Merge pull request #25 from vectorlessflow/dev

Dev

Author: zTgx

README.md

@@ -163,6 +163,105 @@ async fn main() -> vectorless::Result<()> {
 | **Feedback Learning** | Improves from user feedback over time |
 | **Multi-turn Queries** | Handles complex questions with decomposition |
 
+## Configuration
+
+### Zero Configuration (Recommended)
+
+Just set `OPENAI_API_KEY` and you're ready to go:
+
+```bash
+export OPENAI_API_KEY="sk-..."
+```
+
+<details>
+<summary><b>Python</b></summary>
+
+```python
+from vectorless import Engine
+
+# Uses OPENAI_API_KEY from environment
+engine = Engine(workspace="./data")
+```
+
+</details>
+
+<details>
+<summary><b>Rust</b></summary>
+
+```rust
+use vectorless::Engine;
+
+let client = Engine::builder()
+    .with_workspace("./workspace")
+    .build().await?;
+```
+
+</details>
+
+### Environment Variables
+
+| Variable | Description |
+|----------|-------------|
+| `OPENAI_API_KEY` | LLM API key |
+| `VECTORLESS_MODEL` | Default model (e.g., `gpt-4o-mini`) |
+| `VECTORLESS_ENDPOINT` | API endpoint URL |
+| `VECTORLESS_WORKSPACE` | Workspace directory |
+
+### Advanced Configuration
+
+For fine-grained control, use a config file:
+
+```bash
+cp config.toml ./vectorless.toml
+```
+
+<details>
+<summary><b>Python</b></summary>
+
+```python
+from vectorless import Engine
+
+# Use full configuration file
+engine = Engine(config_path="./vectorless.toml")
+
+# Or override specific settings
+engine = Engine(
+    config_path="./vectorless.toml",
+    model="gpt-4o",  # Override model from config
+)
+```
+
+</details>
+
+<details>
+<summary><b>Rust</b></summary>
+
+```rust
+use vectorless::Engine;
+
+// Use full configuration file
+let client = Engine::builder()
+    .with_config_path("./vectorless.toml")
+    .build().await?;
+
+// Or override specific settings
+let client = Engine::builder()
+    .with_config_path("./vectorless.toml")
+    .with_model("gpt-4o", None)  // Override model
+    .build().await?;
+```
+
+</details>
+
+### Configuration Priority
+
+Later overrides earlier:
+
+1. Default configuration
+2. Auto-detected config file (`vectorless.toml`, `config.toml`, `.vectorless.toml`)
+3. Explicit config file (`config_path` / `with_config_path`)
+4. Environment variables
+5. Constructor/builder parameters (highest priority)
 
 ## Architecture
 
@@ -177,7 +276,7 @@ async fn main() -> vectorless::Result<()> {
 
 ## Examples
 
-See the [examples/](examples/) directory.
+See the [examples/](examples/) directory for more usage patterns.
 
 ## Contributing
 

docs/samples/sample.md

@@ -29,53 +29,4 @@ The core module provides fundamental types:
 The parser module handles document parsing:
 - `MarkdownParser` — Parse Markdown files
 - `PdfParser` — Parse PDF files (planned)
-- `HtmlParser` — Parse HTML files (planned)
-
-## Usage Examples
-
-### Basic Usage
-
-```rust
-use vectorless::client::{Vectorless, VectorlessBuilder};
-
-let client = VectorlessBuilder::new()
-    .with_workspace("./workspace")
-    .build()?;
-
-let doc_id = client.index("./document.md").await?;
-```
-
-### Advanced Usage
-
-You can customize the retrieval process:
-
-```rust
-use vectorless::{LlmNavigator, RetrieveOptions};
-
-let retriever = LlmNavigator::with_defaults();
-let options = RetrieveOptions::new()
-    .with_top_k(5)
-    .with_min_score(0.5);
-
-let results = retriever.retrieve(&tree, "What is vectorless?", &options).await?;
-```
-
-## Configuration
-
-The library can be configured via TOML files or programmatically.
-
-### Configuration File
-
-```toml
-[summary]
-model = "gpt-4"
-max_tokens = 200
-
-[retrieval]
-model = "gpt-4"
-top_k = 3
-```
-
-## API Reference
-
-See the API documentation for detailed information about each function and type.
+- `HtmlParser` — Parse HTML files (planned)

examples/python/advanced/README.md

@@ -0,0 +1,45 @@
+# Advanced Example - Full Configuration
+
+Use a configuration file for fine-grained control.
+
+## Setup
+
+```bash
+pip install vectorless
+
+# Copy the example config
+cp ../../../config.toml ./vectorless.toml
+
+# Edit to customize your settings
+vim vectorless.toml
+```
+
+## Run
+
+```bash
+python main.py
+```
+
+## Configuration File Structure
+
+```toml
+[llm]
+api_key = "sk-..."
+
+[llm.summary]
+model = "gpt-4o-mini"
+max_tokens = 200
+
+[llm.retrieval]
+model = "gpt-4o"
+max_tokens = 100
+
+[retrieval]
+top_k = 5
+beam_width = 3
+max_iterations = 10
+
+[storage]
+workspace_dir = "./workspace"
+cache_size = 100
+```

examples/python/advanced/main.py

@@ -0,0 +1,115 @@
+#!/usr/bin/env python3
+"""
+Advanced example - Full Configuration File.
+
+This example demonstrates how to use a full configuration file
+for fine-grained control over all settings.
+
+Usage:
+    cp ../../../config.toml ./vectorless.toml
+    # Edit vectorless.toml to customize settings
+    python main.py
+"""
+
+import os
+from vectorless import Engine, IndexContext
+
+# Path to config file (relative to this script)
+CONFIG_PATH = "./vectorless.toml"
+WORKSPACE = "./workspace"
+
+
+def main():
+    print("=== Vectorless Advanced Example (Full Configuration) ===\n")
+
+    # Check if config file exists
+    if not os.path.exists(CONFIG_PATH):
+        print(f"Error: Config file not found: {CONFIG_PATH}")
+        print("\nCreate it by copying the example:")
+        print(f"  cp ../../../config.toml {CONFIG_PATH}")
+        print("\nThen edit it to customize your settings.")
+        return
+
+    # Create engine with config file
+    engine = Engine(config_path=CONFIG_PATH)
+
+    print(f"✓ Engine created with config file: {CONFIG_PATH}\n")
+
+    # Index a document
+    content = """
+# System Documentation
+
+## Architecture
+
+The system consists of three main components:
+
+1. **Index Pipeline** - Parses documents and builds a navigable tree
+2. **Retrieval Pipeline** - Queries and retrieves relevant content
+3. **Pilot** - LLM-powered navigation guide
+
+## Configuration Options
+
+### LLM Settings
+- `model`: The LLM model to use (e.g., "gpt-4o", "gpt-4o-mini")
+- `endpoint`: API endpoint URL
+- `api_key`: Your API key
+- `temperature`: Generation temperature (0.0 for deterministic)
+
+### Retrieval Settings
+- `top_k`: Number of results to return
+- `max_iterations`: Maximum search iterations
+- `beam_width`: Beam width for multi-path search
+
+### Storage Settings
+- `workspace_dir`: Directory for persisted documents
+- `cache_size`: LRU cache size
+- `compression`: Enable/disable compression
+
+## Performance Tuning
+
+For faster retrieval:
+- Use a smaller model like gpt-4o-mini
+- Reduce max_iterations
+- Enable caching
+
+For higher accuracy:
+- Use a more capable model like gpt-4o
+- Increase beam_width
+- Enable multi-turn decomposition
+"""
+    ctx = IndexContext.from_content(content, name="system_docs", format="markdown")
+    doc_id = engine.index(ctx)
+    print(f"✓ Indexed: {doc_id}\n")
+
+    # Query examples
+    questions = [
+        "What are the main components?",
+        "How can I improve retrieval speed?",
+        "What settings are available?",
+    ]
+
+    for q in questions:
+        result = engine.query(doc_id, q)
+        print(f"Q: {q}")
+        print(f"A: {result.content[:150]}...")
+        print(f"   Score: {result.score:.2f}\n")
+
+    # Cleanup
+    engine.remove(doc_id)
+    print("✓ Cleaned up")
+
+    # Print configuration info
+    print("\n" + "=" * 60)
+    print("Configuration Priority")
+    print("=" * 60)
+    print("""
+1. Default configuration
+2. Auto-detected config file (vectorless.toml, config.toml)
+3. Explicit config file (config_path parameter)
+4. Environment variables (OPENAI_API_KEY, etc.)
+5. Constructor parameters (api_key, model, etc.)
+""")
+
+
+if __name__ == "__main__":
+    main()

examples/python/advanced/pyproject.toml

@@ -0,0 +1,11 @@
+[project]
+name = "vectorless-advanced-example"
+version = "0.1.0"
+requires-python = ">=3.9"
+dependencies = [
+    "vectorless",
+]
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"

examples/python/basic/README.md

@@ -0,0 +1,16 @@
+# Basic Example - Zero Configuration
+
+The simplest way to use Vectorless.
+
+## Setup
+
+```bash
+pip install vectorless
+export OPENAI_API_KEY="sk-..."
+```
+
+## Run
+
+```bash
+python main.py
+```