ws-bom-robot-app 0.0.28__tar.gz → 0.0.30__tar.gz

ws_bom_robot_app-0.0.30/MANIFEST.in

@@ -0,0 +1,2 @@
+include requirements.txt
+include README.md

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: ws_bom_robot_app
-Version: 0.0.28
+Version: 0.0.30
 Summary: A FastAPI application serving ws bom/robot/llm platform ai.
 Home-page: https://github.com/websolutespa/bom
 Author: Websolute Spa
@@ -13,17 +13,17 @@ Description-Content-Type: text/markdown
 Requires-Dist: standardwebhooks==1.0.0
 Requires-Dist: apscheduler==3.11.0
 Requires-Dist: aiofiles==24.1.0
-Requires-Dist: pydantic==2.10.3
-Requires-Dist: pydantic-settings==2.6.1
-Requires-Dist: fastapi[standard]==0.115.5
-Requires-Dist: langchain==0.3.9
-Requires-Dist: langchain-openai==0.2.10
-Requires-Dist: langchain-community==0.3.8
-Requires-Dist: langchain-core==0.3.21
+Requires-Dist: pydantic==2.10.5
+Requires-Dist: pydantic-settings==2.7.1
+Requires-Dist: fastapi[standard]==0.115.6
+Requires-Dist: langchain==0.3.14
+Requires-Dist: langchain-openai==0.3.0
+Requires-Dist: langchain-community==0.3.14
+Requires-Dist: langchain-core==0.3.29
 Requires-Dist: faiss-cpu==1.9.0
-Requires-Dist: unstructured==0.16.11
+Requires-Dist: unstructured==0.16.13
 Requires-Dist: unstructured[image]
-Requires-Dist: unstructured-ingest==0.3.8
+Requires-Dist: unstructured-ingest==0.3.14
 Requires-Dist: unstructured-ingest[azure]
 Requires-Dist: unstructured-ingest[confluence]
 Requires-Dist: unstructured-ingest[dropbox]
@@ -37,7 +37,16 @@ Requires-Dist: unstructured-ingest[sharepoint]
 Requires-Dist: unstructured-ingest[slack]
 Requires-Dist: html5lib==1.1
 Requires-Dist: markdownify==0.14.1
-Requires-Dist: nebuly==0.3.35
+Requires-Dist: nebuly==0.3.36
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
 
 # 🤖 ws-bom-robot-app
 

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["setuptools >= 64"]
+requires = ["setuptools >= 75"]
 build-backend = "setuptools.build_meta"
 
 [tool.pytest.ini_options]

{ws_bom_robot_app-0.0.28/ws_bom_robot_app → ws_bom_robot_app-0.0.30}/requirements.txt

@@ -2,15 +2,15 @@
 standardwebhooks==1.0.0
 apscheduler==3.11.0
 aiofiles==24.1.0
-pydantic==2.10.3
-pydantic-settings==2.6.1
-fastapi[standard]==0.115.5
+pydantic==2.10.5
+pydantic-settings==2.7.1
+fastapi[standard]==0.115.6
 
 #framework
-langchain==0.3.9
-langchain-openai==0.2.10
-langchain-community==0.3.8
-langchain-core==0.3.21
+langchain==0.3.14
+langchain-openai==0.3.0
+langchain-community==0.3.14
+langchain-core==0.3.29
 
 #vector DB
 faiss-cpu==1.9.0
@@ -20,11 +20,11 @@ faiss-cpu==1.9.0
 #loaders
 #python-magic==0.4.27
 #opencv-python-headless==4.10.0.84 #docker specs
-unstructured==0.16.11
+unstructured==0.16.13
 unstructured[image]
 #unstructured[all-docs]==0.16.11
 #langchain_unstructured==0.1.5
-unstructured-ingest==0.3.8
+unstructured-ingest==0.3.14
 unstructured-ingest[azure]
 unstructured-ingest[confluence]
 unstructured-ingest[dropbox]
@@ -42,4 +42,4 @@ html5lib==1.1 #beautifulsoup4 parser
 markdownify==0.14.1 #sitemap
 
 #telemetry
-nebuly==0.3.35
+nebuly==0.3.36

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/setup.py

@@ -1,8 +1,10 @@
 from setuptools import setup, find_packages
 
+_requirements = [line.split('#')[0].strip() for line in open("requirements.txt").readlines() if all([line.strip(), not line.startswith("#")])]
+
 setup(
     name="ws_bom_robot_app",
-    version="0.0.28",
+    version="0.0.30",
     description="A FastAPI application serving ws bom/robot/llm platform ai.",
     long_description=open("README.md", encoding='utf-8').read(),
     long_description_content_type="text/markdown",
@@ -10,7 +12,7 @@ setup(
     author_email="dev@websolute.it",
     url="https://github.com/websolutespa/bom",
     packages=find_packages(),
-    install_requires=[line.split('#')[0].strip() for line in open("ws_bom_robot_app/requirements.txt").readlines() if all([line.strip(), not line.startswith("#")])],
+    install_requires=_requirements,
     classifiers=[
         "Programming Language :: Python :: 3",
         "License :: OSI Approved :: MIT License",
@@ -18,4 +20,3 @@ setup(
     ],
     python_requires=">=3.12",
 )
-

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/ws_bom_robot_app/llm/vector_store/integration/sitemap.py

@@ -1,10 +1,11 @@
-from typing import Any
+from typing import Any, AsyncGenerator, AsyncIterator
 import aiofiles
 import aiofiles.os
 from ws_bom_robot_app.llm.vector_store.integration.base import IntegrationStrategy
 from langchain_community.document_loaders.sitemap import SitemapLoader
 from langchain_community.document_transformers import MarkdownifyTransformer as markdownify
 from langchain_core.documents import Document
+from langchain_core.runnables import run_in_executor
 from bs4 import BeautifulSoup, Tag
 
 class Sitemap(IntegrationStrategy):
@@ -59,7 +60,15 @@ class Sitemap(IntegrationStrategy):
 
     def _remap_if_local(self, url: str) -> str:
         return f"{self.knowledgebase_path}/{url}" if self._is_local(url) else url
-
+    async def alazy_load(self,loader: SitemapLoader) -> AsyncIterator[Document]:
+        """A lazy loader for Documents."""
+        iterator = await run_in_executor(None, loader.lazy_load)
+        done = object()
+        while True:
+            doc = await run_in_executor(None, next, iterator, done)  # type: ignore[call-arg, arg-type]
+            if doc is done:
+                break
+            yield doc  # type: ignore[misc]
     async def load(self) -> list[Document]:
         if (self.__sitemap_url):
             _loader = SitemapLoader(
@@ -68,7 +77,7 @@ class Sitemap(IntegrationStrategy):
                 parsing_function=self._parse,
                 is_local=self._is_local(self.__sitemap_url)
             )
-            _docs = self._output([document async for document in _loader.alazy_load()])
+            _docs = self._output([document async for document in self.alazy_load(_loader)])
             if self._is_local(self.__sitemap_url):
               try:
                   await aiofiles.os.remove(_loader.web_path)

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/ws_bom_robot_app/llm/vector_store/loader/docling.py

@@ -3,12 +3,18 @@ from typing import Iterator, AsyncIterator, Optional
 from langchain_core.document_loaders import BaseLoader
 from langchain_core.documents import Document
 from langchain_core.runnables import run_in_executor
-from docling.document_converter import DocumentConverter, ConversionResult, ConversionStatus
+from docling.document_converter import DocumentConverter, InputFormat, PdfFormatOption
+from docling.datamodel.pipeline_options import PdfPipelineOptions, TableStructureOptions, TableFormerMode
 
 class DoclingLoader(BaseLoader):
   def __init__(self, file_path: str | list[str]) -> None:
       self._file_paths = file_path if isinstance(file_path, list) else [file_path]
-      self._converter = DocumentConverter()
+      self._converter = DocumentConverter(format_options={
+            InputFormat.PDF: PdfFormatOption(
+              pipeline_options=PdfPipelineOptions(
+                table_structure_options=TableStructureOptions(mode=TableFormerMode.ACCURATE)
+              ))
+        })
   def load(self) -> list[Document]:
       """Load data into Document objects."""
       return list(self.lazy_load())

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/ws_bom_robot_app.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: ws_bom_robot_app
-Version: 0.0.28
+Version: 0.0.30
 Summary: A FastAPI application serving ws bom/robot/llm platform ai.
 Home-page: https://github.com/websolutespa/bom
 Author: Websolute Spa
@@ -13,17 +13,17 @@ Description-Content-Type: text/markdown
 Requires-Dist: standardwebhooks==1.0.0
 Requires-Dist: apscheduler==3.11.0
 Requires-Dist: aiofiles==24.1.0
-Requires-Dist: pydantic==2.10.3
-Requires-Dist: pydantic-settings==2.6.1
-Requires-Dist: fastapi[standard]==0.115.5
-Requires-Dist: langchain==0.3.9
-Requires-Dist: langchain-openai==0.2.10
-Requires-Dist: langchain-community==0.3.8
-Requires-Dist: langchain-core==0.3.21
+Requires-Dist: pydantic==2.10.5
+Requires-Dist: pydantic-settings==2.7.1
+Requires-Dist: fastapi[standard]==0.115.6
+Requires-Dist: langchain==0.3.14
+Requires-Dist: langchain-openai==0.3.0
+Requires-Dist: langchain-community==0.3.14
+Requires-Dist: langchain-core==0.3.29
 Requires-Dist: faiss-cpu==1.9.0
-Requires-Dist: unstructured==0.16.11
+Requires-Dist: unstructured==0.16.13
 Requires-Dist: unstructured[image]
-Requires-Dist: unstructured-ingest==0.3.8
+Requires-Dist: unstructured-ingest==0.3.14
 Requires-Dist: unstructured-ingest[azure]
 Requires-Dist: unstructured-ingest[confluence]
 Requires-Dist: unstructured-ingest[dropbox]
@@ -37,7 +37,16 @@ Requires-Dist: unstructured-ingest[sharepoint]
 Requires-Dist: unstructured-ingest[slack]
 Requires-Dist: html5lib==1.1
 Requires-Dist: markdownify==0.14.1
-Requires-Dist: nebuly==0.3.35
+Requires-Dist: nebuly==0.3.36
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
 
 # 🤖 ws-bom-robot-app
 

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/ws_bom_robot_app.egg-info/SOURCES.txt

@@ -1,12 +1,13 @@
+MANIFEST.in
 README.md
 pyproject.toml
+requirements.txt
 setup.py
 ws_bom_robot_app/__init__.py
 ws_bom_robot_app/auth.py
 ws_bom_robot_app/config.py
 ws_bom_robot_app/cron_manager.py
 ws_bom_robot_app/main.py
-ws_bom_robot_app/requirements.txt
 ws_bom_robot_app/task_manager.py
 ws_bom_robot_app/util.py
 ws_bom_robot_app.egg-info/PKG-INFO

{ws_bom_robot_app-0.0.28 → ws_bom_robot_app-0.0.30}/ws_bom_robot_app.egg-info/requires.txt

@@ -1,17 +1,17 @@
 standardwebhooks==1.0.0
 apscheduler==3.11.0
 aiofiles==24.1.0
-pydantic==2.10.3
-pydantic-settings==2.6.1
-fastapi[standard]==0.115.5
-langchain==0.3.9
-langchain-openai==0.2.10
-langchain-community==0.3.8
-langchain-core==0.3.21
+pydantic==2.10.5
+pydantic-settings==2.7.1
+fastapi[standard]==0.115.6
+langchain==0.3.14
+langchain-openai==0.3.0
+langchain-community==0.3.14
+langchain-core==0.3.29
 faiss-cpu==1.9.0
-unstructured==0.16.11
+unstructured==0.16.13
 unstructured[image]
-unstructured-ingest==0.3.8
+unstructured-ingest==0.3.14
 unstructured-ingest[azure]
 unstructured-ingest[confluence]
 unstructured-ingest[dropbox]
@@ -25,4 +25,4 @@ unstructured-ingest[sharepoint]
 unstructured-ingest[slack]
 html5lib==1.1
 markdownify==0.14.1
-nebuly==0.3.35
+nebuly==0.3.36