This commit is contained in:
Oliver Atkinson 2025-01-24 14:46:33 -07:00
commit b454e08b3c
9 changed files with 3501 additions and 0 deletions

2
.gitignore vendored Normal file
View File

@ -0,0 +1,2 @@
/target
*-debug.html

45
.vscode/launch.json vendored Normal file
View File

@ -0,0 +1,45 @@
{
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
"version": "0.2.0",
"configurations": [
{
"type": "lldb",
"request": "launch",
"name": "Debug executable 'search'",
"cargo": {
"args": [
"build",
"--bin=search",
"--package=search"
],
"filter": {
"name": "search",
"kind": "bin"
}
},
"args": [],
"cwd": "${workspaceFolder}"
},
{
"type": "lldb",
"request": "launch",
"name": "Debug unit tests in executable 'search'",
"cargo": {
"args": [
"test",
"--no-run",
"--bin=search",
"--package=search"
],
"filter": {
"name": "search",
"kind": "bin"
}
},
"args": [],
"cwd": "${workspaceFolder}"
}
]
}

3181
Cargo.lock generated Normal file

File diff suppressed because it is too large Load Diff

11
Cargo.toml Normal file
View File

@ -0,0 +1,11 @@
[package]
name = "search"
version = "0.1.0"
edition = "2024"
[dependencies]
reqwest = { version = "0.12.12", features = ["gzip"] }
rocket = "0.5.1"
scraper = "0.22.0"
rocket_dyn_templates = { version = "0.2", features = ["handlebars"] }
serde = { version = "1.0.217", features = ["derive"] }

112
src/main.rs Normal file
View File

@ -0,0 +1,112 @@
/*
html.duckduckgo.com
google.com
*/
use std::fs;
use reqwest::{header::HeaderMap, Client, Method};
use rocket::{form::Form, fs::FileServer, post, routes, FromForm};
use rocket_dyn_templates::{context, Template};
use scraper::{Html, Selector};
use serde::Serialize;
#[derive(Serialize)]
struct SearchResult {
a: String,
desc: String,
title: String,
}
#[derive(FromForm)]
struct Query {
query: String
}
async fn google(client: &Client, search: &str) -> Vec<SearchResult> {
todo!()
}
async fn ddg(client: &Client, search: &str) -> Vec<SearchResult> {
// TODO url encode search
let mut headers = HeaderMap::new();
headers.insert("content-type", "application/x-www-form-urlencoded".parse().unwrap());
headers.insert("Accept-Language", "en-US,en;q=0.9".parse().unwrap());
headers.insert("Referer", "https://html.duckduckgo.com/".parse().unwrap());
let request = client
.request(Method::POST, "https://html.duckduckgo.com/html/")
.headers(headers)
.body(format!("q={}", search))
.build()
.unwrap();
let ddg_dom = client.execute(request)
.await
.unwrap()
.text()
.await
.unwrap();
fs::write("ddg-debug.html", &ddg_dom).unwrap();
let html = Html::parse_document(&ddg_dom);
let search_result = Selector::parse("#links > .results_links > .result__body ").unwrap();
let bot_detected = Selector::parse("anomaly-modal__mask").unwrap();
let count = html.select(&bot_detected).into_iter().count();
if count > 0 {
// we've been found out!
println!("bot");
}
let mut buf = Vec::new();
for result in html.select(&search_result) {
let title_selector = Selector::parse("h2 > a").unwrap();
let title = result.select(&title_selector).next().expect("Failed to get title");
let preview_selector = Selector::parse(".result__snippet").unwrap();
let preview = result.select(&preview_selector).next().expect("Failed to get title");
let a = title.attr("href").unwrap().to_string();
let title = title.inner_html();
let desc = preview.inner_html();
let sr = SearchResult {
a,
title,
desc
};
buf.push(sr);
}
buf
}
#[post("/", data="<input>")]
async fn search(input: Form<Query>) -> Template {
let clean = &input.query.trim();
let client = reqwest::ClientBuilder::new()
.user_agent("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36")
.gzip(true)
.build()
.unwrap();
Template::render("results", context! {
ddg: ddg(&client, clean).await,
last_search: clean
})
}
#[rocket::main]
async fn main() {
rocket::build()
.mount("/", routes![search])
.mount("/", FileServer::from("www"))
.attach(Template::fairing())
.launch()
.await
.unwrap();
}

View File

@ -0,0 +1,44 @@
<!DOCTYPE html>
<html>
<head>
<meta charset='utf-8'>
<meta http-equiv='X-UA-Compatible' content='IE=edge'>
<title>Meta Search</title>
<meta name='viewport' content='width=device-width, initial-scale=1'>
<link rel='stylesheet' type='text/css' media='screen' href='style.css'>
<script src='main.js' defer></script>
</head>
<body>
<div class="search-area">
<h1>Meta Search</h1>
<form method="post">
<input value="{{ last_search }}" name="query" type="text" spellcheck="false" autofocus autocomplete="off">
<button type="submit">🔎</button>
</form>
</div>
<h2>📚 Results</h2>
<div class="carousel">
<div class="result">
<h3>Duck Duck Go</h3>
{{ #each ddg }}
<div class="line-item">
<a href="{{ this.a }}">{{this.title}}</a>
<p>{{ this.desc }}</p>
</div>
{{ /each }}
</div>
<div class="result">
<h3>Google</h3>
{{ #each google }}
<div class="line-item">
<a href="{{ this.a }}">{{this.title}}</a>
<p>{{ this.desc }}</p>
</div>
{{ /each }}
</div>
</div>
</body>
</html>

27
www/index.html Normal file
View File

@ -0,0 +1,27 @@
<!DOCTYPE html>
<html>
<head>
<meta charset='utf-8'>
<meta http-equiv='X-UA-Compatible' content='IE=edge'>
<title>Meta Search</title>
<meta name='viewport' content='width=device-width, initial-scale=1'>
<link rel='stylesheet' type='text/css' media='screen' href='style.css'>
</head>
<body>
<div class="search-area">
<h1>Meta Search</h1>
<form method="post">
<input spellcheck="false" name="query" type="text" autofocus autocomplete="off">
<button type="submit">🔎</button>
</form>
</div>
</body>
</html>
<style>
.search-area {
margin: auto;
width: 50%;
margin-top: 10%;
}
</style>

1
www/main.js Normal file
View File

@ -0,0 +1 @@
console.log("js loaded")

78
www/style.css Normal file
View File

@ -0,0 +1,78 @@
html {
background-color: black;
color: white;
padding: 0;
margin: 0;
font-family: Arial, Helvetica, sans-serif;
}
body {
padding: 0;
margin: 0;
}
a, a:visited {
color: inherit;
}
.search-area {
text-align: center;
}
.result {
width: 80ch;
min-width: 80ch;
height: 75ch;
overflow-y: auto;
border-style: solid;
border-color: #707070;
border-width: 5px;
}
.line-item {
padding: 10px;
background-color: #505050;
}
.line-item:nth-child(odd) {
background-color: #707070;
}
.carousel {
display: flex;
overflow-x: auto;
background-color: #050505;
}
h1 {
font-size: 3rem;
margin: 10px;
}
h2 {
font-size: 2rem;
}
h3 {
text-align: center;
font-size: 2rem;
margin: 10px;
}
form {
font-size: 2rem;
}
input {
font-size: inherit;
}
button {
font-size: inherit;
background-color: inherit;
color: inherit;
border-width: 0;
transition: 200ms;
}
button:hover {
transition: 200ms;
background-color: white;
color: black;
}