Merge branch 'rolling' into qwant

.mergify.yml

@@ -1,13 +1,14 @@
-pull_request_rules:
+queue_rules:
-  - name: Automatic merge on approval
+  - name: default
-    conditions:
+    queue_conditions:
       - "#approved-reviews-by>=2"
       - check-success=build (stable)
       - check-success=CodeFactor
       - check-success=Rust project
-    actions:
+    merge_conditions: []
-      queue:
+    merge_method: squash
-        method: squash
+
+pull_request_rules:
   - name: automatic update of pull requests where more 5 commits behind
     conditions:
       - "#commits-behind>5"
@@ -17,4 +18,8 @@ pull_request_rules:
     conditions:
       - merged
     actions:
-      delete_head_branch: {}
+      delete_head_branch: {}
+  - name: Automatic merge on approval
+    conditions: []
+    actions:
+      queue:

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "websurfx"
-version = "1.17.22"
+version = "1.18.0"
 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"
@@ -19,20 +19,21 @@ reqwest = { version = "0.12.5", default-features = false, features = [
     "gzip",
     "http2",
     "json"
+    "socks",
 ] }
-tokio = { version = "1.32.0", features = [
+tokio = { version = "1.41.0", features = [
     "rt-multi-thread",
     "macros",
     "fs",
     "io-util",
 ], default-features = false }
-serde = { version = "1.0.209", default-features = false, features = ["derive"] }
+serde = { version = "1.0.215", default-features = false, features = ["derive"] }
 serde_json = { version = "1.0.122", default-features = false }
 bincode = {version="1.3.3", default-features=false}
 maud = { version = "0.26.0", default-features = false, features = [
     "actix-web",
 ] }
-scraper = { version = "0.20.0", default-features = false }
+scraper = { version = "0.21.0", default-features = false }
 actix-web = { version = "4.9.0", features = [
     "cookies",
     "macros",
@@ -47,22 +48,22 @@ mlua = { version = "0.9.9", features = [
     "luajit",
     "vendored",
 ], default-features = false }
-redis = { version = "0.27.2", features = [
+redis = { version = "0.27.5", features = [
     "tokio-comp",
     "connection-manager",
     "tcp_nodelay"
 ], default-features = false, optional = true }
-blake3 = { version = "1.5.4", default-features = false }
+blake3 = { version = "1.5.5", default-features = false }
 error-stack = { version = "0.5.0", default-features = false, features = [
     "std",
 ] }
 async-trait = { version = "0.1.80", default-features = false }
-regex = { version = "1.11.0", features = ["perf"], default-features = false }
+regex = { version = "1.11.1", features = ["perf"], default-features = false }
-futures = { version = "0.3.30", default-features = false, features = ["alloc"] }
+futures = { version = "0.3.31", default-features = false, features = ["alloc"] }
 dhat = { version = "0.3.2", optional = true, default-features = false }
 mimalloc = { version = "0.1.43", default-features = false }
 async-once-cell = { version = "0.5.3", default-features = false }
-actix-governor = { version = "0.6.0", default-features = false }
+actix-governor = { version = "0.7.0", default-features = false }
 moka = { version = "0.12.8", optional = true, default-features = false, features = [
     "future",
 ] }
@@ -79,7 +80,7 @@ base64 = { version = "0.21.5", default-features = false, features = [
     "std",
 ], optional = true }
 cfg-if = { version = "1.0.0", default-features = false, optional = true }
-keyword_extraction = { version = "1.4.3", default-features = false, features = [
+keyword_extraction = { version = "1.5.0", default-features = false, features = [
     "tf_idf",
     "rayon",
 ] }
@@ -92,7 +93,7 @@ itertools = {version = "0.13.0", default-features = false}
 [dev-dependencies]
 rusty-hook = { version = "^0.11.2", default-features = false }
 criterion = { version = "0.5.1", default-features = false }
-tempfile = { version = "3.13.0", default-features = false }
+tempfile = { version = "3.14.0", default-features = false }
 
 [build-dependencies]
 lightningcss = { version = "1.0.0-alpha.57", default-features = false, features = [

public/images/close.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="feather feather-x"><line x1="18" y1="6" x2="6" y2="18"></line><line x1="6" y1="6" x2="18" y2="18"></line></svg>

public/static/index.js

@@ -1,34 +1,6 @@
 /**
- * Selects the input element for the search box
+* A function that clears the search input text when the clear button is clicked.
- * @type {HTMLInputElement}
+*/
- */
+function clearSearchText() {
-const searchBox = document.querySelector('input')
+    document.querySelector('.search_bar > input').value = ''
-
-/**
- * Redirects the user to the search results page with the query parameter
- */
-function searchWeb() {
-    const query = searchBox.value.trim()
-    try {
-        let safeSearchLevel = document.querySelector('.search_options select').value
-        if (query) {
-            window.location.href = `search?q=${encodeURIComponent(
-                query,
-            )}&safesearch=${encodeURIComponent(safeSearchLevel)}`
-        }
-    } catch (error) {
-        if (query) {
-            window.location.href = `search?q=${encodeURIComponent(query)}`
-        }
-    }
 }
-
-/**
- * Listens for the 'Enter' key press event on the search box and calls the searchWeb function
- * @param {KeyboardEvent} e - The keyboard event object
- */
-searchBox.addEventListener('keyup', (e) => {
-    if (e.key === 'Enter') {
-        searchWeb()
-    }
-})

public/static/pagination.js

@@ -1,39 +0,0 @@
-/**
- * Navigates to the next page by incrementing the current page number in the URL query string.
- * @returns {void}
- */
-function navigate_forward() {
-    let url = new URL(window.location);
-    let searchParams = url.searchParams;
-
-    let q = searchParams.get('q');
-    let page = parseInt(searchParams.get('page'));
-
-    if (isNaN(page)) {
-        page = 1;
-    } else {
-        page++;
-    }
-
-    window.location.href = `${url.origin}${url.pathname}?q=${encodeURIComponent(q)}&page=${page}`;
-}
-
-/**
- * Navigates to the previous page by decrementing the current page number in the URL query string.
- * @returns {void}
- */
-function navigate_backward() {
-    let url = new URL(window.location);
-    let searchParams = url.searchParams;
-
-    let q = searchParams.get('q');
-    let page = parseInt(searchParams.get('page'));
-
-    if (isNaN(page)) {
-        page = 0;
-    } else if (page > 0) {
-        page--;
-    }
-
-    window.location.href = `${url.origin}${url.pathname}?q=${encodeURIComponent(q)}&page=${page}`;
-}

public/static/search_area_options.js

@@ -1,18 +0,0 @@
-document.addEventListener(
-    'DOMContentLoaded',
-    () => {
-        let url = new URL(window.location)
-        let searchParams = url.searchParams
-
-        let safeSearchLevel = searchParams.get('safesearch')
-
-        if (
-            safeSearchLevel >= 0 &&
-            safeSearchLevel <= 2 &&
-            safeSearchLevel !== null
-        ) {
-            document.querySelector('.search_options select').value = safeSearchLevel
-        }
-    },
-    false,
-)

public/static/themes/simple.css

@@ -73,6 +73,11 @@ button {
   font-size: 1.6rem;
 }
 
+.search_bar input::-webkit-search-results-button,
+.search_bar input::-webkit-search-cancel-button{
+    display: none;
+}
+
 .search_bar input:focus {
   outline: 2px solid var(--foreground-color);
 }
@@ -443,7 +448,7 @@ footer div {
   align-items: center;
 }
 
-.page_navigation button {
+.page_navigation a {
   background: var(--background-color);
   color: var(--foreground-color);
   padding: 1rem;
@@ -452,7 +457,7 @@ footer div {
   border: none;
 }
 
-.page_navigation button:active {
+.page_navigation a:active {
   filter: brightness(1.2);
 }
 

src/config/parser.rs

@@ -6,6 +6,7 @@ use crate::handler::{file_path, FileType};
 use crate::models::parser_models::{AggregatorConfig, RateLimiter, Style};
 use log::LevelFilter;
 use mlua::Lua;
+use reqwest::Proxy;
 use std::{collections::HashMap, fs, thread::available_parallelism};
 
 /// A named struct which stores the parsed config file options.
@@ -48,8 +49,12 @@ pub struct Config {
     pub tcp_connection_keep_alive: u8,
     /// It stores the pool idle connection timeout in seconds.
     pub pool_idle_connection_timeout: u8,
+    /// Url of the proxy to use for outgoing requests.
+    pub proxy: Option<Proxy>,
     /// It stores the number of https connections to keep in the pool.
     pub number_of_https_connections: u8,
+    /// It stores the operating system's TLS certificates for https requests.
+    pub operating_system_tls_certificates: bool,
 }
 
 impl Config {
@@ -120,7 +125,17 @@ impl Config {
             _ => parsed_cet,
         };
 
+        let proxy_opt = globals.get::<_, Option<String>>("proxy")?;
+        let proxy = proxy_opt.and_then(|proxy_str| {
+            Proxy::all(proxy_str).ok().and_then(|_| {
+                log::error!("Invalid proxy url, defaulting to no proxy.");
+                None
+            })
+        });
+
         Ok(Config {
+            operating_system_tls_certificates: globals
+                .get::<_, bool>("operating_system_tls_certificates")?,
             port: globals.get::<_, u16>("port")?,
             binding_ip: globals.get::<_, String>("binding_ip")?,
             style: Style::new(
@@ -151,6 +166,7 @@ impl Config {
             safe_search,
             #[cfg(any(feature = "redis-cache", feature = "memory-cache"))]
             cache_expiry_time,
+            proxy,
         })
     }
 }

src/engines/librex.rs

@@ -30,7 +30,7 @@ impl LibreX {
         Ok(Self {
             parser: SearchResultParser::new(
                 ".text-result-container>p",
-                ".text-result-container",
+                ".text-result-wrapper",
                 ".text-result-wrapper>a>h2",
                 ".text-result-wrapper>a",
                 ".text-result-wrapper>span",

src/lib.rs

@@ -91,7 +91,7 @@ pub fn run(
             .wrap(cors)
             .wrap(Governor::new(
                 &GovernorConfigBuilder::default()
-                    .per_second(config.rate_limiter.time_limit as u64)
+                    .seconds_per_request(config.rate_limiter.time_limit as u64)
                     .burst_size(config.rate_limiter.number_of_requests as u32)
                     .finish()
                     .unwrap(),

src/results/aggregator.rs

@@ -75,7 +75,7 @@ pub async fn aggregate(
     safe_search: u8,
 ) -> Result<SearchResults, Box<dyn std::error::Error>> {
     let client = CLIENT.get_or_init(|| {
-        ClientBuilder::new()
+        let mut cb = ClientBuilder::new()
             .timeout(Duration::from_secs(config.request_timeout as u64)) // Add timeout to request to avoid DDOSing the server
             .pool_idle_timeout(Duration::from_secs(
                 config.pool_idle_connection_timeout as u64,
@@ -83,12 +83,18 @@ pub async fn aggregate(
             .tcp_keepalive(Duration::from_secs(config.tcp_connection_keep_alive as u64))
             .pool_max_idle_per_host(config.number_of_https_connections as usize)
             .connect_timeout(Duration::from_secs(config.request_timeout as u64)) // Add timeout to request to avoid DDOSing the server
+            .use_rustls_tls()
+            .tls_built_in_root_certs(config.operating_system_tls_certificates)
             .https_only(true)
             .gzip(true)
             .brotli(true)
-            .http2_adaptive_window(config.adaptive_window)
+            .http2_adaptive_window(config.adaptive_window);
-            .build()
+
-            .unwrap()
+        if config.proxy.is_some() {
+            cb = cb.proxy(config.proxy.clone().unwrap());
+        }
+
+        cb.build().unwrap()
     });
 
     let user_agent: &str = random_user_agent();
@@ -242,6 +248,7 @@ pub async fn filter_with_lists(
 
     Ok(())
 }
+
 /// Sorts  SearchResults by relevance score.
 /// <br> sort_unstable is used as its faster,stability is not an issue on our side.
 /// For reasons why, check out [`this`](https://rust-lang.github.io/rfcs/1884-unstable-sort.html)
@@ -257,6 +264,7 @@ fn sort_search_results(results: &mut [SearchResult]) {
             .unwrap_or(Ordering::Less)
     })
 }
+
 #[cfg(test)]
 mod tests {
     use super::*;

src/server/routes/search.rs

@@ -85,7 +85,7 @@ pub async fn search(
             let next_page = page + 1;
 
             // Add a random delay before making the request.
-            if config.aggregator.random_delay || !config.debug {
+            if config.aggregator.random_delay || config.debug {
                 let nanos = SystemTime::now().duration_since(UNIX_EPOCH)?.subsec_nanos() as f32;
                 let delay = ((nanos / 1_0000_0000 as f32).floor() as u64) + 1;
                 tokio::time::sleep(Duration::from_secs(delay)).await;
@@ -129,6 +129,7 @@ pub async fn search(
                     &config.style.theme,
                     &config.style.animation,
                     query,
+                    page,
                     &results.0,
                 )
                 .0,

src/templates/partials/bar.rs

@@ -14,9 +14,13 @@ use maud::{html, Markup, PreEscaped};
 /// It returns the compiled html code for the search bar as a result.
 pub fn bar(query: &str) -> Markup {
     html!(
+        (PreEscaped("<form action=\"/search\">"))
         (PreEscaped("<div class=\"search_bar\">"))
-            input type="search" name="search-box" value=(query) placeholder="Type to search";
+            input type="search" name="q" value=(query) placeholder="Type to search";
-            button type="submit" onclick="searchWeb()" {
+            button type="button" onclick="clearSearchText()" {
+                img src="./images/close.svg" alt="Clear button icon for clearing search input text";
+            }
+            button type="submit" {
                 img src="./images/magnifying_glass.svg" alt="Info icon for error box";
             }
     )

src/templates/partials/search_bar.rs

@@ -29,7 +29,7 @@ pub fn search_bar(
             (bar(query))
                 .error_box {
                    @if !engine_errors_info.is_empty(){
-                      button onclick="toggleErrorBox()" class="error_box_toggle_button"{
+                      button type="button" onclick="toggleErrorBox()" class="error_box_toggle_button"{
                          img src="./images/warning.svg" alt="Info icon for error box";
                       }
                       .dropdown_error_box{
@@ -43,7 +43,7 @@ pub fn search_bar(
                       }
                    }
                    @else {
-                      button onclick="toggleErrorBox()" class="error_box_toggle_button"{
+                      button type="button" onclick="toggleErrorBox()" class="error_box_toggle_button"{
                          img src="./images/info.svg" alt="Warning icon for error box";
                       }
                       .dropdown_error_box {
@@ -56,10 +56,10 @@ pub fn search_bar(
             (PreEscaped("</div>"))
             .search_options {
                @if safe_search_level >= 3 {
-                   (PreEscaped("<select name=\"safe_search_levels\" disabled>"))
+                   (PreEscaped("<select name=\"safesearch\" disabled>"))
                }
                @else{
-                   (PreEscaped("<select name=\"safe_search_levels\">"))
+                   (PreEscaped(format!("<select name=\"safesearch\" value=\"{}\">", safe_search_level)))
                }
                @for (idx, name) in SAFE_SEARCH_LEVELS_NAME.iter().enumerate() {
                    @if (safe_search_level as usize) == idx {
@@ -71,6 +71,7 @@ pub fn search_bar(
                }
                (PreEscaped("</select>"))
             }
+            (PreEscaped("</form>"))
         }
     )
 }

src/templates/views/search.rs

@@ -24,6 +24,7 @@ pub fn search(
     theme: &str,
     animation: &Option<String>,
     query: &str,
+    page: u32,
     search_results: &SearchResults,
 ) -> Markup {
     html!(
@@ -108,15 +109,14 @@ pub fn search(
               }
             }
             .page_navigation {
-               button type="button" onclick="navigate_backward()"{
+               a href=(format!("/search?q={}&safesearch={}&page={}", query, search_results.safe_search_level, if page > 1 {page-1} else {1})) {
                    (PreEscaped("&#8592;")) "previous"
                }
-               button type="button" onclick="navigate_forward()"{"next" (PreEscaped("&#8594;"))}
+               a href=(format!("/search?q={}&safesearch={}&page={}", query, search_results.safe_search_level, page+2)) {
+                  "next" (PreEscaped("&#8594;"))}
             }
         }
         script src="static/index.js"{}
-        script src="static/search_area_options.js"{}
-        script src="static/pagination.js"{}
         script src="static/error_box.js"{}
         (footer())
     )

websurfx/config.lua

@@ -19,6 +19,8 @@ rate_limiter = {
 -- Set whether the server will use an adaptive/dynamic HTTPS window size, see https://httpwg.org/specs/rfc9113.html#fc-principles
 https_adaptive_window_size = false
 
+operating_system_tls_certificates = true -- Set whether the server will use operating system's tls certificates alongside rustls certificates while fetching search results from the upstream engines.
+
 number_of_https_connections = 10 -- the number of https connections that should be available in the connection pool.
 -- Set keep-alive timer in seconds; keeps clients connected to the HTTP server, different from the connection to upstream search engines
 client_connection_keep_alive = 120
@@ -76,3 +78,5 @@ upstream_search_engines = {
     Bing = false,
     Qwant = false,
 } -- select the upstream search engines from which the results should be fetched.
+
+proxy = nil -- Proxy to send outgoing requests through. Set to nil to disable.