Spaces:
Runtime error
✨ `LibreX` for the search engine (#429)
Browse files* ✨ feat(engine): provide librex search engine (#318)
* ✨ feat(engine): provide librex search engine (#318)
* ✨ feat(engine): provide librex search engine (#318)
* 🔖 chore(release): bump the app version (#318)
* 🔖 chore(release): bump the app version (#318)
* ✨ feat(engine): provide librex search engine (#318)
* ✨ feat(engine): provide librex search engine (#429)
Co-authored-by: neon_arch <[email protected]>
* ✨ feat(engine): provide librex search engine (#429)
Co-authored-by: neon_arch <[email protected]>
* ✨ feat(engine): provide librex search engine (#429)
Co-authored-by: neon_arch <[email protected]>
* ✨ feat(engine): provide librex search engine (#429)
* ✨ feat(engine): provide librex search engine (#429)
* ✨ feat(engine): provide librex search engine (#429)
* ✨ feat(engine): provide librex search engine (#429)
Co-authored-by: neon_arch <[email protected]>
---------
Co-authored-by: neon_arch <[email protected]>
- Cargo.lock +1 -1
- Cargo.toml +1 -1
- src/engines/librex.rs +111 -0
- src/engines/mod.rs +1 -0
- src/models/engine_models.rs +4 -0
- websurfx/config.lua +1 -0
@@ -4066,7 +4066,7 @@ checksum = "1778a42e8b3b90bff8d0f5032bf22250792889a5cdc752aa0020c84abe3aaf10"
|
|
4066 |
|
4067 |
[[package]]
|
4068 |
name = "websurfx"
|
4069 |
-
version = "1.
|
4070 |
dependencies = [
|
4071 |
"actix-cors",
|
4072 |
"actix-files",
|
|
|
4066 |
|
4067 |
[[package]]
|
4068 |
name = "websurfx"
|
4069 |
+
version = "1.5.0"
|
4070 |
dependencies = [
|
4071 |
"actix-cors",
|
4072 |
"actix-files",
|
@@ -1,6 +1,6 @@
|
|
1 |
[package]
|
2 |
name = "websurfx"
|
3 |
-
version = "1.
|
4 |
edition = "2021"
|
5 |
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."
|
6 |
repository = "https://github.com/neon-mmd/websurfx"
|
|
|
1 |
[package]
|
2 |
name = "websurfx"
|
3 |
+
version = "1.5.0"
|
4 |
edition = "2021"
|
5 |
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."
|
6 |
repository = "https://github.com/neon-mmd/websurfx"
|
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
//! The `librex` module contains the implementation of a search engine for LibreX using the reqwest and scraper libraries.
|
2 |
+
//! It includes a `SearchEngine` trait implementation for interacting with the search engine and retrieving search results.
|
3 |
+
|
4 |
+
use std::collections::HashMap;
|
5 |
+
|
6 |
+
use reqwest::header::HeaderMap;
|
7 |
+
use reqwest::Client;
|
8 |
+
use scraper::Html;
|
9 |
+
|
10 |
+
use crate::models::aggregation_models::SearchResult;
|
11 |
+
use crate::models::engine_models::{EngineError, SearchEngine};
|
12 |
+
|
13 |
+
use error_stack::{Report, Result, ResultExt};
|
14 |
+
|
15 |
+
use super::search_result_parser::SearchResultParser;
|
16 |
+
|
17 |
+
/// Represents the LibreX search engine.
|
18 |
+
pub struct LibreX {
|
19 |
+
/// The parser used to extract search results from HTML documents.
|
20 |
+
parser: SearchResultParser,
|
21 |
+
}
|
22 |
+
|
23 |
+
impl LibreX {
|
24 |
+
/// Creates a new instance of LibreX with a default configuration.
|
25 |
+
///
|
26 |
+
/// # Returns
|
27 |
+
///
|
28 |
+
/// Returns a `Result` containing `LibreX` if successful, otherwise an `EngineError`.
|
29 |
+
pub fn new() -> Result<Self, EngineError> {
|
30 |
+
Ok(Self {
|
31 |
+
parser: SearchResultParser::new(
|
32 |
+
".text-result-container>p",
|
33 |
+
".text-result-container",
|
34 |
+
".text-result-wrapper>a>h2",
|
35 |
+
".text-result-wrapper>a",
|
36 |
+
".text-result-wrapper>span",
|
37 |
+
)?,
|
38 |
+
})
|
39 |
+
}
|
40 |
+
}
|
41 |
+
|
42 |
+
#[async_trait::async_trait]
|
43 |
+
impl SearchEngine for LibreX {
|
44 |
+
/// Retrieves search results from LibreX based on the provided query, page, user agent, and client.
|
45 |
+
///
|
46 |
+
/// # Arguments
|
47 |
+
///
|
48 |
+
/// * `query` - The search query.
|
49 |
+
/// * `page` - The page number for pagination.
|
50 |
+
/// * `user_agent` - The user agent string.
|
51 |
+
/// * `client` - The reqwest client for making HTTP requests.
|
52 |
+
/// * `_safe_search` - A parameter for safe search (not currently used).
|
53 |
+
///
|
54 |
+
/// # Returns
|
55 |
+
///
|
56 |
+
/// Returns a `Result` containing a `HashMap` of search results if successful, otherwise an `EngineError`.
|
57 |
+
/// The `Err` variant is explicit for better documentation.
|
58 |
+
async fn results(
|
59 |
+
&self,
|
60 |
+
query: &str,
|
61 |
+
page: u32,
|
62 |
+
user_agent: &str,
|
63 |
+
client: &Client,
|
64 |
+
_safe_search: u8,
|
65 |
+
) -> Result<HashMap<String, SearchResult>, EngineError> {
|
66 |
+
// Page number can be missing or empty string and so appropriate handling is required
|
67 |
+
// so that upstream server recieves valid page number.
|
68 |
+
let url: String = match page {
|
69 |
+
1 | 0 => {
|
70 |
+
format!("https://search.ahwx.org/search.php?q={query}&p=0&t=10")
|
71 |
+
}
|
72 |
+
_ => {
|
73 |
+
format!(
|
74 |
+
"https://search.ahwx.org/search.php?q={query}&p={}&t=10",
|
75 |
+
page * 10,
|
76 |
+
)
|
77 |
+
}
|
78 |
+
};
|
79 |
+
|
80 |
+
// initializing HeaderMap and adding appropriate headers.
|
81 |
+
let header_map = HeaderMap::try_from(&HashMap::from([
|
82 |
+
("USER_AGENT".to_string(), user_agent.to_string()),
|
83 |
+
("REFERER".to_string(), "https://google.com/".to_string()),
|
84 |
+
("CONTENT_TYPE".to_string(), "application/x-www-form-urlencoded".to_string()),
|
85 |
+
(
|
86 |
+
"COOKIE".to_string(),
|
87 |
+
"theme=amoled; disable_special=on; disable_frontends=on; language=en; number_of_results=10; safe_search=on; save=1".to_string(),
|
88 |
+
),
|
89 |
+
]))
|
90 |
+
.change_context(EngineError::UnexpectedError)?;
|
91 |
+
|
92 |
+
let document: Html = Html::parse_document(
|
93 |
+
&LibreX::fetch_html_from_upstream(self, &url, header_map, client).await?,
|
94 |
+
);
|
95 |
+
|
96 |
+
if self.parser.parse_for_no_results(&document).next().is_some() {
|
97 |
+
return Err(Report::new(EngineError::EmptyResultSet));
|
98 |
+
}
|
99 |
+
|
100 |
+
// scrape all the results from the html
|
101 |
+
self.parser
|
102 |
+
.parse_for_results(&document, |title, url, desc| {
|
103 |
+
Some(SearchResult::new(
|
104 |
+
title.inner_html().trim(),
|
105 |
+
url.inner_html().trim(),
|
106 |
+
desc.inner_html().trim(),
|
107 |
+
&["librex"],
|
108 |
+
))
|
109 |
+
})
|
110 |
+
}
|
111 |
+
}
|
@@ -5,6 +5,7 @@
|
|
5 |
|
6 |
pub mod brave;
|
7 |
pub mod duckduckgo;
|
|
|
8 |
pub mod search_result_parser;
|
9 |
pub mod searx;
|
10 |
pub mod startpage;
|
|
|
5 |
|
6 |
pub mod brave;
|
7 |
pub mod duckduckgo;
|
8 |
+
pub mod librex;
|
9 |
pub mod search_result_parser;
|
10 |
pub mod searx;
|
11 |
pub mod startpage;
|
@@ -158,6 +158,10 @@ impl EngineHandler {
|
|
158 |
let engine = crate::engines::startpage::Startpage::new()?;
|
159 |
("startpage", Box::new(engine))
|
160 |
}
|
|
|
|
|
|
|
|
|
161 |
_ => {
|
162 |
return Err(Report::from(EngineError::NoSuchEngineFound(
|
163 |
engine_name.to_string(),
|
|
|
158 |
let engine = crate::engines::startpage::Startpage::new()?;
|
159 |
("startpage", Box::new(engine))
|
160 |
}
|
161 |
+
"librex" => {
|
162 |
+
let engine = crate::engines::librex::LibreX::new()?;
|
163 |
+
("librex", Box::new(engine))
|
164 |
+
}
|
165 |
_ => {
|
166 |
return Err(Report::from(EngineError::NoSuchEngineFound(
|
167 |
engine_name.to_string(),
|
@@ -54,4 +54,5 @@ upstream_search_engines = {
|
|
54 |
Searx = false,
|
55 |
Brave = false,
|
56 |
Startpage = false,
|
|
|
57 |
} -- select the upstream search engines from which the results should be fetched.
|
|
|
54 |
Searx = false,
|
55 |
Brave = false,
|
56 |
Startpage = false,
|
57 |
+
LibreX = false,
|
58 |
} -- select the upstream search engines from which the results should be fetched.
|