tidying stuff up, dto for search

common/src/utils/config.rs

@@ -2,6 +2,19 @@ use config::{Config, ConfigError, Environment, File};
 use serde::Deserialize;
 use std::env;
 
+/// Selects the embedding backend for vector generation.
+#[derive(Clone, Deserialize, Debug, Default, PartialEq)]
+#[serde(rename_all = "lowercase")]
+pub enum EmbeddingBackend {
+    /// Use OpenAI-compatible API for embeddings.
+    OpenAI,
+    /// Use FastEmbed local embeddings (default).
+    #[default]
+    FastEmbed,
+    /// Use deterministic hashed embeddings (for testing).
+    Hashed,
+}
+
 #[derive(Clone, Deserialize, Debug, PartialEq)]
 #[serde(rename_all = "lowercase")]
 pub enum StorageKind {
@@ -60,6 +73,8 @@ pub struct AppConfig {
     pub fastembed_max_length: Option<usize>,
     #[serde(default)]
     pub retrieval_strategy: Option<String>,
+    #[serde(default)]
+    pub embedding_backend: EmbeddingBackend,
 }
 
 /// Default data directory for persisted assets.
@@ -127,6 +142,7 @@ impl Default for AppConfig {
             fastembed_show_download_progress: None,
             fastembed_max_length: None,
             retrieval_strategy: None,
+            embedding_backend: EmbeddingBackend::default(),
         }
     }
 }

common/src/utils/embedding.rs

@@ -235,6 +235,34 @@ impl EmbeddingProvider {
             },
         })
     }
+
+    /// Creates an embedding provider based on application configuration.
+    ///
+    /// Dispatches to the appropriate constructor based on `config.embedding_backend`:
+    /// - `OpenAI`: Requires a valid OpenAI client
+    /// - `FastEmbed`: Uses local embedding model
+    /// - `Hashed`: Uses deterministic hashed embeddings (for testing)
+    pub async fn from_config(
+        config: &crate::utils::config::AppConfig,
+        openai_client: Option<Arc<Client<async_openai::config::OpenAIConfig>>>,
+    ) -> Result<Self> {
+        use crate::utils::config::EmbeddingBackend;
+
+        match config.embedding_backend {
+            EmbeddingBackend::OpenAI => {
+                let client = openai_client.ok_or_else(|| {
+                    anyhow!("OpenAI embedding backend requires an OpenAI client")
+                })?;
+                // Use defaults that match SystemSettings initial values
+                Self::new_openai(client, "text-embedding-3-small".to_string(), 1536)
+            }
+            EmbeddingBackend::FastEmbed => {
+                // Use nomic-embed-text-v1.5 as the default FastEmbed model
+                Self::new_fastembed(Some("nomic-ai/nomic-embed-text-v1.5".to_string())).await
+            }
+            EmbeddingBackend::Hashed => Self::new_hashed(384),
+        }
+    }
 }
 
 // Helper functions for hashed embeddings