From 8fac4f3e6420a8854ec22a006fddcc98af29ba83 Mon Sep 17 00:00:00 2001
From: Lars Wirzenius <liw@liw.fi>
Date: Mon, 25 Jul 2022 08:45:10 +0300
Subject: feat: exclude unwanted files from site

Sponsored-by: author
---
 riki.md     | 11 +++++++++++
 src/site.rs | 63 ++++++++++++++++++++++++++++++++++++++-----------------------
 2 files changed, 50 insertions(+), 24 deletions(-)

diff --git a/riki.md b/riki.md
index 00637dc..62093ff 100644
--- a/riki.md
+++ b/riki.md
@@ -468,4 +468,15 @@ the site content should be excluded._
 
 ~~~scenario
 given an installed riki
+given file site/index.mdwn from empty
+given file site/img.jpg from empty
+given file site/.git/HEAD from empty
+given file site/index.mdwn~ from empty
+given file site/#index.mdwn# from empty
+when I run riki list site
+then stdout contains "img.jpg"
+then stdout contains "index.mdwn"
+then stdout doesn't contain ".git"
+then stdout doesn't contain "index.mdwn~"
+then stdout doesn't contain "#index.mdwn#"
 ~~~
diff --git a/src/site.rs b/src/site.rs
index 1d1de60..48de58d 100644
--- a/src/site.rs
+++ b/src/site.rs
@@ -2,7 +2,7 @@ use crate::error::SiteError;
 use crate::page::{MarkdownPage, UnprocessedPage, WikitextPage};
 use crate::parser::WikitextParser;
 use crate::token::TokenPatterns;
-use crate::util::{make_path_absolute, make_path_relative_to, make_relative_link, join_subpath};
+use crate::util::{join_subpath, make_path_absolute, make_path_relative_to, make_relative_link};
 use log::{debug, info, trace};
 use std::collections::HashMap;
 use std::path::{Path, PathBuf};
@@ -21,6 +21,9 @@ pub struct Site {
 }
 
 impl Site {
+    const EXCLUDE_SUBSTRINGS: &'static [&'static str] = &[".git"];
+    const EXCLUDE_ENDS: &'static [&'static str] = &[".git", "~", "#"];
+
     pub fn new<P>(srcdir: P, destdir: P) -> Self
     where
         P: AsRef<Path>,
@@ -40,7 +43,8 @@ impl Site {
 
     pub fn scan(&mut self) -> Result<(), SiteError> {
         for filename in Self::all_files(&self.srcdir)? {
-            self.included_files.push(make_path_relative_to(&self.srcdir, &filename));
+            self.included_files
+                .push(make_path_relative_to(&self.srcdir, &filename));
             if Self::is_markdown(&filename) {
                 let page = WikitextPage::read(&self.srcdir, &filename)?;
                 self.add_wikitextpage(page);
@@ -121,14 +125,35 @@ impl Site {
 
     fn all_files(root: &Path) -> Result<Vec<PathBuf>, SiteError> {
         let mut files = vec![];
-        for e in WalkDir::new(root) {
+        for e in WalkDir::new(root)
+        {
             let e = e.map_err(|err| SiteError::WalkDir(root.to_path_buf(), err))?;
             let path = e.path();
-            files.push(path.to_path_buf());
+            if Self::is_excluded(path) {
+                debug!("exclude {}", path.display());
+            } else {
+                debug!("include {}", path.display());
+                files.push(path.to_path_buf());
+            }
         }
         Ok(files)
     }
 
+    fn is_excluded(path: &Path) -> bool {
+        let path = path.to_string_lossy();
+        for pat in Self::EXCLUDE_ENDS {
+            if path.ends_with(pat) {
+                return true;
+            }
+        }
+        for pat in Self::EXCLUDE_SUBSTRINGS {
+            if path.contains(pat) {
+                return true;
+            }
+        }
+        false
+    }
+
     fn is_markdown(path: &Path) -> bool {
         if let Some(ext) = path.extension() {
             ext == "mdwn"
@@ -178,7 +203,11 @@ impl Site {
         // Does a sibling page or file exist?
         if let Some(parent) = page.parent() {
             let path = parent.join(target);
-            trace!("checking for sibling in {}: {}", parent.display(), path.display());
+            trace!(
+                "checking for sibling in {}: {}",
+                parent.display(),
+                path.display()
+            );
             if let Some(path) = self.pages.get(path.as_path()) {
                 trace!("sibling page exists: {}", path.display());
                 return Ok(path.into());
@@ -222,9 +251,7 @@ impl PageSet {
     }
 
     fn normalize(path: &Path) -> String {
-        path.to_str()
-            .expect("path is UTF8")
-            .to_lowercase()
+        path.to_str().expect("path is UTF8").to_lowercase()
     }
 }
 
@@ -272,10 +299,7 @@ mod test {
         let mut site = Site::new(".", ".");
         site.add_wikitextpage(page("/foo/yo"));
         site.process().unwrap();
-        assert_eq!(
-            site.resolve("/foo/bar", "yo").unwrap(),
-            Path::new("yo")
-        );
+        assert_eq!(site.resolve("/foo/bar", "yo").unwrap(), Path::new("yo"));
     }
 
     #[test]
@@ -283,10 +307,7 @@ mod test {
         let mut site = Site::new(".", ".");
         site.add_wikitextpage(page("/foo/yo"));
         site.process().unwrap();
-        assert_eq!(
-            site.resolve("/foo/bar", "YO").unwrap(),
-            Path::new("yo")
-        );
+        assert_eq!(site.resolve("/foo/bar", "YO").unwrap(), Path::new("yo"));
     }
 
     #[test]
@@ -294,10 +315,7 @@ mod test {
         let mut site = Site::new(".", ".");
         site.add_wikitextpage(page("/foo/bar/yo"));
         site.process().unwrap();
-        assert_eq!(
-            site.resolve("/foo/bar", "yo").unwrap(),
-            Path::new("bar/yo")
-        );
+        assert_eq!(site.resolve("/foo/bar", "yo").unwrap(), Path::new("bar/yo"));
     }
 
     #[test]
@@ -306,10 +324,7 @@ mod test {
         site.add_wikitextpage(page("/foo/bar/yo"));
         site.add_wikitextpage(page("/foo/yo"));
         site.process().unwrap();
-        assert_eq!(
-            site.resolve("/foo/bar", "yo").unwrap(),
-            Path::new("bar/yo")
-        );
+        assert_eq!(site.resolve("/foo/bar", "yo").unwrap(), Path::new("bar/yo"));
     }
 
     #[test]
-- 
cgit v1.2.1