refactor: name the modules after what they hold

2026-09-08 12:10:41 +00:00
parent a3c08f6eaa
commit bb0a2cca44
12 changed files with 1014 additions and 989 deletions
--- a/src/commands/link/check.rs
+++ b/src/commands/link/check.rs
@@ -0,0 +1,243 @@
+use fs_err::{read_dir, read_link};
+use std::path::{Path, PathBuf};
+use std::process::ExitCode;
+
+use anyhow::{Result, anyhow};
+
+use super::store::{Repo, resolve, symlink_metadata_opt};
+use crate::cmd::{Cmd, LsFiles};
+use crate::ctx::Ctx;
+
+pub fn check(
+    ctx: &Ctx,
+    paths: &[PathBuf],
+    porcelain: bool,
+    null: bool,
+    exit_code: bool,
+    store: Option<&Path>,
+) -> Result<ExitCode> {
+    let repo = Repo::discover(ctx, store)?;
+    let pathspecs = relative_pathspecs(&repo, paths)?;
+
+    let mut exposed = Vec::new();
+    // git lists untracked and ignored separately
+    for (mark, ignored) in [(UNTRACKED, false), (IGNORED, true)] {
+        for entry in list_others(ctx, &repo, ignored, &pathspecs)? {
+            let rel = PathBuf::from(entry.trim_end_matches('/'));
+
+            // --directory collapses a wholly untracked dir into `dir/`
+            if entry.ends_with('/') {
+                exposed.extend(walk(&repo, &rel, mark)?.1);
+            } else {
+                exposed.extend(classify(&repo, &rel, mark)?);
+            }
+        }
+    }
+
+    exposed.sort_by(|a, b| a.name.cmp(&b.name));
+
+    if porcelain || null {
+        print_porcelain(&exposed, null);
+    } else {
+        print_listing(&repo, &exposed);
+    }
+
+    // git's --exit-code convention: nothing to report is 0, anything is 1
+    if exit_code && !exposed.is_empty() {
+        return Ok(ExitCode::FAILURE);
+    }
+
+    Ok(ExitCode::SUCCESS)
+}
+
+fn print_porcelain(exposed: &[Exposed], null: bool) {
+    let end = if null { '\0' } else { '\n' };
+
+    for item in exposed {
+        match &item.dest {
+            Some(dest) => print!("{} {} -> {}{end}", item.code(), item.name, dest.display()),
+            None => print!("{} {}{end}", item.code(), item.name),
+        }
+    }
+}
+
+fn print_listing(repo: &Repo, exposed: &[Exposed]) {
+    println!("store: {}", repo.store.display());
+
+    if exposed.is_empty() {
+        println!("nothing outside the store, a sandbox would see tracked files only");
+        return;
+    }
+
+    let sections = [
+        (
+            "Untracked paths a sandbox can read:",
+            "  (use \"ahab link add <path>...\" to move them into the store)",
+            Section::Content(UNTRACKED),
+        ),
+        (
+            "Ignored paths a sandbox can read:",
+            "  (use \"ahab link add <path>...\" to move them into the store)",
+            Section::Content(IGNORED),
+        ),
+        (
+            "Symlinks leading outside the store:",
+            "  (their contents are not in the repository either way)",
+            Section::Elsewhere,
+        ),
+    ];
+
+    for (heading, hint, section) in sections {
+        let mut items = exposed.iter().filter(|item| section.holds(item)).peekable();
+        if items.peek().is_none() {
+            continue;
+        }
+
+        println!("\n{heading}\n{hint}");
+        for item in items {
+            match &item.dest {
+                Some(dest) => println!("\t{} -> {}", item.name, dest.display()),
+                None => println!("\t{}", item.name),
+            }
+        }
+    }
+}
+
+enum Section {
+    Content(char),
+    Elsewhere,
+}
+
+impl Section {
+    fn holds(&self, item: &Exposed) -> bool {
+        match self {
+            Self::Content(mark) => item.dest.is_none() && item.mark == *mark,
+            Self::Elsewhere => item.dest.is_some(),
+        }
+    }
+}
+
+// status codes as `git status --porcelain` spells them
+const UNTRACKED: char = '?';
+const IGNORED: char = '!';
+const ELSEWHERE: char = '>';
+
+struct Exposed {
+    mark: char,
+    name: String,
+    dest: Option<PathBuf>,
+}
+
+impl Exposed {
+    fn content(mark: char, name: String) -> Self {
+        Self {
+            mark,
+            name,
+            dest: None,
+        }
+    }
+
+    fn code(&self) -> String {
+        let second = if self.dest.is_some() {
+            ELSEWHERE
+        } else {
+            self.mark
+        };
+        format!("{}{second}", self.mark)
+    }
+}
+
+fn classify(repo: &Repo, rel: &Path, mark: char) -> Result<Option<Exposed>> {
+    let src = repo.root.join(rel);
+    let name = rel.display().to_string();
+
+    let Some(meta) = symlink_metadata_opt(&src)? else {
+        return Ok(None);
+    };
+    if !meta.is_symlink() {
+        return Ok(Some(Exposed::content(mark, name)));
+    }
+
+    let dest = read_link(&src)?;
+    if dest == repo.store.join(rel) {
+        return Ok(None);
+    }
+
+    Ok(Some(Exposed {
+        mark,
+        name,
+        dest: Some(dest),
+    }))
+}
+
+fn walk(repo: &Repo, rel: &Path, mark: char) -> Result<(usize, Vec<Exposed>)> {
+    let dir = repo.root.join(rel);
+    let mut handled = 0;
+    let mut exposed = Vec::new();
+
+    for entry in read_dir(&dir)? {
+        let entry = entry?;
+        let child = rel.join(entry.file_name());
+
+        if entry.file_type()?.is_dir() {
+            let (below, inside) = walk(repo, &child, mark)?;
+            handled += below;
+            exposed.extend(inside);
+            continue;
+        }
+
+        match classify(repo, &child, mark)? {
+            Some(item) => exposed.push(item),
+            None => handled += 1,
+        }
+    }
+
+    // nothing below is in the store, so collapse to one line
+    if handled == 0 && !exposed.is_empty() {
+        let name = format!("{}/", rel.display());
+        return Ok((0, vec![Exposed::content(mark, name)]));
+    }
+
+    Ok((handled, exposed))
+}
+
+fn relative_pathspecs(repo: &Repo, paths: &[PathBuf]) -> Result<Vec<PathBuf>> {
+    let mut specs = Vec::with_capacity(paths.len());
+
+    for path in paths {
+        let abs = resolve(path)?;
+        let rel = abs.strip_prefix(&repo.root).map_err(|_| {
+            anyhow!(
+                "{} is outside the repository {}",
+                abs.display(),
+                repo.root.display()
+            )
+        })?;
+
+        if rel.as_os_str().is_empty() {
+            return Ok(Vec::new());
+        }
+        specs.push(rel.to_path_buf());
+    }
+
+    Ok(specs)
+}
+
+fn list_others(
+    ctx: &Ctx,
+    repo: &Repo,
+    ignored: bool,
+    pathspecs: &[PathBuf],
+) -> Result<Vec<String>> {
+    let mut listing = LsFiles::untracked(&repo.root).limited_to(pathspecs);
+    if ignored {
+        listing = listing.ignored();
+    }
+
+    Ok(listing
+        .capture(ctx)?
+        .split('\0')
+        .filter(|p| !p.is_empty())
+        .map(String::from)
+        .collect())
+}
--- a/src/commands/link/store.rs
+++ b/src/commands/link/store.rs
@@ -0,0 +1,254 @@
+use fs_err as fs;
+use std::env;
+use std::ffi::OsString;
+use std::path::{Path, PathBuf};
+
+use anyhow::{Context, Result, anyhow};
+
+use crate::cmd::{CheckIgnore, Cmd, ConfigGet, LsFiles, RevParse};
+use crate::ctx::Ctx;
+use crate::output::note;
+
+// a checkout with no remote to name it after
+const LOCAL_NAMESPACE: &str = "_local";
+
+pub(super) struct Repo {
+    pub(super) root: PathBuf,
+    pub(super) store: PathBuf,
+    // the configured store root, above the per-repository directories
+    pub(super) base: PathBuf,
+}
+
+impl Repo {
+    pub(super) fn discover(ctx: &Ctx, store: Option<&Path>) -> Result<Self> {
+        let root = git_root(ctx)?;
+        let base = match store {
+            Some(store) => store.to_path_buf(),
+            None => store_root()?,
+        };
+
+        Ok(Self {
+            store: base.join(repo_components(ctx, &root)?),
+            root,
+            base,
+        })
+    }
+    pub(super) fn relative(&self, src: &Path) -> Result<PathBuf> {
+        let rel = src.strip_prefix(&self.root).map_err(|_| {
+            anyhow!(
+                "{} is outside the repository {}",
+                src.display(),
+                self.root.display()
+            )
+        })?;
+
+        // empty means the whole repo
+        if rel.as_os_str().is_empty() {
+            return Err(anyhow!(
+                "refusing to externalize the repository root itself"
+            ));
+        }
+        if rel.starts_with(".git") {
+            return Err(anyhow!("refusing to externalize anything under .git"));
+        }
+
+        Ok(rel.to_path_buf())
+    }
+}
+
+pub(super) fn resolve(path: &Path) -> Result<PathBuf> {
+    let abs = std::path::absolute(path)
+        .with_context(|| format!("resolving absolute path of {}", path.display()))?;
+
+    // the parent must exist so symlinked components resolve like git's toplevel
+    let Some(name) = abs.file_name().map(OsString::from) else {
+        return Ok(fs::canonicalize(&abs)?);
+    };
+    let parent = fs::canonicalize(abs.parent().unwrap_or(Path::new("/")))?;
+    Ok(parent.join(name))
+}
+
+fn git_root(ctx: &Ctx) -> Result<PathBuf> {
+    let root = RevParse
+        .capture(ctx)
+        .map_err(|_| anyhow!("not inside a git repository"))?;
+
+    let root = root.trim().to_string();
+    if root.is_empty() {
+        return Err(anyhow!("git reported an empty repository root"));
+    }
+
+    Ok(fs::canonicalize(&root)?)
+}
+
+fn repo_components(ctx: &Ctx, root: &Path) -> Result<PathBuf> {
+    if let Some(url) = git_origin_url(ctx) {
+        if let Some(components) = components_from_remote(&url) {
+            return Ok(components);
+        }
+        note!("could not parse git remote `{url}`, falling back to the checkout name");
+    }
+
+    let name = root
+        .file_name()
+        .ok_or_else(|| anyhow!("cannot derive a store path for {}", root.display()))?;
+    Ok(Path::new(LOCAL_NAMESPACE).join(sanitize(&name.to_string_lossy())))
+}
+
+fn git_origin_url(ctx: &Ctx) -> Option<String> {
+    let url = ConfigGet {
+        key: "remote.origin.url",
+    }
+    .capture(ctx)
+    .ok()?;
+
+    let url = url.trim().to_string();
+    (!url.is_empty()).then_some(url)
+}
+
+fn components_from_remote(url: &str) -> Option<PathBuf> {
+    let url = url.trim();
+    let url = url.strip_suffix(".git").unwrap_or(url);
+
+    // `scheme://[user@]host[:port]/path`, or scp-like `[user@]host:path`
+    let (authority, path) = match url.split_once("://") {
+        Some((_, after)) => after.split_once('/')?,
+        None => url.split_once(':')?,
+    };
+
+    let host = authority.rsplit_once('@').map_or(authority, |(_, h)| h);
+    let host = host.split_once(':').map_or(host, |(h, _)| h);
+    // a local remote has no host to key on
+    if host.is_empty() {
+        return None;
+    }
+
+    let mut components = PathBuf::from(sanitize(&host.to_lowercase()));
+    let mut depth = 0;
+    for part in path.split('/').filter(|p| !p.is_empty()) {
+        components.push(sanitize(part));
+        depth += 1;
+    }
+    (depth > 0).then_some(components)
+}
+
+fn sanitize(s: &str) -> String {
+    let out: String = s
+        .chars()
+        .map(|c| {
+            if c.is_ascii_alphanumeric() || matches!(c, '-' | '.' | '_') {
+                c
+            } else {
+                '_'
+            }
+        })
+        .collect();
+
+    // `.` and `..` are legal characters but not legal components
+    if out.chars().all(|c| c == '.') {
+        return "_".repeat(out.len());
+    }
+    out
+}
+
+fn store_root() -> Result<PathBuf> {
+    if let Some(xdg) = non_empty_var("XDG_DATA_HOME") {
+        return Ok(PathBuf::from(xdg).join("ahab"));
+    }
+
+    let home = non_empty_var("HOME")
+        .filter(|v| !v.is_empty())
+        .ok_or_else(|| anyhow!("neither XDG_DATA_HOME nor HOME is set"))?;
+
+    Ok(PathBuf::from(home).join(".local/share/ahab"))
+}
+fn non_empty_var(name: &str) -> Option<OsString> {
+    env::var_os(name).filter(|v| !v.is_empty())
+}
+
+pub(super) fn tracked(ctx: &Ctx, repo: &Repo, rel: &Path) -> Result<bool> {
+    let listed = LsFiles::tracked(&repo.root)
+        .limited_to(&[rel])
+        .capture(ctx)?;
+
+    Ok(!listed.is_empty())
+}
+
+pub(super) fn ignored(ctx: &Ctx, repo: &Repo, rel: &Path) -> bool {
+    CheckIgnore::new(&repo.root, rel)
+        .quietly_succeeds(ctx)
+        .unwrap_or(false)
+}
+
+pub(super) fn symlink_metadata_opt(path: &Path) -> Result<Option<std::fs::Metadata>> {
+    // symlink_metadata does not follow the link, so a symlink shows as one
+    match fs::symlink_metadata(path) {
+        Ok(meta) => Ok(Some(meta)),
+        Err(e) if e.kind() == std::io::ErrorKind::NotFound => Ok(None),
+        Err(e) => Err(e.into()),
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::{components_from_remote, sanitize};
+    use std::path::PathBuf;
+
+    #[test]
+    fn parses_every_spelling_of_a_remote() {
+        let cases = [
+            (
+                "git@git.aflabs.org:urnik/afurnik.git",
+                "git.aflabs.org/urnik/afurnik",
+            ),
+            (
+                "https://git.aflabs.org/urnik/afurnik.git",
+                "git.aflabs.org/urnik/afurnik",
+            ),
+            (
+                "https://git.aflabs.org/urnik/afurnik",
+                "git.aflabs.org/urnik/afurnik",
+            ),
+            (
+                "https://git.aflabs.org/urnik/afurnik/",
+                "git.aflabs.org/urnik/afurnik",
+            ),
+            (
+                "ssh://git@git.aflabs.org:22/urnik/afurnik.git",
+                "git.aflabs.org/urnik/afurnik",
+            ),
+            (
+                "git@GIT.Aflabs.org:urnik/AFurnik.git",
+                "git.aflabs.org/urnik/AFurnik",
+            ),
+            (
+                "git@git.aflabs.org:urnik/internal/afurnik.git",
+                "git.aflabs.org/urnik/internal/afurnik",
+            ),
+        ];
+
+        for (url, want) in cases {
+            assert_eq!(
+                components_from_remote(url),
+                Some(PathBuf::from(want)),
+                "url: {url}"
+            );
+        }
+    }
+
+    #[test]
+    fn rejects_remotes_without_a_host() {
+        assert_eq!(components_from_remote("not-a-url"), None);
+        assert_eq!(components_from_remote("/srv/git/afurnik.git"), None);
+        assert_eq!(components_from_remote("file:///srv/git/afurnik.git"), None);
+        assert_eq!(components_from_remote("https://git.aflabs.org/"), None);
+    }
+
+    #[test]
+    fn sanitize_never_yields_a_traversal() {
+        assert_eq!(sanitize(".."), "__");
+        assert_eq!(sanitize("."), "_");
+        assert_eq!(sanitize("a/b"), "a_b");
+        assert_eq!(sanitize(".env"), ".env");
+    }
+}