Skip to content

Index

RefDocInfo dataclass #

Bases: DataClassJsonMixin

用于表示已摄入文档的数据类。

Source code in llama_index/core/storage/docstore/types.py
16
17
18
19
20
21
@dataclass
class RefDocInfo(DataClassJsonMixin):
    """用于表示已摄入文档的数据类。"""

    node_ids: List = field(default_factory=list)
    metadata: Dict[str, Any] = field(default_factory=dict)

BaseDocumentStore #

Bases: ABC

Source code in llama_index/core/storage/docstore/types.py
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
class BaseDocumentStore(ABC):
    # ===== Save/load =====
    def persist(
        self,
        persist_path: str = DEFAULT_PERSIST_PATH,
        fs: Optional[fsspec.AbstractFileSystem] = None,
    ) -> None:
        """将文档存储持久化到文件中。"""

    # ===== Main interface =====
    @property
    @abstractmethod
    def docs(self) -> Dict[str, BaseNode]:
        ...

    @abstractmethod
    def add_documents(
        self,
        docs: Sequence[BaseNode],
        allow_update: bool = True,
        batch_size: int = DEFAULT_BATCH_SIZE,
        store_text: bool = True,
    ) -> None:
        ...

    @abstractmethod
    async def async_add_documents(
        self,
        docs: Sequence[BaseNode],
        allow_update: bool = True,
        batch_size: int = DEFAULT_BATCH_SIZE,
        store_text: bool = True,
    ) -> None:
        ...

    @abstractmethod
    def get_document(self, doc_id: str, raise_error: bool = True) -> Optional[BaseNode]:
        ...

    @abstractmethod
    async def aget_document(
        self, doc_id: str, raise_error: bool = True
    ) -> Optional[BaseNode]:
        ...

    @abstractmethod
    def delete_document(self, doc_id: str, raise_error: bool = True) -> None:
        """从存储中删除一个文档。"""
        ...

    @abstractmethod
    async def adelete_document(self, doc_id: str, raise_error: bool = True) -> None:
        """从存储中删除一个文档。"""
        ...

    @abstractmethod
    def document_exists(self, doc_id: str) -> bool:
        ...

    @abstractmethod
    async def adocument_exists(self, doc_id: str) -> bool:
        ...

    # ===== Hash =====
    @abstractmethod
    def set_document_hash(self, doc_id: str, doc_hash: str) -> None:
        ...

    @abstractmethod
    async def aset_document_hash(self, doc_id: str, doc_hash: str) -> None:
        ...

    @abstractmethod
    def set_document_hashes(self, doc_hashes: Dict[str, str]) -> None:
        ...

    @abstractmethod
    async def aset_document_hashes(self, doc_hashes: Dict[str, str]) -> None:
        ...

    @abstractmethod
    def get_document_hash(self, doc_id: str) -> Optional[str]:
        ...

    @abstractmethod
    async def aget_document_hash(self, doc_id: str) -> Optional[str]:
        ...

    @abstractmethod
    def get_all_document_hashes(self) -> Dict[str, str]:
        ...

    @abstractmethod
    async def aget_all_document_hashes(self) -> Dict[str, str]:
        ...

    # ==== Ref Docs =====
    @abstractmethod
    def get_all_ref_doc_info(self) -> Optional[Dict[str, RefDocInfo]]:
        """获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。"""

    @abstractmethod
    async def aget_all_ref_doc_info(self) -> Optional[Dict[str, RefDocInfo]]:
        """获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。"""

    @abstractmethod
    def get_ref_doc_info(self, ref_doc_id: str) -> Optional[RefDocInfo]:
        """获取给定ref_doc_id的RefDocInfo。"""

    @abstractmethod
    async def aget_ref_doc_info(self, ref_doc_id: str) -> Optional[RefDocInfo]:
        """获取给定ref_doc_id的RefDocInfo。"""

    @abstractmethod
    def delete_ref_doc(self, ref_doc_id: str, raise_error: bool = True) -> None:
        """删除一个ref_doc及其关联的所有节点。"""

    @abstractmethod
    async def adelete_ref_doc(self, ref_doc_id: str, raise_error: bool = True) -> None:
        """删除一个ref_doc及其关联的所有节点。"""

    # ===== Nodes =====
    def get_nodes(
        self, node_ids: List[str], raise_error: bool = True
    ) -> List[BaseNode]:
        """从文档存储中获取节点。

Args:
    node_ids (List[str]): 节点id
    raise_error (bool): 如果未找到节点id,则抛出错误
"""
        return [self.get_node(node_id, raise_error=raise_error) for node_id in node_ids]

    async def aget_nodes(
        self, node_ids: List[str], raise_error: bool = True
    ) -> List[BaseNode]:
        """从文档存储中获取节点。

Args:
    node_ids (List[str]): 节点id
    raise_error (bool): 如果未找到节点id,则抛出错误
"""
        return [
            await self.aget_node(node_id, raise_error=raise_error)
            for node_id in node_ids
        ]

    def get_node(self, node_id: str, raise_error: bool = True) -> BaseNode:
        """从文档存储中获取节点。

Args:
    node_id(str):节点ID
    raise_error(bool):如果未找到node_id,则引发错误
"""
        doc = self.get_document(node_id, raise_error=raise_error)
        if not isinstance(doc, BaseNode):
            raise ValueError(f"Document {node_id} is not a Node.")
        return doc

    async def aget_node(self, node_id: str, raise_error: bool = True) -> BaseNode:
        """从文档存储中获取节点。

Args:
    node_id(str):节点ID
    raise_error(bool):如果未找到node_id,则引发错误
"""
        doc = await self.aget_document(node_id, raise_error=raise_error)
        if not isinstance(doc, BaseNode):
            raise ValueError(f"Document {node_id} is not a Node.")
        return doc

    def get_node_dict(self, node_id_dict: Dict[int, str]) -> Dict[int, BaseNode]:
        """从文档存储中获取节点字典,给定索引到节点ID的映射。

Args:
    node_id_dict(Dict[int, str]):索引到节点ID的映射
"""
        return {
            index: self.get_node(node_id) for index, node_id in node_id_dict.items()
        }

    async def aget_node_dict(self, node_id_dict: Dict[int, str]) -> Dict[int, BaseNode]:
        """从文档存储中获取节点字典,给定索引到节点ID的映射。

Args:
    node_id_dict(Dict[int, str]):索引到节点ID的映射
"""
        return {
            index: await self.aget_node(node_id)
            for index, node_id in node_id_dict.items()
        }

persist #

persist(
    persist_path: str = DEFAULT_PERSIST_PATH,
    fs: Optional[AbstractFileSystem] = None,
) -> None

将文档存储持久化到文件中。

Source code in llama_index/core/storage/docstore/types.py
26
27
28
29
30
31
def persist(
    self,
    persist_path: str = DEFAULT_PERSIST_PATH,
    fs: Optional[fsspec.AbstractFileSystem] = None,
) -> None:
    """将文档存储持久化到文件中。"""

delete_document abstractmethod #

delete_document(
    doc_id: str, raise_error: bool = True
) -> None

从存储中删除一个文档。

Source code in llama_index/core/storage/docstore/types.py
69
70
71
72
@abstractmethod
def delete_document(self, doc_id: str, raise_error: bool = True) -> None:
    """从存储中删除一个文档。"""
    ...

adelete_document abstractmethod async #

adelete_document(
    doc_id: str, raise_error: bool = True
) -> None

从存储中删除一个文档。

Source code in llama_index/core/storage/docstore/types.py
74
75
76
77
@abstractmethod
async def adelete_document(self, doc_id: str, raise_error: bool = True) -> None:
    """从存储中删除一个文档。"""
    ...

get_all_ref_doc_info abstractmethod #

get_all_ref_doc_info() -> Optional[Dict[str, RefDocInfo]]

获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。

Source code in llama_index/core/storage/docstore/types.py
121
122
123
@abstractmethod
def get_all_ref_doc_info(self) -> Optional[Dict[str, RefDocInfo]]:
    """获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。"""

aget_all_ref_doc_info abstractmethod async #

aget_all_ref_doc_info() -> Optional[Dict[str, RefDocInfo]]

获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。

Source code in llama_index/core/storage/docstore/types.py
125
126
127
@abstractmethod
async def aget_all_ref_doc_info(self) -> Optional[Dict[str, RefDocInfo]]:
    """获取所有已摄取文档的 ref_doc_id -> RefDocInfo 的映射。"""

get_ref_doc_info abstractmethod #

get_ref_doc_info(ref_doc_id: str) -> Optional[RefDocInfo]

获取给定ref_doc_id的RefDocInfo。

Source code in llama_index/core/storage/docstore/types.py
129
130
131
@abstractmethod
def get_ref_doc_info(self, ref_doc_id: str) -> Optional[RefDocInfo]:
    """获取给定ref_doc_id的RefDocInfo。"""

aget_ref_doc_info abstractmethod async #

aget_ref_doc_info(ref_doc_id: str) -> Optional[RefDocInfo]

获取给定ref_doc_id的RefDocInfo。

Source code in llama_index/core/storage/docstore/types.py
133
134
135
@abstractmethod
async def aget_ref_doc_info(self, ref_doc_id: str) -> Optional[RefDocInfo]:
    """获取给定ref_doc_id的RefDocInfo。"""

delete_ref_doc abstractmethod #

delete_ref_doc(
    ref_doc_id: str, raise_error: bool = True
) -> None

删除一个ref_doc及其关联的所有节点。

Source code in llama_index/core/storage/docstore/types.py
137
138
139
@abstractmethod
def delete_ref_doc(self, ref_doc_id: str, raise_error: bool = True) -> None:
    """删除一个ref_doc及其关联的所有节点。"""

adelete_ref_doc abstractmethod async #

adelete_ref_doc(
    ref_doc_id: str, raise_error: bool = True
) -> None

删除一个ref_doc及其关联的所有节点。

Source code in llama_index/core/storage/docstore/types.py
141
142
143
@abstractmethod
async def adelete_ref_doc(self, ref_doc_id: str, raise_error: bool = True) -> None:
    """删除一个ref_doc及其关联的所有节点。"""

get_nodes #

get_nodes(
    node_ids: List[str], raise_error: bool = True
) -> List[BaseNode]

从文档存储中获取节点。

Parameters:

Name Type Description Default
node_ids List[str]

节点id

required
raise_error bool

如果未找到节点id,则抛出错误

True
Source code in llama_index/core/storage/docstore/types.py
146
147
148
149
150
151
152
153
154
155
    def get_nodes(
        self, node_ids: List[str], raise_error: bool = True
    ) -> List[BaseNode]:
        """从文档存储中获取节点。

Args:
    node_ids (List[str]): 节点id
    raise_error (bool): 如果未找到节点id,则抛出错误
"""
        return [self.get_node(node_id, raise_error=raise_error) for node_id in node_ids]

aget_nodes async #

aget_nodes(
    node_ids: List[str], raise_error: bool = True
) -> List[BaseNode]

从文档存储中获取节点。

Parameters:

Name Type Description Default
node_ids List[str]

节点id

required
raise_error bool

如果未找到节点id,则抛出错误

True
Source code in llama_index/core/storage/docstore/types.py
157
158
159
160
161
162
163
164
165
166
167
168
169
    async def aget_nodes(
        self, node_ids: List[str], raise_error: bool = True
    ) -> List[BaseNode]:
        """从文档存储中获取节点。

Args:
    node_ids (List[str]): 节点id
    raise_error (bool): 如果未找到节点id,则抛出错误
"""
        return [
            await self.aget_node(node_id, raise_error=raise_error)
            for node_id in node_ids
        ]

get_node #

get_node(
    node_id: str, raise_error: bool = True
) -> BaseNode

从文档存储中获取节点。

Source code in llama_index/core/storage/docstore/types.py
171
172
173
174
175
176
177
178
179
180
181
    def get_node(self, node_id: str, raise_error: bool = True) -> BaseNode:
        """从文档存储中获取节点。

Args:
    node_id(str):节点ID
    raise_error(bool):如果未找到node_id,则引发错误
"""
        doc = self.get_document(node_id, raise_error=raise_error)
        if not isinstance(doc, BaseNode):
            raise ValueError(f"Document {node_id} is not a Node.")
        return doc

aget_node async #

aget_node(
    node_id: str, raise_error: bool = True
) -> BaseNode

从文档存储中获取节点。

Source code in llama_index/core/storage/docstore/types.py
183
184
185
186
187
188
189
190
191
192
193
    async def aget_node(self, node_id: str, raise_error: bool = True) -> BaseNode:
        """从文档存储中获取节点。

Args:
    node_id(str):节点ID
    raise_error(bool):如果未找到node_id,则引发错误
"""
        doc = await self.aget_document(node_id, raise_error=raise_error)
        if not isinstance(doc, BaseNode):
            raise ValueError(f"Document {node_id} is not a Node.")
        return doc

get_node_dict #

get_node_dict(
    node_id_dict: Dict[int, str]
) -> Dict[int, BaseNode]

从文档存储中获取节点字典,给定索引到节点ID的映射。

Source code in llama_index/core/storage/docstore/types.py
195
196
197
198
199
200
201
202
203
    def get_node_dict(self, node_id_dict: Dict[int, str]) -> Dict[int, BaseNode]:
        """从文档存储中获取节点字典,给定索引到节点ID的映射。

Args:
    node_id_dict(Dict[int, str]):索引到节点ID的映射
"""
        return {
            index: self.get_node(node_id) for index, node_id in node_id_dict.items()
        }

aget_node_dict async #

aget_node_dict(
    node_id_dict: Dict[int, str]
) -> Dict[int, BaseNode]

从文档存储中获取节点字典,给定索引到节点ID的映射。

Source code in llama_index/core/storage/docstore/types.py
205
206
207
208
209
210
211
212
213
214
    async def aget_node_dict(self, node_id_dict: Dict[int, str]) -> Dict[int, BaseNode]:
        """从文档存储中获取节点字典,给定索引到节点ID的映射。

Args:
    node_id_dict(Dict[int, str]):索引到节点ID的映射
"""
        return {
            index: await self.aget_node(node_id)
            for index, node_id in node_id_dict.items()
        }