root/Gitcloned-RAG-Anything
1
1
Fork 0
Code Issues Pull Requests Projects Releases Activity

fix lint

Browse Source
This commit is contained in:
zrguo
2025-09-22 10:42:35 +08:00
parent 8e0e05d497
commit 1d48f24b4a
4 changed files with 104 additions and 4863 deletions
Download Patch File Download Diff File Expand all files Collapse all files

122
examples/lmstudio_integration_example.py
View File

@@ -36,13 +36,18 @@ from raganything import RAGAnything, RAGAnythingConfig
from lightrag.utils import EmbeddingFunc
from lightrag.llm.openai import openai_complete_if_cache
LM_BASE_URL = os.getenv('LLM_BINDING_HOST', 'http://localhost:1234/v1')
LM_API_KEY = os.getenv('LLM_BINDING_API_KEY', 'lm-studio')
LM_MODEL_NAME = os.getenv('LLM_MODEL', 'openai/gpt-oss-20b')
LM_EMBED_MODEL = os.getenv('EMBEDDING_MODEL', 'text-embedding-nomic-embed-text-v1.5')
LM_BASE_URL = os.getenv("LLM_BINDING_HOST", "http://localhost:1234/v1")
LM_API_KEY = os.getenv("LLM_BINDING_API_KEY", "lm-studio")
LM_MODEL_NAME = os.getenv("LLM_MODEL", "openai/gpt-oss-20b")
LM_EMBED_MODEL = os.getenv("EMBEDDING_MODEL", "text-embedding-nomic-embed-text-v1.5")
async def lmstudio_llm_model_func(prompt: str, system_prompt: Optional[str] = None,
history_messages: List[Dict] = None, **kwargs) -> str:
async def lmstudio_llm_model_func(
prompt: str,
system_prompt: Optional[str] = None,
history_messages: List[Dict] = None,
**kwargs,
) -> str:
"""Top-level LLM function for LightRAG (pickle-safe)."""
return await openai_complete_if_cache(
model=LM_MODEL_NAME,
@@ -58,6 +63,7 @@ async def lmstudio_llm_model_func(prompt: str, system_prompt: Optional[str] = No
async def lmstudio_embedding_async(texts: List[str]) -> List[List[float]]:
"""Top-level embedding function for LightRAG (pickle-safe)."""
from lightrag.llm.openai import openai_embed
embeddings = await openai_embed(
texts=texts,
model=LM_EMBED_MODEL,
@@ -66,17 +72,19 @@ async def lmstudio_embedding_async(texts: List[str]) -> List[List[float]]:
)
return embeddings.tolist()
class LMStudioRAGIntegration:
"""Integration class for LM Studio with RAG-Anything."""
def __init__(self):
# LM Studio configuration using standard LLM_BINDING variables
self.base_url = os.getenv('LLM_BINDING_HOST', 'http://localhost:1234/v1')
self.api_key = os.getenv('LLM_BINDING_API_KEY', 'lm-studio')
self.model_name = os.getenv('LLM_MODEL', 'openai/gpt-oss-20b')
self.embedding_model = os.getenv('EMBEDDING_MODEL', 'text-embedding-nomic-embed-text-v1.5')
self.base_url = os.getenv("LLM_BINDING_HOST", "http://localhost:1234/v1")
self.api_key = os.getenv("LLM_BINDING_API_KEY", "lm-studio")
self.model_name = os.getenv("LLM_MODEL", "openai/gpt-oss-20b")
self.embedding_model = os.getenv(
"EMBEDDING_MODEL", "text-embedding-nomic-embed-text-v1.5"
)
# RAG-Anything configuration
# Use a fresh working directory each run to avoid legacy doc_status schema conflicts
self.config = RAGAnythingConfig(
@@ -88,7 +96,7 @@ class LMStudioRAGIntegration:
enable_equation_processing=True,
)
print(f"📁 Using working_dir: {self.config.working_dir}")
self.rag = None
async def test_connection(self) -> bool:
@@ -98,16 +106,16 @@ class LMStudioRAGIntegration:
client = AsyncOpenAI(base_url=self.base_url, api_key=self.api_key)
models = await client.models.list()
print(f"✅ Connected successfully! Found {len(models.data)} models")
# Show available models
print("📊 Available models:")
for i, model in enumerate(models.data[:5]):
marker = "🎯" if model.id == self.model_name else " "
print(f"{marker} {i+1}. {model.id}")
if len(models.data) > 5:
print(f" ... and {len(models.data) - 5} more models")
return True
except Exception as e:
print(f"❌ Connection failed: {str(e)}")
@@ -122,7 +130,7 @@ class LMStudioRAGIntegration:
await client.close()
except Exception:
pass
async def test_chat_completion(self) -> bool:
"""Test basic chat functionality."""
try:
@@ -132,14 +140,17 @@ class LMStudioRAGIntegration:
model=self.model_name,
messages=[
{"role": "system", "content": "You are a helpful AI assistant."},
{"role": "user", "content": "Hello! Please confirm you're working and tell me your capabilities."}
{
"role": "user",
"content": "Hello! Please confirm you're working and tell me your capabilities.",
},
],
max_tokens=100,
temperature=0.7
temperature=0.7,
)
result = response.choices[0].message.content.strip()
print(f"✅ Chat test successful!")
print("✅ Chat test successful!")
print(f"Response: {result}")
return True
except Exception as e:
@@ -150,9 +161,9 @@ class LMStudioRAGIntegration:
await client.close()
except Exception:
pass
# Deprecated factory helpers removed to reduce redundancy
def embedding_func_factory(self):
"""Create a completely serializable embedding function."""
return EmbeddingFunc(
@@ -160,11 +171,11 @@ class LMStudioRAGIntegration:
max_token_size=8192, # nomic-embed-text-v1.5 context length
func=lmstudio_embedding_async,
)
async def initialize_rag(self):
"""Initialize RAG-Anything with LM Studio functions."""
print("Initializing RAG-Anything with LM Studio...")
try:
self.rag = RAGAnything(
config=self.config,
@@ -176,6 +187,7 @@ class LMStudioRAGIntegration:
# Older LightRAG versions may not accept this extra field in DocProcessingStatus
async def _noop_mark_multimodal(doc_id: str):
return None
self.rag._mark_multimodal_processing_complete = _noop_mark_multimodal
print("✅ RAG-Anything initialized successfully!")
@@ -183,38 +195,38 @@ class LMStudioRAGIntegration:
except Exception as e:
print(f"❌ RAG initialization failed: {str(e)}")
return False
async def process_document_example(self, file_path: str):
"""Example: Process a document with LM Studio backend."""
if not self.rag:
print("❌ RAG not initialized. Call initialize_rag() first.")
return
try:
print(f"📄 Processing document: {file_path}")
await self.rag.process_document_complete(
file_path=file_path,
output_dir="./output_lmstudio",
parse_method="auto",
display_stats=True
display_stats=True,
)
print("✅ Document processing completed!")
except Exception as e:
print(f"❌ Document processing failed: {str(e)}")
async def query_examples(self):
"""Example queries with different modes."""
if not self.rag:
print("❌ RAG not initialized. Call initialize_rag() first.")
return
# Example queries
queries = [
("What are the main topics in the processed documents?", "hybrid"),
("Summarize any tables or data found in the documents", "local"),
("What images or figures are mentioned?", "global"),
]
print("\n🔍 Running example queries...")
for query, mode in queries:
try:
@@ -223,26 +235,26 @@ class LMStudioRAGIntegration:
print(f"Answer: {result[:200]}...")
except Exception as e:
print(f"❌ Query failed: {str(e)}")
async def simple_query_example(self):
"""Example basic text query with sample content."""
if not self.rag:
print("❌ RAG not initialized")
return
try:
print("\nAdding sample content for testing...")
# Create content list in the format expected by RAGAnything
content_list = [
{
"type": "text",
"text": """LM Studio Integration with RAG-Anything
This integration demonstrates how to connect LM Studio's local AI models with RAG-Anything's document processing capabilities. The system uses:
- LM Studio for local LLM inference
- nomic-embed-text-v1.5 for embeddings (768 dimensions)
- nomic-embed-text-v1.5 for embeddings (768 dimensions)
- RAG-Anything for document processing and retrieval
Key benefits include:
@@ -250,71 +262,73 @@ Key benefits include:
- Performance: Direct API access to local models
- Flexibility: Support for various document formats
- Cost-effective: No external API usage""",
"page_idx": 0
"page_idx": 0,
}
]
# Insert the content list using the correct method
await self.rag.insert_content_list(
content_list=content_list,
file_path="lmstudio_integration_demo.txt",
# Use a unique doc_id to avoid collisions and doc_status reuse across runs
doc_id=f"demo-content-{uuid.uuid4()}",
display_stats=True
display_stats=True,
)
print("✅ Sample content added to knowledge base")
print("\nTesting basic text query...")
# Simple text query example
result = await self.rag.aquery(
"What are the key benefits of this LM Studio integration?",
mode="hybrid"
mode="hybrid",
)
print(f"✅ Query result: {result[:300]}...")
except Exception as e:
print(f"❌ Query failed: {str(e)}")
async def main():
"""Main example function."""
print("=" * 70)
print("LM Studio + RAG-Anything Integration Example")
print("=" * 70)
# Initialize integration
integration = LMStudioRAGIntegration()
# Test connection
if not await integration.test_connection():
return False
print()
if not await integration.test_chat_completion():
return False
# Initialize RAG
print("\n" + "─" * 50)
if not await integration.initialize_rag():
return False
# Example document processing (uncomment and provide a real file path)
# await integration.process_document_example("path/to/your/document.pdf")
# Example queries (uncomment after processing documents)
# await integration.query_examples()
# Example basic query
await integration.simple_query_example()
print("\n" + "=" * 70)
print("Integration example completed successfully!")
print("=" * 70)
return True
if __name__ == "__main__":
print("🚀 Starting LM Studio integration example...")
success = asyncio.run(main())
exit(0 if success else 1)

28
pyproject.toml
View File

@@ -1,16 +1,17 @@
[build-system]
requires = ["setuptools>=45", "wheel"]
requires = ["setuptools>=64", "wheel"]
build-backend = "setuptools.build_meta"
[project]
name = "raganything"
version = "1.2.7"
dynamic = ["version"]
authors = [
{name = "Zirui Guo"}
]
description = "RAGAnything: All-in-One RAG System"
readme = "README.md"
license = { text = "MIT" }
authors = [
{ name = "Zirui Guo" }
]
requires-python = ">=3.10"
classifiers = [
"Development Status :: 4 - Beta",
"Programming Language :: Python :: 3",
@@ -19,7 +20,6 @@ classifiers = [
"Intended Audience :: Developers",
"Topic :: Software Development :: Libraries :: Python Modules",
]
requires-python = ">=3.10"
dependencies = [
"huggingface_hub",
"lightrag-hku",
@@ -31,12 +31,18 @@ dependencies = [
image = ["Pillow>=10.0.0"]
text = ["reportlab>=4.0.0"]
office = [] # Requires LibreOffice (external program)
all = ["Pillow>=10.0.0", "reportlab>=4.0.0"]
markdown = [
"markdown>=3.4.0",
"weasyprint>=60.0",
"pygments>=2.10.0",
]
all = [
"Pillow>=10.0.0",
"reportlab>=4.0.0",
"markdown>=3.4.0",
"weasyprint>=60.0",
"pygments>=2.10.0"
]
[project.urls]
Homepage = "https://github.com/HKUDS/RAG-Anything"
@@ -57,7 +63,13 @@ dev-dependencies = [
]
[tool.setuptools.packages.find]
exclude = ["tests*", "docs*"]
include = ["raganything*"]
[tool.setuptools]
include-package-data = true
[tool.setuptools.dynamic]
version = {attr = "raganything.__version__"}
[tool.ruff]
target-version = "py310"

23
raganything/parser.py
View File

@@ -262,20 +262,29 @@ class Parser:
from reportlab.lib.units import inch
from reportlab.pdfbase import pdfmetrics
from reportlab.pdfbase.ttfonts import TTFont
support_chinese = True
try:
if 'WenQuanYi' not in pdfmetrics.getRegisteredFontNames():
if not Path('/usr/share/fonts/wqy-microhei/wqy-microhei.ttc').exists():
if "WenQuanYi" not in pdfmetrics.getRegisteredFontNames():
if not Path(
"/usr/share/fonts/wqy-microhei/wqy-microhei.ttc"
).exists():
support_chinese = False
logging.warning(
"WenQuanYi font not found at /usr/share/fonts/wqy-microhei/wqy-microhei.ttc. Chinese characters may not render correctly.")
"WenQuanYi font not found at /usr/share/fonts/wqy-microhei/wqy-microhei.ttc. Chinese characters may not render correctly."
)
else:
pdfmetrics.registerFont(
TTFont('WenQuanYi', '/usr/share/fonts/wqy-microhei/wqy-microhei.ttc'))
TTFont(
"WenQuanYi",
"/usr/share/fonts/wqy-microhei/wqy-microhei.ttc",
)
)
except Exception as e:
support_chinese = False
logging.warning(
f"Failed to register WenQuanYi font: {e}. Chinese characters may not render correctly.")
f"Failed to register WenQuanYi font: {e}. Chinese characters may not render correctly."
)
# Create PDF document
doc = SimpleDocTemplate(
@@ -292,8 +301,8 @@ class Parser:
normal_style = styles["Normal"]
heading_style = styles["Heading1"]
if support_chinese:
normal_style.fontName = 'WenQuanYi'
heading_style.fontName = 'WenQuanYi'
normal_style.fontName = "WenQuanYi"
heading_style.fontName = "WenQuanYi"
# Try to register a font that supports Chinese characters
try:

4794
uv.lock generated
View File

File diff suppressed because it is too large Load Diff