dnrops
/
rig
mirror of https://github.com/0xplaygrounds/rig
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix: Update documentation links and optimize document validation logic 

* Fixed Aliyun embedding API documentation link
* Introduced AVG_CHARS_PER_TOKEN constant to improve code readability
* Replaced println with tracing::info to improve logging
This commit is contained in:
tommy 2025-04-19 09:52:25 +08:00
parent b84fcb1af1
commit c97b0c7548
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
rig-core/src/providers/aliyun/embedding.rs
View File

@ -1,7 +1,7 @@
// ================================================================ // ================================================================
//! Aliyun Embedding API Integration //! Aliyun Embedding API Integration
//! Implementation of Aliyun embedding models for text vectorization //! Implementation of Aliyun embedding models for text vectorization
//! From https://help.aliyun.com/zh/model-studio/developer-reference/text-embedding-synchronous-api?spm=a2c4g.11186623.help-menu-2400256.d_2_5_0.58654308woEHy7&scm=20140722.H_2712515._.OR_help-T_cn~zh-V_1#4c08e887c12ru //! From https://help.aliyun.com/zh/model-studio/developer-reference/text-embedding-synchronous-api
// ================================================================ // ================================================================
use serde_json::json; use serde_json::json;
@ -74,6 +74,8 @@ impl EmbeddingModel {
/// * `Ok(())` if validation passes /// * `Ok(())` if validation passes
/// * `Err(EmbeddingError)` with appropriate error message if validation fails /// * `Err(EmbeddingError)` with appropriate error message if validation fails
fn validate_documents(&self, documents: &[String]) -> Result<(), EmbeddingError> { fn validate_documents(&self, documents: &[String]) -> Result<(), EmbeddingError> {
const AVG_CHARS_PER_TOKEN: usize = 4;
if documents.len() > self.max_documents() { if documents.len() > self.max_documents() {
return Err(EmbeddingError::ProviderError(format!( return Err(EmbeddingError::ProviderError(format!(
"Model {} supports maximum {} documents", "Model {} supports maximum {} documents",
@ -83,7 +85,7 @@ impl EmbeddingModel {
} }
for (i, doc) in documents.iter().enumerate() { for (i, doc) in documents.iter().enumerate() {
let estimated_tokens = doc.len() / 4; // Rough estimate: average 4 characters per token let estimated_tokens = doc.len() / AVG_CHARS_PER_TOKEN;
if estimated_tokens > self.max_tokens() { if estimated_tokens > self.max_tokens() {
return Err(EmbeddingError::ProviderError(format!( return Err(EmbeddingError::ProviderError(format!(
"Document #{} exceeds maximum token limit of {}", "Document #{} exceeds maximum token limit of {}",
@ -142,7 +144,7 @@ impl embeddings::EmbeddingModel for EmbeddingModel {
"encoding_format": "float", "encoding_format": "float",
}); });
println!("{}", serde_json::to_string_pretty(&request).unwrap()); tracing::info!("{}", serde_json::to_string_pretty(&request).unwrap());
let response = self let response = self
.client .client