Merge pull request #206 from neon-mmd/change-document-style-with-linter-warnings
Browse files- Cargo.lock +1 -1
- Cargo.toml +1 -1
- src/cache/cacher.rs +4 -7
- src/cache/error.rs +3 -6
- src/cache/mod.rs +3 -0
- src/config/mod.rs +3 -0
- src/config/parser.rs +19 -14
- src/config/parser_models.rs +7 -16
- src/engines/duckduckgo.rs +0 -18
- src/engines/engine_models.rs +58 -10
- src/engines/mod.rs +5 -0
- src/engines/searx.rs +0 -19
- src/handler/mod.rs +4 -0
- src/handler/paths.rs +25 -21
- src/lib.rs +6 -2
- src/results/aggregation_models.rs +31 -21
- src/results/mod.rs +4 -0
- src/results/user_agent.rs +2 -0
- src/server/mod.rs +5 -0
- src/server/routes.rs +29 -17
Cargo.lock
CHANGED
@@ -3797,7 +3797,7 @@ dependencies = [
|
|
3797 |
|
3798 |
[[package]]
|
3799 |
name = "websurfx"
|
3800 |
-
version = "0.20.
|
3801 |
dependencies = [
|
3802 |
"actix-cors",
|
3803 |
"actix-files",
|
|
|
3797 |
|
3798 |
[[package]]
|
3799 |
name = "websurfx"
|
3800 |
+
version = "0.20.1"
|
3801 |
dependencies = [
|
3802 |
"actix-cors",
|
3803 |
"actix-files",
|
Cargo.toml
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
[package]
|
2 |
name = "websurfx"
|
3 |
-
version = "0.20.
|
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 = "0.20.1"
|
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"
|
src/cache/cacher.rs
CHANGED
@@ -10,17 +10,14 @@ use super::error::PoolError;
|
|
10 |
|
11 |
/// A named struct which stores the redis Connection url address to which the client will
|
12 |
/// connect to.
|
13 |
-
///
|
14 |
-
/// # Fields
|
15 |
-
///
|
16 |
-
/// * `connection_pool` - It stores a pool of connections ready to be used.
|
17 |
-
/// * `pool_size` - It stores the size of the connection pool (in other words the number of
|
18 |
-
/// connections that should be stored in the pool).
|
19 |
-
/// * `current_connection` - It stores the index of which connection is being used at the moment.
|
20 |
#[derive(Clone)]
|
21 |
pub struct RedisCache {
|
|
|
22 |
connection_pool: Vec<ConnectionManager>,
|
|
|
|
|
23 |
pool_size: u8,
|
|
|
24 |
current_connection: u8,
|
25 |
}
|
26 |
|
|
|
10 |
|
11 |
/// A named struct which stores the redis Connection url address to which the client will
|
12 |
/// connect to.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
#[derive(Clone)]
|
14 |
pub struct RedisCache {
|
15 |
+
/// It stores a pool of connections ready to be used.
|
16 |
connection_pool: Vec<ConnectionManager>,
|
17 |
+
/// It stores the size of the connection pool (in other words the number of
|
18 |
+
/// connections that should be stored in the pool).
|
19 |
pool_size: u8,
|
20 |
+
/// It stores the index of which connection is being used at the moment.
|
21 |
current_connection: u8,
|
22 |
}
|
23 |
|
src/cache/error.rs
CHANGED
@@ -5,15 +5,12 @@ use std::fmt;
|
|
5 |
use redis::RedisError;
|
6 |
|
7 |
/// A custom error type used for handling redis async pool associated errors.
|
8 |
-
///
|
9 |
-
/// This enum provides variants three different categories of errors:
|
10 |
-
/// * `RedisError` - This variant handles all errors related to `RedisError`,
|
11 |
-
/// * `PoolExhaustionWithConnectionDropError` - This variant handles the error
|
12 |
-
/// which occurs when all the connections in the connection pool return a connection
|
13 |
-
/// dropped redis error.
|
14 |
#[derive(Debug)]
|
15 |
pub enum PoolError {
|
|
|
16 |
RedisError(RedisError),
|
|
|
|
|
17 |
PoolExhaustionWithConnectionDropError,
|
18 |
}
|
19 |
|
|
|
5 |
use redis::RedisError;
|
6 |
|
7 |
/// A custom error type used for handling redis async pool associated errors.
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
#[derive(Debug)]
|
9 |
pub enum PoolError {
|
10 |
+
/// This variant handles all errors related to `RedisError`,
|
11 |
RedisError(RedisError),
|
12 |
+
/// This variant handles the errors which occurs when all the connections
|
13 |
+
/// in the connection pool return a connection dropped redis error.
|
14 |
PoolExhaustionWithConnectionDropError,
|
15 |
}
|
16 |
|
src/cache/mod.rs
CHANGED
@@ -1,2 +1,5 @@
|
|
|
|
|
|
|
|
1 |
pub mod cacher;
|
2 |
pub mod error;
|
|
|
1 |
+
//! This module provides the modules which provide the functionality to cache the aggregated
|
2 |
+
//! results fetched and aggregated from the upstream search engines in a json format.
|
3 |
+
|
4 |
pub mod cacher;
|
5 |
pub mod error;
|
src/config/mod.rs
CHANGED
@@ -1,2 +1,5 @@
|
|
|
|
|
|
|
|
1 |
pub mod parser;
|
2 |
pub mod parser_models;
|
|
|
1 |
+
//! This module provides the modules which handles the functionality to parse the lua config
|
2 |
+
//! and convert the config options into rust readable form.
|
3 |
+
|
4 |
pub mod parser;
|
5 |
pub mod parser_models;
|
src/config/parser.rs
CHANGED
@@ -9,33 +9,33 @@ use mlua::Lua;
|
|
9 |
use std::{collections::HashMap, fs, thread::available_parallelism};
|
10 |
|
11 |
/// A named struct which stores the parsed config file options.
|
12 |
-
///
|
13 |
-
/// # Fields
|
14 |
-
//
|
15 |
-
/// * `port` - It stores the parsed port number option on which the server should launch.
|
16 |
-
/// * `binding_ip` - It stores the parsed ip address option on which the server should launch
|
17 |
-
/// * `style` - It stores the theming options for the website.
|
18 |
-
/// * `redis_url` - It stores the redis connection url address on which the redis
|
19 |
-
/// client should connect.
|
20 |
-
/// * `aggregator` - It stores the option to whether enable or disable production use.
|
21 |
-
/// * `logging` - It stores the option to whether enable or disable logs.
|
22 |
-
/// * `debug` - It stores the option to whether enable or disable debug mode.
|
23 |
-
/// * `upstream_search_engines` - It stores all the engine names that were enabled by the user.
|
24 |
-
/// * `request_timeout` - It stores the time (secs) which controls the server request timeout.
|
25 |
-
/// * `threads` - It stores the number of threads which controls the app will use to run.
|
26 |
#[derive(Clone)]
|
27 |
pub struct Config {
|
|
|
28 |
pub port: u16,
|
|
|
29 |
pub binding_ip: String,
|
|
|
30 |
pub style: Style,
|
|
|
|
|
31 |
pub redis_url: String,
|
|
|
32 |
pub aggregator: AggregatorConfig,
|
|
|
33 |
pub logging: bool,
|
|
|
34 |
pub debug: bool,
|
|
|
35 |
pub upstream_search_engines: Vec<crate::engines::engine_models::EngineHandler>,
|
|
|
36 |
pub request_timeout: u8,
|
|
|
37 |
pub threads: u8,
|
|
|
38 |
pub rate_limiter: RateLimiter,
|
|
|
|
|
39 |
pub safe_search: u8,
|
40 |
}
|
41 |
|
@@ -123,6 +123,11 @@ impl Config {
|
|
123 |
}
|
124 |
|
125 |
/// a helper function that sets the proper logging level
|
|
|
|
|
|
|
|
|
|
|
126 |
fn set_logging_level(debug: bool, logging: bool) {
|
127 |
if let Ok(pkg_env_var) = std::env::var("PKG_ENV") {
|
128 |
if pkg_env_var.to_lowercase() == "dev" {
|
|
|
9 |
use std::{collections::HashMap, fs, thread::available_parallelism};
|
10 |
|
11 |
/// A named struct which stores the parsed config file options.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
#[derive(Clone)]
|
13 |
pub struct Config {
|
14 |
+
/// It stores the parsed port number option on which the server should launch.
|
15 |
pub port: u16,
|
16 |
+
/// It stores the parsed ip address option on which the server should launch
|
17 |
pub binding_ip: String,
|
18 |
+
/// It stores the theming options for the website.
|
19 |
pub style: Style,
|
20 |
+
/// It stores the redis connection url address on which the redis
|
21 |
+
/// client should connect.
|
22 |
pub redis_url: String,
|
23 |
+
/// It stores the option to whether enable or disable production use.
|
24 |
pub aggregator: AggregatorConfig,
|
25 |
+
/// It stores the option to whether enable or disable logs.
|
26 |
pub logging: bool,
|
27 |
+
/// It stores the option to whether enable or disable debug mode.
|
28 |
pub debug: bool,
|
29 |
+
/// It stores all the engine names that were enabled by the user.
|
30 |
pub upstream_search_engines: Vec<crate::engines::engine_models::EngineHandler>,
|
31 |
+
/// It stores the time (secs) which controls the server request timeout.
|
32 |
pub request_timeout: u8,
|
33 |
+
/// It stores the number of threads which controls the app will use to run.
|
34 |
pub threads: u8,
|
35 |
+
/// It stores configuration options for the ratelimiting middleware.
|
36 |
pub rate_limiter: RateLimiter,
|
37 |
+
/// It stores the level of safe search to be used for restricting content in the
|
38 |
+
/// search results.
|
39 |
pub safe_search: u8,
|
40 |
}
|
41 |
|
|
|
123 |
}
|
124 |
|
125 |
/// a helper function that sets the proper logging level
|
126 |
+
///
|
127 |
+
/// # Arguments
|
128 |
+
///
|
129 |
+
/// * `debug` - It takes the option to whether enable or disable debug mode.
|
130 |
+
/// * `logging` - It takes the option to whether enable or disable logs.
|
131 |
fn set_logging_level(debug: bool, logging: bool) {
|
132 |
if let Ok(pkg_env_var) = std::env::var("PKG_ENV") {
|
133 |
if pkg_env_var.to_lowercase() == "dev" {
|
src/config/parser_models.rs
CHANGED
@@ -12,15 +12,12 @@ use serde::{Deserialize, Serialize};
|
|
12 |
/// order to allow the deserializing the json back to struct in aggregate function in
|
13 |
/// aggregator.rs and create a new struct out of it and then serialize it back to json and pass
|
14 |
/// it to the template files.
|
15 |
-
///
|
16 |
-
/// # Fields
|
17 |
-
//
|
18 |
-
/// * `theme` - It stores the parsed theme option used to set a theme for the website.
|
19 |
-
/// * `colorscheme` - It stores the parsed colorscheme option used to set a colorscheme for the
|
20 |
-
/// theme being used.
|
21 |
#[derive(Serialize, Deserialize, Clone, Default)]
|
22 |
pub struct Style {
|
|
|
23 |
pub theme: String,
|
|
|
|
|
24 |
pub colorscheme: String,
|
25 |
}
|
26 |
|
@@ -38,24 +35,18 @@ impl Style {
|
|
38 |
}
|
39 |
|
40 |
/// Configuration options for the aggregator.
|
41 |
-
///
|
42 |
-
/// # Fields
|
43 |
-
///
|
44 |
-
/// * `random_delay` - It stores the option to whether enable or disable random delays between
|
45 |
-
/// requests.
|
46 |
#[derive(Clone)]
|
47 |
pub struct AggregatorConfig {
|
|
|
|
|
48 |
pub random_delay: bool,
|
49 |
}
|
50 |
|
51 |
/// Configuration options for the rate limiter middleware.
|
52 |
-
///
|
53 |
-
/// # Fields
|
54 |
-
///
|
55 |
-
/// * `number_of_requests` - The number of request that are allowed within a provided time limit.
|
56 |
-
/// * `time_limit` - The time limit in which the quantity of requests that should be accepted.
|
57 |
#[derive(Clone)]
|
58 |
pub struct RateLimiter {
|
|
|
59 |
pub number_of_requests: u8,
|
|
|
60 |
pub time_limit: u8,
|
61 |
}
|
|
|
12 |
/// order to allow the deserializing the json back to struct in aggregate function in
|
13 |
/// aggregator.rs and create a new struct out of it and then serialize it back to json and pass
|
14 |
/// it to the template files.
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
#[derive(Serialize, Deserialize, Clone, Default)]
|
16 |
pub struct Style {
|
17 |
+
/// It stores the parsed theme option used to set a theme for the website.
|
18 |
pub theme: String,
|
19 |
+
/// It stores the parsed colorscheme option used to set a colorscheme for the
|
20 |
+
/// theme being used.
|
21 |
pub colorscheme: String,
|
22 |
}
|
23 |
|
|
|
35 |
}
|
36 |
|
37 |
/// Configuration options for the aggregator.
|
|
|
|
|
|
|
|
|
|
|
38 |
#[derive(Clone)]
|
39 |
pub struct AggregatorConfig {
|
40 |
+
/// It stores the option to whether enable or disable random delays between
|
41 |
+
/// requests.
|
42 |
pub random_delay: bool,
|
43 |
}
|
44 |
|
45 |
/// Configuration options for the rate limiter middleware.
|
|
|
|
|
|
|
|
|
|
|
46 |
#[derive(Clone)]
|
47 |
pub struct RateLimiter {
|
48 |
+
/// The number of request that are allowed within a provided time limit.
|
49 |
pub number_of_requests: u8,
|
50 |
+
/// The time limit in which the quantity of requests that should be accepted.
|
51 |
pub time_limit: u8,
|
52 |
}
|
src/engines/duckduckgo.rs
CHANGED
@@ -19,24 +19,6 @@ pub struct DuckDuckGo;
|
|
19 |
|
20 |
#[async_trait::async_trait]
|
21 |
impl SearchEngine for DuckDuckGo {
|
22 |
-
/// This function scrapes results from the upstream engine duckduckgo and puts all the scraped
|
23 |
-
/// results like title, visiting_url (href in html),engine (from which engine it was fetched from)
|
24 |
-
/// and description in a RawSearchResult and then adds that to HashMap whose keys are url and
|
25 |
-
/// values are RawSearchResult struct and then returns it within a Result enum.
|
26 |
-
///
|
27 |
-
/// # Arguments
|
28 |
-
///
|
29 |
-
/// * `query` - Takes the user provided query to query to the upstream search engine with.
|
30 |
-
/// * `page` - Takes an u32 as an argument.
|
31 |
-
/// * `user_agent` - Takes a random user agent string as an argument.
|
32 |
-
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
|
33 |
-
///
|
34 |
-
/// # Errors
|
35 |
-
///
|
36 |
-
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
|
37 |
-
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
|
38 |
-
/// provide results for the requested search query and also returns error if the scraping selector
|
39 |
-
/// or HeaderMap fails to initialize.
|
40 |
async fn results(
|
41 |
&self,
|
42 |
query: &str,
|
|
|
19 |
|
20 |
#[async_trait::async_trait]
|
21 |
impl SearchEngine for DuckDuckGo {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
async fn results(
|
23 |
&self,
|
24 |
query: &str,
|
src/engines/engine_models.rs
CHANGED
@@ -6,19 +6,18 @@ use error_stack::{Result, ResultExt};
|
|
6 |
use std::{collections::HashMap, fmt, time::Duration};
|
7 |
|
8 |
/// A custom error type used for handle engine associated errors.
|
9 |
-
///
|
10 |
-
/// This enum provides variants three different categories of errors:
|
11 |
-
/// * `RequestError` - This variant handles all request related errors like forbidden, not found,
|
12 |
-
/// etc.
|
13 |
-
/// * `EmptyResultSet` - This variant handles the not results found error provide by the upstream
|
14 |
-
/// search engines.
|
15 |
-
/// * `UnexpectedError` - This variant handles all the errors which are unexpected or occur rarely
|
16 |
-
/// and are errors mostly related to failure in initialization of HeaderMap, Selector errors and
|
17 |
-
/// all other errors occurring within the code handling the `upstream search engines`.
|
18 |
#[derive(Debug)]
|
19 |
pub enum EngineError {
|
|
|
|
|
20 |
EmptyResultSet,
|
|
|
|
|
21 |
RequestError,
|
|
|
|
|
|
|
|
|
22 |
UnexpectedError,
|
23 |
}
|
24 |
|
@@ -46,6 +45,23 @@ impl error_stack::Context for EngineError {}
|
|
46 |
/// A trait to define common behavior for all search engines.
|
47 |
#[async_trait::async_trait]
|
48 |
pub trait SearchEngine: Sync + Send {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
async fn fetch_html_from_upstream(
|
50 |
&self,
|
51 |
url: &str,
|
@@ -65,6 +81,24 @@ pub trait SearchEngine: Sync + Send {
|
|
65 |
.change_context(EngineError::RequestError)?)
|
66 |
}
|
67 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
async fn results(
|
69 |
&self,
|
70 |
query: &str,
|
@@ -75,8 +109,12 @@ pub trait SearchEngine: Sync + Send {
|
|
75 |
) -> Result<HashMap<String, SearchResult>, EngineError>;
|
76 |
}
|
77 |
|
|
|
78 |
pub struct EngineHandler {
|
|
|
|
|
79 |
engine: Box<dyn SearchEngine>,
|
|
|
80 |
name: &'static str,
|
81 |
}
|
82 |
|
@@ -87,7 +125,15 @@ impl Clone for EngineHandler {
|
|
87 |
}
|
88 |
|
89 |
impl EngineHandler {
|
90 |
-
///
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
pub fn new(engine_name: &str) -> Option<Self> {
|
92 |
let engine: (&'static str, Box<dyn SearchEngine>) =
|
93 |
match engine_name.to_lowercase().as_str() {
|
@@ -102,6 +148,8 @@ impl EngineHandler {
|
|
102 |
})
|
103 |
}
|
104 |
|
|
|
|
|
105 |
pub fn into_name_engine(self) -> (&'static str, Box<dyn SearchEngine>) {
|
106 |
(self.name, self.engine)
|
107 |
}
|
|
|
6 |
use std::{collections::HashMap, fmt, time::Duration};
|
7 |
|
8 |
/// A custom error type used for handle engine associated errors.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
#[derive(Debug)]
|
10 |
pub enum EngineError {
|
11 |
+
/// This variant handles all request related errors like forbidden, not found,
|
12 |
+
/// etc.
|
13 |
EmptyResultSet,
|
14 |
+
/// This variant handles the not results found error provide by the upstream
|
15 |
+
/// search engines.
|
16 |
RequestError,
|
17 |
+
/// This variant handles all the errors which are unexpected or occur rarely
|
18 |
+
/// and are errors mostly related to failure in initialization of HeaderMap,
|
19 |
+
/// Selector errors and all other errors occurring within the code handling
|
20 |
+
/// the `upstream search engines`.
|
21 |
UnexpectedError,
|
22 |
}
|
23 |
|
|
|
45 |
/// A trait to define common behavior for all search engines.
|
46 |
#[async_trait::async_trait]
|
47 |
pub trait SearchEngine: Sync + Send {
|
48 |
+
/// This helper function fetches/requests the search results from the upstream search engine in
|
49 |
+
/// an html form.
|
50 |
+
///
|
51 |
+
/// # Arguments
|
52 |
+
///
|
53 |
+
/// * `url` - It takes the url of the upstream search engine with the user requested search
|
54 |
+
/// query appended in the search parameters.
|
55 |
+
/// * `header_map` - It takes the http request headers to be sent to the upstream engine in
|
56 |
+
/// order to prevent being detected as a bot. It takes the header as a HeaderMap type.
|
57 |
+
/// * `request_timeout` - It takes the request timeout value as seconds which is used to limit
|
58 |
+
/// the amount of time for each request to remain connected when until the results can be provided
|
59 |
+
/// by the upstream engine.
|
60 |
+
///
|
61 |
+
/// # Error
|
62 |
+
///
|
63 |
+
/// It returns the html data as a string if the upstream engine provides the data as expected
|
64 |
+
/// otherwise it returns a custom `EngineError`.
|
65 |
async fn fetch_html_from_upstream(
|
66 |
&self,
|
67 |
url: &str,
|
|
|
81 |
.change_context(EngineError::RequestError)?)
|
82 |
}
|
83 |
|
84 |
+
/// This function scrapes results from the upstream engine and puts all the scraped results like
|
85 |
+
/// title, visiting_url (href in html),engine (from which engine it was fetched from) and description
|
86 |
+
/// in a RawSearchResult and then adds that to HashMap whose keys are url and values are RawSearchResult
|
87 |
+
/// struct and then returns it within a Result enum.
|
88 |
+
///
|
89 |
+
/// # Arguments
|
90 |
+
///
|
91 |
+
/// * `query` - Takes the user provided query to query to the upstream search engine with.
|
92 |
+
/// * `page` - Takes an u32 as an argument.
|
93 |
+
/// * `user_agent` - Takes a random user agent string as an argument.
|
94 |
+
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
|
95 |
+
///
|
96 |
+
/// # Errors
|
97 |
+
///
|
98 |
+
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
|
99 |
+
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
|
100 |
+
/// provide results for the requested search query and also returns error if the scraping selector
|
101 |
+
/// or HeaderMap fails to initialize.
|
102 |
async fn results(
|
103 |
&self,
|
104 |
query: &str,
|
|
|
109 |
) -> Result<HashMap<String, SearchResult>, EngineError>;
|
110 |
}
|
111 |
|
112 |
+
/// A named struct which stores the engine struct with the name of the associated engine.
|
113 |
pub struct EngineHandler {
|
114 |
+
/// It stores the engine struct wrapped in a box smart pointer as the engine struct implements
|
115 |
+
/// the `SearchEngine` trait.
|
116 |
engine: Box<dyn SearchEngine>,
|
117 |
+
/// It stores the name of the engine to which the struct is associated to.
|
118 |
name: &'static str,
|
119 |
}
|
120 |
|
|
|
125 |
}
|
126 |
|
127 |
impl EngineHandler {
|
128 |
+
/// Parses an engine name into an engine handler.
|
129 |
+
///
|
130 |
+
/// # Arguments
|
131 |
+
///
|
132 |
+
/// * `engine_name` - It takes the name of the engine to which the struct was associated to.
|
133 |
+
///
|
134 |
+
/// # Returns
|
135 |
+
///
|
136 |
+
/// It returns an option either containing the value or a none if the engine is unknown
|
137 |
pub fn new(engine_name: &str) -> Option<Self> {
|
138 |
let engine: (&'static str, Box<dyn SearchEngine>) =
|
139 |
match engine_name.to_lowercase().as_str() {
|
|
|
148 |
})
|
149 |
}
|
150 |
|
151 |
+
/// This function converts the EngineHandler type into a tuple containing the engine name and
|
152 |
+
/// the associated engine struct.
|
153 |
pub fn into_name_engine(self) -> (&'static str, Box<dyn SearchEngine>) {
|
154 |
(self.name, self.engine)
|
155 |
}
|
src/engines/mod.rs
CHANGED
@@ -1,3 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
1 |
pub mod duckduckgo;
|
2 |
pub mod engine_models;
|
3 |
pub mod searx;
|
|
|
1 |
+
//! This module provides different modules which handles the functionlity to fetch results from the
|
2 |
+
//! upstream search engines based on user requested queries. Also provides different models to
|
3 |
+
//! provide a standard functions to be implemented for all the upstream search engine handling
|
4 |
+
//! code. Moreover, it also provides a custom error for the upstream search engine handling code.
|
5 |
+
|
6 |
pub mod duckduckgo;
|
7 |
pub mod engine_models;
|
8 |
pub mod searx;
|
src/engines/searx.rs
CHANGED
@@ -17,25 +17,6 @@ pub struct Searx;
|
|
17 |
|
18 |
#[async_trait::async_trait]
|
19 |
impl SearchEngine for Searx {
|
20 |
-
/// This function scrapes results from the upstream engine duckduckgo and puts all the scraped
|
21 |
-
/// results like title, visiting_url (href in html),engine (from which engine it was fetched from)
|
22 |
-
/// and description in a RawSearchResult and then adds that to HashMap whose keys are url and
|
23 |
-
/// values are RawSearchResult struct and then returns it within a Result enum.
|
24 |
-
///
|
25 |
-
/// # Arguments
|
26 |
-
///
|
27 |
-
/// * `query` - Takes the user provided query to query to the upstream search engine with.
|
28 |
-
/// * `page` - Takes an u32 as an argument.
|
29 |
-
/// * `user_agent` - Takes a random user agent string as an argument.
|
30 |
-
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
|
31 |
-
///
|
32 |
-
/// # Errors
|
33 |
-
///
|
34 |
-
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
|
35 |
-
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
|
36 |
-
/// provide results for the requested search query and also returns error if the scraping selector
|
37 |
-
/// or HeaderMap fails to initialize.
|
38 |
-
|
39 |
async fn results(
|
40 |
&self,
|
41 |
query: &str,
|
|
|
17 |
|
18 |
#[async_trait::async_trait]
|
19 |
impl SearchEngine for Searx {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
async fn results(
|
21 |
&self,
|
22 |
query: &str,
|
src/handler/mod.rs
CHANGED
@@ -1 +1,5 @@
|
|
|
|
|
|
|
|
|
|
1 |
pub mod paths;
|
|
|
1 |
+
//! This module provides modules which provide the functionality to handle paths for different
|
2 |
+
//! files present on different paths and provide one appropriate path on which it is present and
|
3 |
+
//! can be used.
|
4 |
+
|
5 |
pub mod paths;
|
src/handler/paths.rs
CHANGED
@@ -7,42 +7,46 @@ use std::path::Path;
|
|
7 |
use std::sync::OnceLock;
|
8 |
|
9 |
// ------- Constants --------
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
15 |
|
|
|
16 |
#[derive(Hash, PartialEq, Eq, Debug)]
|
17 |
pub enum FileType {
|
|
|
18 |
Config,
|
|
|
19 |
AllowList,
|
|
|
20 |
BlockList,
|
|
|
21 |
Theme,
|
22 |
}
|
23 |
|
|
|
24 |
static FILE_PATHS_FOR_DIFF_FILE_TYPES: OnceLock<HashMap<FileType, Vec<String>>> = OnceLock::new();
|
25 |
|
26 |
-
/// A
|
27 |
-
/// file exists on that path.
|
28 |
///
|
29 |
/// # Error
|
30 |
///
|
31 |
-
/// Returns a
|
32 |
-
///
|
33 |
-
/// 1. `~/.config/websurfx/` if it not present here then it fallbacks to the next one (2)
|
34 |
-
/// 2. `/etc/xdg/websurfx/config.lua` if it is not present here then it fallbacks to the next
|
35 |
-
/// one (3).
|
36 |
-
/// 3. `websurfx/` (under project folder ( or codebase in other words)) if it is not present
|
37 |
-
/// here then it returns an error as mentioned above.
|
38 |
-
|
39 |
-
/// A function which returns an appropriate theme directory path checking if the theme
|
40 |
-
/// directory exists on that path.
|
41 |
///
|
42 |
-
/// #
|
|
|
|
|
|
|
43 |
///
|
44 |
-
/// Returns a `Theme (public) folder not found!!` error if the theme folder is not present under following
|
45 |
-
/// paths which are:
|
46 |
/// 1. `/opt/websurfx` if it not present here then it fallbacks to the next one (2)
|
47 |
/// 2. Under project folder ( or codebase in other words) if it is not present
|
48 |
/// here then it returns an error as mentioned above.
|
@@ -110,6 +114,6 @@ pub fn file_path(file_type: FileType) -> Result<&'static str, Error> {
|
|
110 |
// if no of the configs above exist, return error
|
111 |
Err(Error::new(
|
112 |
std::io::ErrorKind::NotFound,
|
113 |
-
format!("{:?} file not found!!", file_type),
|
114 |
))
|
115 |
}
|
|
|
7 |
use std::sync::OnceLock;
|
8 |
|
9 |
// ------- Constants --------
|
10 |
+
/// The constant holding the name of the theme folder.
|
11 |
+
const PUBLIC_DIRECTORY_NAME: &str = "public";
|
12 |
+
/// The constant holding the name of the common folder.
|
13 |
+
const COMMON_DIRECTORY_NAME: &str = "websurfx";
|
14 |
+
/// The constant holding the name of the config file.
|
15 |
+
const CONFIG_FILE_NAME: &str = "config.lua";
|
16 |
+
/// The constant holding the name of the AllowList text file.
|
17 |
+
const ALLOWLIST_FILE_NAME: &str = "allowlist.txt";
|
18 |
+
/// The constant holding the name of the BlockList text file.
|
19 |
+
const BLOCKLIST_FILE_NAME: &str = "blocklist.txt";
|
20 |
|
21 |
+
/// An enum type which provides different variants to handle paths for various files/folders.
|
22 |
#[derive(Hash, PartialEq, Eq, Debug)]
|
23 |
pub enum FileType {
|
24 |
+
/// This variant handles all the paths associated with the config file.
|
25 |
Config,
|
26 |
+
/// This variant handles all the paths associated with the Allowlist text file.
|
27 |
AllowList,
|
28 |
+
/// This variant handles all the paths associated with the BlockList text file.
|
29 |
BlockList,
|
30 |
+
/// This variant handles all the paths associated with the public folder (Theme folder).
|
31 |
Theme,
|
32 |
}
|
33 |
|
34 |
+
/// A static variable which stores the different filesystem paths for various file/folder types.
|
35 |
static FILE_PATHS_FOR_DIFF_FILE_TYPES: OnceLock<HashMap<FileType, Vec<String>>> = OnceLock::new();
|
36 |
|
37 |
+
/// A function which returns an appropriate path for thr provided file type by checking if the path
|
38 |
+
/// for the given file type exists on that path.
|
39 |
///
|
40 |
/// # Error
|
41 |
///
|
42 |
+
/// Returns a `<File Name> folder/file not found!!` error if the give file_type folder/file is not
|
43 |
+
/// present on the path on which it is being tested.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
///
|
45 |
+
/// # Example
|
46 |
+
///
|
47 |
+
/// If this function is give the file_type of Theme variant then the theme folder is checked by the
|
48 |
+
/// following steps:
|
49 |
///
|
|
|
|
|
50 |
/// 1. `/opt/websurfx` if it not present here then it fallbacks to the next one (2)
|
51 |
/// 2. Under project folder ( or codebase in other words) if it is not present
|
52 |
/// here then it returns an error as mentioned above.
|
|
|
114 |
// if no of the configs above exist, return error
|
115 |
Err(Error::new(
|
116 |
std::io::ErrorKind::NotFound,
|
117 |
+
format!("{:?} file/folder not found!!", file_type),
|
118 |
))
|
119 |
}
|
src/lib.rs
CHANGED
@@ -1,6 +1,10 @@
|
|
1 |
//! This main library module provides the functionality to provide and handle the Tcp server
|
2 |
//! and register all the routes for the `websurfx` meta search engine website.
|
3 |
|
|
|
|
|
|
|
|
|
4 |
pub mod cache;
|
5 |
pub mod config;
|
6 |
pub mod engines;
|
@@ -41,7 +45,7 @@ use handler::paths::{file_path, FileType};
|
|
41 |
/// let server = run(listener,config).expect("Failed to start server");
|
42 |
/// ```
|
43 |
pub fn run(listener: TcpListener, config: Config) -> std::io::Result<Server> {
|
44 |
-
let mut handlebars: Handlebars = Handlebars::new();
|
45 |
|
46 |
let public_folder_path: &str = file_path(FileType::Theme)?;
|
47 |
|
@@ -49,7 +53,7 @@ pub fn run(listener: TcpListener, config: Config) -> std::io::Result<Server> {
|
|
49 |
.register_templates_directory(".html", format!("{}/templates", public_folder_path))
|
50 |
.unwrap();
|
51 |
|
52 |
-
let handlebars_ref: web::Data<Handlebars
|
53 |
|
54 |
let cloned_config_threads_opt: u8 = config.threads;
|
55 |
|
|
|
1 |
//! This main library module provides the functionality to provide and handle the Tcp server
|
2 |
//! and register all the routes for the `websurfx` meta search engine website.
|
3 |
|
4 |
+
#![forbid(unsafe_code, clippy::panic)]
|
5 |
+
#![deny(missing_docs, clippy::missing_docs_in_private_items, clippy::perf)]
|
6 |
+
#![warn(clippy::cognitive_complexity, rust_2018_idioms)]
|
7 |
+
|
8 |
pub mod cache;
|
9 |
pub mod config;
|
10 |
pub mod engines;
|
|
|
45 |
/// let server = run(listener,config).expect("Failed to start server");
|
46 |
/// ```
|
47 |
pub fn run(listener: TcpListener, config: Config) -> std::io::Result<Server> {
|
48 |
+
let mut handlebars: Handlebars<'_> = Handlebars::new();
|
49 |
|
50 |
let public_folder_path: &str = file_path(FileType::Theme)?;
|
51 |
|
|
|
53 |
.register_templates_directory(".html", format!("{}/templates", public_folder_path))
|
54 |
.unwrap();
|
55 |
|
56 |
+
let handlebars_ref: web::Data<Handlebars<'_>> = web::Data::new(handlebars);
|
57 |
|
58 |
let cloned_config_threads_opt: u8 = config.threads;
|
59 |
|
src/results/aggregation_models.rs
CHANGED
@@ -9,20 +9,17 @@ use crate::{config::parser_models::Style, engines::engine_models::EngineError};
|
|
9 |
/// A named struct to store the raw scraped search results scraped search results from the
|
10 |
/// upstream search engines before aggregating it.It derives the Clone trait which is needed
|
11 |
/// to write idiomatic rust using `Iterators`.
|
12 |
-
///
|
13 |
-
/// # Fields
|
14 |
-
///
|
15 |
-
/// * `title` - The title of the search result.
|
16 |
-
/// * `url` - The url which is accessed when clicked on it
|
17 |
/// (href url in html in simple words).
|
18 |
-
|
19 |
-
/// * `engine` - The names of the upstream engines from which this results were provided.
|
20 |
-
#[derive(Clone, Serialize, Deserialize, Debug)]
|
21 |
#[serde(rename_all = "camelCase")]
|
22 |
pub struct SearchResult {
|
|
|
23 |
pub title: String,
|
|
|
24 |
pub url: String,
|
|
|
25 |
pub description: String,
|
|
|
26 |
pub engine: SmallVec<[String; 0]>,
|
27 |
}
|
28 |
|
@@ -64,14 +61,27 @@ impl SearchResult {
|
|
64 |
}
|
65 |
}
|
66 |
|
|
|
67 |
#[derive(Serialize, Deserialize, Clone)]
|
68 |
pub struct EngineErrorInfo {
|
|
|
|
|
69 |
pub error: String,
|
|
|
70 |
pub engine: String,
|
|
|
|
|
71 |
pub severity_color: String,
|
72 |
}
|
73 |
|
74 |
impl EngineErrorInfo {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
75 |
pub fn new(error: &EngineError, engine: &str) -> Self {
|
76 |
Self {
|
77 |
error: match error {
|
@@ -91,25 +101,26 @@ impl EngineErrorInfo {
|
|
91 |
|
92 |
/// A named struct to store, serialize, deserialize the all the search results scraped and
|
93 |
/// aggregated from the upstream search engines.
|
94 |
-
///
|
95 |
-
/// # Fields
|
96 |
-
///
|
97 |
-
/// * `results` - Stores the individual serializable `SearchResult` struct into a vector of
|
98 |
/// `SearchResult` structs.
|
99 |
-
/// * `page_query` - Stores the current pages search query `q` provided in the search url.
|
100 |
-
/// * `style` - Stores the theming options for the website.
|
101 |
-
/// * `engine_errors_info` - Stores the information on which engines failed with their engine name
|
102 |
-
/// and the type of error that caused it.
|
103 |
-
/// * `empty_result_set` - Stores a boolean which indicates that no engines gave a result for the
|
104 |
-
/// given search query.
|
105 |
#[derive(Serialize, Deserialize, Default)]
|
106 |
#[serde(rename_all = "camelCase")]
|
107 |
pub struct SearchResults {
|
|
|
108 |
pub results: Vec<SearchResult>,
|
|
|
109 |
pub page_query: String,
|
|
|
110 |
pub style: Style,
|
|
|
|
|
111 |
pub engine_errors_info: Vec<EngineErrorInfo>,
|
|
|
|
|
|
|
112 |
pub disallowed: bool,
|
|
|
|
|
|
|
113 |
pub filtered: bool,
|
114 |
}
|
115 |
|
@@ -122,9 +133,8 @@ impl SearchResults {
|
|
122 |
/// and stores it into a vector of `SearchResult` structs.
|
123 |
/// * `page_query` - Takes an argument of current page`s search query `q` provided in
|
124 |
/// the search url.
|
125 |
-
/// * `
|
126 |
-
///
|
127 |
-
/// * ``
|
128 |
pub fn new(
|
129 |
results: Vec<SearchResult>,
|
130 |
page_query: &str,
|
|
|
9 |
/// A named struct to store the raw scraped search results scraped search results from the
|
10 |
/// upstream search engines before aggregating it.It derives the Clone trait which is needed
|
11 |
/// to write idiomatic rust using `Iterators`.
|
|
|
|
|
|
|
|
|
|
|
12 |
/// (href url in html in simple words).
|
13 |
+
#[derive(Clone, Serialize, Deserialize)]
|
|
|
|
|
14 |
#[serde(rename_all = "camelCase")]
|
15 |
pub struct SearchResult {
|
16 |
+
/// The title of the search result.
|
17 |
pub title: String,
|
18 |
+
/// The url which is accessed when clicked on it
|
19 |
pub url: String,
|
20 |
+
/// The description of the search result.
|
21 |
pub description: String,
|
22 |
+
/// The names of the upstream engines from which this results were provided.
|
23 |
pub engine: SmallVec<[String; 0]>,
|
24 |
}
|
25 |
|
|
|
61 |
}
|
62 |
}
|
63 |
|
64 |
+
/// A named struct that stores the error info related to the upstream search engines.
|
65 |
#[derive(Serialize, Deserialize, Clone)]
|
66 |
pub struct EngineErrorInfo {
|
67 |
+
/// It stores the error type which occured while fetching the result from a particular search
|
68 |
+
/// engine.
|
69 |
pub error: String,
|
70 |
+
/// It stores the name of the engine that failed to provide the requested search results.
|
71 |
pub engine: String,
|
72 |
+
/// It stores the name of the color to indicate whether how severe the particular error is (In
|
73 |
+
/// other words it indicates the severity of the error/issue).
|
74 |
pub severity_color: String,
|
75 |
}
|
76 |
|
77 |
impl EngineErrorInfo {
|
78 |
+
/// Constructs a new `SearchResult` with the given arguments needed for the struct.
|
79 |
+
///
|
80 |
+
/// # Arguments
|
81 |
+
///
|
82 |
+
/// * `error` - It takes the error type which occured while fetching the result from a particular
|
83 |
+
/// search engine.
|
84 |
+
/// * `engine` - It takes the name of the engine that failed to provide the requested search results.
|
85 |
pub fn new(error: &EngineError, engine: &str) -> Self {
|
86 |
Self {
|
87 |
error: match error {
|
|
|
101 |
|
102 |
/// A named struct to store, serialize, deserialize the all the search results scraped and
|
103 |
/// aggregated from the upstream search engines.
|
|
|
|
|
|
|
|
|
104 |
/// `SearchResult` structs.
|
|
|
|
|
|
|
|
|
|
|
|
|
105 |
#[derive(Serialize, Deserialize, Default)]
|
106 |
#[serde(rename_all = "camelCase")]
|
107 |
pub struct SearchResults {
|
108 |
+
/// Stores the individual serializable `SearchResult` struct into a vector of
|
109 |
pub results: Vec<SearchResult>,
|
110 |
+
/// Stores the current pages search query `q` provided in the search url.
|
111 |
pub page_query: String,
|
112 |
+
/// Stores the theming options for the website.
|
113 |
pub style: Style,
|
114 |
+
/// Stores the information on which engines failed with their engine name
|
115 |
+
/// and the type of error that caused it.
|
116 |
pub engine_errors_info: Vec<EngineErrorInfo>,
|
117 |
+
/// Stores the flag option which holds the check value that the following
|
118 |
+
/// search query was disallowed when the safe search level set to 4 and it
|
119 |
+
/// was present in the `Blocklist` file.
|
120 |
pub disallowed: bool,
|
121 |
+
/// Stores the flag option which holds the check value that the following
|
122 |
+
/// search query was filtered when the safe search level set to 3 and it
|
123 |
+
/// was present in the `Blocklist` file.
|
124 |
pub filtered: bool,
|
125 |
}
|
126 |
|
|
|
133 |
/// and stores it into a vector of `SearchResult` structs.
|
134 |
/// * `page_query` - Takes an argument of current page`s search query `q` provided in
|
135 |
/// the search url.
|
136 |
+
/// * `engine_errors_info` - Takes an array of structs which contains information regarding
|
137 |
+
/// which engines failed with their names, reason and their severity color name.
|
|
|
138 |
pub fn new(
|
139 |
results: Vec<SearchResult>,
|
140 |
page_query: &str,
|
src/results/mod.rs
CHANGED
@@ -1,3 +1,7 @@
|
|
|
|
|
|
|
|
|
|
1 |
pub mod aggregation_models;
|
2 |
pub mod aggregator;
|
3 |
pub mod user_agent;
|
|
|
1 |
+
//! This module provides modules that handle the functionality to aggregate the fetched search
|
2 |
+
//! results from the upstream search engines and filters it if safe search is set to 3 or 4. Also,
|
3 |
+
//! provides various models to aggregate search results into a standardized form.
|
4 |
+
|
5 |
pub mod aggregation_models;
|
6 |
pub mod aggregator;
|
7 |
pub mod user_agent;
|
src/results/user_agent.rs
CHANGED
@@ -4,6 +4,8 @@ use std::sync::OnceLock;
|
|
4 |
|
5 |
use fake_useragent::{Browsers, UserAgents, UserAgentsBuilder};
|
6 |
|
|
|
|
|
7 |
static USER_AGENTS: OnceLock<UserAgents> = OnceLock::new();
|
8 |
|
9 |
/// A function to generate random user agent to improve privacy of the user.
|
|
|
4 |
|
5 |
use fake_useragent::{Browsers, UserAgents, UserAgentsBuilder};
|
6 |
|
7 |
+
/// A static variable which stores the initially build `UserAgents` struct. So as it can be resused
|
8 |
+
/// again and again without the need of reinitializing the `UserAgents` struct.
|
9 |
static USER_AGENTS: OnceLock<UserAgents> = OnceLock::new();
|
10 |
|
11 |
/// A function to generate random user agent to improve privacy of the user.
|
src/server/mod.rs
CHANGED
@@ -1 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
1 |
pub mod routes;
|
|
|
1 |
+
//! This module provides modules that handle the functionality of handling different routes/paths
|
2 |
+
//! for the `websurfx` search engine website. Also it handles the parsing of search parameters in
|
3 |
+
//! the search route. Also, caches the next, current and previous search results in the search
|
4 |
+
//! routes with the help of the redis server.
|
5 |
+
|
6 |
pub mod routes;
|
src/server/routes.rs
CHANGED
@@ -25,17 +25,16 @@ use tokio::join;
|
|
25 |
static REDIS_CACHE: async_once_cell::OnceCell<RedisCache> = async_once_cell::OnceCell::new();
|
26 |
|
27 |
/// A named struct which deserializes all the user provided search parameters and stores them.
|
28 |
-
///
|
29 |
-
/// # Fields
|
30 |
-
///
|
31 |
-
/// * `q` - It stores the search parameter option `q` (or query in simple words)
|
32 |
-
/// of the search url.
|
33 |
-
/// * `page` - It stores the search parameter `page` (or pageno in simple words)
|
34 |
-
/// of the search url.
|
35 |
#[derive(Deserialize)]
|
36 |
struct SearchParams {
|
|
|
|
|
37 |
q: Option<String>,
|
|
|
|
|
38 |
page: Option<u32>,
|
|
|
|
|
39 |
safesearch: Option<u8>,
|
40 |
}
|
41 |
|
@@ -63,17 +62,14 @@ pub async fn not_found(
|
|
63 |
}
|
64 |
|
65 |
/// A named struct which is used to deserialize the cookies fetched from the client side.
|
66 |
-
///
|
67 |
-
/// # Fields
|
68 |
-
///
|
69 |
-
/// * `theme` - It stores the theme name used in the website.
|
70 |
-
/// * `colorscheme` - It stores the colorscheme name used for the website theme.
|
71 |
-
/// * `engines` - It stores the user selected upstream search engines selected from the UI.
|
72 |
#[allow(dead_code)]
|
73 |
#[derive(Deserialize)]
|
74 |
struct Cookie<'a> {
|
|
|
75 |
theme: &'a str,
|
|
|
76 |
colorscheme: &'a str,
|
|
|
77 |
engines: Vec<&'a str>,
|
78 |
}
|
79 |
|
@@ -174,8 +170,21 @@ pub async fn search(
|
|
174 |
}
|
175 |
}
|
176 |
|
177 |
-
/// Fetches the results for a query and page.
|
178 |
-
///
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
179 |
async fn results(
|
180 |
url: String,
|
181 |
config: &Config,
|
@@ -184,6 +193,7 @@ async fn results(
|
|
184 |
req: HttpRequest,
|
185 |
safe_search: u8,
|
186 |
) -> Result<SearchResults, Box<dyn std::error::Error>> {
|
|
|
187 |
let mut redis_cache: RedisCache = REDIS_CACHE
|
188 |
.get_or_init(async {
|
189 |
// Initialize redis cache connection pool only one and store it in the heap.
|
@@ -191,7 +201,6 @@ async fn results(
|
|
191 |
})
|
192 |
.await
|
193 |
.clone();
|
194 |
-
|
195 |
// fetch the cached results json.
|
196 |
let cached_results_json: Result<String, error_stack::Report<crate::cache::error::PoolError>> =
|
197 |
redis_cache.clone().cached_json(&url).await;
|
@@ -223,7 +232,8 @@ async fn results(
|
|
223 |
// UI and use that.
|
224 |
let mut results: SearchResults = match req.cookie("appCookie") {
|
225 |
Some(cookie_value) => {
|
226 |
-
let cookie_value: Cookie =
|
|
|
227 |
|
228 |
let engines: Vec<EngineHandler> = cookie_value
|
229 |
.engines
|
@@ -267,6 +277,8 @@ async fn results(
|
|
267 |
}
|
268 |
}
|
269 |
|
|
|
|
|
270 |
fn is_match_from_filter_list(
|
271 |
file_path: &str,
|
272 |
query: &str,
|
|
|
25 |
static REDIS_CACHE: async_once_cell::OnceCell<RedisCache> = async_once_cell::OnceCell::new();
|
26 |
|
27 |
/// A named struct which deserializes all the user provided search parameters and stores them.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
28 |
#[derive(Deserialize)]
|
29 |
struct SearchParams {
|
30 |
+
/// It stores the search parameter option `q` (or query in simple words)
|
31 |
+
/// of the search url.
|
32 |
q: Option<String>,
|
33 |
+
/// It stores the search parameter `page` (or pageno in simple words)
|
34 |
+
/// of the search url.
|
35 |
page: Option<u32>,
|
36 |
+
/// It stores the search parameter `safesearch` (or safe search level in simple words) of the
|
37 |
+
/// search url.
|
38 |
safesearch: Option<u8>,
|
39 |
}
|
40 |
|
|
|
62 |
}
|
63 |
|
64 |
/// A named struct which is used to deserialize the cookies fetched from the client side.
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
#[allow(dead_code)]
|
66 |
#[derive(Deserialize)]
|
67 |
struct Cookie<'a> {
|
68 |
+
/// It stores the theme name used in the website.
|
69 |
theme: &'a str,
|
70 |
+
/// It stores the colorscheme name used for the website theme.
|
71 |
colorscheme: &'a str,
|
72 |
+
/// It stores the user selected upstream search engines selected from the UI.
|
73 |
engines: Vec<&'a str>,
|
74 |
}
|
75 |
|
|
|
170 |
}
|
171 |
}
|
172 |
|
173 |
+
/// Fetches the results for a query and page. It First checks the redis cache, if that
|
174 |
+
/// fails it gets proper results by requesting from the upstream search engines.
|
175 |
+
///
|
176 |
+
/// # Arguments
|
177 |
+
///
|
178 |
+
/// * `url` - It takes the url of the current page that requested the search results for a
|
179 |
+
/// particular search query.
|
180 |
+
/// * `config` - It takes a parsed config struct.
|
181 |
+
/// * `query` - It takes the page number as u32 value.
|
182 |
+
/// * `req` - It takes the `HttpRequest` struct as a value.
|
183 |
+
///
|
184 |
+
/// # Error
|
185 |
+
///
|
186 |
+
/// It returns the `SearchResults` struct if the search results could be successfully fetched from
|
187 |
+
/// the cache or from the upstream search engines otherwise it returns an appropriate error.
|
188 |
async fn results(
|
189 |
url: String,
|
190 |
config: &Config,
|
|
|
193 |
req: HttpRequest,
|
194 |
safe_search: u8,
|
195 |
) -> Result<SearchResults, Box<dyn std::error::Error>> {
|
196 |
+
// Initialize redis cache connection struct
|
197 |
let mut redis_cache: RedisCache = REDIS_CACHE
|
198 |
.get_or_init(async {
|
199 |
// Initialize redis cache connection pool only one and store it in the heap.
|
|
|
201 |
})
|
202 |
.await
|
203 |
.clone();
|
|
|
204 |
// fetch the cached results json.
|
205 |
let cached_results_json: Result<String, error_stack::Report<crate::cache::error::PoolError>> =
|
206 |
redis_cache.clone().cached_json(&url).await;
|
|
|
232 |
// UI and use that.
|
233 |
let mut results: SearchResults = match req.cookie("appCookie") {
|
234 |
Some(cookie_value) => {
|
235 |
+
let cookie_value: Cookie<'_> =
|
236 |
+
serde_json::from_str(cookie_value.name_value().1)?;
|
237 |
|
238 |
let engines: Vec<EngineHandler> = cookie_value
|
239 |
.engines
|
|
|
277 |
}
|
278 |
}
|
279 |
|
280 |
+
/// A helper function which checks whether the search query contains any keywords which should be
|
281 |
+
/// disallowed/allowed based on the regex based rules present in the blocklist and allowlist files.
|
282 |
fn is_match_from_filter_list(
|
283 |
file_path: &str,
|
284 |
query: &str,
|