ragrig 0.9.10

//! Trait-driven RAG framework with runtime hot-swapping.
//!
//! **Zero native dependencies in the default build.**  `cargo build --release`
//! produces a pure-Rust binary that talks to a local Ollama server for models.
//! No C++ compiler, no `cmake`, no `protoc` required.
//!
//! # Architecture
//!
//! Every pipeline stage is a trait object — swap any agent at runtime
//! without losing your document index or conversation context:
//!
//! | Stage | Trait | Built-in backends |
//! |---|---|---|
//! | Chat | [`agents::Generator`] | Ollama, DeepSeek |
//! | Memory / Rewrite | [`agents::Generator`] | Ollama, DeepSeek |
//! | Ranking | [`store::Ranker`] | RRF fusion, weighted linear, MMR diversity, LLM re-rank |
//! | Embeddings | [`embed::Embedder`] | Ollama, Fastembed (CPU-only), No-op |
//! | Storage | [`store::VectorStore`] | Brute-force (MessagePack), LanceDB |
//! | Parsing | [`parsers::DocumentParser`] | PDF × 5 (incl. vision-VLM), EPUB, DOCX, HTML, Markdown |
//! | Chunking | [`chunker::Chunker`] | Markdown, token, chunkedrs (×4), text-splitter, cognigraph |
//!
//! Every stored chunk records which (parser, chunker, embedder) pipeline
//! produced it ([`PipelineProvenance`]).  Index several pipelines into one
//! store and switch between them at query time with
//! [`PipelineFilter`] / [`RagAgent::set_search_filter`].
//!
//! # Quick Start
//!
//! ```rust,no_run
//! use ragrig::{
//!     ChunkConfig, MarkdownChunker,
//!     embed::EmbedderSpec,
//!     agents::ChatAgentSpec,
//!     parsers::{DocumentParsers, build_parsers},
//!     store::open_store,
//!     vector::{collect_documents, search_similar},
//! };
//! use std::path::Path;
//!
//! # async fn example() -> anyhow::Result<()> {
//! // Build agents from spec enums — swap backends by changing the variant.
//! let embedder = EmbedderSpec::Ollama { model: "nomic-embed-text:latest".into(), request_timeout_secs: None }.build()?;
//! let chat = ChatAgentSpec::Ollama { model: "gemma2:latest".into(), params: Default::default(), request_timeout_secs: None }.build()?;
//!
//! // Open or create the vector store.
//! let folder = Path::new("./my_docs");
//! let store = open_store(folder).await?;
//!
//! // Parse PDFs/EPUBs/DOCXs and index them with the chunker of your choice.
//! let parsers = DocumentParsers::new(build_parsers());
//! let cfg = ChunkConfig::default(); // 1024 tokens, 128 overlap
//! collect_documents(&*embedder, &parsers, &MarkdownChunker, folder, &cfg, &*store).await?;
//!
//! // Search and generate.
//! let results = search_similar(&*embedder, 5, 0.0, &*store, "quantum entanglement").await?;
//! chat.generate("Summarise the following:\n\n...").await?;
//! # Ok(())
//! # }
//! ```
//!
//! See the [repository README](https://github.com/schmettow/ragrig) for the
//! full guide, including the REPL, session persistence, and hot-swap commands.
//!
//! # Feature Flags
//!
//! | Flag | Default | Adds |
//! |---|---|---|
//! | `ollama-embed` | **on** | Embeddings via local Ollama server |
//! | `internal` | **on** | Pure-Rust brute-force vector store |
//! | `internal-embed` | off | Fastembed CPU-only embeddings (requires C compiler) |
//! | `lancedb` | off | LanceDB hybrid vector store (requires protoc, cmake) |
//! | `test-fixtures` | off | Embedded test fixtures for downstream crates |

#![deny(missing_docs)]

pub mod agent;
pub mod agents;
pub mod attach;
pub mod chunker;
pub mod documents;
pub mod embed;
pub mod error;
#[cfg(feature = "test-fixtures")]
pub mod fixtures;
pub mod fs_session_store;
#[cfg(feature = "internal-generate")]
pub mod generate;
pub mod longterm_memory;
pub mod memory;
pub mod parsers;
pub mod prompts;
pub mod store;
pub mod types;
pub mod vector;
mod web;

// --- Re-export all public types ---

pub use types::{
    AttachedDocument, ChatConfig, ChunkConfig, ChunkMeta, ContextSizeMode, DocumentType,
    EmbedConfig, EpubParserBackend, GenerationParams, IndexManifest, MemoryConfig,
    PaperResult, ParseConfig, PdfParserBackend, PipelineFilter, PipelineProvenance,
    RagrigConfig,
};

pub use chunker::{
    ChunkedText, ChunkedrsChunker, ChunkedrsMode, Chunker, MarkdownChunker, TokenChunker,
    available_chunkers,
};
#[cfg(feature = "chunker-textsplitter")]
pub use chunker::{TextSplitterChunker, TextSplitterMode};
#[cfg(feature = "chunker-cognigraph")]
pub use chunker::CognigraphChunker;

pub use attach::{
    AppendAttach, AttachStrategy, NoopAttach, PrependAttach, ReplaceAttach,
};

pub use agents::{ChatAgentSpec, Generator, MutexGenerator, SimpleGenerator};

#[cfg(feature = "internal-embed")]
pub use embed::FastembedEmbedder;
pub use embed::{Embedder, EmbedderSpec, EmbeddingMetadata, NoopEmbedder, OllamaEmbedder};

pub use parsers::{
    DocumentParser, DocumentParsers, VisionPdfParser, build_parsers, build_parsers_without_vision,
    chunk_text, extract_text,
};
#[cfg(feature = "kreuzberg")]
pub use parsers::KreuzbergParser;

pub use fs_session_store::FsSessionStore;
pub use longterm_memory::{
    HistoryStrategy, LogHistory, MemoryStrategyKind, SessionConfig, SessionData, SessionId,
    SessionStore, SummaryHistory, Turn, TurnPairs, TurnPerf, TurnRole,
};

pub use memory::MemoryStrategy;
pub use memory::RewriteMemory;
pub use memory::TranscriptMemory;

pub use prompts::SystemPrompts;

pub use agent::{RagAgent, RagResponse};

pub use store::{
    HybridRrfRanker, LlmReranker, MmrDiversityRanker, Ranker, ScoredChunk, VectorStore,
    WeightedFusionRanker,
};

pub use error::RagrigError;

pub use documents::FileIndexResult;

pub use vector::{
    collect_documents, collect_documents_with_stats, embed_documents, index_folder,
    index_folder_with_chunker, scan_document_files, search_by_document,
};

pub use web::download_and_ingest_url;
pub use web::DEFAULT_MAX_DOWNLOAD_BYTES;