⚙️ 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

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,