Files
dify/api/core/rag/models/document.py
WH-2099 b4c0fb9463 refactor(workflow-file): phase2 replace remaining imports and remove core.file
Complete phase 2 of the file module migration by replacing the remaining repository-wide legacy imports and deleting the temporary core.file compatibility package introduced in phase 1.

What this commit changes
- Replace legacy core.file.* imports with core.workflow.file.* across:
  - controllers
  - core app/agent/datasource/prompt/rag/tools/variables
  - factories, fields, libs, models, services
  - otel parser integration points
  - unit and integration tests that referenced legacy paths
- Migrate residual runtime usages in app/task pipeline paths that still referenced core.file symbols.
- Update tests and model serialization helpers that relied on old module paths.
- Remove the compatibility bridge package entirely:
  - delete core/file/__init__.py
  - delete core/file/constants.py
  - delete core/file/enums.py
  - delete core/file/file_manager.py
  - delete core/file/helpers.py
  - delete core/file/models.py
  - delete core/file/tool_file_parser.py

Verification
- No Python references to core.file remain ( -> empty).
- Targeted regression tests for migrated file primitives and factory/type flows passed:
  - tests/unit_tests/core/test_file.py
  - tests/unit_tests/factories/test_variable_factory.py
  - tests/unit_tests/services/test_variable_truncator.py

Result
- The repository now uses core.workflow.file as the single canonical file namespace.
- The migration is fully split into two commits: phase 1 compatibility + phase 2 full cutover.
2026-02-12 15:34:20 +08:00

171 lines
4.2 KiB
Python

from abc import ABC, abstractmethod
from collections.abc import Sequence
from typing import Any
from pydantic import BaseModel, Field
from core.workflow.file import File
class ChildDocument(BaseModel):
"""Class for storing a piece of text and associated metadata."""
page_content: str
vector: list[float] | None = None
"""Arbitrary metadata about the page content (e.g., source, relationships to other
documents, etc.).
"""
metadata: dict[str, Any] = Field(default_factory=dict)
class AttachmentDocument(BaseModel):
"""Class for storing a piece of text and associated metadata."""
page_content: str
provider: str | None = "dify"
vector: list[float] | None = None
metadata: dict[str, Any] = Field(default_factory=dict)
class Document(BaseModel):
"""Class for storing a piece of text and associated metadata."""
page_content: str
vector: list[float] | None = None
"""Arbitrary metadata about the page content (e.g., source, relationships to other
documents, etc.).
"""
metadata: dict[str, Any] = Field(default_factory=dict)
provider: str | None = "dify"
children: list[ChildDocument] | None = None
attachments: list[AttachmentDocument] | None = None
class GeneralChunk(BaseModel):
"""
General Chunk.
"""
content: str
files: list[File] | None = None
class MultimodalGeneralStructureChunk(BaseModel):
"""
Multimodal General Structure Chunk.
"""
general_chunks: list[GeneralChunk]
class GeneralStructureChunk(BaseModel):
"""
General Structure Chunk.
"""
general_chunks: list[str]
class ParentChildChunk(BaseModel):
"""
Parent Child Chunk.
"""
parent_content: str
child_contents: list[str]
files: list[File] | None = None
class ParentChildStructureChunk(BaseModel):
"""
Parent Child Structure Chunk.
"""
parent_child_chunks: list[ParentChildChunk]
parent_mode: str = "paragraph"
class QAChunk(BaseModel):
"""
QA Chunk.
"""
question: str
answer: str
class QAStructureChunk(BaseModel):
"""
QAStructureChunk.
"""
qa_chunks: list[QAChunk]
class BaseDocumentTransformer(ABC):
"""Abstract base class for document transformation systems.
A document transformation system takes a sequence of Documents and returns a
sequence of transformed Documents.
Example:
.. code-block:: python
class EmbeddingsRedundantFilter(BaseDocumentTransformer, BaseModel):
model_config = ConfigDict(arbitrary_types_allowed=True)
embeddings: Embeddings
similarity_fn: Callable = cosine_similarity
similarity_threshold: float = 0.95
def transform_documents(
self, documents: Sequence[Document], **kwargs: Any
) -> Sequence[Document]:
stateful_documents = get_stateful_documents(documents)
embedded_documents = _get_embeddings_from_stateful_docs(
self.embeddings, stateful_documents
)
included_idxs = _filter_similar_embeddings(
embedded_documents, self.similarity_fn, self.similarity_threshold
)
return [stateful_documents[i] for i in sorted(included_idxs)]
async def atransform_documents(
self, documents: Sequence[Document], **kwargs: Any
) -> Sequence[Document]:
raise NotImplementedError
"""
@abstractmethod
def transform_documents(self, documents: Sequence[Document], **kwargs: Any) -> Sequence[Document]:
"""Transform a list of documents.
Args:
documents: A sequence of Documents to be transformed.
Returns:
A list of transformed Documents.
"""
@abstractmethod
async def atransform_documents(self, documents: Sequence[Document], **kwargs: Any) -> Sequence[Document]:
"""Asynchronously transform a list of documents.
Args:
documents: A sequence of Documents to be transformed.
Returns:
A list of transformed Documents.
"""