Make ignored_content work with nested paths and directories (#1738)

* Make ignored_content work with directories
Just had to remove strip_prefix(path) from file ignore code.
Added tests for subdirectory globbing.

* Add documentation

* add more tests
to confim that simple filename globs still match paths
without strip_prefix

components/config/src/config/mod.rs

@@ -561,21 +561,28 @@ ignored_content = []
         let config_str = r#"
 title = "My site"
 base_url = "example.com"
-ignored_content = ["*.{graphml,iso}", "*.py?"]
+ignored_content = ["*.{graphml,iso}", "*.py?", "**/{target,temp_folder}"]
         "#;
 
         let config = Config::parse(config_str).unwrap();
         let v = config.ignored_content;
-        assert_eq!(v, vec!["*.{graphml,iso}", "*.py?"]);
+        assert_eq!(v, vec!["*.{graphml,iso}", "*.py?", "**/{target,temp_folder}"]);
 
         let g = config.ignored_content_globset.unwrap();
-        assert_eq!(g.len(), 2);
+        assert_eq!(g.len(), 3);
         assert!(g.is_match("foo.graphml"));
+        assert!(g.is_match("foo/bar/foo.graphml"));
         assert!(g.is_match("foo.iso"));
         assert!(!g.is_match("foo.png"));
         assert!(g.is_match("foo.py2"));
         assert!(g.is_match("foo.py3"));
         assert!(!g.is_match("foo.py"));
+        assert!(g.is_match("foo/bar/target"));
+        assert!(g.is_match("foo/bar/baz/temp_folder"));
+        assert!(g.is_match("foo/bar/baz/temp_folder/target"));
+        assert!(g.is_match("temp_folder"));
+        assert!(g.is_match("my/isos/foo.iso"));
+        assert!(g.is_match("content/poetry/zen.py2"));
     }
 
     #[test]

components/library/src/content/mod.rs

@@ -56,10 +56,7 @@ pub fn find_related_assets(path: &Path, config: &Config, recursive: bool) -> Vec
     if let Some(ref globset) = config.ignored_content_globset {
         assets = assets
             .into_iter()
-            .filter(|p| match p.strip_prefix(path) {
-                Err(_) => false,
-                Ok(file) => !globset.is_match(file),
-            })
+            .filter(|p| !globset.is_match(p))
             .collect();
     }
 

components/library/src/content/section.rs

@@ -228,7 +228,7 @@ impl Section {
 
 #[cfg(test)]
 mod tests {
-    use std::fs::{create_dir, File};
+    use std::fs::{create_dir, File, create_dir_all};
     use std::io::Write;
     use std::path::{Path, PathBuf};
 
@@ -268,23 +268,26 @@ mod tests {
     fn section_with_ignored_assets_filters_out_correct_files() {
         let tmp_dir = tempdir().expect("create temp dir");
         let path = tmp_dir.path();
-        create_dir(&path.join("content")).expect("create content temp dir");
-        create_dir(&path.join("content").join("posts")).expect("create posts temp dir");
-        let nested_path = path.join("content").join("posts").join("with-assets");
-        create_dir(&nested_path).expect("create nested temp dir");
-        let mut f = File::create(nested_path.join("_index.md")).unwrap();
+        let article_path = path.join("content/posts/with-assets");
+        create_dir_all(path.join(&article_path).join("foo/bar/baz/quux")).expect("create nested temp dir");
+        create_dir_all(path.join(&article_path).join("foo/baz/quux")).expect("create nested temp dir");
+        let mut f = File::create(article_path.join("_index.md")).unwrap();
         f.write_all(b"+++\nslug=\"hey\"\n+++\n").unwrap();
-        File::create(nested_path.join("example.js")).unwrap();
-        File::create(nested_path.join("graph.jpg")).unwrap();
-        File::create(nested_path.join("fail.png")).unwrap();
+        File::create(article_path.join("example.js")).unwrap();
+        File::create(article_path.join("graph.jpg")).unwrap();
+        File::create(article_path.join("fail.png")).unwrap();
+        File::create(article_path.join("foo/bar/baz/quux/quo.xlsx")).unwrap();
+        File::create(article_path.join("foo/bar/baz/quux/quo.docx")).unwrap();
+        
 
         let mut gsb = GlobSetBuilder::new();
         gsb.add(Glob::new("*.{js,png}").unwrap());
+        gsb.add(Glob::new("foo/**/baz").unwrap());
         let mut config = Config::default();
         config.ignored_content_globset = Some(gsb.build().unwrap());
 
         let res =
-            Section::from_file(nested_path.join("_index.md").as_path(), &config, &PathBuf::new());
+            Section::from_file(article_path.join("_index.md").as_path(), &config, &PathBuf::new());
 
         assert!(res.is_ok());
         let page = res.unwrap();

docs/content/documentation/content/overview.md

@@ -70,13 +70,13 @@ By default, this page's slug will be the directory name and thus its permalink w
 
 It is possible to ignore selected asset files using the
 [ignored_content](@/documentation/getting-started/configuration.md) setting in the config file.
-For example, say that you have an Excel spreadsheet from which you are taking several screenshots and
-then linking to these image files on your website. For maintainability, you want to keep
-the spreadsheet in the same directory as the Markdown file, but you don't want to copy the spreadsheet to
-the public web site. You can achieve this by setting `ignored_content` in the config file:
+For example, say that you have several code files which you are linking to on your website.
+For maintainability, you want to keep your code in the same directory as the Markdown file,
+but you don't want to copy the build folders to the public web site. You can achieve this by setting `ignored_content` in the config file:
 
+(Note of caution: `{Cargo.lock,target}` is _not_ the same as `{Cargo.lock, target}`)
 ```
-ignored_content = ["*.xlsx"]
+ignored_content = ["code_articles/**/{Cargo.lock,target}, *.rs"]
 ```
 
 ## Static assets

docs/content/documentation/getting-started/configuration.md

@@ -52,7 +52,7 @@ minify_html = false
 # directory is processed. Defaults to none, which means that all asset files are
 # copied over to the `public` directory.
 # Example:
-#     ignored_content = ["*.{graphml,xlsx}", "temp.*"]
+#     ignored_content = ["*.{graphml,xlsx}", "temp.*", "**/build_folder"]
 ignored_content = []
 
 # When set to "true", a feed is automatically generated.