Merge pull request #492 from spencerjibz/optimize-caching-code

⚡️ Improve page load by optimizing the caching code

Cargo.lock

@@ -4146,7 +4146,7 @@ checksum = "1778a42e8b3b90bff8d0f5032bf22250792889a5cdc752aa0020c84abe3aaf10"
 
 [[package]]
 name = "websurfx"
-version = "1.9.3"
+version = "1.9.4"
 dependencies = [
  "actix-cors",
  "actix-files",

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "websurfx"
-version = "1.9.3"
+version = "1.9.4"
 edition = "2021"
 description = "An open-source alternative to Searx that provides clean, ad-free, and organic results with incredible speed while keeping privacy and security in mind."
 repository = "https://github.com/neon-mmd/websurfx"

src/cache/cacher.rs

@@ -4,6 +4,7 @@
 use error_stack::Report;
 #[cfg(feature = "memory-cache")]
 use mini_moka::sync::Cache as MokaCache;
+use mini_moka::sync::ConcurrentCacheExt;
 
 #[cfg(feature = "memory-cache")]
 use std::time::Duration;
@@ -61,8 +62,8 @@ pub trait Cacher: Send + Sync {
     /// failure.
     async fn cache_results(
         &mut self,
-        search_results: &SearchResults,
-        url: &str,
+        search_results: &[SearchResults],
+        urls: &[String],
     ) -> Result<(), Report<CacheError>>;
 
     /// A helper function which computes the hash of the url and formats and returns it as string.
@@ -332,14 +333,33 @@ impl Cacher for RedisCache {
 
     async fn cache_results(
         &mut self,
-        search_results: &SearchResults,
-        url: &str,
+        search_results: &[SearchResults],
+        urls: &[String],
     ) -> Result<(), Report<CacheError>> {
         use base64::Engine;
-        let bytes = self.pre_process_search_results(search_results)?;
-        let base64_string = base64::engine::general_purpose::STANDARD_NO_PAD.encode(bytes);
-        let hashed_url_string = self.hash_url(url);
-        self.cache_json(&base64_string, &hashed_url_string).await
+
+        // size of search_results is expected to be equal to size of urls -> key/value pairs  for cache;
+        let search_results_len = search_results.len();
+
+        let mut bytes = Vec::with_capacity(search_results_len);
+
+        for result in search_results {
+            let processed = self.pre_process_search_results(result)?;
+            bytes.push(processed);
+        }
+
+        let base64_strings = bytes
+            .iter()
+            .map(|bytes_vec| base64::engine::general_purpose::STANDARD_NO_PAD.encode(bytes_vec));
+
+        let mut hashed_url_strings = Vec::with_capacity(search_results_len);
+
+        for url in urls {
+            let hash = self.hash_url(url);
+            hashed_url_strings.push(hash);
+        }
+        self.cache_json(base64_strings, hashed_url_strings.into_iter())
+            .await
     }
 }
 /// TryInto implementation for SearchResults from Vec<u8>
@@ -391,12 +411,16 @@ impl Cacher for InMemoryCache {
 
     async fn cache_results(
         &mut self,
-        search_results: &SearchResults,
-        url: &str,
+        search_results: &[SearchResults],
+        urls: &[String],
     ) -> Result<(), Report<CacheError>> {
+        for (url, search_result) in urls.iter().zip(search_results.iter()) {
             let hashed_url_string = self.hash_url(url);
-        let bytes = self.pre_process_search_results(search_results)?;
+            let bytes = self.pre_process_search_results(search_result)?;
             self.cache.insert(hashed_url_string, bytes);
+        }
+
+        self.cache.sync();
         Ok(())
     }
 }
@@ -434,11 +458,13 @@ impl Cacher for HybridCache {
 
     async fn cache_results(
         &mut self,
-        search_results: &SearchResults,
-        url: &str,
+        search_results: &[SearchResults],
+        urls: &[String],
     ) -> Result<(), Report<CacheError>> {
-        self.redis_cache.cache_results(search_results, url).await?;
-        self.memory_cache.cache_results(search_results, url).await?;
+        self.redis_cache.cache_results(search_results, urls).await?;
+        self.memory_cache
+            .cache_results(search_results, urls)
+            .await?;
 
         Ok(())
     }
@@ -460,8 +486,8 @@ impl Cacher for DisabledCache {
 
     async fn cache_results(
         &mut self,
-        _search_results: &SearchResults,
-        _url: &str,
+        _search_results: &[SearchResults],
+        _urls: &[String],
     ) -> Result<(), Report<CacheError>> {
         Ok(())
     }
@@ -519,11 +545,11 @@ impl SharedCache {
     /// on a failure.
     pub async fn cache_results(
         &self,
-        search_results: &SearchResults,
-        url: &str,
+        search_results: &[SearchResults],
+        urls: &[String],
     ) -> Result<(), Report<CacheError>> {
         let mut mut_cache = self.cache.lock().await;
-        mut_cache.cache_results(search_results, url).await
+        mut_cache.cache_results(search_results, urls).await
     }
 }
 

src/cache/redis_cacher.rs

@@ -118,14 +118,18 @@ impl RedisCache {
     /// on a failure.
     pub async fn cache_json(
         &mut self,
-        json_results: &str,
-        key: &str,
+        json_results: impl Iterator<Item = String>,
+        keys: impl Iterator<Item = String>,
     ) -> Result<(), Report<CacheError>> {
         self.current_connection = Default::default();
+        let mut pipeline = redis::Pipeline::with_capacity(3);
 
-        let mut result: Result<(), RedisError> = self.connection_pool
-            [self.current_connection as usize]
-            .set_ex(key, json_results, self.cache_ttl.into())
+        for (key, json_result) in keys.zip(json_results) {
+            pipeline.set_ex(key, json_result, self.cache_ttl.into());
+        }
+
+        let mut result: Result<(), RedisError> = pipeline
+            .query_async(&mut self.connection_pool[self.current_connection as usize])
             .await;
 
         // Code to check whether the current connection being used is dropped with connection error
@@ -145,8 +149,10 @@ impl RedisCache {
                                 CacheError::PoolExhaustionWithConnectionDropError,
                             ));
                         }
-                        result = self.connection_pool[self.current_connection as usize]
-                            .set_ex(key, json_results, 60)
+                        result = pipeline
+                            .query_async(
+                                &mut self.connection_pool[self.current_connection as usize],
+                            )
                             .await;
                         continue;
                     }

src/server/routes/search.rs

@@ -40,6 +40,7 @@ pub async fn search(
     config: web::Data<Config>,
     cache: web::Data<SharedCache>,
 ) -> Result<HttpResponse, Box<dyn std::error::Error>> {
+    use std::sync::Arc;
     let params = web::Query::<SearchParams>::from_query(req.query_string())?;
     match &params.q {
         Some(query) => {
@@ -79,12 +80,50 @@ pub async fn search(
 
             // .max(1) makes sure that the page >= 0.
             let page = params.page.unwrap_or(1).max(1) - 1;
+            let previous_page = page.saturating_sub(1);
+            let next_page = page + 1;
 
-            let (_, results, _) = join!(
-                get_results(page.saturating_sub(1)),
+            let mut results = Arc::new((SearchResults::default(), String::default()));
+            if page != previous_page {
+                let (previous_results, current_results, next_results) = join!(
+                    get_results(previous_page),
                     get_results(page),
-                get_results(page + 1)
+                    get_results(next_page)
                 );
+                let (parsed_previous_results, parsed_next_results) =
+                    (previous_results?, next_results?);
+
+                let (cache_keys, results_list) = (
+                    [
+                        parsed_previous_results.1,
+                        results.1.clone(),
+                        parsed_next_results.1,
+                    ],
+                    [
+                        parsed_previous_results.0,
+                        results.0.clone(),
+                        parsed_next_results.0,
+                    ],
+                );
+
+                results = Arc::new(current_results?);
+
+                tokio::spawn(async move { cache.cache_results(&results_list, &cache_keys).await });
+            } else {
+                let (current_results, next_results) =
+                    join!(get_results(page), get_results(page + 1));
+
+                let parsed_next_results = next_results?;
+
+                results = Arc::new(current_results?);
+
+                let (cache_keys, results_list) = (
+                    [results.1.clone(), parsed_next_results.1.clone()],
+                    [results.0.clone(), parsed_next_results.0],
+                );
+
+                tokio::spawn(async move { cache.cache_results(&results_list, &cache_keys).await });
+            }
 
             Ok(HttpResponse::Ok().content_type(ContentType::html()).body(
                 crate::templates::views::search::search(
@@ -92,7 +131,7 @@ pub async fn search(
                     &config.style.theme,
                     &config.style.animation,
                     query,
-                    &results?,
+                    &results.0,
                 )
                 .0,
             ))
@@ -124,7 +163,7 @@ async fn results(
     query: &str,
     page: u32,
     search_settings: &server_models::Cookie<'_>,
-) -> Result<SearchResults, Box<dyn std::error::Error>> {
+) -> Result<(SearchResults, String), Box<dyn std::error::Error>> {
     // eagerly parse cookie value to evaluate safe search level
     let safe_search_level = search_settings.safe_search_level;
 
@@ -143,7 +182,7 @@ async fn results(
     // check if fetched cache results was indeed fetched or it was an error and if so
     // handle the data accordingly.
     match cached_results {
-        Ok(results) => Ok(results),
+        Ok(results) => Ok((results, cache_key)),
         Err(_) => {
             if safe_search_level == 4 {
                 let mut results: SearchResults = SearchResults::default();
@@ -153,9 +192,11 @@ async fn results(
                 // Return early when query contains disallowed words,
                 if flag {
                     results.set_disallowed();
-                    cache.cache_results(&results, &cache_key).await?;
+                    cache
+                        .cache_results(&[results.clone()], &[cache_key.clone()])
+                        .await?;
                     results.set_safe_search_level(safe_search_level);
-                    return Ok(results);
+                    return Ok((results, cache_key));
                 }
             }
 
@@ -173,7 +214,7 @@ async fn results(
                         &search_settings
                             .engines
                             .iter()
-                            .filter_map(|engine| EngineHandler::new(&engine).ok())
+                            .filter_map(|engine| EngineHandler::new(engine).ok())
                             .collect::<Vec<EngineHandler>>(),
                         config.request_timeout,
                         safe_search_level,
@@ -192,9 +233,11 @@ async fn results(
             {
                 results.set_filtered();
             }
-            cache.cache_results(&results, &cache_key).await?;
+            cache
+                .cache_results(&[results.clone()], &[cache_key.clone()])
+                .await?;
             results.set_safe_search_level(safe_search_level);
-            Ok(results)
+            Ok((results, cache_key))
         }
     }
 }