicked on the text, not the checkbox"""
- # Only toggle items that are checkable and don't have children
if item.flags() & Qt.ItemIsUserCheckable and item.childCount() == 0:
- # Get the rectangle of the checkbox
rect = self.treeWidget.visualItemRect(item)
- checkbox_width = 20 # Approximate width of the checkbox
+ checkbox_width = 20
- # Get current mouse position
mouse_pos = self.treeWidget.mapFromGlobal(self.treeWidget.cursor().pos())
- # Only toggle if click position is not on the checkbox
if mouse_pos.x() > rect.x() + checkbox_width:
- # Toggle the check state
new_state = (
Qt.Unchecked if item.checkState(0) == Qt.Checked else Qt.Checked
)
item.setCheckState(0, new_state)
def update_preview(self, current):
- """Update the preview panel with selected item content"""
if not current:
self.previewEdit.clear()
return
identifier = current.data(0, Qt.UserRole)
- # Special case for the Information item
if identifier == "info:bookinfo":
self._display_book_info()
return
- # Get content based on file type
text = None
if self.file_type == "epub":
- # For EPUB, always use the exact href from the TOC
text = self.content_texts.get(identifier)
- else: # PDF
+ else:
text = self.content_texts.get(identifier)
- # Display content or placeholder text - never remove titles
if text is None:
title = current.text(0)
- # Add title to preview even if no content
self.previewEdit.setPlainText(
f"{title}\n\n(No content available for this item)"
)
@@ -1067,18 +1455,15 @@ class HandlerDialog(QDialog):
self.previewEdit.setPlainText(text)
def _display_book_info(self):
- """Display book metadata and cover image in the preview panel"""
self.previewEdit.clear()
html_content = "<html><body style='font-family: Arial, sans-serif;'>"
- # Add cover image if available
if self.book_metadata["cover_image"]:
try:
image_data = base64.b64encode(self.book_metadata["cover_image"]).decode(
"utf-8"
)
- # Determine image type
image_type = "jpeg"
if self.book_metadata["cover_image"].startswith(b"\x89PNG"):
image_type = "png"
@@ -1095,7 +1480,6 @@ class HandlerDialog(QDialog):
except Exception as e:
html_content += f"<p>Error displaying cover image: {str(e)}</p>"
- # Add title, authors, publisher
if self.book_metadata["title"]:
html_content += (
f"<h2 style='text-align: center;'>{self.book_metadata['title']}</h2>"
@@ -1110,12 +1494,10 @@ class HandlerDialog(QDialog):
html_content += "<hr/>"
- # Add description
if self.book_metadata["description"]:
desc = re.sub(r"<[^>]+>", "", self.book_metadata["description"])
html_content += f"<h3>Description:</h3><p>{desc}</p>"
- # Add file type and page count for PDFs
if self.file_type == "pdf":
page_count = len(self.pdf_doc) if self.pdf_doc else 0
html_content += f"<p>File type: PDF<br>Page count: {page_count}</p>"
@@ -1124,7 +1506,6 @@ class HandlerDialog(QDialog):
self.previewEdit.setHtml(html_content)
def _extract_book_metadata(self):
- """Extract book metadata"""
metadata = {
"title": None,
"authors": [],
@@ -1134,7 +1515,6 @@ class HandlerDialog(QDialog):
}
if self.file_type == "epub":
- # Extract EPUB metadata
title_items = self.book.get_metadata("DC", "title")
if title_items:
metadata["title"] = title_items[0][0]
@@ -1151,7 +1531,6 @@ class HandlerDialog(QDialog):
if publisher_items:
metadata["publisher"] = publisher_items[0][0]
- # Try to find cover image
for item in self.book.get_items_of_type(ebooklib.ITEM_COVER):
metadata["cover_image"] = item.get_content()
break
@@ -1161,8 +1540,7 @@ class HandlerDialog(QDialog):
if "cover" in item.get_name().lower():
metadata["cover_image"] = item.get_content()
break
- else: # PDF
- # Extract PDF metadata
+ else:
pdf_info = self.pdf_doc.metadata
if pdf_info:
metadata["title"] = pdf_info.get("title", None)
@@ -1182,7 +1560,6 @@ class HandlerDialog(QDialog):
metadata["publisher"] = pdf_info.get("creator", None)
- # Try to get cover image from first page
if len(self.pdf_doc) > 0:
try:
pix = self.pdf_doc[0].get_pixmap(matrix=fitz.Matrix(2, 2))
@@ -1193,54 +1570,52 @@ class HandlerDialog(QDialog):
return metadata
def get_selected_text(self):
- """Get selected text and checked identifiers based on file type"""
if self.file_type == "epub":
return self._get_epub_selected_text()
- else: # PDF
+ else:
return self._get_pdf_selected_text()
def _get_epub_selected_text(self):
- """Get selected text from EPUB content"""
- all_checked_hrefs = set()
- chapter_titles = []
+ all_checked_identifiers = set()
+ chapter_texts = []
- # Collect all checked hrefs in tree order to preserve chapter sequence
+ item_order_counter = 0
ordered_checked_items = []
+
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
+ item_order_counter += 1
if item.checkState(0) == Qt.Checked:
- href = item.data(0, Qt.UserRole)
- if href and href != "info:bookinfo":
- all_checked_hrefs.add(href)
- ordered_checked_items.append((item, href))
+ identifier = item.data(0, Qt.UserRole)
+ if identifier and identifier != "info:bookinfo":
+ all_checked_identifiers.add(identifier)
+ ordered_checked_items.append((item_order_counter, item, identifier))
iterator += 1
- # Process checked items in order
- for item, href in ordered_checked_items:
- # Always use the exact href (including fragment) from the TOC
- text = self.content_texts.get(href)
+ ordered_checked_items.sort(key=lambda x: x[0])
+
+ for order, item, identifier in ordered_checked_items:
+ text = self.content_texts.get(identifier)
if text and text.strip():
title = item.text(0)
- title = re.sub(r"^\s*-\s*", "", title).strip()
+ title = re.sub(r"^\s*[-–—]\s*", "", title).strip()
marker = f"<<CHAPTER_MARKER:{title}>>"
- chapter_titles.append((title, marker + "\n" + text))
+ chapter_texts.append(marker + "\n" + text)
- return "\n\n".join([t[1] for t in chapter_titles]), all_checked_hrefs
+ full_text = "\n\n".join(chapter_texts)
+ return full_text, all_checked_identifiers
def _get_pdf_selected_text(self):
- """Get selected text from PDF content"""
all_checked_identifiers = set()
included_text_ids = set()
section_titles = []
all_content = []
- # Check if PDF has no bookmarks
pdf_has_no_bookmarks = (
hasattr(self, "has_pdf_bookmarks") and not self.has_pdf_bookmarks
)
- # Collect all checked identifiers
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
@@ -1250,7 +1625,6 @@ class HandlerDialog(QDialog):
all_checked_identifiers.add(identifier)
iterator += 1
- # For PDFs without bookmarks, collect all content without chapter markers
if pdf_has_no_bookmarks:
sorted_page_ids = sorted(
[id for id in all_checked_identifiers if id.startswith("page_")],
@@ -1264,8 +1638,6 @@ class HandlerDialog(QDialog):
included_text_ids.add(page_id)
return "\n\n".join(all_content), all_checked_identifiers
- # For PDFs with bookmarks, process content with parent-child relationships
- # If only child pages are selected (not parent), use parent's name as chapter marker at first selected child
iterator = QTreeWidgetItemIterator(self.treeWidget)
while iterator.value():
item = iterator.value()
@@ -1273,7 +1645,6 @@ class HandlerDialog(QDialog):
parent_checked = item.checkState(0) == Qt.Checked
parent_id = item.data(0, Qt.UserRole)
parent_title = item.text(0)
- # Gather checked children
checked_children = []
for i in range(item.childCount()):
child = item.child(i)
@@ -1284,7 +1655,6 @@ class HandlerDialog(QDialog):
and child_id not in included_text_ids
):
checked_children.append((child, child_id))
- # If parent is checked, use old logic (parent marker, all content)
if parent_checked and parent_id and parent_id not in included_text_ids:
combined_text = self.content_texts.get(parent_id, "")
for child, child_id in checked_children:
@@ -1297,7 +1667,6 @@ class HandlerDialog(QDialog):
marker = f"<<CHAPTER_MARKER:{title}>>"
section_titles.append((title, marker + "\n" + combined_text))
included_text_ids.add(parent_id)
- # If only children are checked, use parent's name as marker at first child
elif not parent_checked and checked_children:
title = re.sub(r"^\s*-\s*", "", parent_title).strip()
marker = f"<<CHAPTER_MARKER:{title}>>"
@@ -1328,18 +1697,15 @@ class HandlerDialog(QDialog):
return "\n\n".join([t[1] for t in section_titles]), all_checked_identifiers
def on_save_chapters_changed(self, state):
- """Update the save_chapters_separately flag"""
self.save_chapters_separately = bool(state)
self.merge_chapters_checkbox.setEnabled(self.save_chapters_separately)
HandlerDialog._save_chapters_separately = self.save_chapters_separately
def on_merge_chapters_changed(self, state):
- """Update the merge_chapters_at_end flag"""
self.merge_chapters_at_end = bool(state)
HandlerDialog._merge_chapters_at_end = self.merge_chapters_at_end
def get_save_chapters_separately(self):
- """Return whether to save chapters separately"""
return (
self.save_chapters_separately
if self.save_chapters_checkbox.isEnabled()
@@ -1347,11 +1713,9 @@ class HandlerDialog(QDialog):
)
def get_merge_chapters_at_end(self):
- """Return whether to merge chapters at the end"""
return self.merge_chapters_at_end
def on_tree_context_menu(self, pos):
- """Handle context menu on tree items"""
item = self.treeWidget.itemAt(pos)
if (
not item
@@ -1376,7 +1740,6 @@ class HandlerDialog(QDialog):
menu.exec_(self.treeWidget.mapToGlobal(pos))
def closeEvent(self, event):
- """Clean up resources when the dialog is closed"""
if self.pdf_doc is not None:
self.pdf_doc.close()
event.accept()
diff --git a/abogen/gui.py b/abogen/gui.py
index a0c7ff6..3f7ceec 100644
--- a/abogen/gui.py
+++ b/abogen/gui.py
@@ -1445,10 +1445,10 @@ class abogen(QWidget):
# stop loading animation and restore icon on error
if error:
self.loading_movie.stop()
- self.btn_preview.setIcon(self.play_icon)
self._show_error_message_box(
"Loading Error", f"Error loading numpy or KPipeline: {error}"
)
+ self.btn_preview.setIcon(self.play_icon)
self.btn_preview.setEnabled(True)
self.btn_preview.setToolTip("Preview selected voice")
self.voice_combo.setEnabled(True)
@@ -1491,12 +1491,13 @@ class abogen(QWidget):
temp_wav = self.preview_thread.temp_wav
if not temp_wav:
self.loading_movie.stop()
- self.btn_preview.setIcon(self.play_icon)
+
self._show_error_message_box(
"Preview Error", "Preview error: No audio generated."
)
- self.btn_preview.setEnabled(True)
+ self.btn_preview.setIcon(self.play_icon)
self.btn_preview.setToolTip("Preview selected voice")
+ self.btn_preview.setEnabled(True)
self.voice_combo.setEnabled(True)
self.btn_voice_formula_mixer.setEnabled(True) # Re-enable mixer button
self.btn_start.setEnabled(True)
diff --git a/demo/abogen.png b/demo/abogen.png
index 16adbaf..b11c5dd 100644
Binary files a/demo/abogen.png and b/demo/abogen.png differ
diff --git a/pyproject.toml b/pyproject.toml
index 7f626a6..3f5ff16 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -15,7 +15,7 @@ keywords = ["audiobook", "epub", "pdf", "text-to-speech", "subtitle", "tts", "ko
dependencies = [
"PyQt5>=5.15.11",
"kokoro>=0.9.4",
- "ebooklib>=0.18",
+ "ebooklib>=0.19",
"beautifulsoup4>=4.13.4",
"PyMuPDF>=1.25.5",
"platformdirs>=4.3.7",