lumis-ai 0.1.1a1__tar.gz → 0.1.1a3__tar.gz

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lumis-ai
-Version: 0.1.1a1
+Version: 0.1.1a3
 Summary: An AI agent framework for building LLM-powered applications
 Project-URL: Homepage, https://github.com/tareksanger/lumis
 Project-URL: Repository, https://github.com/tareksanger/lumis
@@ -20,11 +20,11 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development :: Libraries
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.11
-Requires-Dist: arxiv>=0.5.0
 Requires-Dist: asgiref>=3.8.1
 Requires-Dist: beautifulsoup4>=4.12.0
 Requires-Dist: deepmerge>=2.0.0
 Requires-Dist: faiss-cpu>=1.7.0
+Requires-Dist: google-genai>=1.61.0
 Requires-Dist: googlesearch-python>=1.2.0
 Requires-Dist: httpx>=0.27.0
 Requires-Dist: huggingface-hub>=0.24.0
@@ -32,7 +32,6 @@ Requires-Dist: langchain-core>=0.3.13
 Requires-Dist: langchain>=0.3.4
 Requires-Dist: lxml>=4.9.0
 Requires-Dist: minify-html>=0.11.0
-Requires-Dist: neo4j>=5.25.0
 Requires-Dist: nltk>=3.9.0
 Requires-Dist: numpy>=1.24
 Requires-Dist: ollama>=0.3.0
@@ -42,14 +41,12 @@ Requires-Dist: pydantic>=2.9.2
 Requires-Dist: pymupdf>=1.22.0
 Requires-Dist: pypdf2>=3.0.0
 Requires-Dist: python-dateutil>=2.8.0
-Requires-Dist: pytrends>=5.1.0
 Requires-Dist: pyvis>=0.3.0
-Requires-Dist: readability-lxml>=3.3.0
+Requires-Dist: readability-lxml>=0.8.0
 Requires-Dist: redis>=5.0.0
 Requires-Dist: requests>=2.28.0
 Requires-Dist: scikit-learn>=1.0.0
 Requires-Dist: sentence-transformers>=2.2.2
-Requires-Dist: tavily-python>=0.3.0
 Requires-Dist: tenacity>=9.0.0
 Requires-Dist: textstat>=0.7.0
 Requires-Dist: tiktoken>=0.5.0
@@ -57,14 +54,15 @@ Requires-Dist: torch>=2.5.0
 Requires-Dist: transformers>=4.0.0
 Requires-Dist: typing-extensions>=4.0.0
 Requires-Dist: wikipedia>=1.4.0
-Requires-Dist: yfinance>=0.2.0
 Provides-Extra: django
 Requires-Dist: django>=4.2; extra == 'django'
+Provides-Extra: search
+Requires-Dist: arxiv>=0.5.0; extra == 'search'
+Requires-Dist: pytrends>=4.5.0; extra == 'search'
+Requires-Dist: tavily-python>=0.3.0; extra == 'search'
+Requires-Dist: yfinance>=0.2.0; extra == 'search'
 Provides-Extra: spacy
-Requires-Dist: coreferee-model-en>=1.0.0; extra == 'spacy'
-Requires-Dist: coreferee>=1.3.0; extra == 'spacy'
-Requires-Dist: en-core-web-lg<3.8.0,>=3.7.0; extra == 'spacy'
-Requires-Dist: en-core-web-trf<3.8.0,>=3.7.0; extra == 'spacy'
+Requires-Dist: pytextrank>=3.2.0; extra == 'spacy'
 Requires-Dist: spacy-llm>=0.7.2; extra == 'spacy'
 Requires-Dist: spacy<3.8.0,>=3.7.0; extra == 'spacy'
 Description-Content-Type: text/markdown
@@ -97,6 +95,7 @@ Some integrations are opt-in to keep the base install lighter:
 | Extra      | What it adds                                  | Install                           |
 |------------|-----------------------------------------------|-----------------------------------|
 | `spacy`    | NER, fact extraction, coreference resolution  | `pip install lumis-ai[spacy]`     |
+| `search`   | arXiv, Google Trends, Yahoo Finance           | `pip install lumis-ai[search]`    |
 | `django`   | Django ORM memory backend                     | `pip install lumis-ai[django]`    |
 
 The `spacy` extra requires language models. After installing, download them:

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/README.md

@@ -26,6 +26,7 @@ Some integrations are opt-in to keep the base install lighter:
 | Extra      | What it adds                                  | Install                           |
 |------------|-----------------------------------------------|-----------------------------------|
 | `spacy`    | NER, fact extraction, coreference resolution  | `pip install lumis-ai[spacy]`     |
+| `search`   | arXiv, Google Trends, Yahoo Finance           | `pip install lumis-ai[search]`    |
 | `django`   | Django ORM memory backend                     | `pip install lumis-ai[django]`    |
 
 The `spacy` extra requires language models. After installing, download them:

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/agents/research_agent/tools/search.py

@@ -7,8 +7,6 @@ from typing import Any, Optional
 from lumis.llm import Gemini
 from lumis.tools import WikipediaSearcher
 from lumis.tools.search.arxiv import ArxivResult, ArxivSearcher
-
-from arxiv import SortCriterion, SortOrder
 from tenacity import retry, retry_if_exception_type, stop_after_attempt, wait_exponential
 
 logger = logging.getLogger(__name__)
@@ -173,7 +171,8 @@ async def search_arxiv(  # noqa: C901
     summary_semaphore = asyncio.Semaphore(4)
 
     try:
-        # Convert string sort parameters to enums
+        from arxiv import SortCriterion, SortOrder
+
         sort_criterion = SortCriterion(sort_by)
         sort_order_enum = SortOrder(sort_order)
 

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/tools/scraper.py

@@ -278,16 +278,6 @@ class WebScrapper:
             raise e
 
         return None
-
-    # async def from_gotenberg(self, url: str, metadata: Optional[dict]) -> Optional[Document]:
-    #     # Same logic as before, but uses injected gb_client
-    #     try:
-    #         response = await self.gb_client.afrom_url(url)
-    #         return await self.parse_pdf(response.content, metadata)
-    #     except Exception as e:
-    #         logger.debug(f"Gotenberg failed for {url}: {e}")
-    #         return None
-
     def _url_with_protocol(self, url: str):
         if not url.startswith("http://") and not url.startswith("https://"):
             url = "https://" + url

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/tools/search/arxiv.py

@@ -7,12 +7,15 @@ import logging
 from typing import Any, Dict, Optional
 import xml.etree.ElementTree as ET
 
-import arxiv
-from arxiv import Client, SortCriterion, SortOrder
+from typing import TYPE_CHECKING
+
 import httpx
 from lumis.core.document import Document
 import pymupdf
 
+if TYPE_CHECKING:
+    from arxiv import SortCriterion, SortOrder
+
 logger = logging.getLogger(__name__)
 
 
@@ -36,8 +39,16 @@ class ArxivResult:
 
 class ArxivSearcher:
     def __init__(self, max_results: int = 10, max_concurrent_pdfs: int = 5) -> None:
+        try:
+            import arxiv
+        except ImportError:
+            raise ImportError(
+                "arxiv is required for ArxivSearcher. "
+                "Install it with: pip install lumis-ai[search]"
+            )
+        self._arxiv = arxiv
         self.max_results: int = max_results
-        self.client = Client()
+        self.client = arxiv.Client()
         self._http_client = httpx.AsyncClient()
         self._result_cache: dict[str, ArxivResult] = {}  # Cache for results by arxiv_id
         self._pdf_semaphore = asyncio.Semaphore(max_concurrent_pdfs)
@@ -46,8 +57,8 @@ class ArxivSearcher:
         self,
         query: str,
         max_results: Optional[int] = None,
-        sort_by: SortCriterion = SortCriterion.SubmittedDate,
-        sort_order: SortOrder = SortOrder.Descending,
+        sort_by: SortCriterion | None = None,
+        sort_order: SortOrder | None = None,
         id_list: Optional[list[str]] = None,
         read_pdfs: bool = False,
     ) -> list[ArxivResult]:
@@ -65,8 +76,13 @@ class ArxivSearcher:
         Returns:
             list of ArxivResult objects
         """
+        if sort_by is None:
+            sort_by = self._arxiv.SortCriterion.SubmittedDate
+        if sort_order is None:
+            sort_order = self._arxiv.SortOrder.Descending
+
         # Create search object with all available options
-        search = arxiv.Search(query=query, max_results=max_results or self.max_results, sort_by=sort_by, sort_order=sort_order, id_list=id_list or [])
+        search = self._arxiv.Search(query=query, max_results=max_results or self.max_results, sort_by=sort_by, sort_order=sort_order, id_list=id_list or [])
 
         results: list[ArxivResult] = []
         pdf_tasks = []

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/tools/search/pytrends.py

@@ -3,7 +3,6 @@ from __future__ import annotations
 from typing import Any, Dict, List
 
 from asgiref.sync import sync_to_async
-from pytrends.request import TrendReq
 
 
 class PyTrends:
@@ -12,6 +11,13 @@ class PyTrends:
     """
 
     def __init__(self, tz: int = 0, retries: int = 3, backoff: float = 0.1):
+        try:
+            from pytrends.request import TrendReq
+        except ImportError:
+            raise ImportError(
+                "pytrends is required for PyTrends. "
+                "Install it with: pip install lumis-ai[search]"
+            )
         self.pytrends = TrendReq(retries=retries, backoff_factor=backoff, tz=tz)  # type: ignore
 
     @sync_to_async

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/tools/search/search_engine_client.py

@@ -8,9 +8,16 @@ from typing import Literal, Optional, Sequence
 
 from googlesearch import search as google_search
 import redis.asyncio as aioredis
-from tavily import TavilyClient, UsageLimitExceededError
 from typing_extensions import TypedDict
 
+try:
+    from tavili import TavilyClient, UsageLimitExceededError
+except ImportError:
+    TavilyClient = None  # type: ignore[assignment,misc]
+
+    class UsageLimitExceededError(Exception):  # type: ignore[no-redef]
+        pass
+
 logger = logging.getLogger(__name__)
 
 

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/lumis/tools/search/yahoofinance.py

@@ -7,7 +7,6 @@ from typing import Any, cast, Optional
 from asgiref.sync import sync_to_async
 import pandas as pd
 from pydantic import BaseModel, Field
-import yfinance as yf
 
 logger = logging.getLogger(__name__)
 
@@ -53,6 +52,16 @@ class YahooFinance:
     with async codebases.
     """
 
+    def __init__(self):
+        try:
+            import yfinance as yf
+        except ImportError:
+            raise ImportError(
+                "yfinance is required for YahooFinance. "
+                "Install it with: pip install lumis-ai[search]"
+            )
+        self._yf = yf
+
     @sync_to_async
     def get_quote(self, ticker: str) -> QuoteData:
         """
@@ -70,7 +79,7 @@ class YahooFinance:
         """
         try:
             logger.debug(f"Fetching quote data for ticker: {ticker}")
-            t = yf.Ticker(ticker)
+            t = self._yf.Ticker(ticker)
             info = t.info
 
             return QuoteData(
@@ -114,7 +123,7 @@ class YahooFinance:
         """
         try:
             logger.debug(f"Fetching historical data for {ticker} with period={period}, interval={interval}")
-            t = yf.Ticker(ticker)
+            t = self._yf.Ticker(ticker)
             hist = t.history(period=period, interval=interval, start=start, end=end)
 
             return [
@@ -149,8 +158,8 @@ class YahooFinance:
         """
         try:
             logger.debug(f"Searching for tickers matching: {query}")
-            # Using yf.Tickers() as get_tickers_by_search is deprecated
-            tickers = yf.Tickers(query)
+            # Using self._yf.Tickers() as get_tickers_by_search is deprecated
+            tickers = self._yf.Tickers(query)
             return [{"symbol": symbol, "name": ticker.info.get("shortName", symbol)} for symbol, ticker in tickers.tickers.items()]
         except Exception as e:
             logger.error(f"Error searching tickers for {query}: {str(e)}")
@@ -172,7 +181,7 @@ class YahooFinance:
         """
         try:
             logger.debug(f"Fetching recommendations for {ticker}")
-            t = yf.Ticker(ticker)
+            t = self._yf.Ticker(ticker)
             if t.recommendations is None:
                 return []
             df = cast(pd.DataFrame, t.recommendations)
@@ -197,7 +206,7 @@ class YahooFinance:
         """
         try:
             logger.debug(f"Fetching major holders for {ticker}")
-            t = yf.Ticker(ticker)
+            t = self._yf.Ticker(ticker)
             if t.major_holders is None:
                 return {}
             df = cast(pd.DataFrame, t.major_holders)

{lumis_ai-0.1.1a1 → lumis_ai-0.1.1a3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "lumis-ai"
-version = "0.1.1a1"
+version = "0.1.1a3"
 description = "An AI agent framework for building LLM-powered applications"
 readme = "README.md"
 license = "MIT"
@@ -22,7 +22,6 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Artificial Intelligence",
 ]
 dependencies = [
-    "arxiv>=0.5.0",
     "asgiref>=3.8.1",
     "beautifulsoup4>=4.12.0",
     "deepmerge>=2.0.0",
@@ -34,7 +33,6 @@ dependencies = [
     "langchain-core>=0.3.13",
     "lxml>=4.9.0",
     "minify-html>=0.11.0",
-    "neo4j>=5.25.0",
     "nltk>=3.9.0",
     "numpy>=1.24",
     "ollama>=0.3.0",
@@ -44,14 +42,12 @@ dependencies = [
     "PyPDF2>=3.0.0",
     "pydantic>=2.9.2",
     "python-dateutil>=2.8.0",
-    "pytrends>=5.1.0",
     "pyvis>=0.3.0",
-    "readability-lxml>=3.3.0",
+    "readability-lxml>=0.8.0",
     "redis>=5.0.0",
     "requests>=2.28.0",
     "scikit-learn>=1.0.0",
     "sentence-transformers>=2.2.2",
-    "tavily-python>=0.3.0",
     "tenacity>=9.0.0",
     "textstat>=0.7.0",
     "tiktoken>=0.5.0",
@@ -59,18 +55,21 @@ dependencies = [
     "transformers>=4.0.0",
     "typing-extensions>=4.0.0",
     "wikipedia>=1.4.0",
-    "yfinance>=0.2.0",
+    "google-genai>=1.61.0",
 ]
 
 [project.optional-dependencies]
 django = [
     "django>=4.2",
 ]
+search = [
+    "arxiv>=0.5.0",
+    "pytrends>=4.5.0",
+    "tavily-python>=0.3.0",
+    "yfinance>=0.2.0",
+]
 spacy = [
-    "coreferee>=1.3.0",
-    "coreferee-model-en>=1.0.0",
-    "en-core-web-lg>=3.7.0,<3.8.0",
-    "en-core-web-trf>=3.7.0,<3.8.0",
+    "pytextrank>=3.2.0",
     "spacy>=3.7.0,<3.8.0",
     "spacy-llm>=0.7.2",
 ]
@@ -94,8 +93,6 @@ dev = [
 torch = [
     { index = "pytorch", marker = "sys_platform != 'darwin'" },
 ]
-en-core-web-trf = { url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_trf-3.7.1/en_core_web_trf-3.7.1-py3-none-any.whl" }
-en-core-web-lg = { url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_lg-3.7.0/en_core_web_lg-3.7.0-py3-none-any.whl" }
 
 [[tool.uv.index]]
 name = "pytorch"