0
0
mirror of https://github.com/neon-mmd/websurfx.git synced 2024-12-22 20:38:22 -05:00

⚙️ refactor: change & add documentation to the code based on the lints (#205)

This commit is contained in:
neon_arch 2023-09-03 19:23:34 +03:00
parent 0d2d449889
commit 049b1c1ddd
16 changed files with 177 additions and 132 deletions

5
src/cache/cacher.rs vendored
View File

@ -6,11 +6,8 @@ use redis::{Client, Commands, Connection};
/// A named struct which stores the redis Connection url address to which the client will
/// connect to.
///
/// # Fields
///
/// * `redis_connection_url` - It stores the redis Connection url address.
pub struct RedisCache {
/// It stores the redis Connection url address.
connection: Connection,
}

3
src/cache/mod.rs vendored
View File

@ -1 +1,4 @@
//! This module provides the modules which provide the functionality to cache the aggregated
//! results fetched and aggregated from the upstream search engines in a json format.
pub mod cacher;

View File

@ -1,2 +1,5 @@
//! This module provides the modules which handles the functionality to parse the lua config
//! and convert the config options into rust readable form.
pub mod parser;
pub mod parser_models;

View File

@ -9,42 +9,36 @@ use rlua::Lua;
use std::{collections::HashMap, fs, thread::available_parallelism};
/// A named struct which stores the parsed config file options.
///
/// # Fields
//
/// * `port` - It stores the parsed port number option on which the server should launch.
/// * `binding_ip` - It stores the parsed ip address option on which the server should launch
/// * `style` - It stores the theming options for the website.
/// * `redis_url` - It stores the redis connection url address on which the redis
/// client should connect.
/// * `aggregator` - It stores the option to whether enable or disable production use.
/// * `logging` - It stores the option to whether enable or disable logs.
/// * `debug` - It stores the option to whether enable or disable debug mode.
/// * `upstream_search_engines` - It stores all the engine names that were enabled by the user.
/// * `request_timeout` - It stores the time (secs) which controls the server request timeout.
/// * `threads` - It stores the number of threads which controls the app will use to run.
#[derive(Clone)]
pub struct Config {
/// It stores the parsed port number option on which the server should launch.
pub port: u16,
/// It stores the parsed ip address option on which the server should launch
pub binding_ip: String,
/// It stores the theming options for the website.
pub style: Style,
/// It stores the redis connection url address on which the redis
/// client should connect.
pub redis_url: String,
/// It stores the option to whether enable or disable production use.
pub aggregator: AggregatorConfig,
/// It stores the option to whether enable or disable logs.
pub logging: bool,
/// It stores the option to whether enable or disable debug mode.
pub debug: bool,
/// It stores all the engine names that were enabled by the user.
pub upstream_search_engines: Vec<crate::engines::engine_models::EngineHandler>,
/// It stores the time (secs) which controls the server request timeout.
pub request_timeout: u8,
/// It stores the number of threads which controls the app will use to run.
pub threads: u8,
}
/// Configuration options for the aggregator.
///
/// # Fields
///
/// * `random_delay` - It stores the option to whether enable or disable random delays between
/// requests.
#[derive(Clone)]
pub struct AggregatorConfig {
/// It stores the option to whether enable or disable random delays between
/// requests.
pub random_delay: bool,
}
@ -115,6 +109,11 @@ impl Config {
}
/// a helper function that sets the proper logging level
///
/// # Arguments
///
/// * `debug` - It takes the option to whether enable or disable debug mode.
/// * `logging` - It takes the option to whether enable or disable logs.
fn set_logging_level(debug: bool, logging: bool) {
if let Ok(pkg_env_var) = std::env::var("PKG_ENV") {
if pkg_env_var.to_lowercase() == "dev" {

View File

@ -12,15 +12,12 @@ use serde::{Deserialize, Serialize};
/// order to allow the deserializing the json back to struct in aggregate function in
/// aggregator.rs and create a new struct out of it and then serialize it back to json and pass
/// it to the template files.
///
/// # Fields
//
/// * `theme` - It stores the parsed theme option used to set a theme for the website.
/// * `colorscheme` - It stores the parsed colorscheme option used to set a colorscheme for the
/// theme being used.
#[derive(Serialize, Deserialize, Clone)]
pub struct Style {
/// It stores the parsed theme option used to set a theme for the website.
pub theme: String,
/// It stores the parsed colorscheme option used to set a colorscheme for the
/// theme being used.
pub colorscheme: String,
}

View File

@ -19,25 +19,7 @@ pub struct DuckDuckGo;
#[async_trait::async_trait]
impl SearchEngine for DuckDuckGo {
/// This function scrapes results from the upstream engine duckduckgo and puts all the scraped
/// results like title, visiting_url (href in html),engine (from which engine it was fetched from)
/// and description in a RawSearchResult and then adds that to HashMap whose keys are url and
/// values are RawSearchResult struct and then returns it within a Result enum.
///
/// # Arguments
///
/// * `query` - Takes the user provided query to query to the upstream search engine with.
/// * `page` - Takes an u32 as an argument.
/// * `user_agent` - Takes a random user agent string as an argument.
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
///
/// # Errors
///
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
/// provide results for the requested search query and also returns error if the scraping selector
/// or HeaderMap fails to initialize.
async fn results(
async fn results(
&self,
query: String,
page: u32,

View File

@ -6,19 +6,18 @@ use error_stack::{IntoReport, Result, ResultExt};
use std::{collections::HashMap, fmt, time::Duration};
/// A custom error type used for handle engine associated errors.
///
/// This enum provides variants three different categories of errors:
/// * `RequestError` - This variant handles all request related errors like forbidden, not found,
/// etc.
/// * `EmptyResultSet` - This variant handles the not results found error provide by the upstream
/// search engines.
/// * `UnexpectedError` - This variant handles all the errors which are unexpected or occur rarely
/// and are errors mostly related to failure in initialization of HeaderMap, Selector errors and
/// all other errors occurring within the code handling the `upstream search engines`.
#[derive(Debug)]
pub enum EngineError {
/// This variant handles all request related errors like forbidden, not found,
/// etc.
EmptyResultSet,
/// This variant handles the not results found error provide by the upstream
/// search engines.
RequestError,
/// This variant handles all the errors which are unexpected or occur rarely
/// and are errors mostly related to failure in initialization of HeaderMap,
/// Selector errors and all other errors occurring within the code handling
/// the `upstream search engines`.
UnexpectedError,
}
@ -46,6 +45,23 @@ impl error_stack::Context for EngineError {}
/// A trait to define common behavior for all search engines.
#[async_trait::async_trait]
pub trait SearchEngine: Sync + Send {
/// This helper function fetches/requests the search results from the upstream search engine in
/// an html form.
///
/// # Arguments
///
/// * `url` - It takes the url of the upstream search engine with the user requested search
/// query appended in the search parameters.
/// * `header_map` - It takes the http request headers to be sent to the upstream engine in
/// order to prevent being detected as a bot. It takes the header as a HeaderMap type.
/// * `request_timeout` - It takes the request timeout value as seconds which is used to limit
/// the amount of time for each request to remain connected when until the results can be provided
/// by the upstream engine.
///
/// # Error
///
/// It returns the html data as a string if the upstream engine provides the data as expected
/// otherwise it returns a custom `EngineError`.
async fn fetch_html_from_upstream(
&self,
url: String,
@ -67,6 +83,24 @@ pub trait SearchEngine: Sync + Send {
.change_context(EngineError::RequestError)?)
}
/// This function scrapes results from the upstream engine and puts all the scraped results like
/// title, visiting_url (href in html),engine (from which engine it was fetched from) and description
/// in a RawSearchResult and then adds that to HashMap whose keys are url and values are RawSearchResult
/// struct and then returns it within a Result enum.
///
/// # Arguments
///
/// * `query` - Takes the user provided query to query to the upstream search engine with.
/// * `page` - Takes an u32 as an argument.
/// * `user_agent` - Takes a random user agent string as an argument.
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
///
/// # Errors
///
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
/// provide results for the requested search query and also returns error if the scraping selector
/// or HeaderMap fails to initialize.
async fn results(
&self,
query: String,
@ -76,8 +110,12 @@ pub trait SearchEngine: Sync + Send {
) -> Result<HashMap<String, SearchResult>, EngineError>;
}
/// A named struct which stores the engine struct with the name of the associated engine.
pub struct EngineHandler {
/// It stores the engine struct wrapped in a box smart pointer as the engine struct implements
/// the `SearchEngine` trait.
engine: Box<dyn SearchEngine>,
/// It stores the name of the engine to which the struct is associated to.
name: &'static str,
}
@ -88,7 +126,15 @@ impl Clone for EngineHandler {
}
impl EngineHandler {
/// parses an engine name into an engine handler, returns none if the engine is unknown
/// Parses an engine name into an engine handler.
///
/// # Arguments
///
/// * `engine_name` - It takes the name of the engine to which the struct was associated to.
///
/// # Returns
///
/// It returns an option either containing the value or a none if the engine is unknown
pub fn new(engine_name: &str) -> Option<Self> {
let engine: (&'static str, Box<dyn SearchEngine>) =
match engine_name.to_lowercase().as_str() {
@ -103,6 +149,8 @@ impl EngineHandler {
})
}
/// This function converts the EngineHandler type into a tuple containing the engine name and
/// the associated engine struct.
pub fn into_name_engine(self) -> (&'static str, Box<dyn SearchEngine>) {
(self.name, self.engine)
}

View File

@ -1,3 +1,8 @@
//! This module provides different modules which handles the functionlity to fetch results from the
//! upstream search engines based on user requested queries. Also provides different models to
//! provide a standard functions to be implemented for all the upstream search engine handling
//! code. Moreover, it also provides a custom error for the upstream search engine handling code.
pub mod duckduckgo;
pub mod engine_models;
pub mod searx;

View File

@ -17,25 +17,6 @@ pub struct Searx;
#[async_trait::async_trait]
impl SearchEngine for Searx {
/// This function scrapes results from the upstream engine duckduckgo and puts all the scraped
/// results like title, visiting_url (href in html),engine (from which engine it was fetched from)
/// and description in a RawSearchResult and then adds that to HashMap whose keys are url and
/// values are RawSearchResult struct and then returns it within a Result enum.
///
/// # Arguments
///
/// * `query` - Takes the user provided query to query to the upstream search engine with.
/// * `page` - Takes an u32 as an argument.
/// * `user_agent` - Takes a random user agent string as an argument.
/// * `request_timeout` - Takes a time (secs) as a value which controls the server request timeout.
///
/// # Errors
///
/// Returns an `EngineErrorKind` if the user is not connected to the internet or if their is failure to
/// reach the above `upstream search engine` page or if the `upstream search engine` is unable to
/// provide results for the requested search query and also returns error if the scraping selector
/// or HeaderMap fails to initialize.
async fn results(
&self,
query: String,

View File

@ -1 +1,5 @@
//! This module provides modules which provide the functionality to handle paths for different
//! files present on different paths and provide one appropriate path on which it is present and
//! can be used.
pub mod paths;

View File

@ -6,20 +6,31 @@ use std::io::Error;
use std::path::Path;
// ------- Constants --------
static PUBLIC_DIRECTORY_NAME: &str = "public";
static COMMON_DIRECTORY_NAME: &str = "websurfx";
static CONFIG_FILE_NAME: &str = "config.lua";
static ALLOWLIST_FILE_NAME: &str = "allowlist.txt";
static BLOCKLIST_FILE_NAME: &str = "blocklist.txt";
/// The constant holding the name of the theme folder.
const PUBLIC_DIRECTORY_NAME: &str = "public";
/// The constant holding the name of the common folder.
const COMMON_DIRECTORY_NAME: &str = "websurfx";
/// The constant holding the name of the config file.
const CONFIG_FILE_NAME: &str = "config.lua";
/// The constant holding the name of the AllowList text file.
const ALLOWLIST_FILE_NAME: &str = "allowlist.txt";
/// The constant holding the name of the BlockList text file.
const BLOCKLIST_FILE_NAME: &str = "blocklist.txt";
/// An enum type which provides different variants to handle paths for various files/folders.
#[derive(Hash, PartialEq, Eq, Debug)]
pub enum FileType {
/// This variant handles all the paths associated with the config file.
Config,
/// This variant handles all the paths associated with the Allowlist text file.
AllowList,
/// This variant handles all the paths associated with the BlockList text file.
BlockList,
/// This variant handles all the paths associated with the public folder (Theme folder).
Theme,
}
/// A static variable which stores the different filesystem paths for various file/folder types.
static FILE_PATHS_FOR_DIFF_FILE_TYPES: once_cell::sync::Lazy<HashMap<FileType, Vec<String>>> =
once_cell::sync::Lazy::new(|| {
HashMap::from([
@ -72,26 +83,19 @@ static FILE_PATHS_FOR_DIFF_FILE_TYPES: once_cell::sync::Lazy<HashMap<FileType, V
])
});
/// A helper function which returns an appropriate config file path checking if the config
/// file exists on that path.
/// A function which returns an appropriate path for thr provided file type by checking if the path
/// for the given file type exists on that path.
///
/// # Error
///
/// Returns a `config file not found!!` error if the config file is not present under following
/// paths which are:
/// 1. `~/.config/websurfx/` if it not present here then it fallbacks to the next one (2)
/// 2. `/etc/xdg/websurfx/config.lua` if it is not present here then it fallbacks to the next
/// one (3).
/// 3. `websurfx/` (under project folder ( or codebase in other words)) if it is not present
/// here then it returns an error as mentioned above.
/// A function which returns an appropriate theme directory path checking if the theme
/// directory exists on that path.
/// Returns a `<File Name> folder/file not found!!` error if the give file_type folder/file is not
/// present on the path on which it is being tested.
///
/// # Error
/// # Example
///
/// If this function is give the file_type of Theme variant then the theme folder is checked by the
/// following steps:
///
/// Returns a `Theme (public) folder not found!!` error if the theme folder is not present under following
/// paths which are:
/// 1. `/opt/websurfx` if it not present here then it fallbacks to the next one (2)
/// 2. Under project folder ( or codebase in other words) if it is not present
/// here then it returns an error as mentioned above.
@ -106,6 +110,6 @@ pub fn file_path(file_type: FileType) -> Result<String, Error> {
// if no of the configs above exist, return error
Err(Error::new(
std::io::ErrorKind::NotFound,
format!("{:?} file not found!!", file_type),
format!("{:?} file/folder not found!!", file_type),
))
}

View File

@ -8,20 +8,17 @@ use crate::{config::parser_models::Style, engines::engine_models::EngineError};
/// A named struct to store the raw scraped search results scraped search results from the
/// upstream search engines before aggregating it.It derives the Clone trait which is needed
/// to write idiomatic rust using `Iterators`.
///
/// # Fields
///
/// * `title` - The title of the search result.
/// * `url` - The url which is accessed when clicked on it
/// (href url in html in simple words).
/// * `description` - The description of the search result.
/// * `engine` - The names of the upstream engines from which this results were provided.
#[derive(Clone, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub struct SearchResult {
/// The title of the search result.
pub title: String,
/// The url which is accessed when clicked on it
pub url: String,
/// The description of the search result.
pub description: String,
/// The names of the upstream engines from which this results were provided.
pub engine: Vec<String>,
}
@ -63,15 +60,27 @@ impl SearchResult {
}
}
///
/// A named struct that stores the error info related to the upstream search engines.
#[derive(Serialize, Deserialize)]
pub struct EngineErrorInfo {
/// It stores the error type which occured while fetching the result from a particular search
/// engine.
pub error: String,
/// It stores the name of the engine that failed to provide the requested search results.
pub engine: String,
/// It stores the name of the color to indicate whether how severe the particular error is (In
/// other words it indicates the severity of the error/issue).
pub severity_color: String,
}
impl EngineErrorInfo {
/// Constructs a new `SearchResult` with the given arguments needed for the struct.
///
/// # Arguments
///
/// * `error` - It takes the error type which occured while fetching the result from a particular
/// search engine.
/// * `engine` - It takes the name of the engine that failed to provide the requested search results.
pub fn new(error: &EngineError, engine: String) -> Self {
Self {
error: match error {
@ -91,23 +100,18 @@ impl EngineErrorInfo {
/// A named struct to store, serialize, deserialize the all the search results scraped and
/// aggregated from the upstream search engines.
///
/// # Fields
///
/// * `results` - Stores the individual serializable `SearchResult` struct into a vector of
/// `SearchResult` structs.
/// * `page_query` - Stores the current pages search query `q` provided in the search url.
/// * `style` - Stores the theming options for the website.
/// * `engine_errors_info` - Stores the information on which engines failed with their engine name
/// and the type of error that caused it.
/// * `empty_result_set` - Stores a boolean which indicates that no engines gave a result for the
/// given search query.
#[derive(Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub struct SearchResults {
/// Stores the individual serializable `SearchResult` struct into a vector of
pub results: Vec<SearchResult>,
/// Stores the current pages search query `q` provided in the search url.
pub page_query: String,
/// Stores the theming options for the website.
pub style: Style,
/// Stores the information on which engines failed with their engine name
/// and the type of error that caused it.
pub engine_errors_info: Vec<EngineErrorInfo>,
}

View File

@ -1,3 +1,7 @@
//! This module provides modules that handle the functionality to aggregate the fetched search
//! results from the upstream search engines and filters it if safe search is set to 3 or 4. Also,
//! provides various models to aggregate search results into a standardized form.
pub mod aggregation_models;
pub mod aggregator;
pub mod user_agent;

View File

@ -2,6 +2,8 @@
use fake_useragent::{Browsers, UserAgents, UserAgentsBuilder};
/// A static variable which stores the initially build `UserAgents` struct. So as it can be resused
/// again and again without the need of reinitializing the `UserAgents` struct.
static USER_AGENTS: once_cell::sync::Lazy<UserAgents> = once_cell::sync::Lazy::new(|| {
UserAgentsBuilder::new()
.cache(false)

View File

@ -1 +1,6 @@
//! This module provides modules that handle the functionality of handling different routes/paths
//! for the `websurfx` search engine website. Also it handles the parsing of search parameters in
//! the search route. Also, caches the next, current and previous search results in the search
//! routes with the help of the redis server.
pub mod routes;

View File

@ -17,16 +17,13 @@ use serde::Deserialize;
use tokio::join;
/// A named struct which deserializes all the user provided search parameters and stores them.
///
/// # Fields
///
/// * `q` - It stores the search parameter option `q` (or query in simple words)
/// of the search url.
/// * `page` - It stores the search parameter `page` (or pageno in simple words)
/// of the search url.
#[derive(Deserialize)]
struct SearchParams {
/// It stores the search parameter option `q` (or query in simple words)
/// of the search url.
q: Option<String>,
/// It stores the search parameter `page` (or pageno in simple words)
/// of the search url.
page: Option<u32>,
}
@ -54,17 +51,14 @@ pub async fn not_found(
}
/// A named struct which is used to deserialize the cookies fetched from the client side.
///
/// # Fields
///
/// * `theme` - It stores the theme name used in the website.
/// * `colorscheme` - It stores the colorscheme name used for the website theme.
/// * `engines` - It stores the user selected upstream search engines selected from the UI.
#[allow(dead_code)]
#[derive(Deserialize)]
struct Cookie {
/// It stores the theme name used in the website.
theme: String,
/// It stores the colorscheme name used for the website theme.
colorscheme: String,
/// It stores the user selected upstream search engines selected from the UI.
engines: Vec<String>,
}
@ -149,8 +143,21 @@ pub async fn search(
}
}
/// Fetches the results for a query and page.
/// First checks the redis cache, if that fails it gets proper results
/// Fetches the results for a query and page. It First checks the redis cache, if that
/// fails it gets proper results by requesting from the upstream search engines.
///
/// # Arguments
///
/// * `url` - It takes the url of the current page that requested the search results for a
/// particular search query.
/// * `config` - It takes a parsed config struct.
/// * `query` - It takes the page number as u32 value.
/// * `req` - It takes the `HttpRequest` struct as a value.
///
/// # Error
///
/// It returns the `SearchResults` struct if the search results could be successfully fetched from
/// the cache or from the upstream search engines otherwise it returns an appropriate error.
async fn results(
url: String,
config: &Config,
@ -158,7 +165,7 @@ async fn results(
page: u32,
req: HttpRequest,
) -> Result<SearchResults, Box<dyn std::error::Error>> {
//Initialize redis cache connection struct
// Initialize redis cache connection struct
let mut redis_cache = RedisCache::new(config.redis_url.clone())?;
// fetch the cached results json.
let cached_results_json = redis_cache.cached_json(&url);