Trait BaseDocumentCompressor 

pub trait BaseDocumentCompressor: Send + Sync {
    // Required method
    fn compress_documents<'life0, 'life1, 'async_trait>(
        &'life0 self,
        documents: Vec<Document>,
        query: &'life1 str,
        callbacks: Option<Callbacks>,
    ) -> Pin<Box<dyn Future<Output = Result<Vec<Document>, Box<dyn Error + Send + Sync>>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait,
             'life1: 'async_trait;

    // Provided method
    fn compress_documents_sync(
        &self,
        documents: Vec<Document>,
        query: &str,
        callbacks: Option<Callbacks>,
    ) -> Result<Vec<Document>, Box<dyn Error + Send + Sync>>
       where Self: Sized { ... }
}

Base trait for document compressors.

This abstraction is primarily used for post-processing of retrieved documents.

Document objects matching a given query are first retrieved. Then the list of documents can be further processed.

For example, one could re-rank the retrieved documents using an LLM.

Users should favor using a RunnableLambda instead of implementing this trait directly when possible.

Example

use agent_chain_core::documents::{BaseDocumentCompressor, Document};
use agent_chain_core::callbacks::Callbacks;
use async_trait::async_trait;

struct MyCompressor {
    threshold: f64,
}

#[async_trait]
impl BaseDocumentCompressor for MyCompressor {
    async fn compress_documents(
        &self,
        documents: Vec<Document>,
        query: &str,
        _callbacks: Option<Callbacks>,
    ) -> Result<Vec<Document>, Box<dyn std::error::Error + Send + Sync>> {
        // Filter documents based on some criteria
        Ok(documents
            .into_iter()
            .filter(|doc| doc.page_content.contains(query))
            .collect())
    }
}

Required Methods

fn compress_documents<'life0, 'life1, 'async_trait>( &'life0 self, documents: Vec<Document>, query: &'life1 str, callbacks: Option<Callbacks>, ) -> Pin<Box<dyn Future<Output = Result<Vec<Document>, Box<dyn Error + Send + Sync>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Compress retrieved documents given the query context.

Arguments

• documents - The retrieved Document objects.
• query - The query context.
• callbacks - Optional Callbacks to run during compression.

Returns

The compressed documents, or an error if compression fails.

Provided Methods

fn compress_documents_sync( &self, documents: Vec<Document>, query: &str, callbacks: Option<Callbacks>, ) -> Result<Vec<Document>, Box<dyn Error + Send + Sync>>

where Self: Sized,

Synchronously compress retrieved documents given the query context.

This is a blocking version of compress_documents. The default implementation runs the async version using a blocking runtime.

Arguments

• documents - The retrieved Document objects.
• query - The query context.
• callbacks - Optional Callbacks to run during compression.

Returns

The compressed documents, or an error if compression fails.

Implementors