bfdb934443
- chunk_text(): char-boundary-safe text chunking at paragraph/sentence breaks (7000 char limit) - embed_memories_batch(): groups messages by room, batch-embeds all texts to reduce Qdrant calls - embed_issue_chunked(): auto-chunks long issue bodies - embed_skill(): upgraded with auto-chunking via chunk_text - TagEmbedInput struct for batch tag embedding - embed_tags_batch() / search_tags() with project isolation - ensure_collections() now creates embed_repo_tag collection
30 lines
1.1 KiB
Rust
30 lines
1.1 KiB
Rust
pub mod client;
|
|
pub mod qdrant;
|
|
pub mod service;
|
|
|
|
pub use client::{EmbedClient, EmbedPayload, EmbedVector, SearchResult};
|
|
pub use qdrant::QdrantClient;
|
|
pub use service::{EmbedMemoryInput, EmbedService, Embeddable, TagEmbedInput};
|
|
|
|
pub async fn new_embed_client(config: &config::AppConfig) -> crate::Result<EmbedClient> {
|
|
let base_url = config
|
|
.get_embed_model_base_url()
|
|
.map_err(|e| crate::AgentError::Internal(e.to_string()))?;
|
|
let api_key = config
|
|
.get_embed_model_api_key()
|
|
.map_err(|e| crate::AgentError::Internal(e.to_string()))?;
|
|
let qdrant_url = config
|
|
.get_qdrant_url()
|
|
.map_err(|e| crate::AgentError::Internal(e.to_string()))?;
|
|
let qdrant_api_key = config.get_qdrant_api_key();
|
|
|
|
let openai = rig::providers::openai::Client::builder()
|
|
.api_key(&api_key)
|
|
.base_url(&base_url)
|
|
.build()
|
|
.map_err(|e| crate::AgentError::Internal(format!("failed to build rig openai client: {}", e)))?;
|
|
|
|
let qdrant = QdrantClient::new(&qdrant_url, qdrant_api_key.as_deref()).await?;
|
|
Ok(EmbedClient::new(openai, qdrant))
|
|
}
|