crates/oxy-core/src/repo/prompt.rs at 4f264d6f20b6 · andymccall/oxygen

# 4f264d6f20b6 oxygen / crates / oxy-core / src / repo / prompt.rs
41.6 KiB · Rust · last changed A verified clean check restamps the manifest; checkout keeps directory times (refs #41) 1 month ago source blame history raw
//! Fast, side-effect-free status for shell prompts.
//!
//! Every snapshot and checkout writes a manifest of the working copy as it
//! was last seen: `size`, `mtime`, and BLAKE3 of each file. `quick_status`
//! walks the working copy and compares against that manifest, reading a
//! file's content only when its size or mtime changed, and records no
//! operation. Ahead/behind counts come from walking the store between the
//! local bookmark and its `origin` tracking state.

use super::*;
use std::time::UNIX_EPOCH;

#[derive(Debug, Clone, Default, PartialEq, Eq)]
pub struct PromptInfo {
    pub bookmark: String,
    /// Working-copy ChangeId prefix.
    pub change_id: String,
    pub heads: usize,
    pub dirty: bool,
    pub conflicted: bool,
    /// Changes local `main` has that `origin/main` lacks, when tracking exists.
    pub ahead: Option<usize>,
    pub behind: Option<usize>,
}

impl PromptInfo {
    /// The glyph for the working copy's state: `‼` conflicted, `✎` dirty,
    /// `✔` clean.
    pub fn state_glyph(&self) -> &'static str {
        if self.conflicted {
            "‼"
        } else if self.dirty {
            "✎"
        } else {
            "✔"
        }
    }

    /// The ANSI colour (SGR parameter) for the state: red, yellow, or cyan.
    pub fn state_color(&self) -> u8 {
        if self.conflicted {
            31
        } else if self.dirty {
            33
        } else {
            36
        }
    }

    /// `format` wrapped in the state colour. `guard` is `("\\[", "\\]")` for
    /// bash prompts, `("%{", "%}")` for zsh, or empty for raw output.
    pub fn format_colored(&self, fmt: &str, guard: (&str, &str)) -> String {
        format!(
            "{}\x1b[{}m{}{}{}\x1b[0m{}",
            guard.0,
            self.state_color(),
            guard.1,
            self.format(fmt),
            guard.0,
            guard.1
        )
    }

    /// Expand `%b` bookmark, `%c` change id, `%h` heads, `%d` dirty marker
    /// (`*`), `%x` conflict marker (`!`), `%a`/`%B` ahead/behind counts
    /// (empty when unknown), `%A`/`%D` `+n`/`-n` when non-zero, `%s` the
    /// state glyph, `%U`/`%O` `↑n`/`↓n` when non-zero, `%i` the repository
    /// icon, `%%` percent.
    pub fn format(&self, fmt: &str) -> String {
        let mut out = String::new();
        let mut chars = fmt.chars();
        while let Some(c) = chars.next() {
            if c != '%' {
                out.push(c);
                continue;
            }
            match chars.next() {
                Some('b') => out.push_str(&self.bookmark),
                Some('c') => out.push_str(&self.change_id),
                Some('h') => out.push_str(&self.heads.to_string()),
                Some('d') => {
                    if self.dirty {
                        out.push('*');
                    }
                }
                Some('x') => {
                    if self.conflicted {
                        out.push('!');
                    }
                }
                Some('a') => out.push_str(&self.ahead.map(|n| n.to_string()).unwrap_or_default()),
                Some('B') => out.push_str(&self.behind.map(|n| n.to_string()).unwrap_or_default()),
                Some('A') => {
                    if let Some(n) = self.ahead.filter(|n| *n > 0) {
                        out.push_str(&format!("+{n}"));
                    }
                }
                Some('D') => {
                    if let Some(n) = self.behind.filter(|n| *n > 0) {
                        out.push_str(&format!("-{n}"));
                    }
                }
                Some('s') => out.push_str(self.state_glyph()),
                Some('U') => {
                    if let Some(n) = self.ahead.filter(|n| *n > 0) {
                        out.push_str(&format!("↑{n}"));
                    }
                }
                Some('O') => {
                    if let Some(n) = self.behind.filter(|n| *n > 0) {
                        out.push_str(&format!("↓{n}"));
                    }
                }
                Some('i') => out.push_str(ICON),
                Some('%') => out.push('%'),
                Some(other) => {
                    out.push('%');
                    out.push(other);
                }
                None => out.push('%'),
            }
        }
        out
    }
}

/// Path → (size, mtime in nanoseconds, BLAKE3 hex) as last seen.
/// One manifest entry: the file's stat, its content hash, and the stored
/// blob for an opaque file.
#[derive(Clone, Debug)]
pub(crate) struct ManifestEntry {
    pub size: u64,
    pub mtime_ns: u128,
    pub hash: [u8; 32],
    pub blob: Option<ObjectRef>,
    /// The graph key (`State::graph_key`) of the base file this text file
    /// was found equal to, rendered, byte for byte.
    pub graph: Option<[u8; 32]>,
    /// Whether the file was executable: a mode change alters neither size
    /// nor time.
    pub exec: bool,
}

#[derive(Clone, Debug, Default)]
pub(crate) struct Manifest {
    /// When the manifest was written, for the racy-stat guard.
    pub written_ns: u128,
    pub entries: BTreeMap<String, ManifestEntry>,
    /// Every directory the walk listed, with its time then; empty when the
    /// manifest was not written from a walk (a checkout), and then the
    /// check lists the tree.
    pub dirs: Vec<(String, u128)>,
    /// Directories holding a repository of their own, as the walk found.
    pub nested: Vec<String>,
}

/// A parsed manifest and the identity (size, time, inode) of the file it
/// came from: reused while the file is the same one.
pub(crate) type CachedManifest = ((u64, u128, u64), (bool, bool, std::sync::Arc<Manifest>));

/// Now, in nanoseconds since the epoch.
fn now_ns() -> u128 {
    std::time::SystemTime::now()
        .duration_since(UNIX_EPOCH)
        .map(|d| d.as_nanos())
        .unwrap_or(0)
}

pub(crate) fn mtime_ns(m: &fs::Metadata) -> u128 {
    m.modified()
        .ok()
        .and_then(|t| t.duration_since(UNIX_EPOCH).ok())
        .map(|d| d.as_nanos())
        .unwrap_or(0)
}

pub const DEFAULT_PROMPT_FORMAT: &str = "%b %c%d%x%A%D";
/// The repository icon for `%i` (an atom, for oxygen).
pub const ICON: &str = "⚛";
/// A format with glyphs: icon, bookmark, change id, state, ahead/behind.
pub const GLYPH_PROMPT_FORMAT: &str = "%i %b %c %s%U%O";

/// The manifest file (oxygen#41): a flat binary layout read in one pass.
/// A magic number; the conflicted and uncommitted flags; when it was
/// written, at a fixed offset so an otherwise equal manifest can be
/// recognised; BLAKE3 of everything else; then each entry in path order:
/// path, size, mtime in nanoseconds, BLAKE3 of the content, the stored
/// blob of an opaque file, the graph key it was found equal to, and
/// whether it is executable. Format 3 puts before the entries every
/// directory the walk listed with its time, and the nested repositories. A file that does not check, or in any other
/// layout (an older manifest), reads as no manifest, and the next snapshot
/// writes it again: it is a cache, written without waiting for the disk.
const MANIFEST_MAGIC: &[u8; 8] = b"oxyman3\0";
/// Where the written time sits: after the magic and the two flags.
const MANIFEST_WRITTEN: std::ops::Range<usize> = 10..26;
/// Where the checksum sits, after the written time.
const MANIFEST_SUM: std::ops::Range<usize> = 26..58;

fn manifest_sum(buf: &[u8]) -> [u8; 32] {
    let mut h = blake3::Hasher::new();
    h.update(&buf[..MANIFEST_SUM.start]);
    h.update_rayon(&buf[MANIFEST_SUM.end..]);
    *h.finalize().as_bytes()
}

fn encode_manifest(
    conflicted: bool,
    uncommitted: bool,
    written_ns: u128,
    dirs: &[(String, u128)],
    nested: &[String],
    entries: &mut dyn Iterator<Item = (&str, ManifestEntry)>,
) -> Vec<u8> {
    let mut w = crate::flat::Writer::default();
    w.fixed(MANIFEST_MAGIC);
    w.u8(conflicted as u8);
    w.u8(uncommitted as u8);
    w.fixed(&written_ns.to_le_bytes());
    w.fixed(&[0; 32]);
    w.count(dirs.len());
    for (d, t) in dirs {
        w.bytes(d.as_bytes());
        w.fixed(&t.to_le_bytes());
    }
    w.count(nested.len());
    for n in nested {
        w.bytes(n.as_bytes());
    }
    for (path, e) in entries {
        w.bytes(path.as_bytes());
        w.u64(e.size);
        w.fixed(&e.mtime_ns.to_le_bytes());
        w.fixed(&e.hash);
        w.opt_object(&e.blob);
        match e.graph {
            None => w.u8(0),
            Some(k) => {
                w.u8(1);
                w.fixed(&k);
            }
        }
        w.u8(e.exec as u8);
    }
    let sum = manifest_sum(&w.0);
    w.0[MANIFEST_SUM].copy_from_slice(&sum);
    w.0
}

fn decode_manifest(buf: &[u8]) -> Option<(bool, bool, Manifest)> {
    let mut r = crate::flat::Reader::new(buf);
    if &r.fixed::<8>().ok()? != MANIFEST_MAGIC {
        return None;
    }
    let conflicted = r.u8().ok()? == 1;
    let uncommitted = r.u8().ok()? == 1;
    let written_ns = u128::from_le_bytes(r.fixed().ok()?);
    if r.fixed::<32>().ok()? != manifest_sum(buf) {
        return None;
    }
    let mut dirs = Vec::new();
    for _ in 0..r.count().ok()? {
        let d = std::str::from_utf8(r.bytes().ok()?).ok()?.to_string();
        dirs.push((d, u128::from_le_bytes(r.fixed().ok()?)));
    }
    let mut nested = Vec::new();
    for _ in 0..r.count().ok()? {
        nested.push(std::str::from_utf8(r.bytes().ok()?).ok()?.to_string());
    }
    let mut entries = Vec::new();
    while !r.at_end() {
        let path = std::str::from_utf8(r.bytes().ok()?).ok()?.to_string();
        let e = ManifestEntry {
            size: r.u64().ok()?,
            mtime_ns: u128::from_le_bytes(r.fixed().ok()?),
            hash: r.fixed().ok()?,
            blob: r.opt_object().ok()?,
            graph: match r.u8().ok()? {
                0 => None,
                _ => Some(r.fixed().ok()?),
            },
            exec: r.u8().ok()? == 1,
        };
        entries.push((path, e));
    }
    // Written in path order, so this builds the map without searching.
    let entries = entries.into_iter().collect();
    Some((
        conflicted,
        uncommitted,
        Manifest {
            written_ns,
            entries,
            dirs,
            nested,
        },
    ))
}

/// A tracked path as it is now: a file's stat, or a symlink's target.
enum Tracked {
    File(super::walk::Stat),
    Link(Vec<u8>),
}

/// The stat of every path in `rels` (relative to `root`; `None` for one
/// gone or neither a file nor a symlink), in the same order. Paths are
/// grouped by directory and each directory is opened once: a stat relative
/// to it resolves one name, where a stat by full path walks every
/// component from the root again (most of a clean status's time, in the
/// kernel's path lookup, on the Linux tree).
fn stat_tracked(root: &Path, rels: &[&String]) -> Vec<Option<Tracked>> {
    use rayon::prelude::*;
    let mut by_dir: HashMap<&str, Vec<(usize, &str)>> = HashMap::new();
    for (i, rel) in rels.iter().enumerate() {
        let (dir, name) = rel.rsplit_once('/').unwrap_or(("", rel.as_str()));
        by_dir.entry(dir).or_default().push((i, name));
    }
    let groups: Vec<(&str, Vec<(usize, &str)>)> = by_dir.into_iter().collect();
    let found: Vec<(usize, Option<Tracked>)> = groups
        .par_iter()
        .flat_map_iter(|(dir, names)| {
            let handle = fs::File::open(root.join(dir)).ok();
            names
                .iter()
                .map(|(i, name)| {
                    let rel = rels[*i];
                    let now = match &handle {
                        Some(d) => stat_in(d, name, || root.join(rel)),
                        None => None,
                    };
                    (*i, now)
                })
                .collect::<Vec<_>>()
        })
        .collect();
    let mut out: Vec<Option<Tracked>> = (0..rels.len()).map(|_| None).collect();
    for (i, now) in found {
        out[i] = now;
    }
    out
}

/// One entry of the open directory `dir`; `path` builds its full path,
/// needed only for a symlink's target.
#[cfg(unix)]
fn stat_in(dir: &fs::File, name: &str, path: impl Fn() -> PathBuf) -> Option<Tracked> {
    use std::os::fd::AsRawFd;
    let c = std::ffi::CString::new(name).ok()?;
    let mut st: libc::stat = unsafe { std::mem::zeroed() };
    // SAFETY: a valid directory descriptor, a NUL-terminated name, and a
    // stat buffer of the right type that outlives the call.
    let r = unsafe {
        libc::fstatat(
            dir.as_raw_fd(),
            c.as_ptr(),
            &mut st,
            libc::AT_SYMLINK_NOFOLLOW,
        )
    };
    if r != 0 {
        return None;
    }
    match st.st_mode & libc::S_IFMT {
        libc::S_IFREG => Some(Tracked::File(super::walk::Stat {
            size: st.st_size as u64,
            // As `mtime_ns` reads it: nanoseconds since the epoch, 0 before it.
            mtime_ns: if st.st_mtime < 0 {
                0
            } else {
                st.st_mtime as u128 * 1_000_000_000 + st.st_mtime_nsec as u128
            },
            exec: st.st_mode & 0o111 != 0,
        })),
        libc::S_IFLNK => {
            let t = fs::read_link(path()).ok()?;
            Some(Tracked::Link(t.to_string_lossy().as_bytes().to_vec()))
        }
        _ => None,
    }
}

#[cfg(not(unix))]
fn stat_in(_dir: &fs::File, _name: &str, path: impl Fn() -> PathBuf) -> Option<Tracked> {
    let p = path();
    let m = fs::symlink_metadata(&p).ok()?;
    if m.file_type().is_symlink() {
        let t = fs::read_link(&p).ok()?;
        Some(Tracked::Link(t.to_string_lossy().as_bytes().to_vec()))
    } else if m.is_file() {
        Some(Tracked::File(super::walk::Stat::of(&m)))
    } else {
        None
    }
}

/// The status cache: a magic number, the working-copy change, the
/// manifest checksum, then each status line (a status byte and a path).
const STATUS_MAGIC: &[u8; 8] = b"oxystat1";

fn status_byte(s: ChangeStatus) -> u8 {
    match s {
        ChangeStatus::Added => 0,
        ChangeStatus::Modified => 1,
        ChangeStatus::Removed => 2,
        ChangeStatus::Renamed => 3,
        ChangeStatus::ModeChanged => 4,
        ChangeStatus::ConflictEdited => 5,
        ChangeStatus::Resolved => 6,
    }
}

fn encode_status_cache(wc: &ChangeHash, stamp: &[u8; 32], statuses: &[PathStatus]) -> Vec<u8> {
    let mut out = Vec::with_capacity(80 + statuses.len() * 40);
    out.extend_from_slice(STATUS_MAGIC);
    out.extend_from_slice(&wc.0);
    out.extend_from_slice(stamp);
    out.extend_from_slice(&(statuses.len() as u32).to_le_bytes());
    for s in statuses {
        out.push(status_byte(s.status));
        out.extend_from_slice(&(s.path.len() as u32).to_le_bytes());
        out.extend_from_slice(s.path.as_bytes());
    }
    out
}

fn decode_status_cache(b: &[u8]) -> Option<(ChangeHash, [u8; 32], Vec<PathStatus>)> {
    let rest = b.strip_prefix(STATUS_MAGIC)?;
    let wc = ObjectRef(rest.get(..32)?.try_into().ok()?);
    let stamp: [u8; 32] = rest.get(32..64)?.try_into().ok()?;
    let n = u32::from_le_bytes(rest.get(64..68)?.try_into().ok()?) as usize;
    let mut at = 68;
    let mut out = Vec::with_capacity(n.min(1 << 16));
    for _ in 0..n {
        let status = match *rest.get(at)? {
            0 => ChangeStatus::Added,
            1 => ChangeStatus::Modified,
            2 => ChangeStatus::Removed,
            3 => ChangeStatus::Renamed,
            4 => ChangeStatus::ModeChanged,
            5 => ChangeStatus::ConflictEdited,
            6 => ChangeStatus::Resolved,
            _ => return None,
        };
        let len = u32::from_le_bytes(rest.get(at + 1..at + 5)?.try_into().ok()?) as usize;
        let path = std::str::from_utf8(rest.get(at + 5..at + 5 + len)?)
            .ok()?
            .to_string();
        out.push(PathStatus { path, status });
        at += 5 + len;
    }
    (at == rest.len()).then_some((wc, stamp, out))
}

impl Repo {
    pub(crate) fn manifest_path(&self) -> PathBuf {
        self.oxy
            .join("workspaces")
            .join(self.workspace.to_hex())
            .join("manifest")
    }

    /// Record the working copy as it is now (after a snapshot, commit, or
    /// checkout), with header lines saying whether the base state is
    /// conflicted and whether the working-copy change holds uncommitted work.
    pub(crate) fn write_manifest(&self, conflicted: bool, uncommitted: bool) -> Result<()> {
        let entries = self.walk_working_copy()?;
        self.write_manifest_entries(conflicted, uncommitted, &entries, &HashMap::new())
    }

    /// Write the manifest from entries already walked; `blobs` names the
    /// stored blob of each opaque path so the next walk can skip reading it.
    pub(crate) fn write_manifest_entries(
        &self,
        conflicted: bool,
        uncommitted: bool,
        entries: &BTreeMap<String, FsEntry>,
        blobs: &HashMap<String, ObjectRef>,
    ) -> Result<()> {
        self.manifest_cache.take();
        self.listing_cache.take();
        let written = std::time::SystemTime::now()
            .duration_since(UNIX_EPOCH)
            .map(|d| d.as_nanos())
            .unwrap_or(0);
        let mut newest_ns: u128 = 0;
        let mut rows = entries.iter().map(|(path, e)| {
            let (size, mtime_ns) = match e.stat {
                Some(known) => known,
                None => match fs::symlink_metadata(self.root.join(path)) {
                    Ok(m) => (m.len(), mtime_ns(&m)),
                    Err(_) => (e.content.len() as u64, 0),
                },
            };
            newest_ns = newest_ns.max(mtime_ns);
            let entry = ManifestEntry {
                size,
                mtime_ns,
                hash: e
                    .content_hash
                    .unwrap_or_else(|| *blake3::hash(&e.content).as_bytes()),
                blob: blobs.get(path).copied().or(e.blob),
                graph: e.graph_key,
                exec: e.exec,
            };
            (path.as_str(), entry)
        });
        // The directories of the walk these entries came from, if any.
        let (dirs, nested) = self.walked_dirs.take().unwrap_or_default();
        let new = encode_manifest(conflicted, uncommitted, written, &dirs, &nested, &mut rows);
        drop(rows);
        let p = self.manifest_path();
        // Leave an unchanged manifest alone when its stamp already trusts
        // every entry (none is within the racy window): rewriting it would
        // change nothing but the file, which editors watching the tree
        // (and subrepos inside it) take as a change.
        if let Ok(old) = fs::read(&p) {
            let (w, sum) = (MANIFEST_WRITTEN, MANIFEST_SUM);
            if old.len() == new.len()
                && old[..w.start] == new[..w.start]
                && old[sum.end..] == new[sum.end..]
            {
                let old_written = u128::from_le_bytes(old[w.clone()].try_into().unwrap());
                if newest_ns + 2_000_000_000 < old_written {
                    return Ok(());
                }
            }
        }
        fs::create_dir_all(p.parent().unwrap())?;
        write_cache(&p, &new)
    }

    /// Change the header flags, and set the graph key of each path in
    /// `keys` (after a commit, the new base's keys of the files it holds as
    /// they are on disk).
    pub(crate) fn rewrite_manifest(
        &self,
        conflicted: bool,
        uncommitted: bool,
        keys: &HashMap<String, [u8; 32]>,
    ) -> Result<()> {
        self.manifest_cache.take();
        self.listing_cache.take();
        let p = self.manifest_path();
        let Some((_, _, m)) = fs::read(&p).ok().and_then(|b| decode_manifest(&b)) else {
            return self.write_manifest(conflicted, uncommitted);
        };
        let mut rows = m.entries.iter().map(|(path, e)| {
            let mut e = e.clone();
            if let Some(k) = keys.get(path) {
                e.graph = Some(*k);
            }
            (path.as_str(), e)
        });
        write_cache(
            &p,
            &encode_manifest(
                conflicted,
                uncommitted,
                m.written_ns,
                &m.dirs,
                &m.nested,
                &mut rows,
            ),
        )
    }

    /// Forget the manifest after an operation that changed the working-copy
    /// change without a snapshot (undo, rewrites); the prompt shows blank
    /// markers until the next snapshot.
    pub(crate) fn drop_manifest(&self) {
        self.manifest_cache.take();
        self.listing_cache.take();
        let _ = fs::remove_file(self.manifest_path());
    }

    /// The manifest, with whether it records a conflict and uncommitted
    /// work; parsed once while the file stays the same one.
    /// Whether the working copy was conflicted when last recorded (the
    /// manifest's flag), without looking at the files.
    pub fn recorded_conflicted(&self) -> Result<bool> {
        Ok(self.read_manifest()?.is_some_and(|(c, _, _)| c))
    }

    pub(crate) fn read_manifest(&self) -> Result<Option<(bool, bool, std::sync::Arc<Manifest>)>> {
        let path = self.manifest_path();
        let id = match fs::metadata(&path) {
            Ok(m) => {
                #[cfg(unix)]
                let ino = std::os::unix::fs::MetadataExt::ino(&m);
                #[cfg(not(unix))]
                let ino = 0;
                (m.len(), mtime_ns(&m), ino)
            }
            Err(e) if e.kind() == std::io::ErrorKind::NotFound => return Ok(None),
            Err(e) => return Err(e.into()),
        };
        if let Some((cached_id, m)) = &*self.manifest_cache.borrow() {
            if *cached_id == id {
                return Ok(Some(m.clone()));
            }
        }
        let Some((c, u, m)) = self.parse_manifest(&path)? else {
            return Ok(None);
        };
        let m = (c, u, std::sync::Arc::new(m));
        *self.manifest_cache.borrow_mut() = Some((id, m.clone()));
        Ok(Some(m))
    }

    fn parse_manifest(&self, path: &std::path::Path) -> Result<Option<(bool, bool, Manifest)>> {
        let buf = match fs::read(path) {
            Ok(b) => b,
            Err(e) if e.kind() == std::io::ErrorKind::NotFound => return Ok(None),
            Err(e) => return Err(e.into()),
        };
        Ok(decode_manifest(&buf))
    }

    /// Whether the working copy is certainly clean, known from the manifest
    /// alone, without loading any history (spec §12.4): the last snapshot
    /// or commit found no uncommitted work and no conflict, there are no
    /// subrepos (whose state the manifest does not hold), and no file was
    /// added, removed or changed since. An entry inside the racy window,
    /// where an equal size and time do not prove equal content, is checked
    /// by hashing the file. `false` means only that a snapshot must decide.
    pub fn clean_by_manifest(&self) -> Result<bool> {
        let _t = crate::timing::phase("status: check the manifest");
        // Why a snapshot must decide, with timings on.
        let not = |why: &str| {
            if crate::timing::enabled() {
                eprintln!("timing status: not clean by the manifest: {why}");
            }
            Ok(false)
        };
        let Some((conflicted, uncommitted, m)) = self.read_manifest()? else {
            return not("no manifest");
        };
        if conflicted || uncommitted || self.root.join(".oxymodules").exists() {
            return not("conflicted, uncommitted or subrepos");
        }
        let started = now_ns();
        if self.disk_differs(&m, true)? {
            return not("a file differs");
        }
        // Verified clean. Entries or directories within the racy window of
        // the manifest's time were hashed or listed just now; restamped with
        // the time this check began, the next check can trust them, as
        // Git restamps its index (anything changed within two seconds of
        // that time is still racy by the same rule).
        // A check that had to list the tree leaves the directories it
        // listed: kept, the next check need not list.
        let window = 2_000_000_000;
        let listed = self.listed_dirs.take();
        let racy = m.dirs.iter().any(|(_, t)| t + window >= m.written_ns)
            || m.entries
                .values()
                .any(|e| e.mtime_ns + window >= m.written_ns);
        if (racy || (m.dirs.is_empty() && listed.is_some())) && started > m.written_ns {
            let (dirs, nested) = listed.unwrap_or_else(|| (m.dirs.clone(), m.nested.clone()));
            self.restamp_manifest(&m, conflicted, uncommitted, started, &dirs, &nested);
        }
        Ok(true)
    }

    /// Rewrite the manifest with a new time and directory list, its entries
    /// unchanged.
    fn restamp_manifest(
        &self,
        m: &Manifest,
        conflicted: bool,
        uncommitted: bool,
        written_ns: u128,
        dirs: &[(String, u128)],
        nested: &[String],
    ) {
        let _t = crate::timing::phase("status: restamp the manifest");
        let mut rows = m.entries.iter().map(|(p, e)| (p.as_str(), e.clone()));
        let bytes = encode_manifest(conflicted, uncommitted, written_ns, dirs, nested, &mut rows);
        self.manifest_cache.take();
        let _ = write_cache(&self.manifest_path(), &bytes);
    }

    /// A manifest entry written within the racy window whose file no
    /// longer matches it. Such entries are hashed, in parallel (a checkout
    /// just wrote them all).
    fn racy_change(&self, m: &Manifest) -> Option<String> {
        use rayon::prelude::*;
        let racy: Vec<(&String, &ManifestEntry)> = m
            .entries
            .iter()
            .filter(|(_, e)| e.mtime_ns + 2_000_000_000 >= m.written_ns)
            .collect();
        let root = &self.root;
        racy.par_iter()
            .find_any(|(rel, e)| {
                let path = root.join(rel.as_str());
                let same = match fs::symlink_metadata(&path) {
                    Ok(md) if md.file_type().is_symlink() => fs::read_link(&path).is_ok_and(|t| {
                        *blake3::hash(t.to_string_lossy().as_bytes()).as_bytes() == e.hash
                    }),
                    Ok(md) if md.is_file() => {
                        md.len() == e.size
                            && fs::read(&path).is_ok_and(|b| *blake3::hash(&b).as_bytes() == e.hash)
                    }
                    _ => false,
                };
                !same
            })
            .map(|(rel, _)| rel.to_string())
    }

    /// The status lines of the last snapshot, when the working copy is as
    /// that snapshot left it: the cache it wrote names the working-copy
    /// change and the manifest it recorded, and every file still matches
    /// that manifest. `None` when a snapshot must decide (no cache, a file
    /// changed, conflicts or subrepos). An empty list is a clean working
    /// copy.
    pub fn status_by_manifest(&self) -> Result<Option<Vec<PathStatus>>> {
        // A listing left by an earlier check describes the files as they
        // were then; this check makes its own.
        self.listing_cache.take();
        let Some((conflicted, uncommitted, m)) = self.read_manifest()? else {
            return Ok(None);
        };
        if !uncommitted {
            return Ok(self.clean_by_manifest()?.then(Vec::new));
        }
        let _t = crate::timing::phase("status: by the status cache");
        if conflicted || self.root.join(".oxymodules").exists() {
            return Ok(None);
        }
        let Some((wc, stamp, statuses)) = fs::read(self.status_cache_path())
            .ok()
            .and_then(|b| decode_status_cache(&b))
        else {
            return Ok(None);
        };
        let now = self.view()?.workspaces.get(&self.workspace).copied();
        if now != Some(wc) || self.manifest_stamp() != Some(stamp) {
            return Ok(None);
        }
        if self.disk_differs(&m, true)? {
            return Ok(None);
        }
        // Test builds check the cache against a snapshot of the files.
        #[cfg(test)]
        {
            let snap = self.snapshot_status("status cache check")?;
            let key = |v: &[PathStatus]| {
                let mut k: Vec<(String, u8)> = v
                    .iter()
                    .map(|p| (p.path.clone(), status_byte(p.status)))
                    .collect();
                k.sort();
                k
            };
            assert_eq!(
                key(&statuses),
                key(&snap.statuses),
                "status cache disagrees with a snapshot"
            );
        }
        Ok(Some(statuses))
    }

    fn status_cache_path(&self) -> PathBuf {
        self.manifest_path().with_file_name("status")
    }

    /// The manifest file's checksum, which names its content.
    fn manifest_stamp(&self) -> Option<[u8; 32]> {
        use std::io::Read;
        let mut head = [0u8; MANIFEST_SUM.end];
        fs::File::open(self.manifest_path())
            .and_then(|mut f| f.read_exact(&mut head))
            .ok()?;
        head[MANIFEST_SUM].try_into().ok()
    }

    /// Record `statuses` as the status of working-copy change `wc` against
    /// the manifest as it is now; with no uncommitted work there is nothing
    /// to record, and the cache goes.
    pub(crate) fn write_status_cache(&self, wc: Option<&ChangeHash>, statuses: &[PathStatus]) {
        let path = self.status_cache_path();
        let (Some(wc), Some(stamp)) = (wc, self.manifest_stamp()) else {
            let _ = fs::remove_file(&path);
            return;
        };
        let _ = write_cache(&path, &encode_status_cache(wc, &stamp, statuses));
    }

    /// Whether the working copy differs from the last manifest. Without a
    /// manifest the answer is `None`.
    pub fn quick_dirty(&self) -> Result<Option<bool>> {
        self.quick_dirty_keeping(false)
    }

    /// `quick_dirty`; with `keep`, the listing it made is left for the
    /// next walk of the working copy.
    fn quick_dirty_keeping(&self, keep: bool) -> Result<Option<bool>> {
        let Some((_, uncommitted, manifest)) = self.read_manifest()? else {
            return Ok(None);
        };
        if uncommitted {
            return Ok(Some(true));
        }
        Ok(Some(self.disk_differs(&manifest, keep)?))
    }

    /// Whether the working copy differs from `manifest`: a file added,
    /// gone or changed. With `keep`, the listing made is left for the next
    /// walk of the working copy when it does.
    fn disk_differs(&self, manifest: &Manifest, keep: bool) -> Result<bool> {
        // A file written within the racy window can change without its size
        // or time moving (a coarse clock): those entries are hashed. One
        // that changed still lets the directory check below hand its stats
        // to the snapshot that follows.
        let racy = self.racy_change(manifest).is_some();
        if let Some(differs) = self.disk_differs_by_dirs_keeping(manifest, keep, racy) {
            return Ok(differs || racy);
        }
        if racy {
            return Ok(true);
        }
        self.disk_differs_listing(manifest, keep)
    }

    /// [`Repo::disk_differs`] without listing the tree, when the manifest
    /// knows its directories: a directory's time moves when an entry is
    /// added, removed or renamed in it, so with every directory's time as
    /// recorded (and none within the racy window) the files are the
    /// manifest's, and each is checked by its stat. `None` when a
    /// directory changed or the manifest has none: the tree must be listed.
    #[cfg(test)]
    pub(crate) fn disk_differs_by_dirs(&self, manifest: &Manifest) -> Option<bool> {
        self.disk_differs_by_dirs_keeping(manifest, false, false)
    }

    /// [`Repo::disk_differs_by_dirs`]; with `keep`, when a file differs
    /// (or `racy` says one does), the stats taken become the listing of the
    /// snapshot that follows: no directory's entries changed, so the
    /// manifest's files are the tree's.
    fn disk_differs_by_dirs_keeping(
        &self,
        manifest: &Manifest,
        keep: bool,
        racy: bool,
    ) -> Option<bool> {
        use rayon::prelude::*;
        if manifest.dirs.is_empty() || !self.config.sparse_paths.is_empty() {
            return None;
        }
        let _t = crate::timing::phase("status: directories and files by stat");
        let root = &self.root;
        let dirs_same = manifest.dirs.par_iter().all(|(rel, t)| {
            *t + 2_000_000_000 < manifest.written_ns
                && fs::metadata(root.join(rel)).is_ok_and(|m| m.is_dir() && mtime_ns(&m) == *t)
        });
        if !dirs_same {
            return None;
        }
        let rels: Vec<&String> = manifest.entries.keys().collect();
        let now = stat_tracked(root, &rels);
        let differs_from = |me: &ManifestEntry, rel: &str, now: &Tracked| match now {
            Tracked::Link(t) => *blake3::hash(t).as_bytes() != me.hash,
            Tracked::File(m) => {
                m.size != me.size
                    || m.exec != me.exec
                    || (m.mtime_ns != me.mtime_ns
                        && !fs::read(root.join(rel))
                            .is_ok_and(|b| *blake3::hash(&b).as_bytes() == me.hash))
            }
        };
        // A path gone or no longer a file: its directory moved, unless
        // within a clock tick; list the tree to be sure.
        if now.iter().any(Option::is_none) {
            return None;
        }
        // The entries in the order `rels` was taken in: no lookup by path.
        let entries: Vec<&ManifestEntry> = manifest.entries.values().collect();
        let differs = rels
            .par_iter()
            .zip(entries.par_iter())
            .zip(now.par_iter())
            .any(|((rel, me), n)| differs_from(me, rel, n.as_ref().unwrap()));
        if keep && (differs || racy) {
            let mut listing = super::walk::Listing {
                dirs: manifest.dirs.clone(),
                nested: manifest.nested.iter().map(|n| root.join(n)).collect(),
                ..Default::default()
            };
            for (rel, n) in rels.into_iter().zip(now) {
                match n.unwrap() {
                    Tracked::File(m) => listing.files.push((rel.clone(), root.join(rel), m)),
                    Tracked::Link(t) => listing.symlinks.push((rel.clone(), t)),
                }
            }
            *self.listing_cache.borrow_mut() = Some(listing);
        }
        if !differs {
            // Nested repositories the manifest has no entry for are not
            // subrepos: nothing in them is tracked (oxygen#43).
            *self.untracked_nested.borrow_mut() = manifest
                .nested
                .iter()
                .filter(|rel| !manifest.entries.contains_key(*rel))
                .cloned()
                .collect();
        }
        Some(differs)
    }

    /// [`Repo::disk_differs`] by listing the tree.
    fn disk_differs_listing(&self, manifest: &Manifest, keep: bool) -> Result<bool> {
        use rayon::prelude::*;
        let sparse = &self.config.sparse_paths;
        let in_scope = |p: &str| {
            sparse.is_empty()
                || sparse
                    .iter()
                    .any(|s| p == s || p.starts_with(&format!("{s}/")))
        };
        // A working copy that cannot be listed counts as changed.
        // Known files are not matched against the ignore rules: with no
        // uncommitted work (the clean check), the manifest's files are the
        // base's, and a prompt's answer is the same either way.
        let known = |p: &str| manifest.entries.contains_key(p);
        let Ok(listing) = super::walk::list_knowing(&self.root, &in_scope, &known) else {
            return Ok(true);
        };
        let differs = || -> Result<bool> {
            let links_differ = listing.symlinks.iter().any(|(rel, target)| {
                manifest
                    .entries
                    .get(rel)
                    .is_none_or(|me| *blake3::hash(target).as_bytes() != me.hash)
            });
            if links_differ {
                return Ok(true);
            }
            // Each file is checked on its own, in parallel, stopping at the
            // first difference; an unreadable one counts as different.
            let files_differ = listing.files.par_iter().any(|(rel, path, m)| {
                manifest.entries.get(rel).is_none_or(|me| {
                    m.size != me.size
                        || m.exec != me.exec
                        || (m.mtime_ns != me.mtime_ns
                            && !fs::read(path)
                                .is_ok_and(|b| *blake3::hash(&b).as_bytes() == me.hash))
                })
            });
            if files_differ {
                return Ok(true);
            }
            let seen = listing.files.len() + listing.symlinks.len();
            if seen == manifest.entries.len() {
                return Ok(false);
            }
            // Manifest entries the walk skipped: tracked files that ignore
            // rules hide (Linux tracks some its own .gitignore names). Found
            // by merging the two sorted lists; they count as present when
            // still on disk and unchanged, checked on every core.
            let mut listed: Vec<&str> = listing
                .files
                .iter()
                .map(|(r, _, _)| r.as_str())
                .chain(listing.symlinks.iter().map(|(r, _)| r.as_str()))
                .collect();
            listed.sort_unstable();
            let mut at = listed.iter().peekable();
            let skipped: Vec<(&String, &ManifestEntry)> = manifest
                .entries
                .iter()
                .filter(|(rel, _)| {
                    while at.next_if(|l| **l < rel.as_str()).is_some() {}
                    at.peek().is_none_or(|l| **l != rel.as_str())
                })
                .collect();
            let root = &self.root;
            Ok(skipped.par_iter().any(|(rel, me)| {
                let path = root.join(rel.as_str());
                let Ok(m) = fs::symlink_metadata(&path) else {
                    return true;
                };
                !m.file_type().is_symlink()
                    && (m.len() != me.size
                        || super::is_executable(&m) != me.exec
                        || (mtime_ns(&m) != me.mtime_ns
                            && !fs::read(&path)
                                .is_ok_and(|b| *blake3::hash(&b).as_bytes() == me.hash)))
            }))
        };
        let dirty = differs()?;
        if !dirty {
            // Nested repositories the manifest has no entry for are not
            // subrepos: nothing in them is tracked (oxygen#43).
            let mut nested: Vec<String> = listing
                .nested
                .iter()
                .map(|d| self.rel_path(d))
                .filter(|rel| !manifest.entries.contains_key(rel))
                .collect();
            nested.sort();
            *self.untracked_nested.borrow_mut() = nested;
            // A complete list of the directories, each timed before it was
            // read, for a restamp to keep (see `clean_by_manifest`).
            *self.listed_dirs.borrow_mut() = Some((
                listing.dirs.clone(),
                listing.nested.iter().map(|d| self.rel_path(d)).collect(),
            ));
        }
        if dirty && keep {
            *self.listing_cache.borrow_mut() = Some(listing);
        }
        Ok(dirty)
    }

    /// Prompt information without recording anything.
    pub fn quick_status(&self) -> Result<PromptInfo> {
        let view = self.view()?;
        let bookmark = self.config.bookmark.clone();
        let heads = self.heads_of(&view, &bookmark)?;
        let wc_hash = view.workspaces.get(&self.workspace).copied();
        let change_id = match wc_hash {
            Some(h) => self.store.get_change_meta(&h)?.change_id.short(8),
            None => String::new(),
        };
        // Conflictedness was recorded by the last snapshot; without a
        // manifest (never snapshotted) both flags are unknown and shown clear.
        let manifest = self.read_manifest()?;
        let conflicted = manifest.as_ref().map(|(c, _, _)| *c).unwrap_or(false);
        let dirty = self.quick_dirty()?.unwrap_or(false);
        let (ahead, behind) = match view.remotes.get("origin").and_then(|b| b.get(&bookmark)) {
            Some(st) => {
                let remote_heads = self.store.get_state(st)?.heads;
                let ahead = self.negotiate(&remote_heads, &heads).map(|v| v.len()).ok();
                let behind = self.negotiate(&heads, &remote_heads).map(|v| v.len()).ok();
                (ahead, behind)
            }
            None => (None, None),
        };
        Ok(PromptInfo {
            bookmark,
            change_id,
            heads: heads.len(),
            dirty,
            conflicted,
            ahead,
            behind,
        })
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn placeholders_and_glyphs() {
        let clean = PromptInfo {
            bookmark: "main".into(),
            change_id: "658990aa".into(),
            heads: 1,
            dirty: false,
            conflicted: false,
            ahead: Some(0),
            behind: None,
        };
        assert_eq!(clean.format(DEFAULT_PROMPT_FORMAT), "main 658990aa");
        assert_eq!(clean.format(GLYPH_PROMPT_FORMAT), "⚛ main 658990aa ✔");
        let busy = PromptInfo {
            dirty: true,
            ahead: Some(2),
            behind: Some(3),
            ..clean.clone()
        };
        assert_eq!(busy.format(DEFAULT_PROMPT_FORMAT), "main 658990aa*+2-3");
        assert_eq!(busy.format(GLYPH_PROMPT_FORMAT), "⚛ main 658990aa ✎↑2↓3");
        let bad = PromptInfo {
            conflicted: true,
            ..busy.clone()
        };
        assert_eq!(bad.format("%s%x%d %h %a/%B %%"), "‼!* 1 2/3 %");
        assert_eq!(bad.state_color(), 31);
        assert_eq!(
            busy.format_colored("%b", ("\\[", "\\]")),
            "\\[\x1b[33m\\]main\\[\x1b[0m\\]"
        );
        assert_eq!(clean.format_colored("%b", ("", "")), "\x1b[36mmain\x1b[0m");
        assert_eq!(clean.format("%q"), "%q");
    }
}
`/`	Search
`g` `c`	Code
`g` `h`	History
`g` `b`	Bookmarks
`g` `t`	Trust
`y`	Permalink to this state (files and trees)
`?`	This sheet
Keyboard
