notionary 0.1.25__py3-none-any.whl → 0.1.26__py3-none-any.whl

notionary/page/markdown_to_notion_converter.py

@@ -7,180 +7,157 @@ from notionary.elements.registry.block_element_registry_builder import (
 
 
 class MarkdownToNotionConverter:
+    """Converts Markdown text to Notion API block format with support for pipe syntax for nested structures."""
+
     SPACER_MARKER = "<!-- spacer -->"
-    MULTILINE_CONTENT_MARKER = "<!-- REMOVED_MULTILINE_CONTENT -->"
-    TOGGLE_MARKER = "<!-- toggle_content -->"
-    TOGGLE_MARKER_PREFIX = "<!-- toggle_"
-    TOGGLE_MARKER_SUFFIX = " -->"
+    TOGGLE_ELEMENT_TYPES = ["ToggleElement", "ToggleableHeadingElement"]
+    PIPE_CONTENT_PATTERN = r"^\|\s?(.*)$"
 
     def __init__(self, block_registry: Optional[BlockElementRegistry] = None):
-        """
-        Initialize the MarkdownToNotionConverter.
-
-        Args:
-            block_registry: Optional registry of Notion block elements
-        """
+        """Initialize the converter with an optional custom block registry."""
         self._block_registry = (
             block_registry or BlockElementRegistryBuilder().create_full_registry()
         )
 
-        self._setup_element_callbacks()
-
-    def _setup_element_callbacks(self) -> None:
-        """Registriert den Converter als Callback für Elemente, die ihn benötigen."""
+    def convert(self, markdown_text: str) -> List[Dict[str, Any]]:
+        """Convert markdown text to Notion API block format."""
+        if not markdown_text:
+            return []
 
-        for element in self._block_registry.get_elements():
-            if hasattr(element, "set_converter_callback"):
-                element.set_converter_callback(self.convert)
+        # Collect all blocks with their positions in the text
+        all_blocks_with_positions = self._collect_all_blocks_with_positions(
+            markdown_text
+        )
 
-    def convert(self, markdown_text: str) -> List[Dict[str, Any]]:
-        """
-        Convert markdown text to Notion API block format.
+        # Sort all blocks by their position in the text
+        all_blocks_with_positions.sort(key=lambda x: x[0])
 
-        Args:
-            markdown_text: The markdown text to convert
+        # Extract just the blocks without position information
+        blocks = [block for _, _, block in all_blocks_with_positions]
 
-        Returns:
-            List of Notion blocks
-        """
-        if not markdown_text:
-            return []
+        # Process spacing between blocks
+        return self._process_block_spacing(blocks)
 
-        # We'll process all blocks in order, preserving their original positions
+    def _collect_all_blocks_with_positions(
+        self, markdown_text: str
+    ) -> List[Tuple[int, int, Dict[str, Any]]]:
+        """Collect all blocks with their positions in the text."""
         all_blocks = []
 
-        # First, identify all toggle blocks
-        toggle_blocks = self._identify_toggle_blocks(markdown_text)
-
-        # If we have toggles, process them and extract positions
-        if toggle_blocks:
-            all_blocks.extend(toggle_blocks)
+        # Process toggleable elements first (both Toggle and ToggleableHeading)
+        toggleable_blocks = self._identify_toggleable_blocks(markdown_text)
 
         # Process other multiline elements
-        multiline_blocks = self._identify_multiline_blocks(markdown_text, toggle_blocks)
-        if multiline_blocks:
-            all_blocks.extend(multiline_blocks)
-
-        # Process remaining text line by line
-        line_blocks = self._process_text_lines(
-            markdown_text, toggle_blocks + multiline_blocks
+        multiline_blocks = self._identify_multiline_blocks(
+            markdown_text, toggleable_blocks
         )
-        if line_blocks:
-            all_blocks.extend(line_blocks)
 
-        # Sort all blocks by their position in the text
-        all_blocks.sort(key=lambda x: x[0])
+        # Process remaining text line by line
+        processed_blocks = toggleable_blocks + multiline_blocks
+        line_blocks = self._process_text_lines(markdown_text, processed_blocks)
 
-        # Extract just the blocks without position information
-        blocks = [block for _, _, block in all_blocks]
+        # Combine all blocks
+        all_blocks.extend(toggleable_blocks)
+        all_blocks.extend(multiline_blocks)
+        all_blocks.extend(line_blocks)
 
-        # Process spacing between blocks
-        return self._process_block_spacing(blocks)
+        return all_blocks
 
-    def _identify_toggle_blocks(
+    def _identify_toggleable_blocks(
         self, text: str
     ) -> List[Tuple[int, int, Dict[str, Any]]]:
-        """
-        Identify all toggle blocks in the text without replacing them.
+        """Identify all toggleable blocks (Toggle and ToggleableHeading) in the text."""
+        toggleable_blocks = []
+
+        # Find all toggleable elements
+        toggleable_elements = self._get_toggleable_elements()
 
-        Args:
-            text: The text to process
+        if not toggleable_elements:
+            return []
+
+        # Process each toggleable element type
+        for element in toggleable_elements:
+            if hasattr(element, "find_matches"):
+                # Find matches with context awareness
+                matches = element.find_matches(text, self.convert, context_aware=True)
+                if matches:
+                    toggleable_blocks.extend(matches)
 
-        Returns:
-            List of (start_pos, end_pos, block) tuples
-        """
-        # Find toggle element in registry
-        toggle_element = None
+        return toggleable_blocks
+
+    def _get_toggleable_elements(self):
+        """Return all toggleable elements from the registry."""
+        toggleable_elements = []
         for element in self._block_registry.get_elements():
             if (
                 element.is_multiline()
                 and hasattr(element, "match_markdown")
-                and element.__name__ == "ToggleElement"
+                and element.__name__ in self.TOGGLE_ELEMENT_TYPES
             ):
-                toggle_element = element
-                break
-
-        if not toggle_element:
-            return []
-
-        # Use the find_matches method with context awareness
-        # Pass the converter's convert method as a callback to process nested content
-        toggle_blocks = toggle_element.find_matches(
-            text, self.convert, context_aware=True
-        )
-        return toggle_blocks
+                toggleable_elements.append(element)
+        return toggleable_elements
 
     def _identify_multiline_blocks(
         self, text: str, exclude_blocks: List[Tuple[int, int, Dict[str, Any]]]
     ) -> List[Tuple[int, int, Dict[str, Any]]]:
-        """
-        Identify all multiline blocks (except toggle blocks) without altering the text.
-
-        Args:
-            text: The text to process
-            exclude_blocks: Blocks to exclude (e.g., already identified toggle blocks)
-
-        Returns:
-            List of (start_pos, end_pos, block) tuples
-        """
-        # Get all multiline elements except ToggleElement
-        multiline_elements = [
-            element
-            for element in self._block_registry.get_multiline_elements()
-            if element.__name__ != "ToggleElement"
-        ]
+        """Identify all multiline blocks (except toggleable blocks)."""
+        # Get all multiline elements except toggleable ones
+        multiline_elements = self._get_non_toggleable_multiline_elements()
 
         if not multiline_elements:
             return []
 
-        # Create a set of ranges to exclude
-        exclude_ranges = set()
-        for start, end, _ in exclude_blocks:
-            exclude_ranges.update(range(start, end + 1))
+        # Create set of positions to exclude
+        excluded_ranges = self._create_excluded_position_set(exclude_blocks)
 
         multiline_blocks = []
         for element in multiline_elements:
             if not hasattr(element, "find_matches"):
                 continue
 
-            # Find all matches for this element
-            if hasattr(element, "set_converter_callback"):
-                matches = element.find_matches(text, self.convert)
-            else:
-                matches = element.find_matches(text)
+            matches = element.find_matches(text)
 
             if not matches:
                 continue
 
-            # Add only blocks that don't overlap with excluded ranges
-            for start, end, block in matches:
-                # Check if this block overlaps with any excluded range
-                if any(start <= i <= end for i in exclude_ranges):
+            # Add blocks that don't overlap with excluded positions
+            for start_pos, end_pos, block in matches:
+                if self._overlaps_with_excluded_positions(
+                    start_pos, end_pos, excluded_ranges
+                ):
                     continue
-                multiline_blocks.append((start, end, block))
+                multiline_blocks.append((start_pos, end_pos, block))
 
         return multiline_blocks
 
+    def _get_non_toggleable_multiline_elements(self):
+        """Get multiline elements that are not toggleable elements."""
+        return [
+            element
+            for element in self._block_registry.get_multiline_elements()
+            if element.__name__ not in self.TOGGLE_ELEMENT_TYPES
+        ]
+
+    def _create_excluded_position_set(self, exclude_blocks):
+        """Create a set of positions to exclude based on block ranges."""
+        excluded_positions = set()
+        for start_pos, end_pos, _ in exclude_blocks:
+            excluded_positions.update(range(start_pos, end_pos + 1))
+        return excluded_positions
+
+    def _overlaps_with_excluded_positions(self, start_pos, end_pos, excluded_positions):
+        """Check if a range overlaps with any excluded positions."""
+        return any(pos in excluded_positions for pos in range(start_pos, end_pos + 1))
+
     def _process_text_lines(
         self, text: str, exclude_blocks: List[Tuple[int, int, Dict[str, Any]]]
     ) -> List[Tuple[int, int, Dict[str, Any]]]:
-        """
-        Process text line by line, excluding ranges already processed.
-
-        Args:
-            text: The text to process
-            exclude_blocks: Blocks to exclude (e.g., already identified toggle and multiline blocks)
-
-        Returns:
-            List of (start_pos, end_pos, block) tuples
-        """
+        """Process text line by line, excluding already processed ranges and handling pipe syntax lines."""
         if not text:
             return []
 
-        # Create a set of excluded positions
-        exclude_positions = set()
-        for start, end, _ in exclude_blocks:
-            exclude_positions.update(range(start, end + 1))
+        # Create set of excluded positions
+        excluded_positions = self._create_excluded_position_set(exclude_blocks)
 
         line_blocks = []
         lines = text.split("\n")
@@ -194,210 +171,245 @@ class MarkdownToNotionConverter:
             line_length = len(line) + 1  # +1 for newline
             line_end = current_pos + line_length - 1
 
-            # Skip lines that are part of excluded blocks
-            if any(current_pos <= pos <= line_end for pos in exclude_positions):
+            # Skip excluded lines and pipe syntax lines (they're part of toggleable content)
+            if self._overlaps_with_excluded_positions(
+                current_pos, line_end, excluded_positions
+            ) or self._is_pipe_syntax_line(line):
                 current_pos += line_length
                 continue
 
-            # Check for spacer marker
-            if self._is_spacer_marker(line):
-                line_blocks.append(
-                    (
-                        current_pos,
-                        current_pos + line_length - 1,
-                        self._create_empty_paragraph(),
-                    )
-                )
-                current_pos += line_length
-                continue
+            processed = self._process_line(
+                line,
+                current_pos,
+                line_end,
+                line_blocks,
+                current_paragraph,
+                paragraph_start,
+                in_todo_sequence,
+            )
 
-            # Process todos first to keep them grouped
-            todo_block = self._extract_todo_item(line)
-            if todo_block:
-                self._handle_todo_item(
-                    todo_block,
-                    line_length,
-                    current_pos,
-                    current_paragraph,
-                    paragraph_start,
-                    line_blocks,
-                    in_todo_sequence,
-                )
-                in_todo_sequence = True
-                current_pos += line_length
-                continue
+            current_pos = processed["current_pos"]
+            current_paragraph = processed["current_paragraph"]
+            paragraph_start = processed["paragraph_start"]
+            in_todo_sequence = processed["in_todo_sequence"]
 
-            if in_todo_sequence:
-                in_todo_sequence = False
+        # Process remaining paragraph
+        self._process_paragraph(
+            current_paragraph, paragraph_start, current_pos, line_blocks
+        )
 
-            if not line.strip():
-                self._process_paragraph_if_present(
-                    current_paragraph, paragraph_start, current_pos, line_blocks
-                )
-                current_paragraph = []
-                current_pos += line_length
-                continue
+        return line_blocks
 
-            special_block = self._extract_special_block(line)
-            if special_block:
-                self._process_paragraph_if_present(
-                    current_paragraph, paragraph_start, current_pos, line_blocks
-                )
-                line_blocks.append(
-                    (current_pos, current_pos + line_length - 1, special_block)
-                )
-                current_paragraph = []
-                current_pos += line_length
-                continue
+    def _is_pipe_syntax_line(self, line: str) -> bool:
+        """Check if a line uses pipe syntax (for nested content)."""
+        import re
 
-            # Handle as part of paragraph
-            if not current_paragraph:
-                paragraph_start = current_pos
-            current_paragraph.append(line)
-            current_pos += line_length
+        return bool(re.match(self.PIPE_CONTENT_PATTERN, line))
 
-        # Process any remaining paragraph content
-        self._process_paragraph_if_present(
-            current_paragraph, paragraph_start, current_pos, line_blocks
-        )
+    def _process_line(
+        self,
+        line: str,
+        current_pos: int,
+        line_end: int,
+        line_blocks: List[Tuple[int, int, Dict[str, Any]]],
+        current_paragraph: List[str],
+        paragraph_start: int,
+        in_todo_sequence: bool,
+    ) -> Dict[str, Any]:
+        """Process a single line of text."""
+        line_length = len(line) + 1  # +1 for newline
+
+        # Check for spacer
+        if self._is_spacer_line(line):
+            line_blocks.append((current_pos, line_end, self._create_empty_paragraph()))
+            return self._update_line_state(
+                current_pos + line_length,
+                current_paragraph,
+                paragraph_start,
+                in_todo_sequence,
+            )
 
-        return line_blocks
+        # Handle todo items
+        todo_block = self._extract_todo_item(line)
+        if todo_block:
+            return self._process_todo_line(
+                todo_block,
+                current_pos,
+                line_end,
+                line_blocks,
+                current_paragraph,
+                paragraph_start,
+                in_todo_sequence,
+                line_length,
+            )
 
-    def _is_spacer_marker(self, line: str) -> bool:
-        """Check if a line is a spacer marker."""
-        return line.strip() == self.SPACER_MARKER
+        if in_todo_sequence:
+            in_todo_sequence = False
 
-    def _extract_todo_item(self, line: str) -> Optional[Dict[str, Any]]:
-        """
-        Try to extract a todo item from a line.
+        # Handle empty lines
+        if not line.strip():
+            self._process_paragraph(
+                current_paragraph, paragraph_start, current_pos, line_blocks
+            )
+            return self._update_line_state(
+                current_pos + line_length, [], paragraph_start, False
+            )
 
-        Returns:
-            Todo block if line is a todo item, None otherwise
-        """
-        for element in self._block_registry.get_elements():
-            if (
-                not element.is_multiline()
-                and hasattr(element, "match_markdown")
-                and element.__name__ == "TodoElement"
-                and element.match_markdown(line)
-            ):
-                return element.markdown_to_notion(line)
-        return None
+        # Handle special blocks
+        special_block = self._extract_special_block(line)
+        if special_block:
+            self._process_paragraph(
+                current_paragraph, paragraph_start, current_pos, line_blocks
+            )
+            line_blocks.append((current_pos, line_end, special_block))
+            return self._update_line_state(
+                current_pos + line_length, [], paragraph_start, False
+            )
+
+        # Handle as paragraph
+        if not current_paragraph:
+            paragraph_start = current_pos
+        current_paragraph.append(line)
 
-    def _handle_todo_item(
+        return self._update_line_state(
+            current_pos + line_length,
+            current_paragraph,
+            paragraph_start,
+            in_todo_sequence,
+        )
+
+    def _is_spacer_line(self, line: str) -> bool:
+        """Check if a line is a spacer marker."""
+        return line.strip() == self.SPACER_MARKER
+
+    def _process_todo_line(
         self,
         todo_block: Dict[str, Any],
-        line_length: int,
         current_pos: int,
+        line_end: int,
+        line_blocks: List[Tuple[int, int, Dict[str, Any]]],
         current_paragraph: List[str],
         paragraph_start: int,
-        line_blocks: List[Tuple[int, int, Dict[str, Any]]],
         in_todo_sequence: bool,
-    ) -> None:
-        """Handle a todo item line."""
-        # If we were building a paragraph, finish it before starting todos
+        line_length: int,
+    ) -> Dict[str, Any]:
+        """Process a line that contains a todo item."""
+        # Finish paragraph if needed
         if not in_todo_sequence and current_paragraph:
-            self._process_paragraph_if_present(
+            self._process_paragraph(
                 current_paragraph, paragraph_start, current_pos, line_blocks
             )
-            current_paragraph.clear()
 
-        line_blocks.append((current_pos, current_pos + line_length - 1, todo_block))
+        line_blocks.append((current_pos, line_end, todo_block))
+
+        return self._update_line_state(
+            current_pos + line_length, [], paragraph_start, True
+        )
+
+    def _update_line_state(
+        self,
+        current_pos: int,
+        current_paragraph: List[str],
+        paragraph_start: int,
+        in_todo_sequence: bool,
+    ) -> Dict[str, Any]:
+        """Update and return the state after processing a line."""
+        return {
+            "current_pos": current_pos,
+            "current_paragraph": current_paragraph,
+            "paragraph_start": paragraph_start,
+            "in_todo_sequence": in_todo_sequence,
+        }
+
+    def _extract_todo_item(self, line: str) -> Optional[Dict[str, Any]]:
+        """Extract a todo item from a line if possible."""
+        todo_elements = [
+            element
+            for element in self._block_registry.get_elements()
+            if not element.is_multiline() and element.__name__ == "TodoElement"
+        ]
+
+        for element in todo_elements:
+            if element.match_markdown(line):
+                return element.markdown_to_notion(line)
+        return None
 
     def _extract_special_block(self, line: str) -> Optional[Dict[str, Any]]:
-        """
-        Try to extract a special block (not paragraph) from a line.
+        """Extract a special block (not paragraph) from a line if possible."""
+        non_multiline_elements = [
+            element
+            for element in self._block_registry.get_elements()
+            if not element.is_multiline()
+        ]
 
-        Returns:
-            Block if line is a special block, None otherwise
-        """
-        for element in self._block_registry.get_elements():
-            if (
-                not element.is_multiline()
-                and hasattr(element, "match_markdown")
-                and element.match_markdown(line)
-            ):
+        for element in non_multiline_elements:
+            if element.match_markdown(line):
                 block = element.markdown_to_notion(line)
                 if block and block.get("type") != "paragraph":
                     return block
         return None
 
-    def _process_paragraph_if_present(
+    def _process_paragraph(
         self,
         paragraph_lines: List[str],
         start_pos: int,
         end_pos: int,
         blocks: List[Tuple[int, int, Dict[str, Any]]],
     ) -> None:
-        """
-        Process a paragraph and add it to the blocks list if valid.
-
-        Args:
-            paragraph_lines: Lines that make up the paragraph
-            start_pos: Starting position of the paragraph
-            end_pos: Ending position of the paragraph
-            blocks: List to add the processed paragraph block to
-        """
+        """Process a paragraph and add it to blocks if valid."""
         if not paragraph_lines:
             return
 
         paragraph_text = "\n".join(paragraph_lines)
         block = self._block_registry.markdown_to_notion(paragraph_text)
 
-        if not block:
-            return
-
-        blocks.append((start_pos, end_pos, block))
+        if block:
+            blocks.append((start_pos, end_pos, block))
 
     def _process_block_spacing(
         self, blocks: List[Dict[str, Any]]
     ) -> List[Dict[str, Any]]:
-        """
-        Process blocks and add spacing only where no explicit spacer is present.
-
-        Args:
-            blocks: List of Notion blocks
-
-        Returns:
-            List of Notion blocks with processed spacing
-        """
+        """Add spacing between blocks where needed."""
         if not blocks:
             return blocks
 
         final_blocks = []
-        i = 0
 
-        while i < len(blocks):
-            current_block = blocks[i]
+        for block_index, current_block in enumerate(blocks):
             final_blocks.append(current_block)
 
-            # Check if this is a multiline element that needs spacing
+            # Only add spacing after multiline blocks
             if not self._is_multiline_block_type(current_block.get("type")):
-                i += 1
                 continue
 
-            # Check if the next block is already a spacer
-            if i + 1 < len(blocks) and self._is_empty_paragraph(blocks[i + 1]):
-                # Next block is already a spacer, don't add another
-                pass
-            else:
-                # No explicit spacer found, add one automatically
+            # Check if we need to add a spacer
+            if self._needs_spacer_after_block(blocks, block_index):
                 final_blocks.append(self._create_empty_paragraph())
 
-            i += 1
-
         return final_blocks
 
-    def _is_multiline_block_type(self, block_type: str) -> bool:
-        """
-        Check if a block type corresponds to a multiline element.
+    def _needs_spacer_after_block(
+        self, blocks: List[Dict[str, Any]], block_index: int
+    ) -> bool:
+        """Determine if we need to add a spacer after the current block."""
+        # Check if this is the last block (no need for spacer)
+        if block_index + 1 >= len(blocks):
+            return False
+
+        # Check if next block is already a spacer
+        next_block = blocks[block_index + 1]
+        if self._is_empty_paragraph(next_block):
+            return False
 
-        Args:
-            block_type: The type of block to check
+        # No spacer needed
+        return True
 
-        Returns:
-            True if the block type is a multiline element, False otherwise
-        """
+    def _create_empty_paragraph(self):
+        """Create an empty paragraph block."""
+        return {"type": "paragraph", "paragraph": {"rich_text": []}}
+
+    def _is_multiline_block_type(self, block_type: str) -> bool:
+        """Check if a block type corresponds to a multiline element."""
         if not block_type:
             return False
 
@@ -416,26 +428,9 @@ class MarkdownToNotionConverter:
         return False
 
     def _is_empty_paragraph(self, block: Dict[str, Any]) -> bool:
-        """
-        Check if a block is an empty paragraph.
-
-        Args:
-            block: The block to check
-
-        Returns:
-            True if it's an empty paragraph, False otherwise
-        """
+        """Check if a block is an empty paragraph."""
         if block.get("type") != "paragraph":
             return False
 
         rich_text = block.get("paragraph", {}).get("rich_text", [])
         return not rich_text or len(rich_text) == 0
-
-    def _create_empty_paragraph(self) -> Dict[str, Any]:
-        """
-        Create an empty paragraph block.
-
-        Returns:
-            Empty paragraph block dictionary
-        """
-        return {"type": "paragraph", "paragraph": {"rich_text": []}}