jstrong
/
zola
1
0
Fork 0
Code Issues 0 Pull Requests 0 Releases 0 Wiki Activity
Browse Source

Add anchor existence checking to link_checker component (#786) 

* Add anchor existant checking to link_checker component

* Oops, forgot some changes

* Drop scraper dependency and rework tests

* Handle name attributes
index-subcmd
Neil Kistner Vincent Prouillet 5 years ago
parent
096fefe7ed
commit
52c2b74b39
3 changed files with 103 additions and 1 deletions
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. +1
    -0
      Cargo.lock
  2. +2
    -0
      components/link_checker/Cargo.toml
  3. +100
    -1
      components/link_checker/src/lib.rs

+ 1
- 0
Cargo.lock View File

@@ -1223,6 +1223,7 @@ dependencies = [
name = "link_checker" name = "link_checker"
version = "0.1.0" version = "0.1.0"
dependencies = [ dependencies = [
"errors 0.1.0",
"lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)", "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
"reqwest 0.9.20 (registry+https://github.com/rust-lang/crates.io-index)", "reqwest 0.9.20 (registry+https://github.com/rust-lang/crates.io-index)",
] ]


+ 2
- 0
components/link_checker/Cargo.toml View File

@@ -6,3 +6,5 @@ authors = ["Vincent Prouillet <prouillet.vincent@gmail.com>"]
[dependencies] [dependencies]
reqwest = "0.9" reqwest = "0.9"
lazy_static = "1" lazy_static = "1"

errors = { path = "../errors" }

+ 100
- 1
components/link_checker/src/lib.rs View File

@@ -2,8 +2,13 @@ extern crate reqwest;
#[macro_use] #[macro_use]
extern crate lazy_static; extern crate lazy_static;


extern crate errors;

use reqwest::header::{HeaderMap, ACCEPT}; use reqwest::header::{HeaderMap, ACCEPT};
use reqwest::StatusCode; use reqwest::StatusCode;

use errors::Result;

use std::collections::HashMap; use std::collections::HashMap;
use std::error::Error; use std::error::Error;
use std::sync::{Arc, RwLock}; use std::sync::{Arc, RwLock};
@@ -62,6 +67,12 @@ pub fn check_url(url: &str) -> LinkResult {


// Need to actually do the link checking // Need to actually do the link checking
let res = match client.get(url).headers(headers).send() { let res = match client.get(url).headers(headers).send() {
Ok(ref mut response) if has_anchor(url) => {
match check_page_for_anchor(url, response.text()) {
Ok(_) => LinkResult { code: Some(response.status()), error: None },
Err(e) => LinkResult { code: None, error: Some(e.to_string()) },
}
}
Ok(response) => LinkResult { code: Some(response.status()), error: None }, Ok(response) => LinkResult { code: Some(response.status()), error: None },
Err(e) => LinkResult { code: None, error: Some(e.description().to_string()) }, Err(e) => LinkResult { code: None, error: Some(e.description().to_string()) },
}; };
@@ -70,9 +81,37 @@ pub fn check_url(url: &str) -> LinkResult {
res res
} }


fn has_anchor(url: &str) -> bool {
match url.find('#') {
Some(index) => match url.get(index..=index + 1) {
Some("#/") | Some("#!") | None => false,
Some(_) => true,
},
None => false,
}
}

fn check_page_for_anchor(url: &str, body: reqwest::Result<String>) -> Result<()> {
let body = body.unwrap();
let index = url.find('#').unwrap();
let anchor = url.get(index + 1..).unwrap();
let checks: [String; 4] = [
format!(" id='{}'", anchor),
format!(r#" id="{}""#, anchor),
format!(" name='{}'", anchor),
format!(r#" name="{}""#, anchor),
];

if checks.iter().any(|check| body[..].contains(&check[..])) {
Ok(())
} else {
Err(errors::Error::from(format!("Anchor `#{}` not found on page", anchor)))
}
}

#[cfg(test)] #[cfg(test)]
mod tests { mod tests {
use super::{check_url, LINKS};
use super::{check_page_for_anchor, check_url, has_anchor, LINKS};


#[test] #[test]
fn can_validate_ok_links() { fn can_validate_ok_links() {
@@ -91,4 +130,64 @@ mod tests {
assert!(res.code.is_none()); assert!(res.code.is_none());
assert!(res.error.is_some()); assert!(res.error.is_some());
} }

#[test]
fn can_validate_anchors() {
let url = "https://doc.rust-lang.org/std/iter/trait.Iterator.html#method.collect";
let body = "<body><h3 id='method.collect'>collect</h3></body>".to_string();
let res = check_page_for_anchor(url, Ok(body));
assert!(res.is_ok());
}

#[test]
fn can_validate_anchors_with_other_quotes() {
let url = "https://doc.rust-lang.org/std/iter/trait.Iterator.html#method.collect";
let body = r#"<body><h3 id="method.collect">collect</h3></body>"#.to_string();
let res = check_page_for_anchor(url, Ok(body));
assert!(res.is_ok());
}

#[test]
fn can_validate_anchors_with_name_attr() {
let url = "https://doc.rust-lang.org/std/iter/trait.Iterator.html#method.collect";
let body = r#"<body><h3 name="method.collect">collect</h3></body>"#.to_string();
let res = check_page_for_anchor(url, Ok(body));
assert!(res.is_ok());
}

#[test]
fn can_fail_when_anchor_not_found() {
let url = "https://doc.rust-lang.org/std/iter/trait.Iterator.html#me";
let body = "<body><h3 id='method.collect'>collect</h3></body>".to_string();
let res = check_page_for_anchor(url, Ok(body));
assert!(res.is_err());
}

#[test]
fn can_check_url_for_anchor() {
let url = "https://doc.rust-lang.org/std/index.html#the-rust-standard-library";
let res = has_anchor(url);
assert_eq!(res, true);
}

#[test]
fn will_return_false_when_no_anchor() {
let url = "https://doc.rust-lang.org/std/index.html";
let res = has_anchor(url);
assert_eq!(res, false);
}

#[test]
fn will_return_false_when_has_router_url() {
let url = "https://doc.rust-lang.org/#/std";
let res = has_anchor(url);
assert_eq!(res, false);
}

#[test]
fn will_return_false_when_has_router_url_alt() {
let url = "https://doc.rust-lang.org/#!/std";
let res = has_anchor(url);
assert_eq!(res, false);
}
} }

Write Preview
Loading…
Cancel
Save