//! 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>));

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,

        },

    ))

}

/// 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)

        });

<<<<<<< oxy conflict 1 of 1, side 1 of 2: change 484e854e "The manifest keeps directory times: status lists the tree only when a directory changed (refs #41)"

        // The directories of the walk these entries came from, if any.

        let (dirs, nested) = self.walked_dirs.take().unwrap_or_default();

======= oxy conflict 1 of 1, side 2 of 2: change c04ec0a5 "Timing phases for writing the manifest (refs #41)"

        let encoding = crate::timing::phase("manifest: rows and encoding");

>>>>>>> oxy conflict 1 of 1 end

        let new = encode_manifest(conflicted, uncommitted, written, &dirs, &nested, &mut rows);

        drop(rows);

        drop(encoding);

        if crate::timing::enabled() {

            let unstated = entries.values().filter(|e| e.stat.is_none()).count();

            let unhashed = entries

                .values()

                .filter(|e| e.content_hash.is_none())

                .count();

            eprintln!(

                "timing manifest: {} rows, {unstated} without a stat, {unhashed} without a hash",

                entries.len()

            );

        }

        let _t = crate::timing::phase("manifest: compare and write");

        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.

    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 {