Add chunking domain objects and DocumentChunker port

ChunkingOptions and ChunkResult value objects, document_json/chunks_json
fields on AnalysisJob, and DocumentChunker protocol for adapter injection.

document-parser/domain/models.py

@@ -42,6 +42,8 @@ class AnalysisJob:
     content_markdown: str | None = None
     content_html: str | None = None
     pages_json: str | None = None
+    document_json: str | None = None
+    chunks_json: str | None = None
     error_message: str | None = None
     started_at: datetime | None = None
     completed_at: datetime | None = None
@@ -55,12 +57,19 @@ class AnalysisJob:
         self.started_at = _utcnow()
 
     def mark_completed(
-        self, markdown: str, html: str, pages_json: str,
+        self,
+        markdown: str,
+        html: str,
+        pages_json: str,
+        document_json: str | None = None,
+        chunks_json: str | None = None,
     ) -> None:
         self.status = AnalysisStatus.COMPLETED
         self.content_markdown = markdown
         self.content_html = html
         self.pages_json = pages_json
+        self.document_json = document_json
+        self.chunks_json = chunks_json
         self.completed_at = _utcnow()
 
     def mark_failed(self, error: str) -> None:

document-parser/domain/ports.py

@@ -9,7 +9,12 @@ from __future__ import annotations
 from typing import TYPE_CHECKING, Protocol
 
 if TYPE_CHECKING:
-    from domain.value_objects import ConversionOptions, ConversionResult
+    from domain.value_objects import (
+        ChunkingOptions,
+        ChunkResult,
+        ConversionOptions,
+        ConversionResult,
+    )
 
 
 class DocumentConverter(Protocol):
@@ -20,5 +25,20 @@ class DocumentConverter(Protocol):
     """
 
     async def convert(
-        self, file_path: str, options: ConversionOptions,
+        self,
+        file_path: str,
+        options: ConversionOptions,
     ) -> ConversionResult: ...
+
+
+class DocumentChunker(Protocol):
+    """Port for document chunking.
+
+    Takes a serialized DoclingDocument (JSON) and returns chunks.
+    """
+
+    async def chunk(
+        self,
+        document_json: str,
+        options: ChunkingOptions,
+    ) -> list[ChunkResult]: ...

document-parser/domain/value_objects.py

@@ -49,3 +49,23 @@ class ConversionResult:
     content_html: str
     pages: list[PageDetail]
     skipped_items: int = 0
+    document_json: str | None = None
+
+
+@dataclass
+class ChunkingOptions:
+    chunker_type: str = "hybrid"  # "hybrid", "hierarchical", "page"
+    max_tokens: int = 512
+    merge_peers: bool = True
+    repeat_table_header: bool = True
+
+    def is_default(self) -> bool:
+        return self == ChunkingOptions()
+
+
+@dataclass
+class ChunkResult:
+    text: str
+    headings: list[str] = field(default_factory=list)
+    source_page: int | None = None
+    token_count: int = 0