mirror of
https://github.com/neon-mmd/websurfx.git
synced 2024-11-22 14:08:23 -05:00
Merge branch 'rolling' into rolling
This commit is contained in:
commit
f39755eedd
2
Cargo.lock
generated
2
Cargo.lock
generated
@ -4066,7 +4066,7 @@ checksum = "1778a42e8b3b90bff8d0f5032bf22250792889a5cdc752aa0020c84abe3aaf10"
|
||||
|
||||
[[package]]
|
||||
name = "websurfx"
|
||||
version = "1.3.6"
|
||||
version = "1.4.1"
|
||||
dependencies = [
|
||||
"actix-cors",
|
||||
"actix-files",
|
||||
|
@ -1,6 +1,6 @@
|
||||
[package]
|
||||
name = "websurfx"
|
||||
version = "1.3.6"
|
||||
version = "1.4.1"
|
||||
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"
|
||||
|
@ -326,6 +326,9 @@ footer {
|
||||
font-size: 1.2rem;
|
||||
padding: 1rem;
|
||||
color: var(--color-five);
|
||||
display: flex;
|
||||
gap: 1rem;
|
||||
justify-content: right;
|
||||
}
|
||||
|
||||
/* Styles for the 404 page */
|
||||
|
@ -7,3 +7,4 @@ pub mod brave;
|
||||
pub mod duckduckgo;
|
||||
pub mod search_result_parser;
|
||||
pub mod searx;
|
||||
pub mod startpage;
|
||||
|
96
src/engines/startpage.rs
Normal file
96
src/engines/startpage.rs
Normal file
@ -0,0 +1,96 @@
|
||||
//! The `duckduckgo` module handles the scraping of results from the duckduckgo search engine
|
||||
//! by querying the upstream duckduckgo search engine with user provided query and with a page
|
||||
//! number if provided.
|
||||
|
||||
use std::collections::HashMap;
|
||||
|
||||
use reqwest::header::HeaderMap;
|
||||
use reqwest::Client;
|
||||
use scraper::Html;
|
||||
|
||||
use crate::models::aggregation_models::SearchResult;
|
||||
|
||||
use crate::models::engine_models::{EngineError, SearchEngine};
|
||||
|
||||
use error_stack::{Report, Result, ResultExt};
|
||||
|
||||
use super::search_result_parser::SearchResultParser;
|
||||
|
||||
/// A new Startpage engine type defined in-order to implement the `SearchEngine` trait which allows to
|
||||
/// reduce code duplication as well as allows to create vector of different search engines easily.
|
||||
pub struct Startpage {
|
||||
/// The parser, used to interpret the search result.
|
||||
parser: SearchResultParser,
|
||||
}
|
||||
|
||||
impl Startpage {
|
||||
/// Creates the Startpage parser.
|
||||
pub fn new() -> Result<Self, EngineError> {
|
||||
Ok(Self {
|
||||
parser: SearchResultParser::new(
|
||||
".no-results",
|
||||
".w-gl__result__main",
|
||||
".w-gl__result-second-line-container>.w-gl__result-title>h3",
|
||||
".w-gl__result-url",
|
||||
".w-gl__description",
|
||||
)?,
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
#[async_trait::async_trait]
|
||||
impl SearchEngine for Startpage {
|
||||
async fn results(
|
||||
&self,
|
||||
query: &str,
|
||||
page: u32,
|
||||
user_agent: &str,
|
||||
client: &Client,
|
||||
_safe_search: u8,
|
||||
) -> Result<HashMap<String, SearchResult>, EngineError> {
|
||||
// Page number can be missing or empty string and so appropriate handling is required
|
||||
// so that upstream server recieves valid page number.
|
||||
let url: String = match page {
|
||||
1 | 0 => {
|
||||
format!("https://startpage.com/do/dsearch?q={query}&num=10&start=0")
|
||||
}
|
||||
_ => {
|
||||
format!(
|
||||
"https://startpage.com/do/dsearch?q={query}&num=10&start={}",
|
||||
page * 10,
|
||||
)
|
||||
}
|
||||
};
|
||||
|
||||
// initializing HeaderMap and adding appropriate headers.
|
||||
let header_map = HeaderMap::try_from(&HashMap::from([
|
||||
("USER_AGENT".to_string(), user_agent.to_string()),
|
||||
("REFERER".to_string(), "https://google.com/".to_string()),
|
||||
(
|
||||
"CONTENT_TYPE".to_string(),
|
||||
"application/x-www-form-urlencoded".to_string(),
|
||||
),
|
||||
("COOKIE".to_string(), "preferences=connect_to_serverEEE0N1Ndate_timeEEEworldN1Ndisable_family_filterEEE0N1Ndisable_open_in_new_windowEEE0N1Nenable_post_methodEEE1N1Nenable_proxy_safety_suggestEEE1N1Nenable_stay_controlEEE0N1Ninstant_answersEEE1N1Nlang_homepageEEEs%2Fnight%2FenN1NlanguageEEEenglishN1Nlanguage_uiEEEenglishN1Nnum_of_resultsEEE10N1Nsearch_results_regionEEEallN1NsuggestionsEEE1N1Nwt_unitEEEcelsius".to_string()),
|
||||
]))
|
||||
.change_context(EngineError::UnexpectedError)?;
|
||||
|
||||
let document: Html = Html::parse_document(
|
||||
&Startpage::fetch_html_from_upstream(self, &url, header_map, client).await?,
|
||||
);
|
||||
|
||||
if self.parser.parse_for_no_results(&document).next().is_some() {
|
||||
return Err(Report::new(EngineError::EmptyResultSet));
|
||||
}
|
||||
|
||||
// scrape all the results from the html
|
||||
self.parser
|
||||
.parse_for_results(&document, |title, url, desc| {
|
||||
Some(SearchResult::new(
|
||||
title.inner_html().trim(),
|
||||
&format!("{}", url.inner_html().trim()),
|
||||
desc.inner_html().trim(),
|
||||
&["startpage"],
|
||||
))
|
||||
})
|
||||
}
|
||||
}
|
@ -154,6 +154,10 @@ impl EngineHandler {
|
||||
let engine = crate::engines::brave::Brave::new()?;
|
||||
("brave", Box::new(engine))
|
||||
}
|
||||
"startpage" => {
|
||||
let engine = crate::engines::startpage::Startpage::new()?;
|
||||
("startpage", Box::new(engine))
|
||||
}
|
||||
_ => {
|
||||
return Err(Report::from(EngineError::NoSuchEngineFound(
|
||||
engine_name.to_string(),
|
||||
|
@ -53,4 +53,5 @@ upstream_search_engines = {
|
||||
DuckDuckGo = true,
|
||||
Searx = false,
|
||||
Brave = false,
|
||||
Startpage = false,
|
||||
} -- select the upstream search engines from which the results should be fetched.
|
||||
|
Loading…
Reference in New Issue
Block a user