Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve test suite #16

Merged
merged 4 commits into from
Dec 13, 2023
Merged

Improve test suite #16

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
327e267
Add test for page that returns 404
tsengia Dec 13, 2023
b88f59e
Fix case when page returns HTTP error code
tsengia Dec 13, 2023
d8507fd
Add tests for empty or missing href attributes
tsengia Dec 13, 2023
2c4e14e
Formatting changes
tsengia Dec 13, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions src/algo.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,11 @@ pub async fn visit_page(

// Record the HTTP status code
page.status_code = Some(response.status());
if !response.status().is_success() {
println!("Found bad link! {}", url);
page.good = Some(false);
return false;
}

// Attempt to get the Content-Type of the page
let (parse_html, content_type) = check_content_type(&response);
Expand Down
169 changes: 169 additions & 0 deletions src/tests.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,3 +87,172 @@ async fn test_two_pages() {
// Make sure there are two pages in the page map
assert_eq!(map.len(), 2);
}

#[tokio::test]
async fn test_missing_page() {
let mut server = Server::new();

let url = server.url();
let parsed_url = Url::parse(url.as_str()).unwrap();

let mock = server.mock("GET", "/")
.with_status(201)
.with_header("content-type", "text/html")
.with_body("<!DOCTYPE html><html><body><a href=\"page2.html\" >This points to a missing page!</a></body></html>")
.create();

let missing_page_mock = server.mock("GET", "/page2.html").with_status(404).create();

let mut spider_crab = SpiderCrab::new(&[url.as_str()]);

let success = spider_crab.visit_website(url.as_str()).await;

// Make sure the HTTP request was made to the first page
mock.assert();

// Make sure the HTTP request was made to the missing page
missing_page_mock.assert();

// Make sure that visit _website() returned false
assert!(!success);

let graph = &spider_crab.graph;
// Make sure that the page graph contains two pages
assert_eq!(graph.node_count(), 2);

// Make sure there is only one link in the page graph
assert_eq!(graph.edge_count(), 1);

let map = &spider_crab.map;

// Make sure that the page map contains the mock page
assert!(map.contains_key(&parsed_url));

// Make sure there are two pages in the page map
assert_eq!(map.len(), 2);
}

#[tokio::test]
async fn test_missing_href() {
let mut server = Server::new();

let url = server.url();
let parsed_url = Url::parse(url.as_str()).unwrap();

let mock = server.mock("GET", "/")
.with_status(201)
.with_header("content-type", "text/html")
.with_body("<!DOCTYPE html><html><body><a>This link doesn't have an href attribute!</a></body></html>")
.create();

let mut spider_crab = SpiderCrab::new(&[url.as_str()]);

let success = spider_crab.visit_website(url.as_str()).await;

// Make sure the HTTP request was made to the first page
mock.assert();

// Make sure that visit _website() returned false
assert!(!success);

let graph = &spider_crab.graph;
// Make sure that the page graph contains one page
assert_eq!(graph.node_count(), 1);

// Make sure there are no links in the page graph
assert_eq!(graph.edge_count(), 0);

let map = &spider_crab.map;

// Make sure that the page map contains the mock page
assert!(map.contains_key(&parsed_url));

// Make sure there is only one page in the page map
assert_eq!(map.len(), 1);
}

#[tokio::test]
async fn test_empty_href() {
let mut server = Server::new();

let url = server.url();
let parsed_url = Url::parse(url.as_str()).unwrap();

let mock = server.mock("GET", "/")
.with_status(201)
.with_header("content-type", "text/html")
.with_body("<!DOCTYPE html><html><body><a href=\"\">This link's href attribute is empty!</a></body></html>")
.create();

let mut spider_crab = SpiderCrab::new(&[url.as_str()]);

let success = spider_crab.visit_website(url.as_str()).await;

// Make sure the HTTP request was made to the first page
mock.assert();

// Make sure that visit _website() returned false
assert!(!success);

let graph = &spider_crab.graph;
// Make sure that the page graph contains one page
assert_eq!(graph.node_count(), 1);

// Make sure there are no links in the page graph
assert_eq!(graph.edge_count(), 0);

let map = &spider_crab.map;

// Make sure that the page map contains the mock page
assert!(map.contains_key(&parsed_url));

// Make sure there is only one page in the page map
assert_eq!(map.len(), 1);
}

#[tokio::test]
async fn test_empty_href_in_second_page() {
let mut server = Server::new();

let url = server.url();
let parsed_url = Url::parse(url.as_str()).unwrap();

let mock = server.mock("GET", "/")
.with_status(201)
.with_header("content-type", "text/html")
.with_body("<!DOCTYPE html><html><body><a href=\"pageB.html\">This is a link to page B.</a></body></html>")
.create();

let mock_page_b = server.mock("GET", "/pageB.html")
.with_status(201)
.with_header("content-type", "text/html")
.with_body("<!DOCTYPE html><html><body><a href=\"\">This link has an empty href attribute!</a></body></html>")
.create();

let mut spider_crab = SpiderCrab::new(&[url.as_str()]);

let success = spider_crab.visit_website(url.as_str()).await;

// Make sure the HTTP request was made to the first page
mock.assert();
mock_page_b.assert();

// Make sure that visit _website() returned false
assert!(!success);

let graph = &spider_crab.graph;
// Make sure that the page graph contains two pages
assert_eq!(graph.node_count(), 2);

// Make sure there are is only one link in the graph
assert_eq!(graph.edge_count(), 1);

let map = &spider_crab.map;

// Make sure that the page map contains the mock page
assert!(map.contains_key(&parsed_url));
assert!(map.contains_key(&parsed_url.join("pageB.html").unwrap()));

// Make sure there are two pages in the page map
assert_eq!(map.len(), 2);
}