-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathschema.py
510 lines (390 loc) · 14 KB
/
schema.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
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
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
"""Common schema objects."""
from __future__ import annotations
from abc import ABC, abstractmethod
from typing import (
Any,
Dict,
List,
Optional,
Sequence,
TypeVar,
Union,
)
import uuid
from dotagent.load.serializable import Serializable
from pydantic import BaseModel, Extra, Field, root_validator
from enum import auto, Enum
from hashlib import sha256
DEFAULT_TEXT_NODE_TMPL = "{metadata_str}\n\n{content}"
DEFAULT_METADATA_TMPL = "{key}: {value}"
RUN_KEY = "__run"
class BaseMessage(Serializable):
"""Message object."""
content: str
additional_kwargs: dict = Field(default_factory=dict)
@property
@abstractmethod
def type(self) -> str:
"""Type of the message, used for serialization."""
class BaseMemory(Serializable, ABC):
"""Base interface for memory in chains."""
class Config:
"""Configuration for this pydantic object."""
extra = Extra.forbid
arbitrary_types_allowed = True
@property
@abstractmethod
def memory_variables(self) -> List[str]:
"""Input keys this memory class will load dynamically."""
@abstractmethod
def load_memory_variables(self, inputs: Dict[str, Any]) -> Dict[str, Any]:
"""Return key-value pairs given the text input to the chain.
If None, return all memories
"""
@abstractmethod
def save_context(self, inputs: Dict[str, Any], outputs: Dict[str, str]) -> None:
"""Save the context of this model run to memory."""
@abstractmethod
def clear(self) -> None:
"""Clear memory contents."""
class Document(Serializable):
"""Interface for interacting with a document."""
page_content: str
metadata: dict = Field(default_factory=dict)
class ObjectType(str, Enum):
TEXT = auto()
IMAGE = auto()
INDEX = auto()
DOCUMENT = auto()
class MetadataMode(str, Enum):
ALL = auto()
EMBED = auto()
LLM = auto()
NONE = auto()
class NodeRelationship(str, Enum):
"""Node relationships used in `BaseNode` class.
Attributes:
SOURCE: The node is the source document.
PREVIOUS: The node is the previous node in the document.
NEXT: The node is the next node in the document.
PARENT: The node is the parent node in the document.
CHILD: The node is a child node in the document.
"""
SOURCE = auto()
PREVIOUS = auto()
NEXT = auto()
PARENT = auto()
CHILD = auto()
class RelatedNodeInfo(BaseModel):
node_id: str
node_type: Optional[ObjectType] = None
metadata: Dict[str, Any] = Field(default_factory=dict)
hash: Optional[str] = None
RelatedNodeType = Union[RelatedNodeInfo, List[RelatedNodeInfo]]
class BaseNode(BaseModel):
"""Base node Object.
Generic abstract interface for retrievable nodes
"""
class Config:
allow_population_by_field_name = True
id_: str = Field(
default_factory=lambda: str(uuid.uuid4()), description="Unique ID of the node."
)
embedding: Optional[List[float]] = Field(
default=None, description="Embedding of the node."
)
""""
metadata fields
- injected as part of the text shown to LLMs as context
- injected as part of the text for generating embeddings
- used by vector DBs for metadata filtering
"""
metadata: Dict[str, Any] = Field(
default_factory=dict,
description="A flat dictionary of metadata fields",
alias="extra_info",
)
excluded_embed_metadata_keys: List[str] = Field(
default_factory=list,
description="Metadata keys that are exluded from text for the embed model.",
)
excluded_llm_metadata_keys: List[str] = Field(
default_factory=list,
description="Metadata keys that are exluded from text for the LLM.",
)
relationships: Dict[NodeRelationship, RelatedNodeType] = Field(
default_factory=dict,
description="A mapping of relationships to other node information.",
)
hash: str = Field(default="", description="Hash of the node content.")
@classmethod
@abstractmethod
def get_type(cls) -> str:
"""Get Object type."""
@abstractmethod
def get_content(self, metadata_mode: MetadataMode = MetadataMode.ALL) -> str:
"""Get object content."""
@abstractmethod
def get_metadata_str(self, mode: MetadataMode = MetadataMode.ALL) -> str:
"""Metadata string."""
@abstractmethod
def set_content(self, value: Any) -> None:
"""Set the content of the node."""
@property
def node_id(self) -> str:
return self.id_
@property
def source_node(self) -> Optional[RelatedNodeInfo]:
"""Source object node.
Extracted from the relationships field.
"""
if NodeRelationship.SOURCE not in self.relationships:
return None
relation = self.relationships[NodeRelationship.SOURCE]
if isinstance(relation, list):
raise ValueError("Source object must be a single RelatedNodeInfo object")
return relation
@property
def prev_node(self) -> Optional[RelatedNodeInfo]:
"""Prev node."""
if NodeRelationship.PREVIOUS not in self.relationships:
return None
relation = self.relationships[NodeRelationship.PREVIOUS]
if not isinstance(relation, RelatedNodeInfo):
raise ValueError("Previous object must be a single RelatedNodeInfo object")
return relation
@property
def next_node(self) -> Optional[RelatedNodeInfo]:
"""Next node."""
if NodeRelationship.NEXT not in self.relationships:
return None
relation = self.relationships[NodeRelationship.NEXT]
if not isinstance(relation, RelatedNodeInfo):
raise ValueError("Next object must be a single RelatedNodeInfo object")
return relation
@property
def parent_node(self) -> Optional[RelatedNodeInfo]:
"""Parent node."""
if NodeRelationship.PARENT not in self.relationships:
return None
relation = self.relationships[NodeRelationship.PARENT]
if not isinstance(relation, RelatedNodeInfo):
raise ValueError("Parent object must be a single RelatedNodeInfo object")
return relation
@property
def child_nodes(self) -> Optional[List[RelatedNodeInfo]]:
"""Child nodes."""
if NodeRelationship.CHILD not in self.relationships:
return None
relation = self.relationships[NodeRelationship.PARENT]
if not isinstance(relation, list):
raise ValueError("Child objects must be a list of RelatedNodeInfo objects.")
return relation
@property
def ref_doc_id(self) -> Optional[str]:
"""Deprecated: Get ref doc id."""
source_node = self.source_node
if source_node is None:
return None
return source_node.node_id
@property
def extra_info(self) -> Dict[str, Any]:
"""TODO: DEPRECATED: Extra info."""
return self.metadata
def get_embedding(self) -> List[float]:
"""Get embedding.
Errors if embedding is None.
"""
if self.embedding is None:
raise ValueError("embedding not set.")
return self.embedding
def as_related_node_info(self) -> RelatedNodeInfo:
"""Get node as RelatedNodeInfo."""
return RelatedNodeInfo(
node_id=self.node_id, metadata=self.metadata, hash=self.hash
)
class TextNode(BaseNode):
text: str = Field(default="", description="Text content of the node.")
start_char_idx: Optional[int] = Field(
default=None, description="Start char index of the node."
)
end_char_idx: Optional[int] = Field(
default=None, description="End char index of the node."
)
text_template: str = Field(
default=DEFAULT_TEXT_NODE_TMPL,
description=(
"Template for how text is formatted, with {content} and "
"{metadata_str} placeholders."
),
)
metadata_template: str = Field(
default=DEFAULT_METADATA_TMPL,
description=(
"Template for how metadata is formatted, with {key} and "
"{value} placeholders."
),
)
metadata_seperator: str = Field(
default="\n",
description="Seperator between metadata fields when converting to string.",
)
@root_validator
def _check_hash(cls, values: dict) -> dict:
"""Generate a hash to represent the node."""
text = values.get("text", "")
metadata = values.get("metadata", {})
doc_identity = str(text) + str(metadata)
values["hash"] = str(
sha256(doc_identity.encode("utf-8", "surrogatepass")).hexdigest()
)
return values
@classmethod
def get_type(cls) -> str:
"""Get Object type."""
return ObjectType.TEXT
def get_content(self, metadata_mode: MetadataMode = MetadataMode.NONE) -> str:
"""Get object content."""
metadata_str = self.get_metadata_str(mode=metadata_mode).strip()
return self.text_template.format(
content=self.text, metadata_str=metadata_str
).strip()
def get_metadata_str(self, mode: MetadataMode = MetadataMode.ALL) -> str:
"""metadata info string."""
if mode == MetadataMode.NONE:
return ""
usable_metadata_keys = set(self.metadata.keys())
if mode == MetadataMode.LLM:
for key in self.excluded_llm_metadata_keys:
if key in usable_metadata_keys:
usable_metadata_keys.remove(key)
elif mode == MetadataMode.EMBED:
for key in self.excluded_embed_metadata_keys:
if key in usable_metadata_keys:
usable_metadata_keys.remove(key)
return self.metadata_seperator.join(
[
self.metadata_template.format(key=key, value=str(value))
for key, value in self.metadata.items()
if key in usable_metadata_keys
]
)
def set_content(self, value: str) -> None:
"""Set the content of the node."""
self.text = value
def get_node_info(self) -> Dict[str, Any]:
"""Get node info."""
return {"start": self.start_char_idx, "end": self.end_char_idx}
def get_text(self) -> str:
return self.get_content(metadata_mode=MetadataMode.NONE)
@property
def node_info(self) -> Dict[str, Any]:
"""Deprecated: Get node info."""
return self.get_node_info()
Node = TextNode
class DocumentNode(TextNode):
"""Generic interface for a data document.
This document connects to data sources.
"""
# TODO: A lot of backwards compatibility logic here, clean up
id_: str = Field(
default_factory=lambda: str(uuid.uuid4()),
description="Unique ID of the node.",
alias="doc_id",
)
_compat_fields = {"doc_id": "id_", "extra_info": "metadata"}
@classmethod
def get_type(cls) -> str:
"""Get Document type."""
return ObjectType.DOCUMENT
@property
def doc_id(self) -> str:
"""Get document ID."""
return self.id_
def get_doc_id(self) -> str:
"""TODO: Deprecated: Get document ID."""
return self.id_
def __setattr__(self, name: str, value: object) -> None:
if name in self._compat_fields:
name = self._compat_fields[name]
super().__setattr__(name, value)
def to_langchain_format(self) -> Document:
"""Convert struct to LangChain document format."""
metadata = self.metadata or {}
return Document(page_content=self.text, metadata=metadata)
@classmethod
def from_langchain_format(cls, doc: Document) -> "Document":
"""Convert struct from LangChain document format."""
return cls(text=doc.page_content, metadata=doc.metadata)
@classmethod
def example(cls) -> "Document":
document = Document(
text="SAMPLE_TEXT",
metadata={"filename": "README.md", "category": "codebase"},
)
return document
class BaseRetriever(ABC):
@abstractmethod
def get_relevant_documents(self, query: str) -> List[Document]:
"""Get documents relevant for a query.
Args:
query: string to find relevant documents for
Returns:
List of relevant documents
"""
@abstractmethod
async def aget_relevant_documents(self, query: str) -> List[Document]:
"""Get documents relevant for a query.
Args:
query: string to find relevant documents for
Returns:
List of relevant documents
"""
# For backwards compatibility
Memory = BaseMemory
T = TypeVar("T")
class BaseDocumentTransformer(ABC):
"""Base interface for transforming documents."""
@abstractmethod
def transform_documents(
self, documents: Sequence[Document], **kwargs: Any
) -> Sequence[Document]:
"""Transform a list of documents."""
@abstractmethod
async def atransform_documents(
self, documents: Sequence[Document], **kwargs: Any
) -> Sequence[Document]:
"""Asynchronously transform a list of documents."""
class AgentBoxStatus(BaseModel):
"""
Represents the status of a AgentBox instance.
"""
status: str
def __str__(self):
return self.status
def __repr__(self):
return f"Status({self.status})"
def __eq__(self, other):
return self.__str__() == other.__str__()
class AgentBoxOutput(BaseModel):
"""
Represents the code execution output of a AgentBox instance.
"""
type: str
content: str
def __str__(self):
return self.content
def __repr__(self):
return f"{self.type}({self.content})"
def __eq__(self, other):
return self.__str__() == other.__str__()
class AgentBoxFile(BaseModel):
"""
Represents a file returned from a AgentBox instance.
"""
name: str
content: Optional[bytes] = None
def __str__(self):
return self.name
def __repr__(self):
return f"File({self.name})"