0
0
mirror of https://github.com/neon-mmd/websurfx.git synced 2024-11-22 05:58:21 -05:00

⚙️ refactor: add several optimizations to the engine code (#180)(#178)

This commit is contained in:
neon_arch 2023-08-27 20:59:08 +03:00
parent 4ccd0486e7
commit f5f0488954
2 changed files with 36 additions and 75 deletions

View File

@ -4,14 +4,14 @@
use std::collections::HashMap;
use reqwest::header::{HeaderMap, CONTENT_TYPE, COOKIE, REFERER, USER_AGENT};
use reqwest::header::HeaderMap;
use scraper::{Html, Selector};
use crate::results::aggregation_models::SearchResult;
use super::engine_models::{EngineError, SearchEngine};
use error_stack::{IntoReport, Report, Result, ResultExt};
use error_stack::{Report, Result, ResultExt};
/// A new DuckDuckGo 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.
@ -39,9 +39,9 @@ impl SearchEngine for DuckDuckGo {
/// or HeaderMap fails to initialize.
async fn results(
&self,
query: String,
query: &str,
page: u32,
user_agent: String,
user_agent: &str,
request_timeout: u8,
) -> Result<HashMap<String, SearchResult>, EngineError> {
// Page number can be missing or empty string and so appropriate handling is required
@ -61,38 +61,19 @@ impl SearchEngine for DuckDuckGo {
};
// initializing HeaderMap and adding appropriate headers.
let mut header_map = HeaderMap::new();
header_map.insert(
USER_AGENT,
user_agent
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
header_map.insert(
REFERER,
"https://google.com/"
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
header_map.insert(
CONTENT_TYPE,
"application/x-www-form-urlencoded"
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
header_map.insert(
COOKIE,
"kl=wt-wt"
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
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(), "kl=wt-wt".to_string()),
]))
.change_context(EngineError::UnexpectedError)?;
let document: Html = Html::parse_document(
&DuckDuckGo::fetch_html_from_upstream(self, url, header_map, request_timeout).await?,
&DuckDuckGo::fetch_html_from_upstream(self, &url, header_map, request_timeout).await?,
);
let no_result: Selector = Selector::parse(".no-results")
@ -126,8 +107,7 @@ impl SearchEngine for DuckDuckGo {
.next()
.unwrap()
.inner_html()
.trim()
.to_string(),
.trim(),
format!(
"https://{}",
result
@ -136,15 +116,15 @@ impl SearchEngine for DuckDuckGo {
.unwrap()
.inner_html()
.trim()
),
)
.as_str(),
result
.select(&result_desc)
.next()
.unwrap()
.inner_html()
.trim()
.to_string(),
vec!["duckduckgo".to_string()],
.trim(),
&["duckduckgo"],
)
})
.map(|search_result| (search_result.url.clone(), search_result))

View File

@ -2,14 +2,14 @@
//! by querying the upstream searx search engine instance with user provided query and with a page
//! number if provided.
use reqwest::header::{HeaderMap, CONTENT_TYPE, COOKIE, REFERER, USER_AGENT};
use reqwest::header::HeaderMap;
use scraper::{Html, Selector};
use std::collections::HashMap;
use crate::results::aggregation_models::SearchResult;
use super::engine_models::{EngineError, SearchEngine};
use error_stack::{IntoReport, Report, Result, ResultExt};
use error_stack::{Report, Result, ResultExt};
/// A new Searx 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.
@ -38,9 +38,9 @@ impl SearchEngine for Searx {
async fn results(
&self,
query: String,
query: &str,
page: u32,
user_agent: String,
user_agent: &str,
request_timeout: u8,
) -> Result<HashMap<String, SearchResult>, EngineError> {
// Page number can be missing or empty string and so appropriate handling is required
@ -51,32 +51,16 @@ impl SearchEngine for Searx {
};
// initializing headers and adding appropriate headers.
let mut header_map = HeaderMap::new();
header_map.insert(
USER_AGENT,
user_agent
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
header_map.insert(
REFERER,
"https://google.com/"
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
header_map.insert(
CONTENT_TYPE,
"application/x-www-form-urlencoded"
.parse()
.into_report()
.change_context(EngineError::UnexpectedError)?,
);
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 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(), "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".to_string())
]))
.change_context(EngineError::UnexpectedError)?;
let document: Html = Html::parse_document(
&Searx::fetch_html_from_upstream(self, url, header_map, request_timeout).await?,
&Searx::fetch_html_from_upstream(self, &url, header_map, request_timeout).await?,
);
let no_result: Selector = Selector::parse("#urls>.dialog-error>p")
@ -117,24 +101,21 @@ impl SearchEngine for Searx {
.next()
.unwrap()
.inner_html()
.trim()
.to_string(),
.trim(),
result
.select(&result_url)
.next()
.unwrap()
.value()
.attr("href")
.unwrap()
.to_string(),
.unwrap(),
result
.select(&result_desc)
.next()
.unwrap()
.inner_html()
.trim()
.to_string(),
vec!["searx".to_string()],
.trim(),
&["searx"],
)
})
.map(|search_result| (search_result.url.clone(), search_result))