fabricatio 0.2.4.dev1__cp312-cp312-win_amd64.whl → 0.2.4.dev3__cp312-cp312-win_amd64.whl

fabricatio/__init__.py

@@ -3,7 +3,7 @@
 from importlib.util import find_spec
 
 from fabricatio._rust_instances import template_manager
-from fabricatio.actions import ExtractArticleEssence
+from fabricatio.actions.article import ExtractArticleEssence
 from fabricatio.core import env
 from fabricatio.fs import magika
 from fabricatio.journal import logger
@@ -44,6 +44,8 @@ __all__ = [
 
 
 if find_spec("pymilvus"):
+    from fabricatio.actions.rag import InjectToDB
     from fabricatio.capabilities.rag import RAG
+    from fabricatio.workflows.articles import StoreArticle
 
-    __all__ += ["RAG"]
+    __all__ += ["RAG", "InjectToDB", "StoreArticle"]

fabricatio/actions/article.py

@@ -11,11 +11,16 @@ from fabricatio.models.task import Task
 
 
 class ExtractArticleEssence(Action):
-    """Extract the essence of article(s)."""
+    """Extract the essence of article(s) in text format from the paths specified in the task dependencies.
+
+    Notes:
+        This action is designed to extract vital information from articles with Markdown format, which is pure text, and
+        which is converted from pdf files using `magic-pdf` from the `MinerU` project, see https://github.com/opendatalab/MinerU
+    """
 
     name: str = "extract article essence"
     """The name of the action."""
-    description: str = "Extract the essence of an article. output as json"
+    description: str = "Extract the essence of article(s) from the paths specified in the task dependencies."
     """The description of the action."""
 
     output_key: str = "article_essence"

fabricatio/actions/rag.py

@@ -0,0 +1,25 @@
+"""Inject data into the database."""
+
+from typing import List, Optional, Unpack
+
+from fabricatio.capabilities.rag import RAG
+from fabricatio.models.action import Action
+from fabricatio.models.generic import PrepareVectorization
+
+
+class InjectToDB(Action, RAG):
+    """Inject data into the database."""
+
+    output_key: str = "collection_name"
+
+    async def _execute[T: PrepareVectorization](
+        self, to_inject: T | List[T], collection_name: Optional[str] = "my_collection", **cxt: Unpack
+    ) -> str:
+        if not isinstance(to_inject, list):
+            to_inject = [to_inject]
+
+        await self.view(collection_name, create=True).consume_string(
+            [t.prepare_vectorization(self.embedding_max_sequence_length) for t in to_inject],
+        )
+
+        return collection_name

fabricatio/capabilities/rag.py

@@ -13,10 +13,16 @@ from typing import Any, Callable, Dict, List, Optional, Self, Union, Unpack, ove
 from fabricatio._rust_instances import template_manager
 from fabricatio.config import configs
 from fabricatio.journal import logger
-from fabricatio.models.kwargs_types import CollectionSimpleConfigKwargs, EmbeddingKwargs, FetchKwargs, LLMKwargs
+from fabricatio.models.kwargs_types import (
+    ChooseKwargs,
+    CollectionSimpleConfigKwargs,
+    EmbeddingKwargs,
+    FetchKwargs,
+    LLMKwargs,
+)
 from fabricatio.models.usages import EmbeddingUsage
 from fabricatio.models.utils import MilvusData
-from more_itertools.recipes import flatten
+from more_itertools.recipes import flatten, unique
 from pydantic import Field, PrivateAttr
 
 
@@ -221,9 +227,9 @@ class RAG(EmbeddingUsage):
 
         # Step 2: Flatten the search results
         flattened_results = flatten(search_results)
-
+        unique_results = unique(flattened_results, key=itemgetter("id"))
         # Step 3: Sort by distance (descending)
-        sorted_results = sorted(flattened_results, key=itemgetter("distance"), reverse=True)
+        sorted_results = sorted(unique_results, key=itemgetter("distance"), reverse=True)
 
         logger.debug(f"Searched similarities: {[t['distance'] for t in sorted_results]}")
         # Step 4: Extract the entities
@@ -308,3 +314,21 @@ class RAG(EmbeddingUsage):
             f"{rendered}\n\n{extra_system_message}",
             **kwargs,
         )
+
+    async def arefined_query(self, question: List[str] | str, **kwargs: Unpack[ChooseKwargs]) -> List[str]:
+        """Refines the given question using a template.
+
+        Args:
+            question (List[str] | str): The question to be refined.
+            **kwargs (Unpack[ChooseKwargs]): Additional keyword arguments for the refinement process.
+
+        Returns:
+            List[str]: A list of refined questions.
+        """
+        return await self.aliststr(
+            template_manager.render_template(
+                configs.templates.refined_query_template,
+                {"question": [question] if isinstance(question, str) else question},
+            ),
+            **kwargs,
+        )

fabricatio/config.py

@@ -209,6 +209,9 @@ class TemplateConfig(BaseModel):
     liststr_template: str = Field(default="liststr")
     """The name of the liststr template which will be used to display a list of strings."""
 
+    refined_query_template: str = Field(default="refined_query")
+    """The name of the refined query template which will be used to refine a query."""
+
 
 class MagikaConfig(BaseModel):
     """Magika configuration class."""

fabricatio/fs/__init__.py

@@ -1,5 +1,17 @@
 """FileSystem manipulation module for Fabricatio."""
 
-from fabricatio.fs.readers import magika
+from fabricatio.fs.curd import copy_file, create_directory, delete_directory, delete_file, dump_text, move_file, tree
+from fabricatio.fs.readers import magika, safe_json_read, safe_text_read
 
-__all__ = ["magika"]
+__all__ = [
+    "copy_file",
+    "create_directory",
+    "delete_directory",
+    "delete_file",
+    "dump_text",
+    "magika",
+    "move_file",
+    "safe_json_read",
+    "safe_text_read",
+    "tree",
+]

fabricatio/models/action.py

@@ -3,7 +3,7 @@
 import traceback
 from abc import abstractmethod
 from asyncio import Queue, create_task
-from typing import Any, Dict, Self, Tuple, Type, Union, Unpack
+from typing import Any, Dict, Self, Tuple, Type, Union, Unpack, final
 
 from fabricatio.capabilities.rating import GiveRating
 from fabricatio.capabilities.task import HandleTask, ProposeTask
@@ -17,11 +17,26 @@ from pydantic import Field, PrivateAttr
 class Action(HandleTask, ProposeTask, GiveRating):
     """Class that represents an action to be executed in a workflow."""
 
+    name: str = Field(default="")
+    """The name of the action."""
+    description: str = Field(default="")
+    """The description of the action."""
     personality: str = Field(default="")
     """The personality of whom the action belongs to."""
     output_key: str = Field(default="")
     """The key of the output data."""
 
+    @final
+    def model_post_init(self, __context: Any) -> None:
+        """Initialize the action by setting the name if not provided.
+
+        Args:
+            __context: The context to be used for initialization.
+        """
+        self.name = self.name or self.__class__.__name__
+
+        self.description = self.description or self.__class__.__doc__ or ""
+
     @abstractmethod
     async def _execute(self, **cxt: Unpack) -> Any:
         """Execute the action with the provided arguments.
@@ -34,6 +49,7 @@ class Action(HandleTask, ProposeTask, GiveRating):
         """
         pass
 
+    @final
     async def act(self, cxt: Dict[str, Any]) -> Dict[str, Any]:
         """Perform the action by executing it and setting the output data.
 

fabricatio/workflows/articles.py

@@ -0,0 +1,11 @@
+"""Store article essence in the database."""
+
+from fabricatio.actions.article import ExtractArticleEssence
+from fabricatio.actions.rag import InjectToDB
+from fabricatio.models.action import WorkFlow
+
+StoreArticle = WorkFlow(
+    name="Extract Article Essence",
+    description="Extract the essence of an article in the given path, and store it in the database.",
+    steps=(ExtractArticleEssence(output_key="to_inject"), InjectToDB(output_key="task_output")),
+)

{fabricatio-0.2.4.dev1.dist-info → fabricatio-0.2.4.dev3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fabricatio
-Version: 0.2.4.dev1
+Version: 0.2.4.dev3
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Rust
 Classifier: Programming Language :: Python :: 3.12

{fabricatio-0.2.4.dev1.dist-info → fabricatio-0.2.4.dev3.dist-info}/RECORD

@@ -1,20 +1,20 @@
-fabricatio-0.2.4.dev1.dist-info/METADATA,sha256=_gk2kBD669jBzVql6w4SSnaMHDkJlQ393i-7QyJBock,8861
-fabricatio-0.2.4.dev1.dist-info/WHEEL,sha256=tpW5AN9B-9qsM9WW2FXG2r193YXiqexDadpKp0A2daI,96
-fabricatio-0.2.4.dev1.dist-info/licenses/LICENSE,sha256=do7J7EiCGbq0QPbMAL_FqLYufXpHnCnXBOuqVPwSV8Y,1088
-fabricatio/actions/article.py,sha256=8T_ZqNs8OPxL2nTQ5D0YsOMQ7R5_TBf7oM5dbPmMbe8,1654
-fabricatio/actions/__init__.py,sha256=CjlbRaRqqKWPxqZEtk7tQt3Mul17CY9O4alo6VjWGNk,130
+fabricatio-0.2.4.dev3.dist-info/METADATA,sha256=ylm4KBEqXPqU0YJT7HpEaLhwMuWEPU3YaDcTMsleHEE,8861
+fabricatio-0.2.4.dev3.dist-info/WHEEL,sha256=tpW5AN9B-9qsM9WW2FXG2r193YXiqexDadpKp0A2daI,96
+fabricatio-0.2.4.dev3.dist-info/licenses/LICENSE,sha256=do7J7EiCGbq0QPbMAL_FqLYufXpHnCnXBOuqVPwSV8Y,1088
+fabricatio/actions/article.py,sha256=xrY04lIbq2ol_eW0kw0l9GmZsOXSlTCpjuR38t_bzEE,2026
+fabricatio/actions/rag.py,sha256=lZfw9SZ8oxbWPK_bvWsEpVkWJbGP8HUnlNbxLh11Wdg,821
 fabricatio/capabilities/propose.py,sha256=nahXjB6_nP0Fru880oh_9oINrjrL0Qs-SLHA-d3CFUE,1769
-fabricatio/capabilities/rag.py,sha256=paq2zUOfw6whIBFkKDo1Kg5Ft5YXgWiJBNKq-6uGhuU,13295
+fabricatio/capabilities/rag.py,sha256=UWRUQtlhGUT9Zz3c5iErWjpiTr4LWUV184uMSMhrC_g,14139
 fabricatio/capabilities/rating.py,sha256=PcUpKxPfVO-vitgA6py1xg9iLJZdf7Fru--18ZUFKKA,14026
 fabricatio/capabilities/task.py,sha256=s6FiC9Wg_l-qSa2LgsoKV9f6wXZN6Q_FlWn3XbSnrys,4618
-fabricatio/config.py,sha256=yPJjrGhvoWyXfLBTJp2pMvHt9F-0uskE-va7QDo0rxI,13525
+fabricatio/config.py,sha256=dbVwH1ZEkBSt1wiO7q4gl6U5_72y6AKkcwswPYXEQCc,13681
 fabricatio/core.py,sha256=VQ_JKgUGIy2gZ8xsTBZCdr_IP7wC5aPg0_bsOmjQ588,6458
 fabricatio/decorators.py,sha256=uzsP4tFKQNjDHBkofsjjoJA0IUAaYOtt6YVedoyOqlo,6551
 fabricatio/fs/curd.py,sha256=faMstgGUiQ4k2AW3OXfvvWWTldTtKXco7QINYaMjmyA,3981
 fabricatio/fs/readers.py,sha256=Pz1-cdZYtmqr032dsroImlkFXAd0kCYY_9qVpD4UrG4,1045
-fabricatio/fs/__init__.py,sha256=lWcKYg0v3mv2LnnSegOQaTtlVDODU0vtw_s6iKU5IqQ,122
+fabricatio/fs/__init__.py,sha256=9yH1-3mdWf2isbCqbaQO7xs1f6eoiPn4uKUAETNRk0A,465
 fabricatio/journal.py,sha256=siqimKF0M_QaaOCMxtjr_BJVNyUIAQWILzE9Q4T6-7c,781
-fabricatio/models/action.py,sha256=HVU0cAOuiBszRJrOskFASwt1yxSO7HCP4epS03p92WU,5947
+fabricatio/models/action.py,sha256=yj3Tt_UWGnkoVQNCMOBIqxDeyfXFEQp1oS4xd-nDXCY,6495
 fabricatio/models/events.py,sha256=pt-WkFhhA5SXmp6-3Vb_o_7I5xbKoTCJ22GAK7YYwpA,4101
 fabricatio/models/extra.py,sha256=0v5v0a_xcXcDTeIDmwtzUO68HjrmVvLHM6ALWvl9h6w,4328
 fabricatio/models/generic.py,sha256=3S16T7nZEDhDF-VLSmRPa7NEtapZjU5WQHosoisqbTw,11843
@@ -30,9 +30,10 @@ fabricatio/toolboxes/arithmetic.py,sha256=WLqhY-Pikv11Y_0SGajwZx3WhsLNpHKf9drzAq
 fabricatio/toolboxes/fs.py,sha256=YkNgon5-bvCiPVEND9971W-6wj8btKNL6nGry2otn9I,498
 fabricatio/toolboxes/task.py,sha256=kU4a501awIDV7GwNDuSlK3_Ym-5OhCp5sS-insTmUmQ,269
 fabricatio/toolboxes/__init__.py,sha256=b13KmASO8q5fBLwew964fn9oH86ER5g-S1PgA4fZ_xs,482
+fabricatio/workflows/articles.py,sha256=gZlmC2tS1YnSajZft6jN0VTSj6rcuT8367nBYD8edoI,473
 fabricatio/_rust.pyi,sha256=0wCqtwWkVxxoqprvk8T27T8QYKIAKHS7xgsmdMNjQKc,1756
 fabricatio/_rust_instances.py,sha256=dl0-yZ4UvT5g20tQgnPJpmqtkjFGXNG_YK4eLfi_ugQ,279
-fabricatio/__init__.py,sha256=lk9xEYWoiGdsIPjTCdoZ85tz0a7SDjXm6ORaxGx-dYw,1405
-fabricatio/_rust.cp312-win_amd64.pyd,sha256=OyB3QxCCDgBSoyWf_sseCWdRhXrQ9HqBrb7Vk7QEN7Y,1256448
-fabricatio-0.2.4.dev1.data/scripts/tdown.exe,sha256=aRm7EGhU3Mr6brN3iWId7kCKojnttdHnjI0M7Ap2vr0,3397632
-fabricatio-0.2.4.dev1.dist-info/RECORD,,
+fabricatio/__init__.py,sha256=lIjAvhmypC8kZNpymjmYByJtZjzJNXDmNzY36x1o7h8,1554
+fabricatio/_rust.cp312-win_amd64.pyd,sha256=awkdolid46gR8FEGI9Hxx04KPmXs35jZdXaauhazzoM,1277952
+fabricatio-0.2.4.dev3.data/scripts/tdown.exe,sha256=OhB53d4Eswjsw9TEcNsQCkd5D32BVEME6yRGsti1PTU,3397632
+fabricatio-0.2.4.dev3.dist-info/RECORD,,

fabricatio/actions/__init__.py

@@ -1,5 +0,0 @@
-"""module for actions."""
-
-from fabricatio.actions.article import ExtractArticleEssence
-
-__all__ = ["ExtractArticleEssence"]