notionary 0.4.0__py3-none-any.whl → 0.4.2__py3-none-any.whl

notionary/page/content/markdown/nodes/quote.py

@@ -2,7 +2,7 @@ from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
 from notionary.page.content.markdown.nodes.container import ContainerNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class QuoteMarkdownNode(ContainerNode):
@@ -10,7 +10,7 @@ class QuoteMarkdownNode(ContainerNode):
         self,
         text: str,
         children: list[MarkdownNode] | None = None,
-        syntax_registry: SyntaxRegistry | None = None,
+        syntax_registry: SyntaxDefinitionRegistry | None = None,
     ) -> None:
         super().__init__(syntax_registry=syntax_registry)
         self.text = text

notionary/page/content/markdown/nodes/space.py

@@ -1,11 +1,11 @@
 from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class SpaceMarkdownNode(MarkdownNode):
-    def __init__(self, syntax_registry: SyntaxRegistry | None = None) -> None:
+    def __init__(self, syntax_registry: SyntaxDefinitionRegistry | None = None) -> None:
         super().__init__(syntax_registry=syntax_registry)
 
     @override

notionary/page/content/markdown/nodes/table.py

@@ -1,12 +1,15 @@
 from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class TableMarkdownNode(MarkdownNode):
     def __init__(
-        self, headers: list[str], rows: list[list[str]], syntax_registry: SyntaxRegistry | None = None
+        self,
+        headers: list[str],
+        rows: list[list[str]],
+        syntax_registry: SyntaxDefinitionRegistry | None = None,
     ) -> None:
         super().__init__(syntax_registry=syntax_registry)
         self._validate_input(headers, rows)
@@ -36,10 +39,10 @@ class TableMarkdownNode(MarkdownNode):
         return f"{delimiter} {joined_cells} {delimiter}"
 
     def _build_separator_row(self) -> str:
-        table_syntax = self._syntax_registry.get_table_syntax()
         col_count = len(self.headers)
-        separators = [table_syntax.end_delimiter] * col_count
-        return self._format_row(separators)
+        separators = ["-"] * col_count
+        row = self._format_row(separators)
+        return row
 
     def _build_data_rows(self) -> list[str]:
         return [self._format_row(row) for row in self.rows]

notionary/page/content/markdown/nodes/table_of_contents.py

@@ -1,11 +1,11 @@
 from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class TableOfContentsMarkdownNode(MarkdownNode):
-    def __init__(self, syntax_registry: SyntaxRegistry | None = None) -> None:
+    def __init__(self, syntax_registry: SyntaxDefinitionRegistry | None = None) -> None:
         super().__init__(syntax_registry=syntax_registry)
 
     @override

notionary/page/content/markdown/nodes/todo.py

@@ -2,7 +2,7 @@ from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
 from notionary.page.content.markdown.nodes.container import ContainerNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class TodoMarkdownNode(ContainerNode):
@@ -14,7 +14,7 @@ class TodoMarkdownNode(ContainerNode):
         checked: bool = False,
         marker: str = "-",
         children: list[MarkdownNode] | None = None,
-        syntax_registry: SyntaxRegistry | None = None,
+        syntax_registry: SyntaxDefinitionRegistry | None = None,
     ):
         super().__init__(syntax_registry=syntax_registry)
         self.text = text
@@ -24,9 +24,13 @@ class TodoMarkdownNode(ContainerNode):
 
     @override
     def to_markdown(self) -> str:
-        validated_marker = self._get_validated_marker()
-        checkbox_state = self._get_checkbox_state()
-        result = f"{validated_marker}{checkbox_state} {self.text}"
+        # Get the appropriate syntax based on checked state
+        if self.checked:
+            todo_syntax = self._syntax_registry.get_todo_done_syntax()
+        else:
+            todo_syntax = self._syntax_registry.get_todo_syntax()
+
+        result = f"{todo_syntax.start_delimiter} {self.text}"
         result += self.render_children()
         return result
 
@@ -34,5 +38,9 @@ class TodoMarkdownNode(ContainerNode):
         return self.marker if self.marker == self.VALID_MARKER else self.VALID_MARKER
 
     def _get_checkbox_state(self) -> str:
-        todo_syntax = self._syntax_registry.get_todo_syntax()
-        return todo_syntax.end_delimiter if self.checked else todo_syntax.start_delimiter
+        if self.checked:
+            todo_done_syntax = self._syntax_registry.get_todo_done_syntax()
+            return todo_done_syntax.start_delimiter
+        else:
+            todo_syntax = self._syntax_registry.get_todo_syntax()
+            return todo_syntax.start_delimiter

notionary/page/content/markdown/nodes/toggle.py

@@ -2,7 +2,7 @@ from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
 from notionary.page.content.markdown.nodes.container import ContainerNode
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class ToggleMarkdownNode(ContainerNode):
@@ -10,7 +10,7 @@ class ToggleMarkdownNode(ContainerNode):
         self,
         title: str,
         children: list[MarkdownNode] | None = None,
-        syntax_registry: SyntaxRegistry | None = None,
+        syntax_registry: SyntaxDefinitionRegistry | None = None,
     ):
         super().__init__(syntax_registry=syntax_registry)
         self.title = title

notionary/page/content/markdown/nodes/video.py

@@ -1,8 +1,10 @@
 from typing import override
 
 from notionary.page.content.markdown.nodes.base import MarkdownNode
-from notionary.page.content.markdown.nodes.mixins.caption import CaptionMarkdownNodeMixin
-from notionary.page.content.syntax import SyntaxRegistry
+from notionary.page.content.markdown.nodes.mixins.caption import (
+    CaptionMarkdownNodeMixin,
+)
+from notionary.page.content.syntax.definition import SyntaxDefinitionRegistry
 
 
 class VideoMarkdownNode(MarkdownNode, CaptionMarkdownNodeMixin):
@@ -10,7 +12,7 @@ class VideoMarkdownNode(MarkdownNode, CaptionMarkdownNodeMixin):
         self,
         url: str,
         caption: str | None = None,
-        syntax_registry: SyntaxRegistry | None = None,
+        syntax_registry: SyntaxDefinitionRegistry | None = None,
     ) -> None:
         super().__init__(syntax_registry=syntax_registry)
         self.url = url
@@ -19,5 +21,7 @@ class VideoMarkdownNode(MarkdownNode, CaptionMarkdownNodeMixin):
     @override
     def to_markdown(self) -> str:
         video_syntax = self._syntax_registry.get_video_syntax()
-        base_markdown = f"{video_syntax.start_delimiter}{self.url}{video_syntax.end_delimiter}"
+        base_markdown = (
+            f"{video_syntax.start_delimiter}{self.url}{video_syntax.end_delimiter}"
+        )
         return self._append_caption_to_markdown(base_markdown, self.caption)

notionary/page/content/markdown/structured_output/__init__.py

@@ -0,0 +1,73 @@
+from .models import (
+    AnyMarkdownNode,
+    AudioSchema,
+    BookmarkSchema,
+    BreadcrumbSchema,
+    BulletedListItemSchema,
+    BulletedListSchema,
+    CalloutSchema,
+    CodeSchema,
+    ColumnSchema,
+    ColumnsSchema,
+    DividerSchema,
+    EmbedSchema,
+    EquationSchema,
+    FileSchema,
+    Heading1Schema,
+    Heading2Schema,
+    Heading3Schema,
+    ImageSchema,
+    MarkdownDocumentSchema,
+    MarkdownNodeSchema,
+    MermaidSchema,
+    NumberedListItemSchema,
+    NumberedListSchema,
+    ParagraphSchema,
+    PdfSchema,
+    QuoteSchema,
+    SpaceSchema,
+    TableOfContentsSchema,
+    TableSchema,
+    TodoListSchema,
+    TodoSchema,
+    ToggleSchema,
+    VideoSchema,
+)
+from .service import StructuredOutputMarkdownConverter
+
+__all__ = [
+    "AnyMarkdownNode",
+    "AudioSchema",
+    "BookmarkSchema",
+    "BreadcrumbSchema",
+    "BulletedListItemSchema",
+    "BulletedListSchema",
+    "CalloutSchema",
+    "CodeSchema",
+    "ColumnSchema",
+    "ColumnsSchema",
+    "DividerSchema",
+    "EmbedSchema",
+    "EquationSchema",
+    "FileSchema",
+    "Heading1Schema",
+    "Heading2Schema",
+    "Heading3Schema",
+    "ImageSchema",
+    "MarkdownDocumentSchema",
+    "MarkdownNodeSchema",
+    "MermaidSchema",
+    "NumberedListItemSchema",
+    "NumberedListSchema",
+    "ParagraphSchema",
+    "PdfSchema",
+    "QuoteSchema",
+    "SpaceSchema",
+    "StructuredOutputMarkdownConverter",
+    "TableOfContentsSchema",
+    "TableSchema",
+    "TodoListSchema",
+    "TodoSchema",
+    "ToggleSchema",
+    "VideoSchema",
+]

notionary/page/content/markdown/structured_output/models.py

@@ -0,0 +1,391 @@
+from __future__ import annotations
+
+from enum import StrEnum
+from typing import TYPE_CHECKING, Annotated, Literal
+
+from pydantic import BaseModel, ConfigDict, Field
+
+from notionary.blocks.enums import CodingLanguage
+
+if TYPE_CHECKING:
+    from notionary.page.content.markdown.structured_output.service import (
+        StructuredOutputMarkdownConverter,
+    )
+
+
+class MarkdownNodeType(StrEnum):
+    PARAGRAPH = "paragraph"
+    HEADING_1 = "heading_1"
+    HEADING_2 = "heading_2"
+    HEADING_3 = "heading_3"
+    SPACE = "space"
+    DIVIDER = "divider"
+    QUOTE = "quote"
+    BULLETED_LIST = "bulleted_list"
+    BULLETED_LIST_ITEM = "bulleted_list_item"
+    NUMBERED_LIST = "numbered_list"
+    NUMBERED_LIST_ITEM = "numbered_list_item"
+    TODO = "todo"
+    TODO_LIST = "todo_list"
+    CALLOUT = "callout"
+    TOGGLE = "toggle"
+    IMAGE = "image"
+    VIDEO = "video"
+    AUDIO = "audio"
+    FILE = "file"
+    PDF = "pdf"
+    BOOKMARK = "bookmark"
+    EMBED = "embed"
+    CODE = "code"
+    MERMAID = "mermaid"
+    TABLE = "table"
+    BREADCRUMB = "breadcrumb"
+    EQUATION = "equation"
+    TABLE_OF_CONTENTS = "table_of_contents"
+    COLUMNS = "columns"
+
+
+class MarkdownNodeSchema(BaseModel):
+    type: MarkdownNodeType
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        raise NotImplementedError(
+            f"{self.__class__.__name__} must implement process_with()"
+        )
+
+
+class ParagraphSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.PARAGRAPH] = MarkdownNodeType.PARAGRAPH
+    text: str = Field(description="The paragraph text content")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_paragraph(self)
+
+
+class Heading1Schema(MarkdownNodeSchema):
+    type: Literal["heading_1"] = "heading_1"
+    text: str = Field(description="The heading 1 text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional child nodes"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_heading_1(self)
+
+
+class Heading2Schema(MarkdownNodeSchema):
+    type: Literal["heading_2"] = "heading_2"
+    text: str = Field(description="The heading 2 text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional child nodes"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_heading_2(self)
+
+
+class Heading3Schema(MarkdownNodeSchema):
+    type: Literal["heading_3"] = "heading_3"
+    text: str = Field(description="The heading 3 text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional child nodes"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_heading_3(self)
+
+
+class SpaceSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.SPACE] = MarkdownNodeType.SPACE
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_space()
+
+
+class DividerSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.DIVIDER] = MarkdownNodeType.DIVIDER
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_divider()
+
+
+class QuoteSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.QUOTE] = MarkdownNodeType.QUOTE
+    text: str = Field(description="The quote text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional child nodes"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_quote(self)
+
+
+class BulletedListItemSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.BULLETED_LIST_ITEM] = (
+        MarkdownNodeType.BULLETED_LIST_ITEM
+    )
+    text: str = Field(description="The bullet point text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional nested content"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_bulleted_list_item(self)
+
+
+class BulletedListSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.BULLETED_LIST] = MarkdownNodeType.BULLETED_LIST
+    items: list[BulletedListItemSchema] = Field(
+        description="List of BulletedListItemSchema objects. Each item must have 'type', 'text', and optionally 'children'"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_bulleted_list(self)
+
+
+class NumberedListItemSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.NUMBERED_LIST_ITEM] = (
+        MarkdownNodeType.NUMBERED_LIST_ITEM
+    )
+    text: str = Field(description="The numbered item text")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional nested content"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_numbered_list_item(self)
+
+
+class NumberedListSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.NUMBERED_LIST] = MarkdownNodeType.NUMBERED_LIST
+    items: list[NumberedListItemSchema] = Field(
+        description="List of NumberedListItemSchema objects. Each item must have 'type', 'text', and optionally 'children'"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_numbered_list(self)
+
+
+class TodoSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.TODO] = MarkdownNodeType.TODO
+    text: str = Field(description="The todo item text")
+    checked: bool = Field(default=False, description="Whether the todo is completed")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional nested content"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_todo(self)
+
+
+class TodoListSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.TODO_LIST] = MarkdownNodeType.TODO_LIST
+    items: list[TodoSchema] = Field(
+        description="List of TodoSchema objects. Each item must have 'type', 'text', 'checked', and optionally 'children'"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_todo_list(self)
+
+
+class CalloutSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.CALLOUT] = MarkdownNodeType.CALLOUT
+    text: str = Field(description="The callout text")
+    emoji: str | None = Field(default=None, description="Optional emoji icon")
+    children: list[MarkdownNodeSchema] | None = Field(
+        default=None, description="Optional child nodes"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_callout(self)
+
+
+class ToggleSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.TOGGLE] = MarkdownNodeType.TOGGLE
+    title: str = Field(description="The toggle title")
+    children: list[MarkdownNodeSchema] = Field(description="Content inside the toggle")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_toggle(self)
+
+
+class ImageSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.IMAGE] = MarkdownNodeType.IMAGE
+    url: str = Field(description="Image URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_image(self)
+
+
+class VideoSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.VIDEO] = MarkdownNodeType.VIDEO
+    url: str = Field(description="Video URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_video(self)
+
+
+class AudioSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.AUDIO] = MarkdownNodeType.AUDIO
+    url: str = Field(description="Audio URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_audio(self)
+
+
+class FileSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.FILE] = MarkdownNodeType.FILE
+    url: str = Field(description="File URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_file(self)
+
+
+class PdfSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.PDF] = MarkdownNodeType.PDF
+    url: str = Field(description="PDF URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_pdf(self)
+
+
+class BookmarkSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.BOOKMARK] = MarkdownNodeType.BOOKMARK
+    url: str = Field(description="Bookmark URL")
+    title: str | None = Field(default=None, description="Optional title")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_bookmark(self)
+
+
+class EmbedSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.EMBED] = MarkdownNodeType.EMBED
+    url: str = Field(description="Embed URL")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_embed(self)
+
+
+class CodeSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.CODE] = MarkdownNodeType.CODE
+    code: str = Field(description="Code content")
+    language: CodingLanguage | None = Field(
+        default=None, description="Programming language"
+    )
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_code(self)
+
+
+class MermaidSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.MERMAID] = MarkdownNodeType.MERMAID
+    diagram: str = Field(description="Mermaid diagram code")
+    caption: str | None = Field(default=None, description="Optional caption")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_mermaid(self)
+
+
+class TableSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.TABLE] = MarkdownNodeType.TABLE
+    headers: list[str] = Field(description="Table header row")
+    rows: list[list[str]] = Field(description="Table data rows")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_table(self)
+
+
+class BreadcrumbSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.BREADCRUMB] = MarkdownNodeType.BREADCRUMB
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_breadcrumb(self)
+
+
+class EquationSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.EQUATION] = MarkdownNodeType.EQUATION
+    expression: str = Field(description="LaTeX equation expression")
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_equation(self)
+
+
+class TableOfContentsSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.TABLE_OF_CONTENTS] = (
+        MarkdownNodeType.TABLE_OF_CONTENTS
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_table_of_contents(self)
+
+
+class ColumnSchema(BaseModel):
+    """Single column in a multi-column layout."""
+
+    width_ratio: float | None = Field(
+        default=None,
+        description="Relative width of this column (e.g., 0.5 for half width). If not specified, columns are equal width",
+    )
+    children: list[MarkdownNodeSchema] = Field(
+        description="Content inside this column. Can contain any markdown nodes"
+    )
+
+
+class ColumnsSchema(MarkdownNodeSchema):
+    type: Literal[MarkdownNodeType.COLUMNS] = MarkdownNodeType.COLUMNS
+    columns: list[ColumnSchema] = Field(
+        description="List of columns in this layout. Each column contains its own content"
+    )
+
+    def process_with(self, processor: StructuredOutputMarkdownConverter) -> None:
+        processor._process_columns(self)
+
+
+type AnyMarkdownNode = Annotated[
+    Heading1Schema
+    | Heading2Schema
+    | Heading3Schema
+    | ParagraphSchema
+    | SpaceSchema
+    | DividerSchema
+    | QuoteSchema
+    | BulletedListSchema
+    | BulletedListItemSchema
+    | NumberedListSchema
+    | NumberedListItemSchema
+    | TodoSchema
+    | TodoListSchema
+    | CalloutSchema
+    | ToggleSchema
+    | ImageSchema
+    | VideoSchema
+    | AudioSchema
+    | FileSchema
+    | PdfSchema
+    | BookmarkSchema
+    | EmbedSchema
+    | CodeSchema
+    | MermaidSchema
+    | TableSchema
+    | BreadcrumbSchema
+    | EquationSchema
+    | TableOfContentsSchema
+    | ColumnsSchema,
+    Field(discriminator="type"),
+]
+
+
+class MarkdownDocumentSchema(BaseModel):
+    model_config = ConfigDict(extra="forbid")
+
+    nodes: list[AnyMarkdownNode] = Field(
+        description="Ordered list of top-level markdown nodes in the document. Each node can contain nested children"
+    )