notionary 0.2.19__py3-none-any.whl → 0.2.22__py3-none-any.whl

notionary/page/{content/page_content_writer.py → page_content_deleting_service.py}

@@ -1,76 +1,51 @@
 from typing import Optional
 
-from notionary.blocks import BlockRegistry
-from notionary.blocks.shared.block_client import NotionBlockClient
-from notionary.models.notion_block_response import Block
-from notionary.page.content.markdown_whitespace_processor import (
-    MarkdownWhitespaceProcessor,
-)
-from notionary.page.content.notion_text_length_utils import fix_blocks_content_length
-from notionary.page.formatting.markdown_to_notion_converter import (
-    MarkdownToNotionConverter,
-)
-
+from notionary.blocks.client import NotionBlockClient
+from notionary.blocks.models import Block
+from notionary.blocks.registry.block_registry import BlockRegistry
+from notionary.page.reader.page_content_retriever import PageContentRetriever
 from notionary.util import LoggingMixin
 
 
-class PageContentWriter(LoggingMixin):
+class PageContentDeletingService(LoggingMixin):
+    """Service responsible for deleting page content and blocks."""
+
     def __init__(self, page_id: str, block_registry: BlockRegistry):
         self.page_id = page_id
         self.block_registry = block_registry
         self._block_client = NotionBlockClient()
+        self._content_retriever = PageContentRetriever(block_registry=block_registry)
 
-        self._markdown_to_notion_converter = MarkdownToNotionConverter(
-            block_registry=block_registry
-        )
-
-    async def append_markdown(self, markdown_text: str, append_divider=True) -> bool:
-        """Append markdown text to a Notion page, automatically handling content length limits."""
-        if append_divider:
-            markdown_text = markdown_text + "---\n"
-
-        markdown_text = self._process_markdown_whitespace(markdown_text)
-
-        try:
-            blocks = self._markdown_to_notion_converter.convert(markdown_text)
-
-            fixed_blocks = fix_blocks_content_length(blocks)
-
-            result = await self._block_client.append_block_children(
-                block_id=self.page_id, children=fixed_blocks
-            )
-            self.logger.debug("Append block children result: %r", result)
-            return bool(result)
-        except Exception as e:
-            import traceback
-
-            self.logger.error(
-                "Error appending markdown: %s\nTraceback:\n%s",
-                str(e),
-                traceback.format_exc(),
-            )
-            return False
-
-    async def clear_page_content(self) -> bool:
-        """Clear all content of the page."""
+    async def clear_page_content(self) -> Optional[str]:
+        """Clear all content of the page and return deleted content as markdown."""
         try:
             children_response = await self._block_client.get_block_children(
                 block_id=self.page_id
             )
 
             if not children_response or not children_response.results:
-                return True
+                return None
+
+            # Use PageContentRetriever for sophisticated markdown conversion
+            deleted_content = self._content_retriever._convert_blocks_to_markdown(
+                children_response.results, indent_level=0
+            )
 
+            # Delete blocks
             success = True
             for block in children_response.results:
                 block_success = await self._delete_block_with_children(block)
                 if not block_success:
                     success = False
 
-            return success
-        except Exception as e:
-            self.logger.error("Error clearing page content: %s", str(e))
-            return False
+            if not success:
+                self.logger.warning("Some blocks could not be deleted")
+
+            return deleted_content if deleted_content else None
+
+        except Exception:
+            self.logger.error("Error clearing page content", exc_info=True)
+            return None
 
     async def _delete_block_with_children(self, block: Block) -> bool:
         """Delete a block and all its children recursively."""
@@ -140,12 +115,3 @@ class PageContentWriter(LoggingMixin):
         else:
             self.logger.warning("Block %s was not properly archived/deleted", block.id)
             return False
-
-    def _process_markdown_whitespace(self, markdown_text: str) -> str:
-        """Process markdown text to normalize whitespace while preserving code blocks."""
-        lines = markdown_text.split("\n")
-        if not lines:
-            return ""
-
-        processor = MarkdownWhitespaceProcessor()
-        return processor.process_lines(lines)

notionary/page/page_content_writer.py

@@ -0,0 +1,177 @@
+from typing import Callable, Optional, Union
+
+from notionary.blocks.client import NotionBlockClient
+from notionary.blocks.divider import DividerElement
+from notionary.blocks.registry.block_registry import BlockRegistry
+from notionary.blocks.table_of_contents import TableOfContentsElement
+from notionary.markdown.markdown_builder import MarkdownBuilder
+from notionary.page.writer.markdown_to_notion_converter import MarkdownToNotionConverter
+from notionary.util import LoggingMixin
+
+
+class PageContentWriter(LoggingMixin):
+    def __init__(self, page_id: str, block_registry: BlockRegistry):
+        self.page_id = page_id
+        self.block_registry = block_registry
+        self._block_client = NotionBlockClient()
+
+        self._markdown_to_notion_converter = MarkdownToNotionConverter(
+            block_registry=block_registry
+        )
+
+    async def append_markdown(
+        self,
+        content: Union[str, Callable[[MarkdownBuilder], MarkdownBuilder]],
+        *,
+        append_divider: bool = True,
+        prepend_table_of_contents: bool = False,
+    ) -> Optional[str]:
+        """
+        Append markdown content to a Notion page using either text or builder callback.
+        """
+
+        if isinstance(content, str):
+            final_markdown = content
+        elif callable(content):
+            builder = MarkdownBuilder()
+            content(builder)
+            final_markdown = builder.build()
+        else:
+            raise ValueError(
+                "content must be either a string or a callable that takes a MarkdownBuilder"
+            )
+
+        # Add optional components
+        if prepend_table_of_contents:
+            self._ensure_table_of_contents_exists_in_registry()
+            final_markdown = "[toc]\n\n" + final_markdown
+
+        if append_divider:
+            self._ensure_divider_exists_in_registry()
+            final_markdown = final_markdown + "\n\n---\n"
+
+        processed_markdown = self._process_markdown_whitespace(final_markdown)
+
+        try:
+            blocks = await self._markdown_to_notion_converter.convert(
+                processed_markdown
+            )
+
+            result = await self._block_client.append_block_children(
+                block_id=self.page_id, children=blocks
+            )
+
+            if result:
+                self.logger.debug("Successfully appended %d blocks", len(blocks))
+                return processed_markdown
+            else:
+                self.logger.error("Failed to append blocks")
+                return None
+
+        except Exception as e:
+            self.logger.error("Error appending markdown: %s", str(e), exc_info=True)
+            return None
+
+    def _process_markdown_whitespace(self, markdown_text: str) -> str:
+        """Process markdown text to normalize whitespace while preserving code blocks."""
+        lines = markdown_text.split("\n")
+        if not lines:
+            return ""
+
+        return self._process_whitespace_lines(lines)
+
+    def _process_whitespace_lines(self, lines: list[str]) -> str:
+        """Process all lines and return the processed markdown."""
+        processed_lines = []
+        in_code_block = False
+        current_code_block = []
+
+        for line in lines:
+            processed_lines, in_code_block, current_code_block = (
+                self._process_single_line(
+                    line, processed_lines, in_code_block, current_code_block
+                )
+            )
+
+        return "\n".join(processed_lines)
+
+    def _process_single_line(
+        self,
+        line: str,
+        processed_lines: list[str],
+        in_code_block: bool,
+        current_code_block: list[str],
+    ) -> tuple[list[str], bool, list[str]]:
+        """Process a single line and return updated state."""
+        if self._is_code_block_marker(line):
+            return self._handle_code_block_marker(
+                line, processed_lines, in_code_block, current_code_block
+            )
+        if in_code_block:
+            current_code_block.append(line)
+            return processed_lines, in_code_block, current_code_block
+        else:
+            processed_lines.append(line.lstrip())
+            return processed_lines, in_code_block, current_code_block
+
+    def _handle_code_block_marker(
+        self,
+        line: str,
+        processed_lines: list[str],
+        in_code_block: bool,
+        current_code_block: list[str],
+    ) -> tuple[list[str], bool, list[str]]:
+        """Handle code block start/end markers."""
+        if not in_code_block:
+            return self._start_code_block(line, processed_lines)
+        else:
+            return self._end_code_block(processed_lines, current_code_block)
+
+    def _start_code_block(
+        self, line: str, processed_lines: list[str]
+    ) -> tuple[list[str], bool, list[str]]:
+        """Start a new code block."""
+        processed_lines.append(self._normalize_code_block_start(line))
+        return processed_lines, True, []
+
+    def _end_code_block(
+        self, processed_lines: list[str], current_code_block: list[str]
+    ) -> tuple[list[str], bool, list[str]]:
+        """End the current code block."""
+        processed_lines.extend(self._normalize_code_block_content(current_code_block))
+        processed_lines.append("```")
+        return processed_lines, False, []
+
+    def _is_code_block_marker(self, line: str) -> bool:
+        """Check if line is a code block marker."""
+        return line.lstrip().startswith("```")
+
+    def _normalize_code_block_start(self, line: str) -> str:
+        """Normalize code block opening marker."""
+        language = line.lstrip().replace("```", "", 1).strip()
+        return "```" + language
+
+    def _normalize_code_block_content(self, code_lines: list[str]) -> list[str]:
+        """Normalize code block indentation."""
+        if not code_lines:
+            return []
+
+        # Find minimum indentation from non-empty lines
+        non_empty_lines = [line for line in code_lines if line.strip()]
+        if not non_empty_lines:
+            return [""] * len(code_lines)
+
+        min_indent = min(len(line) - len(line.lstrip()) for line in non_empty_lines)
+        if min_indent == 0:
+            return code_lines
+
+        # Remove common indentation
+        return ["" if not line.strip() else line[min_indent:] for line in code_lines]
+
+    def _ensure_table_of_contents_exists_in_registry(self) -> None:
+        """Ensure TableOfContents is registered in the block registry."""
+        self.block_registry.register(TableOfContentsElement)
+
+    def _ensure_divider_exists_in_registry(self) -> None:
+        """Ensure DividerBlock is registered in the block registry."""
+        self.block_registry.register(DividerElement)

notionary/page/page_context.py

@@ -0,0 +1,65 @@
+# notionary/blocks/context/page_context.py
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Optional
+from dataclasses import dataclass
+from contextvars import ContextVar
+
+if TYPE_CHECKING:
+    from notionary.database.client import NotionDatabaseClient
+    from notionary.file_upload import NotionFileUploadClient
+
+
+@dataclass(frozen=True)
+class PageContextProvider:
+    """Context object that provides dependencies for block conversion operations."""
+
+    page_id: str
+    database_client: NotionDatabaseClient
+    file_upload_client: NotionFileUploadClient
+
+
+# Context variable
+_page_context: ContextVar[Optional[PageContextProvider]] = ContextVar(
+    "page_context", default=None
+)
+
+
+def get_page_context() -> PageContextProvider:
+    """Get current page context or raise if not available."""
+    context = _page_context.get()
+    if context is None:
+        raise RuntimeError(
+            "No page context available. Use 'async with page_context(...)'"
+        )
+    return context
+
+
+def get_page_context_optional() -> Optional[PageContextProvider]:
+    """Get current page context or None if not available."""
+    return _page_context.get()
+
+
+class page_context:
+    """Async-only context manager for page operations."""
+
+    def __init__(self, provider: PageContextProvider):
+        self.provider = provider
+        self._token = None
+
+    def _set_context(self) -> PageContextProvider:
+        """Helper to set context and return provider."""
+        self._token = _page_context.set(self.provider)
+        return self.provider
+
+    def _reset_context(self) -> None:
+        """Helper to reset context."""
+        if self._token is not None:
+            _page_context.reset(self._token)
+
+    async def __aenter__(self) -> PageContextProvider:
+        return self._set_context()
+
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        self._reset_context()
+        return False

notionary/page/reader/handler/__init__.py

@@ -0,0 +1,19 @@
+from .base_block_renderer import BlockHandler
+from .block_rendering_context import BlockRenderingContext
+from .column_list_renderer import ColumnListRenderer
+from .column_renderer import ColumnRenderer
+from .line_renderer import LineRenderer
+from .numbered_list_renderer import NumberedListRenderer
+from .toggle_renderer import ToggleRenderer
+from .toggleable_heading_renderer import ToggleableHeadingRenderer
+
+__all__ = [
+    "BlockHandler",
+    "BlockRenderingContext",
+    "ColumnListRenderer",
+    "ColumnRenderer",
+    "LineRenderer",
+    "NumberedListRenderer",
+    "ToggleRenderer",
+    "ToggleableHeadingRenderer",
+]

notionary/page/reader/handler/base_block_renderer.py

@@ -0,0 +1,44 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+from typing import Optional
+
+from notionary.page.reader.handler.block_rendering_context import BlockRenderingContext
+
+
+class BlockHandler(ABC):
+    """Abstract base class for block handlers."""
+
+    def __init__(self):
+        self._next_handler: Optional[BlockHandler] = None
+
+    def set_next(self, handler: BlockHandler) -> BlockHandler:
+        """Set the next handler in the chain."""
+        self._next_handler = handler
+        return handler
+
+    async def handle(self, context: BlockRenderingContext) -> None:
+        """Handle the block or pass to next handler."""
+        if self._can_handle(context):
+            await self._process(context)
+        elif self._next_handler:
+            await self._next_handler.handle(context)
+
+    @abstractmethod
+    def _can_handle(self, context: BlockRenderingContext) -> bool:
+        """Check if this handler can process the current block."""
+        pass
+
+    @abstractmethod
+    async def _process(self, context: BlockRenderingContext) -> None:
+        """Process the block and update context."""
+        pass
+
+    def _indent_text(self, text: str, spaces: int = 4) -> str:
+        """Indent each line of text with specified number of spaces."""
+        if not text:
+            return text
+
+        indent = " " * spaces
+        lines = text.split("\n")
+        return "\n".join(f"{indent}{line}" if line.strip() else line for line in lines)

notionary/page/reader/handler/block_processing_context.py

@@ -0,0 +1,35 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Optional
+
+from notionary.blocks.models import Block
+from notionary.blocks.registry.block_registry import BlockRegistry
+
+
+@dataclass
+class BlockProcessingContext:
+    """Context for processing blocks during markdown conversion."""
+
+    block: Block
+    indent_level: int
+    block_registry: BlockRegistry
+
+    # Result
+    markdown_result: Optional[str] = None
+    children_result: Optional[str] = None
+    was_processed: bool = False
+
+    def has_children(self) -> bool:
+        """Check if block has children that need processing."""
+        return (
+            self.block.has_children
+            and self.block.children is not None
+            and len(self.block.children) > 0
+        )
+
+    def get_children_blocks(self) -> list[Block]:
+        """Get the children blocks safely."""
+        if self.has_children():
+            return self.block.children
+        return []

notionary/page/reader/handler/block_rendering_context.py

@@ -0,0 +1,48 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Callable, Optional
+
+from notionary.blocks.models import Block
+from notionary.blocks.registry.block_registry import BlockRegistry
+
+
+@dataclass
+class BlockRenderingContext:
+    """Context for processing blocks during markdown conversion."""
+
+    block: Block
+    indent_level: int
+    block_registry: BlockRegistry
+    convert_children_callback: Optional[Callable[[list[Block], int], str]] = None
+
+    # For batch processing
+    all_blocks: Optional[list[Block]] = None
+    current_block_index: Optional[int] = None
+    blocks_consumed: int = 0
+
+    # Result
+    markdown_result: Optional[str] = None
+    children_result: Optional[str] = None
+    was_processed: bool = False
+
+    def has_children(self) -> bool:
+        """Check if block has children that need processing."""
+        return (
+            self.block.has_children
+            and self.block.children is not None
+            and len(self.block.children) > 0
+        )
+
+    def get_children_blocks(self) -> list[Block]:
+        """Get the children blocks safely."""
+        if self.has_children():
+            return self.block.children
+        return []
+
+    def convert_children_to_markdown(self, indent_level: int = 0) -> str:
+        """Convert children blocks to markdown using the callback."""
+        if not self.has_children() or not self.convert_children_callback:
+            return ""
+
+        return self.convert_children_callback(self.get_children_blocks(), indent_level)

notionary/page/reader/handler/column_list_renderer.py

@@ -0,0 +1,51 @@
+from notionary.blocks.column.column_list_element import ColumnListElement
+from notionary.page.reader.handler import BlockHandler, BlockRenderingContext
+
+
+class ColumnListRenderer(BlockHandler):
+    """Handles column list blocks with their column children."""
+
+    def _can_handle(self, context: BlockRenderingContext) -> bool:
+        return ColumnListElement.match_notion(context.block)
+
+    def _process(self, context: BlockRenderingContext) -> None:
+        # Create column list start line
+        column_list_start = "::: columns"
+
+        # Apply indentation if needed
+        if context.indent_level > 0:
+            column_list_start = self._indent_text(
+                column_list_start, spaces=context.indent_level * 4
+            )
+
+        # Process children if they exist
+        children_markdown = ""
+        if context.has_children():
+            # Import here to avoid circular dependency
+            from notionary.page.reader.page_content_retriever import (
+                PageContentRetriever,
+            )
+
+            # Create a temporary retriever to process children
+            retriever = PageContentRetriever(context.block_registry)
+            children_markdown = retriever._convert_blocks_to_markdown(
+                context.get_children_blocks(),
+                indent_level=0,  # No indentation for content inside column lists
+            )
+
+        # Create column list end line
+        column_list_end = ":::"
+        if context.indent_level > 0:
+            column_list_end = self._indent_text(
+                column_list_end, spaces=context.indent_level * 4
+            )
+
+        # Combine column list with children content
+        if children_markdown:
+            context.markdown_result = (
+                f"{column_list_start}\n{children_markdown}\n{column_list_end}"
+            )
+        else:
+            context.markdown_result = f"{column_list_start}\n{column_list_end}"
+
+        context.was_processed = True

notionary/page/reader/handler/column_renderer.py

@@ -0,0 +1,60 @@
+from notionary.blocks.column.column_element import ColumnElement
+from notionary.page.reader.handler import BlockHandler, BlockRenderingContext
+
+
+class ColumnRenderer(BlockHandler):
+    """Handles individual column blocks with their children content."""
+
+    def _can_handle(self, context: BlockRenderingContext) -> bool:
+        return ColumnElement.match_notion(context.block)
+
+    def _process(self, context: BlockRenderingContext) -> None:
+        # Get the column start line with potential width ratio
+        column_start = self._extract_column_start(context.block)
+
+        # Apply indentation if needed
+        if context.indent_level > 0:
+            column_start = self._indent_text(
+                column_start, spaces=context.indent_level * 4
+            )
+
+        # Process children if they exist
+        children_markdown = ""
+        if context.has_children():
+            # Import here to avoid circular dependency
+            from notionary.page.reader.page_content_retriever import (
+                PageContentRetriever,
+            )
+
+            # Create a temporary retriever to process children
+            retriever = PageContentRetriever(context.block_registry)
+            children_markdown = retriever._convert_blocks_to_markdown(
+                context.get_children_blocks(),
+                indent_level=0,  # No indentation for content inside columns
+            )
+
+        # Create column end line
+        column_end = ":::"
+        if context.indent_level > 0:
+            column_end = self._indent_text(column_end, spaces=context.indent_level * 4)
+
+        # Combine column with children content
+        if children_markdown:
+            context.markdown_result = (
+                f"{column_start}\n{children_markdown}\n{column_end}"
+            )
+        else:
+            context.markdown_result = f"{column_start}\n{column_end}"
+
+        context.was_processed = True
+
+    def _extract_column_start(self, block) -> str:
+        """Extract column start line with potential width ratio."""
+        if not block.column:
+            return "::: column"
+
+        width_ratio = block.column.width_ratio
+        if width_ratio:
+            return f"::: column {width_ratio}"
+        else:
+            return "::: column"