Add --skip-external-links CLI flag to check subcommand (#2756)

* Add `--skip-external-links` CLI flag to `check` subcommand

* Add a test for checks with external links skipped

* Add a description of `--skip-external-links` flag

components/site/src/lib.rs

@@ -65,6 +65,8 @@ pub struct Site {
     include_drafts: bool,
     build_mode: BuildMode,
     shortcode_definitions: HashMap<String, ShortcodeDefinition>,
+    /// Whether to check external links
+    check_external_links: bool,
 }
 
 impl Site {
@@ -108,6 +110,7 @@ impl Site {
             library: Arc::new(RwLock::new(Library::default())),
             build_mode: BuildMode::Disk,
             shortcode_definitions,
+            check_external_links: true,
         };
 
         Ok(site)
@@ -126,6 +129,11 @@ impl Site {
         self.include_drafts = true;
     }
 
+    /// Set the site checker to skip external links check.
+    pub fn skip_external_links_check(&mut self) {
+        self.check_external_links = false;
+    }
+
     /// The index sections are ALWAYS at those paths
     /// There are one index section for the default language + 1 per language
     fn index_section_paths(&self) -> Vec<(PathBuf, Option<&str>)> {
@@ -344,7 +352,7 @@ impl Site {
         }
 
         // check external links, log the results, and error out if needed
-        if self.config.is_in_check_mode() {
+        if self.config.is_in_check_mode() && self.check_external_links {
             let external_link_messages = link_checking::check_external_links(self);
             if !external_link_messages.is_empty() {
                 let messages: Vec<String> = external_link_messages

components/site/tests/site.rs

@@ -851,6 +851,32 @@ fn panics_on_invalid_external_domain() {
     site.load().expect("link check test_site");
 }
 
+#[test]
+fn external_links_ignored_on_check() {
+    let (mut site, _tmp_dir, _public) = build_site("test_site");
+
+    // remove the invalid domain skip prefix
+    let i = site
+        .config
+        .link_checker
+        .skip_prefixes
+        .iter()
+        .position(|prefix| prefix == "http://invaliddomain")
+        .unwrap();
+    site.config.link_checker.skip_prefixes.remove(i);
+
+    // confirm the invalid domain skip prefix was removed
+    assert_eq!(site.config.link_checker.skip_prefixes, vec!["http://[2001:db8::]/"]);
+
+    // set a flag to skip external links check
+    site.skip_external_links_check();
+
+    // check the test site with all external links (including invalid domain) skipped, which should
+    // not cause a panic
+    site.config.enable_check_mode();
+    site.load().expect("link check test_site");
+}
+
 #[test]
 fn can_find_site_and_page_authors() {
     let mut path = env::current_dir().unwrap().parent().unwrap().parent().unwrap().to_path_buf();

docs/content/documentation/getting-started/cli-usage.md

@@ -114,6 +114,8 @@ The check subcommand will try to build all pages just like the build command wou
 results to disk. Additionally, it will also check all external links in Markdown files by trying to fetch
 them (links in the template files are not checked).
 
+You can skip link checking for all the external links by `--skip-external-links` flag.
+
 By default, drafts are not loaded. If you wish to include them, pass the `--drafts` flag.
 
 ## Colored output

src/cli.rs

@@ -100,6 +100,9 @@ pub enum Command {
         /// Include drafts when loading the site
         #[clap(long)]
         drafts: bool,
+        /// Skip external links
+        #[clap(long)]
+        skip_external_links: bool,
     },
 
     /// Generate shell completion

src/cmd/check.rs

@@ -11,6 +11,7 @@ pub fn check(
     base_path: Option<&str>,
     base_url: Option<&str>,
     include_drafts: bool,
+    skip_external_links: bool,
 ) -> Result<()> {
     let bp = base_path.map(PathBuf::from).unwrap_or_else(|| PathBuf::from(root_dir));
     let mut site = Site::new(bp, config_file)?;
@@ -22,6 +23,9 @@ pub fn check(
     if include_drafts {
         site.include_drafts();
     }
+    if skip_external_links {
+        site.skip_external_links_check();
+    }
     site.load()?;
     messages::check_site_summary(&site);
     messages::warn_about_ignored_pages(&site);

src/main.rs

@@ -121,11 +121,11 @@ fn main() {
                 std::process::exit(1);
             }
         }
-        Command::Check { drafts } => {
+        Command::Check { drafts, skip_external_links } => {
             console::info("Checking site...");
             let start = Instant::now();
             let (root_dir, config_file) = get_config_file_path(&cli_dir, &cli.config);
-            match cmd::check(&root_dir, &config_file, None, None, drafts) {
+            match cmd::check(&root_dir, &config_file, None, None, drafts, skip_external_links) {
                 Ok(()) => messages::report_elapsed_time(start),
                 Err(e) => {
                     messages::unravel_errors("Failed to check the site", &e);