Error handling (#24)

* Add test when url is invalid and panic

* Initial error handling

ref #22

* Rename ErrorKind::HttpClient => ErrorKind::Http

* Implement std::error::Error and rename to Error

Author: spk

src/http/reqwest/async_reqwest.rs

@@ -1,24 +1,26 @@
 use reqwest::{Client, Request};
-use reqwest::{Method, Error};
+use reqwest::Method;
+use reqwest::Error as ReqwestError;
 use reqwest::header::HeaderValue;
 use url::{Origin, Url};
 use reqwest::header::USER_AGENT;
 use crate::http::{RobotsTxtClient, DEFAULT_USER_AGENT};
 use crate::parser::{ParseResult, parse_fetched_robots_txt};
 use crate::model::FetchedRobotsTxt;
+use crate::model::{Error, ErrorKind};
 use std::pin::Pin;
 use futures::task::{Context, Poll};
 use futures::Future;
 use futures::future::TryFutureExt;
 use futures::future::ok as future_ok;
 
-type FetchFuture = Box<dyn Future<Output=Result<(ResponseInfo, String), Error>>>;
+type FetchFuture = Box<dyn Future<Output=Result<(ResponseInfo, String), ReqwestError>>>;
 
 impl RobotsTxtClient for Client {
-    type Result = RobotsTxtResponse;
+    type Result = Result<RobotsTxtResponse, Error>;
     fn fetch_robots_txt(&self, origin: Origin) -> Self::Result {
         let url = format!("{}/robots.txt", origin.unicode_serialization());
-        let url = Url::parse(&url).expect("Unable to parse robots.txt url");
+        let url = Url::parse(&url).map_err(|err| Error {kind: ErrorKind::Url(err)})?;
         let mut request = Request::new(Method::GET, url);
         let _ = request.headers_mut().insert(USER_AGENT, HeaderValue::from_static(DEFAULT_USER_AGENT));
         let response = self
@@ -29,11 +31,11 @@ impl RobotsTxtClient for Client {
                     return future_ok((response_info, response_text));
                 });
             });
-        let response: Pin<Box<dyn Future<Output=Result<(ResponseInfo, String), Error>>>> = Box::pin(response);
-        return RobotsTxtResponse {
+        let response: Pin<Box<dyn Future<Output=Result<(ResponseInfo, String), ReqwestError>>>> = Box::pin(response);
+        Ok(RobotsTxtResponse {
             origin,
             response,
-        }
+        })
     }
 }
 
@@ -55,7 +57,7 @@ impl RobotsTxtResponse {
 }
 
 impl Future for RobotsTxtResponse {
-    type Output = Result<ParseResult<FetchedRobotsTxt>, Error>;
+    type Output = Result<ParseResult<FetchedRobotsTxt>, ReqwestError>;
 
     fn poll(self: Pin<&mut Self>, cx: &mut Context) -> Poll<Self::Output> {
         let self_mut = self.get_mut();
@@ -73,4 +75,4 @@ impl Future for RobotsTxtResponse {
             },
         }
     }
-}
+}

src/http/reqwest/sync_reqwest.rs

@@ -1,23 +1,24 @@
 use reqwest::blocking::{Client, Request};
-use reqwest::{Method, Error};
+use reqwest::Method;
 use reqwest::header::HeaderValue;
 use url::{Origin, Url};
 use reqwest::header::USER_AGENT;
 use crate::http::{RobotsTxtClient, DEFAULT_USER_AGENT};
 use crate::parser::{ParseResult, parse_fetched_robots_txt};
 use crate::model::FetchedRobotsTxt;
+use crate::model::{Error, ErrorKind};
 
 impl RobotsTxtClient for Client {
     type Result = Result<ParseResult<FetchedRobotsTxt>, Error>;
     fn fetch_robots_txt(&self, origin: Origin) -> Self::Result {
         let url = format!("{}/robots.txt", origin.unicode_serialization());
-        let url = Url::parse(&url).expect("Unable to parse robots.txt url");
+        let url = Url::parse(&url).map_err(|err| Error {kind: ErrorKind::Url(err)})?;
         let mut request = Request::new(Method::GET, url);
         let _ = request.headers_mut().insert(USER_AGENT, HeaderValue::from_static(DEFAULT_USER_AGENT));
-        let response = self.execute(request)?;
+        let response = self.execute(request).map_err(|err| Error {kind: ErrorKind::Http(err)})?;
         let status_code = response.status().as_u16();
-        let text = response.text()?;
+        let text = response.text().map_err(|err| Error {kind: ErrorKind::Http(err)})?;
         let robots_txt = parse_fetched_robots_txt(origin, status_code, &text);
         return Ok(robots_txt);
     }
-}
+}

src/model.rs

@@ -14,4 +14,6 @@ pub (crate) use self::fetched_robots_txt::FetchedRobotsTxtContainer;
 mod fetched_robots_txt;
 pub use self::robots_txt::RobotsTxt;
 mod path;
-pub (crate) use self::path::Path;
+pub (crate) use self::path::Path;
+mod errors;
+pub use self::errors::{Error, ErrorKind};

src/model/errors.rs

@@ -0,0 +1,23 @@
+use std::fmt;
+
+#[derive(Debug)]
+pub struct Error {
+    pub kind: ErrorKind,
+}
+
+#[derive(Debug)]
+pub enum ErrorKind {
+    Url(url::ParseError),
+    Http(reqwest::Error),
+}
+
+impl fmt::Display for Error {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self.kind {
+            ErrorKind::Url(ref err) => err.fmt(f),
+            ErrorKind::Http(ref err) => err.fmt(f),
+        }
+    }
+}
+
+impl std::error::Error for Error {}

tests/test_reqwest_async.rs

@@ -3,14 +3,28 @@ use robotparser::service::RobotsTxtService;
 use reqwest::Client;
 use url::Url;
 use tokio::runtime::Runtime;
+use url::{Host, Origin};
 
 #[test]
 fn test_reqwest_async() {
     let mut runtime = Runtime::new().unwrap();
     let client = Client::new();
     let robots_txt_url = Url::parse("https://www.python.org/robots.txt").unwrap();
-    let robots_txt_response = runtime.block_on(client.fetch_robots_txt(robots_txt_url.origin()));
+    let robots_txt_response = runtime.block_on(client.fetch_robots_txt(robots_txt_url.origin()).unwrap());
     let robots_txt = robots_txt_response.unwrap().get_result();
     let fetch_url = Url::parse("https://www.python.org/robots.txt").unwrap();
     assert!(robots_txt.can_fetch("*", &fetch_url));
-}
+    let fetch_url = Url::parse("http://www.python.org/webstats/").unwrap();
+    assert!(!robots_txt.can_fetch("*", &fetch_url));
+}
+
+#[test]
+fn test_reqwest_blocking_panic_url() {
+    let client = Client::new();
+    let host = Host::Domain("python.org::".into());
+    let origin = Origin::Tuple("https".into(), host, 80);
+    match client.fetch_robots_txt(origin) {
+        Ok(_) => assert!(false),
+        Err(_) => assert!(true)
+    }
+}

tests/test_reqwest_blocking.rs

@@ -2,6 +2,7 @@ use robotparser::http::RobotsTxtClient;
 use robotparser::service::RobotsTxtService;
 use reqwest::blocking::Client;
 use url::Url;
+use url::{Host, Origin};
 
 #[test]
 fn test_reqwest_blocking() {
@@ -10,4 +11,17 @@ fn test_reqwest_blocking() {
     let robots_txt = client.fetch_robots_txt(robots_txt_url.origin()).unwrap().get_result();
     let fetch_url = Url::parse("https://www.python.org/robots.txt").unwrap();
     assert!(robots_txt.can_fetch("*", &fetch_url));
+    let fetch_url = Url::parse("https://www.python.org/webstats/").unwrap();
+    assert!(!robots_txt.can_fetch("*", &fetch_url));
+}
+
+#[test]
+fn test_reqwest_blocking_panic_url() {
+    let client = Client::new();
+    let host = Host::Domain("python.org::".into());
+    let origin = Origin::Tuple("https".into(), host, 80);
+    match client.fetch_robots_txt(origin) {
+        Ok(_) => assert!(false),
+        Err(_) => assert!(true)
+    }
 }