fabricatio 0.2.3.dev2__cp312-cp312-win_amd64.whl → 0.2.4.dev0__cp312-cp312-win_amd64.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
fabricatio/__init__.py CHANGED
@@ -1,6 +1,9 @@
1
1
  """Fabricatio is a Python library for building llm app using event-based agent structure."""
2
2
 
3
+ from importlib.util import find_spec
4
+
3
5
  from fabricatio._rust_instances import template_manager
6
+ from fabricatio.actions import ExtractArticleEssence
4
7
  from fabricatio.core import env
5
8
  from fabricatio.fs import magika
6
9
  from fabricatio.journal import logger
@@ -18,6 +21,7 @@ __all__ = [
18
21
  "Capture",
19
22
  "CodeBlockCapture",
20
23
  "Event",
24
+ "ExtractArticleEssence",
21
25
  "JsonCapture",
22
26
  "Message",
23
27
  "Messages",
@@ -35,3 +39,9 @@ __all__ = [
35
39
  "task_toolbox",
36
40
  "template_manager",
37
41
  ]
42
+
43
+
44
+ if find_spec("pymilvus"):
45
+ from fabricatio.capabilities.rag import RAG
46
+
47
+ __all__ += ["RAG"]
Binary file
@@ -1,5 +1,5 @@
1
1
  """module for actions."""
2
2
 
3
- from fabricatio.actions.transmission import PublishTask
3
+ from fabricatio.actions.article import ExtractArticleEssence
4
4
 
5
- __all__ = ["PublishTask"]
5
+ __all__ = ["ExtractArticleEssence"]
@@ -0,0 +1,127 @@
1
+ """Actions for transmitting tasks to targets."""
2
+
3
+ from os import PathLike
4
+ from pathlib import Path
5
+ from typing import Callable, List
6
+
7
+ from pydantic import BaseModel, Field
8
+ from pydantic.config import ConfigDict
9
+
10
+ from fabricatio.journal import logger
11
+ from fabricatio.models.action import Action
12
+ from fabricatio.models.generic import ProposedAble
13
+ from fabricatio.models.task import Task
14
+
15
+
16
+ class Equation(BaseModel):
17
+ """Structured representation of mathematical equations (including their physical or conceptual meanings)."""
18
+
19
+ model_config = ConfigDict(use_attribute_docstrings=True)
20
+
21
+ description: str = Field(...)
22
+ """A concise explanation of the equation's meaning, purpose, and relevance in the context of the research."""
23
+
24
+ latex_code: str = Field(...)
25
+ """The LaTeX code used to represent the equation in a publication-ready format."""
26
+
27
+
28
+ class Figure(BaseModel):
29
+ """Structured representation of figures (including their academic significance and explanatory captions)."""
30
+
31
+ model_config = ConfigDict(use_attribute_docstrings=True)
32
+
33
+ description: str = Field(...)
34
+ """A detailed explanation of the figure's content and its role in conveying key insights."""
35
+
36
+ figure_caption: str = Field(...)
37
+ """The caption accompanying the figure, summarizing its main points and academic value."""
38
+
39
+
40
+ class ArticleEssence(ProposedAble):
41
+ """Structured representation of the core elements of an academic paper(providing a comprehensive digital profile of the paper's essential information)."""
42
+
43
+ # Basic Metadata
44
+ title: str = Field(...)
45
+ """The full title of the paper, including any subtitles if applicable."""
46
+
47
+ authors: List[str] = Field(default_factory=list)
48
+ """A list of the paper's authors, typically in the order of contribution."""
49
+
50
+ keywords: List[str] = Field(default_factory=list)
51
+ """A list of keywords that summarize the paper's focus and facilitate indexing."""
52
+
53
+ publication_year: int = Field(None)
54
+ """The year in which the paper was published."""
55
+
56
+ # Core Content Elements
57
+ domain: List[str] = Field(default_factory=list)
58
+ """The research domains or fields addressed by the paper (e.g., ['Natural Language Processing', 'Computer Vision'])."""
59
+
60
+ abstract: str = Field(...)
61
+ """A structured abstract that outlines the research problem, methodology, and conclusions in three distinct sections."""
62
+
63
+ core_contributions: List[str] = Field(default_factory=list)
64
+ """Key academic contributions that distinguish the paper from prior work in the field."""
65
+
66
+ technical_novelty: List[str] = Field(default_factory=list)
67
+ """Specific technical innovations introduced by the research, listed as individual points."""
68
+
69
+ # Academic Achievements Showcase
70
+ highlighted_equations: List[Equation] = Field(default_factory=list)
71
+ """Core mathematical equations that represent breakthroughs in the field, accompanied by explanations of their physical or conceptual significance."""
72
+
73
+ highlighted_algorithms: List[str] = Field(default_factory=list)
74
+ """Pseudocode for key algorithms, annotated to highlight innovative components."""
75
+
76
+ highlighted_figures: List[Figure] = Field(default_factory=list)
77
+ """Critical diagrams or illustrations, each accompanied by a caption explaining their academic importance."""
78
+
79
+ highlighted_tables: List[str] = Field(default_factory=list)
80
+ """Important data tables, annotated to indicate statistical significance or other notable findings."""
81
+
82
+ # Academic Discussion Dimensions
83
+ research_problem: str = Field("")
84
+ """A clearly defined research question or problem addressed by the study."""
85
+
86
+ limitations: List[str] = Field(default_factory=list)
87
+ """An analysis of the methodological or experimental limitations of the research."""
88
+
89
+ future_work: List[str] = Field(default_factory=list)
90
+ """Suggestions for potential directions or topics for follow-up studies."""
91
+
92
+ impact_analysis: str = Field("")
93
+ """An assessment of the paper's potential influence on the development of the field."""
94
+
95
+
96
+ class ExtractArticleEssence(Action):
97
+ """Extract the essence of article(s)."""
98
+
99
+ name: str = "extract article essence"
100
+ """The name of the action."""
101
+ description: str = "Extract the essence of an article. output as json"
102
+ """The description of the action."""
103
+
104
+ output_key: str = "article_essence"
105
+ """The key of the output data."""
106
+
107
+ async def _execute[P: PathLike | str](
108
+ self,
109
+ task_input: Task,
110
+ reader: Callable[[P], str] = lambda p: Path(p).read_text(encoding="utf-8"),
111
+ **_,
112
+ ) -> List[ArticleEssence]:
113
+ if not await self.ajudge(
114
+ f"= Task\n{task_input.briefing}\n\n\n= Role\n{self.briefing}",
115
+ affirm_case="The task does not violate the role, and could be approved since the file dependencies are specified.",
116
+ deny_case="The task does violate the role, and could not be approved.",
117
+ ):
118
+ logger.info(err := "Task not approved.")
119
+ raise RuntimeError(err)
120
+
121
+ # trim the references
122
+ contents = ["References".join(c.split("References")[:-1]) for c in map(reader, task_input.dependencies)]
123
+ return await self.propose(
124
+ ArticleEssence,
125
+ contents,
126
+ system_message=f"# your personal briefing: \n{self.briefing}",
127
+ )
@@ -0,0 +1,55 @@
1
+ """A module for the task capabilities of the Fabricatio library."""
2
+
3
+ from typing import List, Type, Unpack, overload
4
+
5
+ from fabricatio.models.generic import ProposedAble
6
+ from fabricatio.models.kwargs_types import GenerateKwargs
7
+ from fabricatio.models.usages import LLMUsage
8
+
9
+
10
+ class Propose[M: ProposedAble](LLMUsage):
11
+ """A class that proposes an Obj based on a prompt."""
12
+
13
+ @overload
14
+ async def propose(
15
+ self,
16
+ cls: Type[M],
17
+ prompt: List[str],
18
+ **kwargs: Unpack[GenerateKwargs],
19
+ ) -> List[M]: ...
20
+
21
+ @overload
22
+ async def propose(
23
+ self,
24
+ cls: Type[M],
25
+ prompt: str,
26
+ **kwargs: Unpack[GenerateKwargs],
27
+ ) -> M: ...
28
+
29
+ async def propose(
30
+ self,
31
+ cls: Type[M],
32
+ prompt: List[str] | str,
33
+ **kwargs: Unpack[GenerateKwargs],
34
+ ) -> List[M] | M:
35
+ """Asynchronously proposes a task based on a given prompt and parameters.
36
+
37
+ Parameters:
38
+ cls: The class type of the task to be proposed.
39
+ prompt: The prompt text for proposing a task, which is a string that must be provided.
40
+ **kwargs: The keyword arguments for the LLM (Large Language Model) usage.
41
+
42
+ Returns:
43
+ A Task object based on the proposal result.
44
+ """
45
+ if isinstance(prompt, str):
46
+ return await self.aask_validate(
47
+ question=cls.create_json_prompt(prompt),
48
+ validator=cls.instantiate_from_string,
49
+ **kwargs,
50
+ )
51
+ return await self.aask_validate_batch(
52
+ questions=[cls.create_json_prompt(p) for p in prompt],
53
+ validator=cls.instantiate_from_string,
54
+ **kwargs,
55
+ )
@@ -1,89 +1,146 @@
1
1
  """A module for the RAG (Retrieval Augmented Generation) model."""
2
2
 
3
+ try:
4
+ from pymilvus import MilvusClient
5
+ except ImportError as e:
6
+ raise RuntimeError("pymilvus is not installed. Have you installed `fabricatio[rag]` instead of `fabricatio`") from e
7
+ from functools import lru_cache
3
8
  from operator import itemgetter
4
9
  from os import PathLike
5
10
  from pathlib import Path
6
- from typing import Any, Callable, Dict, List, Optional, Self, Union
11
+ from typing import Any, Callable, Dict, List, Optional, Self, Union, Unpack, overload
7
12
 
13
+ from fabricatio._rust_instances import template_manager
8
14
  from fabricatio.config import configs
9
- from fabricatio.models.usages import LLMUsage
15
+ from fabricatio.journal import logger
16
+ from fabricatio.models.kwargs_types import CollectionSimpleConfigKwargs, EmbeddingKwargs, FetchKwargs, LLMKwargs
17
+ from fabricatio.models.usages import EmbeddingUsage
10
18
  from fabricatio.models.utils import MilvusData
11
19
  from more_itertools.recipes import flatten
20
+ from pydantic import Field, PrivateAttr
12
21
 
13
- try:
14
- from pymilvus import MilvusClient
15
- except ImportError as e:
16
- raise RuntimeError("pymilvus is not installed. Have you installed `fabricatio[rag]` instead of `fabricatio`") from e
17
- from pydantic import PrivateAttr
22
+
23
+ @lru_cache(maxsize=None)
24
+ def create_client(uri: str, token: str = "", timeout: Optional[float] = None) -> MilvusClient:
25
+ """Create a Milvus client."""
26
+ return MilvusClient(
27
+ uri=uri,
28
+ token=token,
29
+ timeout=timeout,
30
+ )
18
31
 
19
32
 
20
- class Rag(LLMUsage):
33
+ class RAG(EmbeddingUsage):
21
34
  """A class representing the RAG (Retrieval Augmented Generation) model."""
22
35
 
23
- _client: MilvusClient = PrivateAttr(
24
- default=MilvusClient(
25
- uri=configs.rag.milvus_uri.unicode_string(),
26
- token=configs.rag.milvus_token.get_secret_value(),
27
- timeout=configs.rag.milvus_timeout,
28
- ),
29
- )
30
- _target_collection: Optional[str] = PrivateAttr(default=None)
36
+ target_collection: Optional[str] = Field(default=None)
37
+ """The name of the collection being viewed."""
38
+
39
+ _client: Optional[MilvusClient] = PrivateAttr(None)
40
+ """The Milvus client used for the RAG model."""
31
41
 
32
42
  @property
33
43
  def client(self) -> MilvusClient:
34
- """The Milvus client."""
44
+ """Return the Milvus client."""
45
+ if self._client is None:
46
+ raise RuntimeError("Client is not initialized. Have you called `self.init_client()`?")
35
47
  return self._client
36
48
 
37
- def view(self, collection_name: str, create: bool = False) -> Self:
49
+ def init_client(
50
+ self,
51
+ milvus_uri: Optional[str] = None,
52
+ milvus_token: Optional[str] = None,
53
+ milvus_timeout: Optional[float] = None,
54
+ ) -> Self:
55
+ """Initialize the Milvus client."""
56
+ self._client = create_client(
57
+ uri=milvus_uri or (self.milvus_uri or configs.rag.milvus_uri).unicode_string(),
58
+ token=milvus_token
59
+ or (token.get_secret_value() if (token := (self.milvus_token or configs.rag.milvus_token)) else ""),
60
+ timeout=milvus_timeout or self.milvus_timeout,
61
+ )
62
+ return self
63
+
64
+ @overload
65
+ async def pack(
66
+ self, input_text: List[str], subject: Optional[str] = None, **kwargs: Unpack[EmbeddingKwargs]
67
+ ) -> List[MilvusData]: ...
68
+ @overload
69
+ async def pack(
70
+ self, input_text: str, subject: Optional[str] = None, **kwargs: Unpack[EmbeddingKwargs]
71
+ ) -> MilvusData: ...
72
+
73
+ async def pack(
74
+ self, input_text: List[str] | str, subject: Optional[str] = None, **kwargs: Unpack[EmbeddingKwargs]
75
+ ) -> List[MilvusData] | MilvusData:
76
+ """Asynchronously generates MilvusData objects for the given input text.
77
+
78
+ Args:
79
+ input_text (List[str] | str): A string or list of strings to generate embeddings for.
80
+ subject (Optional[str]): The subject of the input text. Defaults to None.
81
+ **kwargs (Unpack[EmbeddingKwargs]): Additional keyword arguments for embedding.
82
+
83
+ Returns:
84
+ List[MilvusData] | MilvusData: The generated MilvusData objects.
85
+ """
86
+ if isinstance(input_text, str):
87
+ return MilvusData(vector=await self.vectorize(input_text, **kwargs), text=input_text, subject=subject)
88
+ vecs = await self.vectorize(input_text, **kwargs)
89
+ return [
90
+ MilvusData(
91
+ vector=vec,
92
+ text=text,
93
+ subject=subject,
94
+ )
95
+ for text, vec in zip(input_text, vecs, strict=True)
96
+ ]
97
+
98
+ def view(
99
+ self, collection_name: Optional[str], create: bool = False, **kwargs: Unpack[CollectionSimpleConfigKwargs]
100
+ ) -> Self:
38
101
  """View the specified collection.
39
102
 
40
103
  Args:
41
104
  collection_name (str): The name of the collection.
42
105
  create (bool): Whether to create the collection if it does not exist.
106
+ **kwargs (Unpack[CollectionSimpleConfigKwargs]): Additional keyword arguments for collection configuration.
43
107
  """
44
- if create and self._client.has_collection(collection_name):
45
- self._client.create_collection(collection_name)
108
+ if create and collection_name and not self._client.has_collection(collection_name):
109
+ kwargs["dimension"] = kwargs.get("dimension") or self.milvus_dimensions or configs.rag.milvus_dimensions
110
+ self._client.create_collection(collection_name, auto_id=True, **kwargs)
111
+ logger.info(f"Creating collection {collection_name}")
46
112
 
47
- self._target_collection = collection_name
113
+ self.target_collection = collection_name
48
114
  return self
49
115
 
50
- def quit_view(self) -> Self:
116
+ def quit_viewing(self) -> Self:
51
117
  """Quit the current view.
52
118
 
53
119
  Returns:
54
120
  Self: The current instance, allowing for method chaining.
55
121
  """
56
- self._target_collection = None
57
- return self
58
-
59
- @property
60
- def viewing_collection(self) -> Optional[str]:
61
- """Get the name of the collection being viewed.
62
-
63
- Returns:
64
- Optional[str]: The name of the collection being viewed.
65
- """
66
- return self._target_collection
122
+ return self.view(None)
67
123
 
68
124
  @property
69
- def safe_viewing_collection(self) -> str:
125
+ def safe_target_collection(self) -> str:
70
126
  """Get the name of the collection being viewed, raise an error if not viewing any collection.
71
127
 
72
128
  Returns:
73
129
  str: The name of the collection being viewed.
74
130
  """
75
- if self._target_collection is None:
131
+ if self.target_collection is None:
76
132
  raise RuntimeError("No collection is being viewed. Have you called `self.view()`?")
77
- return self._target_collection
133
+ return self.target_collection
78
134
 
79
135
  def add_document[D: Union[Dict[str, Any], MilvusData]](
80
- self, data: D | List[D], collection_name: Optional[str] = None
136
+ self, data: D | List[D], collection_name: Optional[str] = None, flush: bool = False
81
137
  ) -> Self:
82
138
  """Adds a document to the specified collection.
83
139
 
84
140
  Args:
85
141
  data (Union[Dict[str, Any], MilvusData] | List[Union[Dict[str, Any], MilvusData]]): The data to be added to the collection.
86
142
  collection_name (Optional[str]): The name of the collection. If not provided, the currently viewed collection is used.
143
+ flush (bool): Whether to flush the collection after insertion.
87
144
 
88
145
  Returns:
89
146
  Self: The current instance, allowing for method chaining.
@@ -92,11 +149,19 @@ class Rag(LLMUsage):
92
149
  data = data.prepare_insertion()
93
150
  if isinstance(data, list):
94
151
  data = [d.prepare_insertion() if isinstance(d, MilvusData) else d for d in data]
95
- self._client.insert(collection_name or self.safe_viewing_collection, data)
152
+ c_name = collection_name or self.safe_target_collection
153
+ self._client.insert(c_name, data)
154
+
155
+ if flush:
156
+ logger.debug(f"Flushing collection {c_name}")
157
+ self._client.flush(c_name)
96
158
  return self
97
159
 
98
- def consume(
99
- self, source: PathLike, reader: Callable[[PathLike], MilvusData], collection_name: Optional[str] = None
160
+ async def consume_file(
161
+ self,
162
+ source: List[PathLike] | PathLike,
163
+ reader: Callable[[PathLike], str] = lambda path: Path(path).read_text(encoding="utf-8"),
164
+ collection_name: Optional[str] = None,
100
165
  ) -> Self:
101
166
  """Consume a file and add its content to the collection.
102
167
 
@@ -108,8 +173,21 @@ class Rag(LLMUsage):
108
173
  Returns:
109
174
  Self: The current instance, allowing for method chaining.
110
175
  """
111
- data = reader(Path(source))
112
- self.add_document(data, collection_name or self.safe_viewing_collection)
176
+ if not isinstance(source, list):
177
+ source = [source]
178
+ return await self.consume_string([reader(s) for s in source], collection_name)
179
+
180
+ async def consume_string(self, text: List[str] | str, collection_name: Optional[str] = None) -> Self:
181
+ """Consume a string and add it to the collection.
182
+
183
+ Args:
184
+ text (List[str] | str): The text to be added to the collection.
185
+ collection_name (Optional[str]): The name of the collection. If not provided, the currently viewed collection is used.
186
+
187
+ Returns:
188
+ Self: The current instance, allowing for method chaining.
189
+ """
190
+ self.add_document(await self.pack(text), collection_name or self.safe_target_collection, flush=True)
113
191
  return self
114
192
 
115
193
  async def afetch_document(
@@ -117,6 +195,7 @@ class Rag(LLMUsage):
117
195
  vecs: List[List[float]],
118
196
  desired_fields: List[str] | str,
119
197
  collection_name: Optional[str] = None,
198
+ similarity_threshold: float = 0.37,
120
199
  result_per_query: int = 10,
121
200
  ) -> List[Dict[str, Any]] | List[Any]:
122
201
  """Fetch data from the collection.
@@ -125,6 +204,7 @@ class Rag(LLMUsage):
125
204
  vecs (List[List[float]]): The vectors to search for.
126
205
  desired_fields (List[str] | str): The fields to retrieve.
127
206
  collection_name (Optional[str]): The name of the collection. If not provided, the currently viewed collection is used.
207
+ similarity_threshold (float): The threshold for similarity, only results above this threshold will be returned.
128
208
  result_per_query (int): The number of results to return per query.
129
209
 
130
210
  Returns:
@@ -132,8 +212,9 @@ class Rag(LLMUsage):
132
212
  """
133
213
  # Step 1: Search for vectors
134
214
  search_results = self._client.search(
135
- collection_name or self.safe_viewing_collection,
215
+ collection_name or self.safe_target_collection,
136
216
  vecs,
217
+ search_params={"radius": similarity_threshold},
137
218
  output_fields=desired_fields if isinstance(desired_fields, list) else [desired_fields],
138
219
  limit=result_per_query,
139
220
  )
@@ -144,6 +225,7 @@ class Rag(LLMUsage):
144
225
  # Step 3: Sort by distance (descending)
145
226
  sorted_results = sorted(flattened_results, key=itemgetter("distance"), reverse=True)
146
227
 
228
+ logger.debug(f"Searched similarities: {[t['distance'] for t in sorted_results]}")
147
229
  # Step 4: Extract the entities
148
230
  resp = [result["entity"] for result in sorted_results]
149
231
 
@@ -155,25 +237,74 @@ class Rag(LLMUsage):
155
237
  self,
156
238
  query: List[str] | str,
157
239
  collection_name: Optional[str] = None,
158
- result_per_query: int = 10,
159
240
  final_limit: int = 20,
241
+ **kwargs: Unpack[FetchKwargs],
160
242
  ) -> List[str]:
161
243
  """Retrieve data from the collection.
162
244
 
163
245
  Args:
164
246
  query (List[str] | str): The query to be used for retrieval.
165
247
  collection_name (Optional[str]): The name of the collection. If not provided, the currently viewed collection is used.
166
- result_per_query (int): The number of results to be returned per query.
167
248
  final_limit (int): The final limit on the number of results to return.
249
+ **kwargs (Unpack[FetchKwargs]): Additional keyword arguments for retrieval.
168
250
 
169
251
  Returns:
170
252
  List[str]: A list of strings containing the retrieved data.
171
253
  """
172
254
  if isinstance(query, str):
173
255
  query = [query]
174
- return await self.afetch_document(
175
- vecs=(await self.vectorize(query)),
176
- desired_fields="text",
177
- collection_name=collection_name,
178
- result_per_query=result_per_query,
256
+ return (
257
+ await self.afetch_document(
258
+ vecs=(await self.vectorize(query)),
259
+ desired_fields="text",
260
+ collection_name=collection_name,
261
+ **kwargs,
262
+ )
179
263
  )[:final_limit]
264
+
265
+ async def aask_retrieved(
266
+ self,
267
+ question: str | List[str],
268
+ query: List[str] | str,
269
+ collection_name: Optional[str] = None,
270
+ extra_system_message: str = "",
271
+ result_per_query: int = 10,
272
+ final_limit: int = 20,
273
+ similarity_threshold: float = 0.37,
274
+ **kwargs: Unpack[LLMKwargs],
275
+ ) -> str:
276
+ """Asks a question by retrieving relevant documents based on the provided query.
277
+
278
+ This method performs document retrieval using the given query, then asks the
279
+ specified question using the retrieved documents as context.
280
+
281
+ Args:
282
+ question (str | List[str]): The question or list of questions to be asked.
283
+ query (List[str] | str): The query or list of queries used for document retrieval.
284
+ collection_name (Optional[str]): The name of the collection to retrieve documents from.
285
+ If not provided, the currently viewed collection is used.
286
+ extra_system_message (str): An additional system message to be included in the prompt.
287
+ result_per_query (int): The number of results to return per query. Default is 10.
288
+ final_limit (int): The maximum number of retrieved documents to consider. Default is 20.
289
+ similarity_threshold (float): The threshold for similarity, only results above this threshold will be returned.
290
+ **kwargs (Unpack[LLMKwargs]): Additional keyword arguments passed to the underlying `aask` method.
291
+
292
+ Returns:
293
+ str: A string response generated after asking with the context of retrieved documents.
294
+ """
295
+ docs = await self.aretrieve(
296
+ query,
297
+ collection_name,
298
+ final_limit,
299
+ result_per_query=result_per_query,
300
+ similarity_threshold=similarity_threshold,
301
+ )
302
+
303
+ rendered = template_manager.render_template(configs.templates.retrieved_display_template, {"docs": docs[::-1]})
304
+
305
+ logger.debug(f"Retrieved Documents: \n{rendered}")
306
+ return await self.aask(
307
+ question,
308
+ f"{rendered}\n\n{extra_system_message}",
309
+ **kwargs,
310
+ )
@@ -5,21 +5,21 @@ from typing import Any, Dict, List, Optional, Tuple, Unpack
5
5
 
6
6
  import orjson
7
7
  from fabricatio._rust_instances import template_manager
8
+ from fabricatio.capabilities.propose import Propose
8
9
  from fabricatio.config import configs
9
10
  from fabricatio.models.generic import WithBriefing
10
11
  from fabricatio.models.kwargs_types import ChooseKwargs, ValidateKwargs
11
12
  from fabricatio.models.task import Task
12
13
  from fabricatio.models.tool import Tool, ToolExecutor
13
- from fabricatio.models.usages import LLMUsage, ToolBoxUsage
14
+ from fabricatio.models.usages import ToolBoxUsage
14
15
  from fabricatio.parser import JsonCapture, PythonCapture
15
16
  from loguru import logger
16
- from pydantic import ValidationError
17
17
 
18
18
 
19
- class ProposeTask(WithBriefing, LLMUsage):
19
+ class ProposeTask(WithBriefing, Propose):
20
20
  """A class that proposes a task based on a prompt."""
21
21
 
22
- async def propose[T](
22
+ async def propose_task[T](
23
23
  self,
24
24
  prompt: str,
25
25
  **kwargs: Unpack[ValidateKwargs],
@@ -34,27 +34,10 @@ class ProposeTask(WithBriefing, LLMUsage):
34
34
  A Task object based on the proposal result.
35
35
  """
36
36
  if not prompt:
37
- err = f"{self.name}: Prompt must be provided."
38
- logger.error(err)
37
+ logger.error(err := f"{self.name}: Prompt must be provided.")
39
38
  raise ValueError(err)
40
39
 
41
- def _validate_json(response: str) -> None | Task:
42
- try:
43
- cap = JsonCapture.capture(response)
44
- logger.debug(f"Response: \n{response}")
45
- logger.info(f"Captured JSON: \n{cap}")
46
- return Task.model_validate_json(cap)
47
- except ValidationError as e:
48
- logger.error(f"Failed to parse task from JSON: {e}")
49
- return None
50
-
51
- template_data = {"prompt": prompt, "json_example": Task.json_example()}
52
- return await self.aask_validate(
53
- question=template_manager.render_template(configs.templates.propose_task_template, template_data),
54
- validator=_validate_json,
55
- system_message=f"# your personal briefing: \n{self.briefing}",
56
- **kwargs,
57
- )
40
+ return await self.propose(Task, prompt, system_message=f"# your personal briefing: \n{self.briefing}", **kwargs)
58
41
 
59
42
 
60
43
  class HandleTask(WithBriefing, ToolBoxUsage):