Merge branch 'rolling' into patch-searx-results-on-first-page

docs/configuration.md

@@ -29,6 +29,7 @@ Some of the configuration options provided in the file are stated below. These a
 - **port:** Port number on which server should be launched.
 - **binding_ip_addr:** IP address on the which server should be launched.
 - **production_use:** Whether to use production mode or not (in other words this option should be used if it is to be used to host it on the server to provide a service to a large number of users). If production_use is set to true. There will be a random delay before sending the request to the search engines, this is to prevent DDoSing the upstream search engines from a large number of simultaneous requests. This is newly added option and hence is only available in the **edge version**.
+- **request_timeout:** Timeout for the search requests sent to the upstream search engines to be fetched (value in seconds).
 
 ## Website
 

docs/installation.md

@@ -86,6 +86,7 @@ binding_ip_addr = "0.0.0.0" --ip address on the which server should be launched.
 production_use = false -- whether to use production mode or not (in other words this option should be used if it is to be used to host it on the server to provide a service to a large number of users)
 -- if production_use is set to true
 -- There will be a random delay before sending the request to the search engines, this is to prevent DDoSing the upstream search engines from a large number of simultaneous requests.
+request_timeout = 60 -- timeout for the search requests sent to the upstream search engines to be fetched (value in seconds).
 
 -- ### Website ###
 -- The different colorschemes provided are:

src/config/parser.rs

@@ -22,6 +22,7 @@ static CONFIG_FILE_NAME: &str = "config.lua";
 /// * `logging` - It stores the option to whether enable or disable logs.
 /// * `debug` - It stores the option to whether enable or disable debug mode.
 /// * `upstream_search_engines` - It stores all the engine names that were enabled by the user.
+/// * `request_timeout` - It stores the time (secs) which controls the server request timeout.
 #[derive(Clone)]
 pub struct Config {
     pub port: u16,
@@ -32,6 +33,7 @@ pub struct Config {
     pub logging: bool,
     pub debug: bool,
     pub upstream_search_engines: Vec<String>,
+    pub request_timeout: u8,
 }
 
 /// Configuration options for the aggregator.
@@ -80,6 +82,7 @@ impl Config {
                     .into_iter()
                     .filter_map(|(key, value)| value.then_some(key))
                     .collect(),
+                request_timeout: globals.get::<_, u8>("request_timeout")?,
             })
         })
     }

src/engines/duckduckgo.rs

@@ -29,6 +29,7 @@ impl SearchEngine for DuckDuckGo {
     /// * `query` - Takes the user provided query to query to the upstream search engine with.
     /// * `page` - Takes an u32 as an argument.
     /// * `user_agent` - Takes a random user agent string as an argument.
+    /// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
     ///
     /// # Errors
     ///
@@ -41,6 +42,7 @@ impl SearchEngine for DuckDuckGo {
         query: String,
         page: u32,
         user_agent: String,
+        request_timeout: u8,
     ) -> Result<HashMap<String, RawSearchResult>, EngineError> {
         // Page number can be missing or empty string and so appropriate handling is required
         // so that upstream server recieves valid page number.
@@ -90,7 +92,7 @@ impl SearchEngine for DuckDuckGo {
         );
 
         let document: Html = Html::parse_document(
-            &DuckDuckGo::fetch_html_from_upstream(self, url, header_map).await?,
+            &DuckDuckGo::fetch_html_from_upstream(self, url, header_map, request_timeout).await?,
         );
 
         let no_result: Selector = Selector::parse(".no-results")

src/engines/engine_models.rs

@@ -50,11 +50,12 @@ pub trait SearchEngine {
         &self,
         url: String,
         header_map: reqwest::header::HeaderMap,
+        request_timeout: u8,
     ) -> Result<String, EngineError> {
         // fetch the html from upstream search engine
         Ok(reqwest::Client::new()
             .get(url)
-            .timeout(Duration::from_secs(30)) // Add timeout to request to avoid DDOSing the server
+            .timeout(Duration::from_secs(request_timeout as u64)) // Add timeout to request to avoid DDOSing the server
             .headers(header_map) // add spoofed headers to emulate human behaviour
             .send()
             .await
@@ -71,5 +72,6 @@ pub trait SearchEngine {
         query: String,
         page: u32,
         user_agent: String,
+        request_timeout: u8,
     ) -> Result<HashMap<String, RawSearchResult>, EngineError>;
 }

src/engines/searx.rs

@@ -27,6 +27,7 @@ impl SearchEngine for Searx {
     /// * `query` - Takes the user provided query to query to the upstream search engine with.
     /// * `page` - Takes an u32 as an argument.
     /// * `user_agent` - Takes a random user agent string as an argument.
+    /// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
     ///
     /// # Errors
     ///
@@ -40,6 +41,7 @@ impl SearchEngine for Searx {
         query: String,
         page: u32,
         user_agent: String,
+        request_timeout: u8,
     ) -> Result<HashMap<String, RawSearchResult>, EngineError> {
         // Page number can be missing or empty string and so appropriate handling is required
         // so that upstream server recieves valid page number.
@@ -73,8 +75,9 @@ impl SearchEngine for Searx {
         );
         header_map.insert(COOKIE, "categories=general; language=auto; locale=en; autocomplete=duckduckgo; image_proxy=1; method=POST; safesearch=2; theme=simple; results_on_new_tab=1; doi_resolver=oadoi.org; simple_style=auto; center_alignment=1; query_in_title=1; infinite_scroll=0; disabled_engines=; enabled_engines=\"archive is__general\\054yep__general\\054curlie__general\\054currency__general\\054ddg definitions__general\\054wikidata__general\\054duckduckgo__general\\054tineye__general\\054lingva__general\\054startpage__general\\054yahoo__general\\054wiby__general\\054marginalia__general\\054alexandria__general\\054wikibooks__general\\054wikiquote__general\\054wikisource__general\\054wikiversity__general\\054wikivoyage__general\\054dictzone__general\\054seznam__general\\054mojeek__general\\054naver__general\\054wikimini__general\\054brave__general\\054petalsearch__general\\054goo__general\"; disabled_plugins=; enabled_plugins=\"searx.plugins.hostname_replace\\054searx.plugins.oa_doi_rewrite\\054searx.plugins.vim_hotkeys\"; tokens=; maintab=on; enginetab=on".parse().into_report().change_context(EngineError::UnexpectedError)?);
 
-        let document: Html =
-            Html::parse_document(&Searx::fetch_html_from_upstream(self, url, header_map).await?);
+        let document: Html = Html::parse_document(
+            &Searx::fetch_html_from_upstream(self, url, header_map, request_timeout).await?,
+        );
 
         let no_result: Selector = Selector::parse("#urls>.dialog-error>p")
             .map_err(|_| Report::new(EngineError::UnexpectedError))

src/results/aggregator.rs

@@ -51,6 +51,7 @@ type FutureVec = Vec<JoinHandle<Result<HashMap<String, RawSearchResult>, Report<
 /// * `random_delay` - Accepts a boolean value to add a random delay before making the request.
 /// * `debug` - Accepts a boolean value to enable or disable debug mode option.
 /// * `upstream_search_engines` - Accepts a vector of search engine names which was selected by the
+/// * `request_timeout` - Accepts a time (secs) as a value which controls the server request timeout.
 /// user through the UI or the config file.
 ///
 /// # Error
@@ -64,6 +65,7 @@ pub async fn aggregate(
     random_delay: bool,
     debug: bool,
     upstream_search_engines: Vec<String>,
+    request_timeout: u8,
 ) -> Result<SearchResults, Box<dyn std::error::Error>> {
     let user_agent: String = random_user_agent();
     let mut result_map: HashMap<String, RawSearchResult> = HashMap::new();
@@ -92,9 +94,11 @@ pub async fn aggregate(
         .map(|search_engine| {
             let query: String = query.clone();
             let user_agent: String = user_agent.clone();
-            tokio::spawn(
-                async move { search_engine.results(query, page, user_agent.clone()).await },
-            )
+            tokio::spawn(async move {
+                search_engine
+                    .results(query, page, user_agent.clone(), request_timeout)
+                    .await
+            })
         })
         .collect();
 

src/server/routes.rs

@@ -146,6 +146,7 @@ async fn results(
                         config.aggregator.random_delay,
                         config.debug,
                         cookie_value.engines,
+                        config.request_timeout,
                     )
                     .await?
                 }
@@ -156,6 +157,7 @@ async fn results(
                         config.aggregator.random_delay,
                         config.debug,
                         config.upstream_search_engines.clone(),
+                        config.request_timeout,
                     )
                     .await?
                 }

websurfx/config.lua

@@ -8,6 +8,7 @@ binding_ip = "127.0.0.1" --ip address on the which server should be launched.
 production_use = false -- whether to use production mode or not (in other words this option should be used if it is to be used to host it on the server to provide a service to a large number of users (more than one))
 -- if production_use is set to true
 -- There will be a random delay before sending the request to the search engines, this is to prevent DDoSing the upstream search engines from a large number of simultaneous requests.
+request_timeout = 30 -- timeout for the search requests sent to the upstream search engines to be fetched (value in seconds).
 
 -- ### Website ###
 -- The different colorschemes provided are: