rust-dirstate: introduce intermediate struct for dirstate-v2 data

This is passed often as a long tuple that is not easy to know the form of, so
we refactor everything in this struct.

This also renames `wdir_tracked` to follow the Python `wc_tracked`, even though
the on-disk format uses `WDIR_TRACKED`.
I think a single naming scheme is better, but we can't easily break the Python
impl now because of extensions, so this is low-effort enough and facilitates
grepping.

Differential Revision: https://phab.mercurial-scm.org/D12498

rust/hg-core/src/dirstate/entry.rs

@@ -1,711 +1,716 b''
 use crate::dirstate_tree::on_disk::DirstateV2ParseError;
 use crate::errors::HgError;
 use bitflags::bitflags;
 use std::convert::{TryFrom, TryInto};
 use std::fs;
 use std::io;
 use std::time::{SystemTime, UNIX_EPOCH};
 
 #[derive(Copy, Clone, Debug, Eq, PartialEq)]
 pub enum EntryState {
     Normal,
     Added,
     Removed,
     Merged,
 }
 
 /// `size` and `mtime.seconds` are truncated to 31 bits.
 ///
 /// TODO: double-check status algorithm correctness for files
 /// larger than 2 GiB or modified after 2038.
 #[derive(Debug, Copy, Clone)]
 pub struct DirstateEntry {
     pub(crate) flags: Flags,
     mode_size: Option<(u32, u32)>,
     mtime: Option<TruncatedTimestamp>,
 }
 
 bitflags! {
     pub(crate) struct Flags: u8 {
         const WDIR_TRACKED = 1 << 0;
         const P1_TRACKED = 1 << 1;
         const P2_INFO = 1 << 2;
         const HAS_FALLBACK_EXEC = 1 << 3;
         const FALLBACK_EXEC = 1 << 4;
         const HAS_FALLBACK_SYMLINK = 1 << 5;
         const FALLBACK_SYMLINK = 1 << 6;
     }
 }
 
 /// A Unix timestamp with nanoseconds precision
 #[derive(Debug, Copy, Clone)]
 pub struct TruncatedTimestamp {
     truncated_seconds: u32,
     /// Always in the `0 .. 1_000_000_000` range.
     nanoseconds: u32,
     /// TODO this should be in DirstateEntry, but the current code needs
     /// refactoring to use DirstateEntry instead of TruncatedTimestamp for
     /// comparison.
     pub second_ambiguous: bool,
 }
 
 impl TruncatedTimestamp {
     /// Constructs from a timestamp potentially outside of the supported range,
     /// and truncate the seconds components to its lower 31 bits.
     ///
     /// Panics if the nanoseconds components is not in the expected range.
     pub fn new_truncate(
         seconds: i64,
         nanoseconds: u32,
         second_ambiguous: bool,
     ) -> Self {
         assert!(nanoseconds < NSEC_PER_SEC);
         Self {
             truncated_seconds: seconds as u32 & RANGE_MASK_31BIT,
             nanoseconds,
             second_ambiguous,
         }
     }
 
     /// Construct from components. Returns an error if they are not in the
     /// expcted range.
     pub fn from_already_truncated(
         truncated_seconds: u32,
         nanoseconds: u32,
         second_ambiguous: bool,
     ) -> Result<Self, DirstateV2ParseError> {
         if truncated_seconds & !RANGE_MASK_31BIT == 0
             && nanoseconds < NSEC_PER_SEC
         {
             Ok(Self {
                 truncated_seconds,
                 nanoseconds,
                 second_ambiguous,
             })
         } else {
             Err(DirstateV2ParseError)
         }
     }
 
     /// Returns a `TruncatedTimestamp` for the modification time of `metadata`.
     ///
     /// Propagates errors from `std` on platforms where modification time
     /// is not available at all.
     pub fn for_mtime_of(metadata: &fs::Metadata) -> io::Result<Self> {
         #[cfg(unix)]
         {
             use std::os::unix::fs::MetadataExt;
             let seconds = metadata.mtime();
             // i64 -> u32 with value always in the `0 .. NSEC_PER_SEC` range
             let nanoseconds = metadata.mtime_nsec().try_into().unwrap();
             Ok(Self::new_truncate(seconds, nanoseconds, false))
         }
         #[cfg(not(unix))]
         {
             metadata.modified().map(Self::from)
         }
     }
 
     /// Like `for_mtime_of`, but may return `None` or a value with
     /// `second_ambiguous` set if the mtime is not "reliable".
     ///
     /// A modification time is reliable if it is older than `boundary` (or
     /// sufficiently in the future).
     ///
     /// Otherwise a concurrent modification might happens with the same mtime.
     pub fn for_reliable_mtime_of(
         metadata: &fs::Metadata,
         boundary: &Self,
     ) -> io::Result<Option<Self>> {
         let mut mtime = Self::for_mtime_of(metadata)?;
         // If the mtime of the ambiguous file is younger (or equal) to the
         // starting point of the `status` walk, we cannot garantee that
         // another, racy, write will not happen right after with the same mtime
         // and we cannot cache the information.
         //
         // However if the mtime is far away in the future, this is likely some
         // mismatch between the current clock and previous file system
         // operation. So mtime more than one days in the future are considered
         // fine.
         let reliable = if mtime.truncated_seconds == boundary.truncated_seconds
         {
             mtime.second_ambiguous = true;
             mtime.nanoseconds != 0
                 && boundary.nanoseconds != 0
                 && mtime.nanoseconds < boundary.nanoseconds
         } else {
             // `truncated_seconds` is less than 2**31,
             // so this does not overflow `u32`:
             let one_day_later = boundary.truncated_seconds + 24 * 3600;
             mtime.truncated_seconds < boundary.truncated_seconds
                 || mtime.truncated_seconds > one_day_later
         };
         if reliable {
             Ok(Some(mtime))
         } else {
             Ok(None)
         }
     }
 
     /// The lower 31 bits of the number of seconds since the epoch.
     pub fn truncated_seconds(&self) -> u32 {
         self.truncated_seconds
     }
 
     /// The sub-second component of this timestamp, in nanoseconds.
     /// Always in the `0 .. 1_000_000_000` range.
     ///
     /// This timestamp is after `(seconds, 0)` by this many nanoseconds.
     pub fn nanoseconds(&self) -> u32 {
         self.nanoseconds
     }
 
     /// Returns whether two timestamps are equal modulo 2**31 seconds.
     ///
     /// If this returns `true`, the original values converted from `SystemTime`
     /// or given to `new_truncate` were very likely equal. A false positive is
     /// possible if they were exactly a multiple of 2**31 seconds apart (around
     /// 68 years). This is deemed very unlikely to happen by chance, especially
     /// on filesystems that support sub-second precision.
     ///
     /// If someone is manipulating the modification times of some files to
     /// intentionally make `hg status` return incorrect results, not truncating
     /// wouldn’t help much since they can set exactly the expected timestamp.
     ///
     /// Sub-second precision is ignored if it is zero in either value.
     /// Some APIs simply return zero when more precision is not available.
     /// When comparing values from different sources, if only one is truncated
     /// in that way, doing a simple comparison would cause many false
     /// negatives.
     pub fn likely_equal(self, other: Self) -> bool {
         if self.truncated_seconds != other.truncated_seconds {
             false
         } else if self.nanoseconds == 0 || other.nanoseconds == 0 {
             if self.second_ambiguous {
                 false
             } else {
                 true
             }
         } else {
             self.nanoseconds == other.nanoseconds
         }
     }
 
     pub fn likely_equal_to_mtime_of(
         self,
         metadata: &fs::Metadata,
     ) -> io::Result<bool> {
         Ok(self.likely_equal(Self::for_mtime_of(metadata)?))
     }
 }
 
 impl From<SystemTime> for TruncatedTimestamp {
     fn from(system_time: SystemTime) -> Self {
         // On Unix, `SystemTime` is a wrapper for the `timespec` C struct:
         // https://www.gnu.org/software/libc/manual/html_node/Time-Types.html#index-struct-timespec
         // We want to effectively access its fields, but the Rust standard
         // library does not expose them. The best we can do is:
         let seconds;
         let nanoseconds;
         match system_time.duration_since(UNIX_EPOCH) {
             Ok(duration) => {
                 seconds = duration.as_secs() as i64;
                 nanoseconds = duration.subsec_nanos();
             }
             Err(error) => {
                 // `system_time` is before `UNIX_EPOCH`.
                 // We need to undo this algorithm:
                 // https://github.com/rust-lang/rust/blob/6bed1f0bc3cc50c10aab26d5f94b16a00776b8a5/library/std/src/sys/unix/time.rs#L40-L41
                 let negative = error.duration();
                 let negative_secs = negative.as_secs() as i64;
                 let negative_nanos = negative.subsec_nanos();
                 if negative_nanos == 0 {
                     seconds = -negative_secs;
                     nanoseconds = 0;
                 } else {
                     // For example if `system_time` was 4.3 seconds before
                     // the Unix epoch we get a Duration that represents
                     // `(-4, -0.3)` but we want `(-5, +0.7)`:
                     seconds = -1 - negative_secs;
                     nanoseconds = NSEC_PER_SEC - negative_nanos;
                 }
             }
         };
         Self::new_truncate(seconds, nanoseconds, false)
     }
 }
 
 const NSEC_PER_SEC: u32 = 1_000_000_000;
 pub const RANGE_MASK_31BIT: u32 = 0x7FFF_FFFF;
 
 pub const MTIME_UNSET: i32 = -1;
 
 /// A `DirstateEntry` with a size of `-2` means that it was merged from the
 /// other parent. This allows revert to pick the right status back during a
 /// merge.
 pub const SIZE_FROM_OTHER_PARENT: i32 = -2;
 /// A special value used for internal representation of special case in
 /// dirstate v1 format.
 pub const SIZE_NON_NORMAL: i32 = -1;
 
+#[derive(Debug, Default, Copy, Clone)]
+pub struct DirstateV2Data {
+    pub wc_tracked: bool,
+    pub p1_tracked: bool,
+    pub p2_info: bool,
+    pub mode_size: Option<(u32, u32)>,
+    pub mtime: Option<TruncatedTimestamp>,
+    pub fallback_exec: Option<bool>,
+    pub fallback_symlink: Option<bool>,
+}
+
 impl DirstateEntry {
-    pub fn from_v2_data(
-        wdir_tracked: bool,
-        p1_tracked: bool,
-        p2_info: bool,
-        mode_size: Option<(u32, u32)>,
-        mtime: Option<TruncatedTimestamp>,
-        fallback_exec: Option<bool>,
-        fallback_symlink: Option<bool>,
-    ) -> Self {
+    pub fn from_v2_data(v2_data: DirstateV2Data) -> Self {
+        let DirstateV2Data {
+            wc_tracked,
+            p1_tracked,
+            p2_info,
+            mode_size,
+            mtime,
+            fallback_exec,
+            fallback_symlink,
+        } = v2_data;
         if let Some((mode, size)) = mode_size {
             // TODO: return an error for out of range values?
             assert!(mode & !RANGE_MASK_31BIT == 0);
             assert!(size & !RANGE_MASK_31BIT == 0);
         }
         let mut flags = Flags::empty();
-        flags.set(Flags::WDIR_TRACKED, wdir_tracked);
+        flags.set(Flags::WDIR_TRACKED, wc_tracked);
         flags.set(Flags::P1_TRACKED, p1_tracked);
         flags.set(Flags::P2_INFO, p2_info);
         if let Some(exec) = fallback_exec {
             flags.insert(Flags::HAS_FALLBACK_EXEC);
             if exec {
                 flags.insert(Flags::FALLBACK_EXEC);
             }
         }
         if let Some(exec) = fallback_symlink {
             flags.insert(Flags::HAS_FALLBACK_SYMLINK);
             if exec {
                 flags.insert(Flags::FALLBACK_SYMLINK);
             }
         }
         Self {
             flags,
             mode_size,
             mtime,
         }
     }
 
     pub fn from_v1_data(
         state: EntryState,
         mode: i32,
         size: i32,
         mtime: i32,
     ) -> Self {
         match state {
             EntryState::Normal => {
                 if size == SIZE_FROM_OTHER_PARENT {
                     Self {
                         // might be missing P1_TRACKED
                         flags: Flags::WDIR_TRACKED | Flags::P2_INFO,
                         mode_size: None,
                         mtime: None,
                     }
                 } else if size == SIZE_NON_NORMAL {
                     Self {
                         flags: Flags::WDIR_TRACKED | Flags::P1_TRACKED,
                         mode_size: None,
                         mtime: None,
                     }
                 } else if mtime == MTIME_UNSET {
                     // TODO: return an error for negative values?
                     let mode = u32::try_from(mode).unwrap();
                     let size = u32::try_from(size).unwrap();
                     Self {
                         flags: Flags::WDIR_TRACKED | Flags::P1_TRACKED,
                         mode_size: Some((mode, size)),
                         mtime: None,
                     }
                 } else {
                     // TODO: return an error for negative values?
                     let mode = u32::try_from(mode).unwrap();
                     let size = u32::try_from(size).unwrap();
                     let mtime = u32::try_from(mtime).unwrap();
                     let mtime = TruncatedTimestamp::from_already_truncated(
                         mtime, 0, false,
                     )
                     .unwrap();
                     Self {
                         flags: Flags::WDIR_TRACKED | Flags::P1_TRACKED,
                         mode_size: Some((mode, size)),
                         mtime: Some(mtime),
                     }
                 }
             }
             EntryState::Added => Self {
                 flags: Flags::WDIR_TRACKED,
                 mode_size: None,
                 mtime: None,
             },
             EntryState::Removed => Self {
                 flags: if size == SIZE_NON_NORMAL {
                     Flags::P1_TRACKED | Flags::P2_INFO
                 } else if size == SIZE_FROM_OTHER_PARENT {
                     // We don’t know if P1_TRACKED should be set (file history)
                     Flags::P2_INFO
                 } else {
                     Flags::P1_TRACKED
                 },
                 mode_size: None,
                 mtime: None,
             },
             EntryState::Merged => Self {
                 flags: Flags::WDIR_TRACKED
                     | Flags::P1_TRACKED // might not be true because of rename ?
                     | Flags::P2_INFO, // might not be true because of rename ?
                 mode_size: None,
                 mtime: None,
             },
         }
     }
 
     /// Creates a new entry in "removed" state.
     ///
     /// `size` is expected to be zero, `SIZE_NON_NORMAL`, or
     /// `SIZE_FROM_OTHER_PARENT`
     pub fn new_removed(size: i32) -> Self {
         Self::from_v1_data(EntryState::Removed, 0, size, 0)
     }
 
     pub fn new_tracked() -> Self {
-        Self::from_v2_data(true, false, false, None, None, None, None)
+        let data = DirstateV2Data {
+            wc_tracked: true,
+            ..Default::default()
+        };
+        Self::from_v2_data(data)
     }
 
     pub fn tracked(&self) -> bool {
         self.flags.contains(Flags::WDIR_TRACKED)
     }
 
     pub fn p1_tracked(&self) -> bool {
         self.flags.contains(Flags::P1_TRACKED)
     }
 
     fn in_either_parent(&self) -> bool {
         self.flags.intersects(Flags::P1_TRACKED | Flags::P2_INFO)
     }
 
     pub fn removed(&self) -> bool {
         self.in_either_parent() && !self.flags.contains(Flags::WDIR_TRACKED)
     }
 
     pub fn p2_info(&self) -> bool {
         self.flags.contains(Flags::WDIR_TRACKED | Flags::P2_INFO)
     }
 
     pub fn added(&self) -> bool {
         self.flags.contains(Flags::WDIR_TRACKED) && !self.in_either_parent()
     }
 
     pub fn maybe_clean(&self) -> bool {
         if !self.flags.contains(Flags::WDIR_TRACKED) {
             false
         } else if !self.flags.contains(Flags::P1_TRACKED) {
             false
         } else if self.flags.contains(Flags::P2_INFO) {
             false
         } else {
             true
         }
     }
 
     pub fn any_tracked(&self) -> bool {
         self.flags.intersects(
             Flags::WDIR_TRACKED | Flags::P1_TRACKED | Flags::P2_INFO,
         )
     }
 
-    /// Returns `(wdir_tracked, p1_tracked, p2_info, mode_size, mtime)`
-    pub(crate) fn v2_data(
-        &self,
-    ) -> (
-        bool,
-        bool,
-        bool,
-        Option<(u32, u32)>,
-        Option<TruncatedTimestamp>,
-        Option<bool>,
-        Option<bool>,
-    ) {
+    pub(crate) fn v2_data(&self) -> DirstateV2Data {
         if !self.any_tracked() {
             // TODO: return an Option instead?
             panic!("Accessing v2_data of an untracked DirstateEntry")
         }
-        let wdir_tracked = self.flags.contains(Flags::WDIR_TRACKED);
+        let wc_tracked = self.flags.contains(Flags::WDIR_TRACKED);
         let p1_tracked = self.flags.contains(Flags::P1_TRACKED);
         let p2_info = self.flags.contains(Flags::P2_INFO);
         let mode_size = self.mode_size;
         let mtime = self.mtime;
-        (
-            wdir_tracked,
+        DirstateV2Data {
+            wc_tracked,
             p1_tracked,
             p2_info,
             mode_size,
             mtime,
-            self.get_fallback_exec(),
-            self.get_fallback_symlink(),
-        )
+            fallback_exec: self.get_fallback_exec(),
+            fallback_symlink: self.get_fallback_symlink(),
+        }
     }
 
     fn v1_state(&self) -> EntryState {
         if !self.any_tracked() {
             // TODO: return an Option instead?
             panic!("Accessing v1_state of an untracked DirstateEntry")
         }
         if self.removed() {
             EntryState::Removed
         } else if self
             .flags
             .contains(Flags::WDIR_TRACKED | Flags::P1_TRACKED | Flags::P2_INFO)
         {
             EntryState::Merged
         } else if self.added() {
             EntryState::Added
         } else {
             EntryState::Normal
         }
     }
 
     fn v1_mode(&self) -> i32 {
         if let Some((mode, _size)) = self.mode_size {
             i32::try_from(mode).unwrap()
         } else {
             0
         }
     }
 
     fn v1_size(&self) -> i32 {
         if !self.any_tracked() {
             // TODO: return an Option instead?
             panic!("Accessing v1_size of an untracked DirstateEntry")
         }
         if self.removed()
             && self.flags.contains(Flags::P1_TRACKED | Flags::P2_INFO)
         {
             SIZE_NON_NORMAL
         } else if self.flags.contains(Flags::P2_INFO) {
             SIZE_FROM_OTHER_PARENT
         } else if self.removed() {
             0
         } else if self.added() {
             SIZE_NON_NORMAL
         } else if let Some((_mode, size)) = self.mode_size {
             i32::try_from(size).unwrap()
         } else {
             SIZE_NON_NORMAL
         }
     }
 
     fn v1_mtime(&self) -> i32 {
         if !self.any_tracked() {
             // TODO: return an Option instead?
             panic!("Accessing v1_mtime of an untracked DirstateEntry")
         }
         if self.removed() {
             0
         } else if self.flags.contains(Flags::P2_INFO) {
             MTIME_UNSET
         } else if !self.flags.contains(Flags::P1_TRACKED) {
             MTIME_UNSET
         } else if let Some(mtime) = self.mtime {
             if mtime.second_ambiguous {
                 MTIME_UNSET
             } else {
                 i32::try_from(mtime.truncated_seconds()).unwrap()
             }
         } else {
             MTIME_UNSET
         }
     }
 
     // TODO: return `Option<EntryState>`? None when `!self.any_tracked`
     pub fn state(&self) -> EntryState {
         self.v1_state()
     }
 
     // TODO: return Option?
     pub fn mode(&self) -> i32 {
         self.v1_mode()
     }
 
     // TODO: return Option?
     pub fn size(&self) -> i32 {
         self.v1_size()
     }
 
     // TODO: return Option?
     pub fn mtime(&self) -> i32 {
         self.v1_mtime()
     }
 
     pub fn get_fallback_exec(&self) -> Option<bool> {
         if self.flags.contains(Flags::HAS_FALLBACK_EXEC) {
             Some(self.flags.contains(Flags::FALLBACK_EXEC))
         } else {
             None
         }
     }
 
     pub fn set_fallback_exec(&mut self, value: Option<bool>) {
         match value {
             None => {
                 self.flags.remove(Flags::HAS_FALLBACK_EXEC);
                 self.flags.remove(Flags::FALLBACK_EXEC);
             }
             Some(exec) => {
                 self.flags.insert(Flags::HAS_FALLBACK_EXEC);
                 if exec {
                     self.flags.insert(Flags::FALLBACK_EXEC);
                 }
             }
         }
     }
 
     pub fn get_fallback_symlink(&self) -> Option<bool> {
         if self.flags.contains(Flags::HAS_FALLBACK_SYMLINK) {
             Some(self.flags.contains(Flags::FALLBACK_SYMLINK))
         } else {
             None
         }
     }
 
     pub fn set_fallback_symlink(&mut self, value: Option<bool>) {
         match value {
             None => {
                 self.flags.remove(Flags::HAS_FALLBACK_SYMLINK);
                 self.flags.remove(Flags::FALLBACK_SYMLINK);
             }
             Some(symlink) => {
                 self.flags.insert(Flags::HAS_FALLBACK_SYMLINK);
                 if symlink {
                     self.flags.insert(Flags::FALLBACK_SYMLINK);
                 }
             }
         }
     }
 
     pub fn truncated_mtime(&self) -> Option<TruncatedTimestamp> {
         self.mtime
     }
 
     pub fn drop_merge_data(&mut self) {
         if self.flags.contains(Flags::P2_INFO) {
             self.flags.remove(Flags::P2_INFO);
             self.mode_size = None;
             self.mtime = None;
         }
     }
 
     pub fn set_possibly_dirty(&mut self) {
         self.mtime = None
     }
 
     pub fn set_clean(
         &mut self,
         mode: u32,
         size: u32,
         mtime: TruncatedTimestamp,
     ) {
         let size = size & RANGE_MASK_31BIT;
         self.flags.insert(Flags::WDIR_TRACKED | Flags::P1_TRACKED);
         self.mode_size = Some((mode, size));
         self.mtime = Some(mtime);
     }
 
     pub fn set_tracked(&mut self) {
         self.flags.insert(Flags::WDIR_TRACKED);
         // `set_tracked` is replacing various `normallookup` call. So we mark
         // the files as needing lookup
         //
         // Consider dropping this in the future in favor of something less
         // broad.
         self.mtime = None;
     }
 
     pub fn set_untracked(&mut self) {
         self.flags.remove(Flags::WDIR_TRACKED);
         self.mode_size = None;
         self.mtime = None;
     }
 
     /// Returns `(state, mode, size, mtime)` for the puprose of serialization
     /// in the dirstate-v1 format.
     ///
     /// This includes marker values such as `mtime == -1`. In the future we may
     /// want to not represent these cases that way in memory, but serialization
     /// will need to keep the same format.
     pub fn v1_data(&self) -> (u8, i32, i32, i32) {
         (
             self.v1_state().into(),
             self.v1_mode(),
             self.v1_size(),
             self.v1_mtime(),
         )
     }
 
     pub(crate) fn is_from_other_parent(&self) -> bool {
         self.state() == EntryState::Normal
             && self.size() == SIZE_FROM_OTHER_PARENT
     }
 
     // TODO: other platforms
     #[cfg(unix)]
     pub fn mode_changed(
         &self,
         filesystem_metadata: &std::fs::Metadata,
     ) -> bool {
         let dirstate_exec_bit = (self.mode() as u32 & EXEC_BIT_MASK) != 0;
         let fs_exec_bit = has_exec_bit(filesystem_metadata);
         dirstate_exec_bit != fs_exec_bit
     }
 
     /// Returns a `(state, mode, size, mtime)` tuple as for
     /// `DirstateMapMethods::debug_iter`.
     pub fn debug_tuple(&self) -> (u8, i32, i32, i32) {
         (self.state().into(), self.mode(), self.size(), self.mtime())
     }
 }
 
 impl EntryState {
     pub fn is_tracked(self) -> bool {
         use EntryState::*;
         match self {
             Normal | Added | Merged => true,
             Removed => false,
         }
     }
 }
 
 impl TryFrom<u8> for EntryState {
     type Error = HgError;
 
     fn try_from(value: u8) -> Result<Self, Self::Error> {
         match value {
             b'n' => Ok(EntryState::Normal),
             b'a' => Ok(EntryState::Added),
             b'r' => Ok(EntryState::Removed),
             b'm' => Ok(EntryState::Merged),
             _ => Err(HgError::CorruptedRepository(format!(
                 "Incorrect dirstate entry state {}",
                 value
             ))),
         }
     }
 }
 
 impl Into<u8> for EntryState {
     fn into(self) -> u8 {
         match self {
             EntryState::Normal => b'n',
             EntryState::Added => b'a',
             EntryState::Removed => b'r',
             EntryState::Merged => b'm',
         }
     }
 }
 
 const EXEC_BIT_MASK: u32 = 0o100;
 
 pub fn has_exec_bit(metadata: &std::fs::Metadata) -> bool {
     // TODO: How to handle executable permissions on Windows?
     use std::os::unix::fs::MetadataExt;
     (metadata.mode() & EXEC_BIT_MASK) != 0
 }

rust/hg-core/src/dirstate_tree/on_disk.rs

@@ -1,843 +1,843 b''
 //! The "version 2" disk representation of the dirstate
 //!
 //! See `mercurial/helptext/internals/dirstate-v2.txt`
 
-use crate::dirstate::TruncatedTimestamp;
+use crate::dirstate::{DirstateV2Data, TruncatedTimestamp};
 use crate::dirstate_tree::dirstate_map::{self, DirstateMap, NodeRef};
 use crate::dirstate_tree::path_with_basename::WithBasename;
 use crate::errors::HgError;
 use crate::utils::hg_path::HgPath;
 use crate::DirstateEntry;
 use crate::DirstateError;
 use crate::DirstateParents;
 use bitflags::bitflags;
 use bytes_cast::unaligned::{U16Be, U32Be};
 use bytes_cast::BytesCast;
 use format_bytes::format_bytes;
 use rand::Rng;
 use std::borrow::Cow;
 use std::convert::{TryFrom, TryInto};
 use std::fmt::Write;
 
 /// Added at the start of `.hg/dirstate` when the "v2" format is used.
 /// This a redundant sanity check more than an actual "magic number" since
 /// `.hg/requires` already governs which format should be used.
 pub const V2_FORMAT_MARKER: &[u8; 12] = b"dirstate-v2\n";
 
 /// Keep space for 256-bit hashes
 const STORED_NODE_ID_BYTES: usize = 32;
 
 /// … even though only 160 bits are used for now, with SHA-1
 const USED_NODE_ID_BYTES: usize = 20;
 
 pub(super) const IGNORE_PATTERNS_HASH_LEN: usize = 20;
 pub(super) type IgnorePatternsHash = [u8; IGNORE_PATTERNS_HASH_LEN];
 
 /// Must match constants of the same names in `mercurial/dirstateutils/v2.py`
 const TREE_METADATA_SIZE: usize = 44;
 const NODE_SIZE: usize = 44;
 
 /// Make sure that size-affecting changes are made knowingly
 #[allow(unused)]
 fn static_assert_size_of() {
     let _ = std::mem::transmute::<TreeMetadata, [u8; TREE_METADATA_SIZE]>;
     let _ = std::mem::transmute::<DocketHeader, [u8; TREE_METADATA_SIZE + 81]>;
     let _ = std::mem::transmute::<Node, [u8; NODE_SIZE]>;
 }
 
 // Must match `HEADER` in `mercurial/dirstateutils/docket.py`
 #[derive(BytesCast)]
 #[repr(C)]
 struct DocketHeader {
     marker: [u8; V2_FORMAT_MARKER.len()],
     parent_1: [u8; STORED_NODE_ID_BYTES],
     parent_2: [u8; STORED_NODE_ID_BYTES],
 
     metadata: TreeMetadata,
 
     /// Counted in bytes
     data_size: Size,
 
     uuid_size: u8,
 }
 
 pub struct Docket<'on_disk> {
     header: &'on_disk DocketHeader,
     pub uuid: &'on_disk [u8],
 }
 
 /// Fields are documented in the *Tree metadata in the docket file*
 /// section of `mercurial/helptext/internals/dirstate-v2.txt`
 #[derive(BytesCast)]
 #[repr(C)]
 pub struct TreeMetadata {
     root_nodes: ChildNodes,
     nodes_with_entry_count: Size,
     nodes_with_copy_source_count: Size,
     unreachable_bytes: Size,
     unused: [u8; 4],
 
     /// See *Optional hash of ignore patterns* section of
     /// `mercurial/helptext/internals/dirstate-v2.txt`
     ignore_patterns_hash: IgnorePatternsHash,
 }
 
 /// Fields are documented in the *The data file format*
 /// section of `mercurial/helptext/internals/dirstate-v2.txt`
 #[derive(BytesCast)]
 #[repr(C)]
 pub(super) struct Node {
     full_path: PathSlice,
 
     /// In bytes from `self.full_path.start`
     base_name_start: PathSize,
 
     copy_source: OptPathSlice,
     children: ChildNodes,
     pub(super) descendants_with_entry_count: Size,
     pub(super) tracked_descendants_count: Size,
     flags: U16Be,
     size: U32Be,
     mtime: PackedTruncatedTimestamp,
 }
 
 bitflags! {
     #[repr(C)]
     struct Flags: u16 {
         const WDIR_TRACKED = 1 << 0;
         const P1_TRACKED = 1 << 1;
         const P2_INFO = 1 << 2;
         const MODE_EXEC_PERM = 1 << 3;
         const MODE_IS_SYMLINK = 1 << 4;
         const HAS_FALLBACK_EXEC = 1 <<  5;
         const FALLBACK_EXEC = 1 <<  6;
         const HAS_FALLBACK_SYMLINK = 1 <<  7;
         const FALLBACK_SYMLINK = 1 <<  8;
         const EXPECTED_STATE_IS_MODIFIED = 1 << 9;
         const HAS_MODE_AND_SIZE = 1 <<10;
         const HAS_MTIME = 1 <<11;
         const MTIME_SECOND_AMBIGUOUS = 1 << 12;
         const DIRECTORY = 1 <<13;
         const ALL_UNKNOWN_RECORDED = 1 <<14;
         const ALL_IGNORED_RECORDED = 1 <<15;
     }
 }
 
 /// Duration since the Unix epoch
 #[derive(BytesCast, Copy, Clone)]
 #[repr(C)]
 struct PackedTruncatedTimestamp {
     truncated_seconds: U32Be,
     nanoseconds: U32Be,
 }
 
 /// Counted in bytes from the start of the file
 ///
 /// NOTE: not supporting `.hg/dirstate` files larger than 4 GiB.
 type Offset = U32Be;
 
 /// Counted in number of items
 ///
 /// NOTE: we choose not to support counting more than 4 billion nodes anywhere.
 type Size = U32Be;
 
 /// Counted in bytes
 ///
 /// NOTE: we choose not to support file names/paths longer than 64 KiB.
 type PathSize = U16Be;
 
 /// A contiguous sequence of `len` times `Node`, representing the child nodes
 /// of either some other node or of the repository root.
 ///
 /// Always sorted by ascending `full_path`, to allow binary search.
 /// Since nodes with the same parent nodes also have the same parent path,
 /// only the `base_name`s need to be compared during binary search.
 #[derive(BytesCast, Copy, Clone)]
 #[repr(C)]
 struct ChildNodes {
     start: Offset,
     len: Size,
 }
 
 /// A `HgPath` of `len` bytes
 #[derive(BytesCast, Copy, Clone)]
 #[repr(C)]
 struct PathSlice {
     start: Offset,
     len: PathSize,
 }
 
 /// Either nothing if `start == 0`, or a `HgPath` of `len` bytes
 type OptPathSlice = PathSlice;
 
 /// Unexpected file format found in `.hg/dirstate` with the "v2" format.
 ///
 /// This should only happen if Mercurial is buggy or a repository is corrupted.
 #[derive(Debug)]
 pub struct DirstateV2ParseError;
 
 impl From<DirstateV2ParseError> for HgError {
     fn from(_: DirstateV2ParseError) -> Self {
         HgError::corrupted("dirstate-v2 parse error")
     }
 }
 
 impl From<DirstateV2ParseError> for crate::DirstateError {
     fn from(error: DirstateV2ParseError) -> Self {
         HgError::from(error).into()
     }
 }
 
 impl TreeMetadata {
     pub fn as_bytes(&self) -> &[u8] {
         BytesCast::as_bytes(self)
     }
 }
 
 impl<'on_disk> Docket<'on_disk> {
     /// Generate the identifier for a new data file
     ///
     /// TODO: support the `HGTEST_UUIDFILE` environment variable.
     /// See `mercurial/revlogutils/docket.py`
     pub fn new_uid() -> String {
         const ID_LENGTH: usize = 8;
         let mut id = String::with_capacity(ID_LENGTH);
         let mut rng = rand::thread_rng();
         for _ in 0..ID_LENGTH {
             // One random hexadecimal digit.
             // `unwrap` never panics because `impl Write for String`
             // never returns an error.
             write!(&mut id, "{:x}", rng.gen_range(0..16)).unwrap();
         }
         id
     }
 
     pub fn serialize(
         parents: DirstateParents,
         tree_metadata: TreeMetadata,
         data_size: u64,
         uuid: &[u8],
     ) -> Result<Vec<u8>, std::num::TryFromIntError> {
         let header = DocketHeader {
             marker: *V2_FORMAT_MARKER,
             parent_1: parents.p1.pad_to_256_bits(),
             parent_2: parents.p2.pad_to_256_bits(),
             metadata: tree_metadata,
             data_size: u32::try_from(data_size)?.into(),
             uuid_size: uuid.len().try_into()?,
         };
         let header = header.as_bytes();
         let mut docket = Vec::with_capacity(header.len() + uuid.len());
         docket.extend_from_slice(header);
         docket.extend_from_slice(uuid);
         Ok(docket)
     }
 
     pub fn parents(&self) -> DirstateParents {
         use crate::Node;
         let p1 = Node::try_from(&self.header.parent_1[..USED_NODE_ID_BYTES])
             .unwrap()
             .clone();
         let p2 = Node::try_from(&self.header.parent_2[..USED_NODE_ID_BYTES])
             .unwrap()
             .clone();
         DirstateParents { p1, p2 }
     }
 
     pub fn tree_metadata(&self) -> &[u8] {
         self.header.metadata.as_bytes()
     }
 
     pub fn data_size(&self) -> usize {
         // This `unwrap` could only panic on a 16-bit CPU
         self.header.data_size.get().try_into().unwrap()
     }
 
     pub fn data_filename(&self) -> String {
         String::from_utf8(format_bytes!(b"dirstate.{}", self.uuid)).unwrap()
     }
 }
 
 pub fn read_docket(
     on_disk: &[u8],
 ) -> Result<Docket<'_>, DirstateV2ParseError> {
     let (header, uuid) =
         DocketHeader::from_bytes(on_disk).map_err(|_| DirstateV2ParseError)?;
     let uuid_size = header.uuid_size as usize;
     if header.marker == *V2_FORMAT_MARKER && uuid.len() == uuid_size {
         Ok(Docket { header, uuid })
     } else {
         Err(DirstateV2ParseError)
     }
 }
 
 pub(super) fn read<'on_disk>(
     on_disk: &'on_disk [u8],
     metadata: &[u8],
 ) -> Result<DirstateMap<'on_disk>, DirstateV2ParseError> {
     if on_disk.is_empty() {
         return Ok(DirstateMap::empty(on_disk));
     }
     let (meta, _) = TreeMetadata::from_bytes(metadata)
         .map_err(|_| DirstateV2ParseError)?;
     let dirstate_map = DirstateMap {
         on_disk,
         root: dirstate_map::ChildNodes::OnDisk(read_nodes(
             on_disk,
             meta.root_nodes,
         )?),
         nodes_with_entry_count: meta.nodes_with_entry_count.get(),
         nodes_with_copy_source_count: meta.nodes_with_copy_source_count.get(),
         ignore_patterns_hash: meta.ignore_patterns_hash,
         unreachable_bytes: meta.unreachable_bytes.get(),
     };
     Ok(dirstate_map)
 }
 
 impl Node {
     pub(super) fn full_path<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<&'on_disk HgPath, DirstateV2ParseError> {
         read_hg_path(on_disk, self.full_path)
     }
 
     pub(super) fn base_name_start<'on_disk>(
         &self,
     ) -> Result<usize, DirstateV2ParseError> {
         let start = self.base_name_start.get();
         if start < self.full_path.len.get() {
             let start = usize::try_from(start)
                 // u32 -> usize, could only panic on a 16-bit CPU
                 .expect("dirstate-v2 base_name_start out of bounds");
             Ok(start)
         } else {
             Err(DirstateV2ParseError)
         }
     }
 
     pub(super) fn base_name<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<&'on_disk HgPath, DirstateV2ParseError> {
         let full_path = self.full_path(on_disk)?;
         let base_name_start = self.base_name_start()?;
         Ok(HgPath::new(&full_path.as_bytes()[base_name_start..]))
     }
 
     pub(super) fn path<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<dirstate_map::NodeKey<'on_disk>, DirstateV2ParseError> {
         Ok(WithBasename::from_raw_parts(
             Cow::Borrowed(self.full_path(on_disk)?),
             self.base_name_start()?,
         ))
     }
 
     pub(super) fn has_copy_source<'on_disk>(&self) -> bool {
         self.copy_source.start.get() != 0
     }
 
     pub(super) fn copy_source<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<Option<&'on_disk HgPath>, DirstateV2ParseError> {
         Ok(if self.has_copy_source() {
             Some(read_hg_path(on_disk, self.copy_source)?)
         } else {
             None
         })
     }
 
     fn flags(&self) -> Flags {
         Flags::from_bits_truncate(self.flags.get())
     }
 
     fn has_entry(&self) -> bool {
         self.flags().intersects(
             Flags::WDIR_TRACKED | Flags::P1_TRACKED | Flags::P2_INFO,
         )
     }
 
     pub(super) fn node_data(
         &self,
     ) -> Result<dirstate_map::NodeData, DirstateV2ParseError> {
         if self.has_entry() {
             Ok(dirstate_map::NodeData::Entry(self.assume_entry()?))
         } else if let Some(mtime) = self.cached_directory_mtime()? {
             Ok(dirstate_map::NodeData::CachedDirectory { mtime })
         } else {
             Ok(dirstate_map::NodeData::None)
         }
     }
 
     pub(super) fn cached_directory_mtime(
         &self,
     ) -> Result<Option<TruncatedTimestamp>, DirstateV2ParseError> {
         // For now we do not have code to handle the absence of
         // ALL_UNKNOWN_RECORDED, so we ignore the mtime if the flag is
         // unset.
         if self.flags().contains(Flags::DIRECTORY)
             && self.flags().contains(Flags::HAS_MTIME)
             && self.flags().contains(Flags::ALL_UNKNOWN_RECORDED)
         {
             Ok(Some(self.mtime()?))
         } else {
             Ok(None)
         }
     }
 
     fn synthesize_unix_mode(&self) -> u32 {
         let file_type = if self.flags().contains(Flags::MODE_IS_SYMLINK) {
             libc::S_IFLNK
         } else {
             libc::S_IFREG
         };
         let permisions = if self.flags().contains(Flags::MODE_EXEC_PERM) {
             0o755
         } else {
             0o644
         };
         (file_type | permisions).into()
     }
 
     fn mtime(&self) -> Result<TruncatedTimestamp, DirstateV2ParseError> {
         let mut m: TruncatedTimestamp = self.mtime.try_into()?;
         if self.flags().contains(Flags::MTIME_SECOND_AMBIGUOUS) {
             m.second_ambiguous = true;
         }
         Ok(m)
     }
 
     fn assume_entry(&self) -> Result<DirstateEntry, DirstateV2ParseError> {
         // TODO: convert through raw bits instead?
-        let wdir_tracked = self.flags().contains(Flags::WDIR_TRACKED);
+        let wc_tracked = self.flags().contains(Flags::WDIR_TRACKED);
         let p1_tracked = self.flags().contains(Flags::P1_TRACKED);
         let p2_info = self.flags().contains(Flags::P2_INFO);
         let mode_size = if self.flags().contains(Flags::HAS_MODE_AND_SIZE)
             && !self.flags().contains(Flags::EXPECTED_STATE_IS_MODIFIED)
         {
             Some((self.synthesize_unix_mode(), self.size.into()))
         } else {
             None
         };
         let mtime = if self.flags().contains(Flags::HAS_MTIME)
             && !self.flags().contains(Flags::DIRECTORY)
             && !self.flags().contains(Flags::EXPECTED_STATE_IS_MODIFIED)
         {
             Some(self.mtime()?)
         } else {
             None
         };
         let fallback_exec = if self.flags().contains(Flags::HAS_FALLBACK_EXEC)
         {
             Some(self.flags().contains(Flags::FALLBACK_EXEC))
         } else {
             None
         };
         let fallback_symlink =
             if self.flags().contains(Flags::HAS_FALLBACK_SYMLINK) {
                 Some(self.flags().contains(Flags::FALLBACK_SYMLINK))
             } else {
                 None
             };
-        Ok(DirstateEntry::from_v2_data(
-            wdir_tracked,
+        Ok(DirstateEntry::from_v2_data(DirstateV2Data {
+            wc_tracked,
             p1_tracked,
             p2_info,
             mode_size,
             mtime,
             fallback_exec,
             fallback_symlink,
-        ))
+        }))
     }
 
     pub(super) fn entry(
         &self,
     ) -> Result<Option<DirstateEntry>, DirstateV2ParseError> {
         if self.has_entry() {
             Ok(Some(self.assume_entry()?))
         } else {
             Ok(None)
         }
     }
 
     pub(super) fn children<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<&'on_disk [Node], DirstateV2ParseError> {
         read_nodes(on_disk, self.children)
     }
 
     pub(super) fn to_in_memory_node<'on_disk>(
         &self,
         on_disk: &'on_disk [u8],
     ) -> Result<dirstate_map::Node<'on_disk>, DirstateV2ParseError> {
         Ok(dirstate_map::Node {
             children: dirstate_map::ChildNodes::OnDisk(
                 self.children(on_disk)?,
             ),
             copy_source: self.copy_source(on_disk)?.map(Cow::Borrowed),
             data: self.node_data()?,
             descendants_with_entry_count: self
                 .descendants_with_entry_count
                 .get(),
             tracked_descendants_count: self.tracked_descendants_count.get(),
         })
     }
 
     fn from_dirstate_entry(
         entry: &DirstateEntry,
     ) -> (Flags, U32Be, PackedTruncatedTimestamp) {
-        let (
-            wdir_tracked,
+        let DirstateV2Data {
+            wc_tracked,
             p1_tracked,
             p2_info,
-            mode_size_opt,
-            mtime_opt,
+            mode_size: mode_size_opt,
+            mtime: mtime_opt,
             fallback_exec,
             fallback_symlink,
-        ) = entry.v2_data();
-        // TODO: convert throug raw flag bits instead?
+        } = entry.v2_data();
+        // TODO: convert through raw flag bits instead?
         let mut flags = Flags::empty();
-        flags.set(Flags::WDIR_TRACKED, wdir_tracked);
+        flags.set(Flags::WDIR_TRACKED, wc_tracked);
         flags.set(Flags::P1_TRACKED, p1_tracked);
         flags.set(Flags::P2_INFO, p2_info);
         let size = if let Some((m, s)) = mode_size_opt {
             let exec_perm = m & (libc::S_IXUSR as u32) != 0;
             let is_symlink = m & (libc::S_IFMT as u32) == libc::S_IFLNK as u32;
             flags.set(Flags::MODE_EXEC_PERM, exec_perm);
             flags.set(Flags::MODE_IS_SYMLINK, is_symlink);
             flags.insert(Flags::HAS_MODE_AND_SIZE);
             s.into()
         } else {
             0.into()
         };
         let mtime = if let Some(m) = mtime_opt {
             flags.insert(Flags::HAS_MTIME);
             if m.second_ambiguous {
                 flags.insert(Flags::MTIME_SECOND_AMBIGUOUS);
             };
             m.into()
         } else {
             PackedTruncatedTimestamp::null()
         };
         if let Some(f_exec) = fallback_exec {
             flags.insert(Flags::HAS_FALLBACK_EXEC);
             if f_exec {
                 flags.insert(Flags::FALLBACK_EXEC);
             }
         }
         if let Some(f_symlink) = fallback_symlink {
             flags.insert(Flags::HAS_FALLBACK_SYMLINK);
             if f_symlink {
                 flags.insert(Flags::FALLBACK_SYMLINK);
             }
         }
         (flags, size, mtime)
     }
 }
 
 fn read_hg_path(
     on_disk: &[u8],
     slice: PathSlice,
 ) -> Result<&HgPath, DirstateV2ParseError> {
     read_slice(on_disk, slice.start, slice.len.get()).map(HgPath::new)
 }
 
 fn read_nodes(
     on_disk: &[u8],
     slice: ChildNodes,
 ) -> Result<&[Node], DirstateV2ParseError> {
     read_slice(on_disk, slice.start, slice.len.get())
 }
 
 fn read_slice<T, Len>(
     on_disk: &[u8],
     start: Offset,
     len: Len,
 ) -> Result<&[T], DirstateV2ParseError>
 where
     T: BytesCast,
     Len: TryInto<usize>,
 {
     // Either `usize::MAX` would result in "out of bounds" error since a single
     // `&[u8]` cannot occupy the entire addess space.
     let start = start.get().try_into().unwrap_or(std::usize::MAX);
     let len = len.try_into().unwrap_or(std::usize::MAX);
     on_disk
         .get(start..)
         .and_then(|bytes| T::slice_from_bytes(bytes, len).ok())
         .map(|(slice, _rest)| slice)
         .ok_or_else(|| DirstateV2ParseError)
 }
 
 pub(crate) fn for_each_tracked_path<'on_disk>(
     on_disk: &'on_disk [u8],
     metadata: &[u8],
     mut f: impl FnMut(&'on_disk HgPath),
 ) -> Result<(), DirstateV2ParseError> {
     let (meta, _) = TreeMetadata::from_bytes(metadata)
         .map_err(|_| DirstateV2ParseError)?;
     fn recur<'on_disk>(
         on_disk: &'on_disk [u8],
         nodes: ChildNodes,
         f: &mut impl FnMut(&'on_disk HgPath),
     ) -> Result<(), DirstateV2ParseError> {
         for node in read_nodes(on_disk, nodes)? {
             if let Some(entry) = node.entry()? {
                 if entry.state().is_tracked() {
                     f(node.full_path(on_disk)?)
                 }
             }
             recur(on_disk, node.children, f)?
         }
         Ok(())
     }
     recur(on_disk, meta.root_nodes, &mut f)
 }
 
 /// Returns new data and metadata, together with whether that data should be
 /// appended to the existing data file whose content is at
 /// `dirstate_map.on_disk` (true), instead of written to a new data file
 /// (false).
 pub(super) fn write(
     dirstate_map: &DirstateMap,
     can_append: bool,
 ) -> Result<(Vec<u8>, TreeMetadata, bool), DirstateError> {
     let append = can_append && dirstate_map.write_should_append();
 
     // This ignores the space for paths, and for nodes without an entry.
     // TODO: better estimate? Skip the `Vec` and write to a file directly?
     let size_guess = std::mem::size_of::<Node>()
         * dirstate_map.nodes_with_entry_count as usize;
 
     let mut writer = Writer {
         dirstate_map,
         append,
         out: Vec::with_capacity(size_guess),
     };
 
     let root_nodes = writer.write_nodes(dirstate_map.root.as_ref())?;
 
     let meta = TreeMetadata {
         root_nodes,
         nodes_with_entry_count: dirstate_map.nodes_with_entry_count.into(),
         nodes_with_copy_source_count: dirstate_map
             .nodes_with_copy_source_count
             .into(),
         unreachable_bytes: dirstate_map.unreachable_bytes.into(),
         unused: [0; 4],
         ignore_patterns_hash: dirstate_map.ignore_patterns_hash,
     };
     Ok((writer.out, meta, append))
 }
 
 struct Writer<'dmap, 'on_disk> {
     dirstate_map: &'dmap DirstateMap<'on_disk>,
     append: bool,
     out: Vec<u8>,
 }
 
 impl Writer<'_, '_> {
     fn write_nodes(
         &mut self,
         nodes: dirstate_map::ChildNodesRef,
     ) -> Result<ChildNodes, DirstateError> {
         // Reuse already-written nodes if possible
         if self.append {
             if let dirstate_map::ChildNodesRef::OnDisk(nodes_slice) = nodes {
                 let start = self.on_disk_offset_of(nodes_slice).expect(
                     "dirstate-v2 OnDisk nodes not found within on_disk",
                 );
                 let len = child_nodes_len_from_usize(nodes_slice.len());
                 return Ok(ChildNodes { start, len });
             }
         }
 
         // `dirstate_map::ChildNodes::InMemory` contains a `HashMap` which has
         // undefined iteration order. Sort to enable binary search in the
         // written file.
         let nodes = nodes.sorted();
         let nodes_len = nodes.len();
 
         // First accumulate serialized nodes in a `Vec`
         let mut on_disk_nodes = Vec::with_capacity(nodes_len);
         for node in nodes {
             let children =
                 self.write_nodes(node.children(self.dirstate_map.on_disk)?)?;
             let full_path = node.full_path(self.dirstate_map.on_disk)?;
             let full_path = self.write_path(full_path.as_bytes());
             let copy_source = if let Some(source) =
                 node.copy_source(self.dirstate_map.on_disk)?
             {
                 self.write_path(source.as_bytes())
             } else {
                 PathSlice {
                     start: 0.into(),
                     len: 0.into(),
                 }
             };
             on_disk_nodes.push(match node {
                 NodeRef::InMemory(path, node) => {
                     let (flags, size, mtime) = match &node.data {
                         dirstate_map::NodeData::Entry(entry) => {
                             Node::from_dirstate_entry(entry)
                         }
                         dirstate_map::NodeData::CachedDirectory { mtime } => {
                             // we currently never set a mtime if unknown file
                             // are present.
                             // So if we have a mtime for a directory, we know
                             // they are no unknown
                             // files and we
                             // blindly set ALL_UNKNOWN_RECORDED.
                             //
                             // We never set ALL_IGNORED_RECORDED since we
                             // don't track that case
                             // currently.
                             let mut flags = Flags::DIRECTORY
                                 | Flags::HAS_MTIME
                                 | Flags::ALL_UNKNOWN_RECORDED;
                             if mtime.second_ambiguous {
                                 flags.insert(Flags::MTIME_SECOND_AMBIGUOUS)
                             }
                             (flags, 0.into(), (*mtime).into())
                         }
                         dirstate_map::NodeData::None => (
                             Flags::DIRECTORY,
                             0.into(),
                             PackedTruncatedTimestamp::null(),
                         ),
                     };
                     Node {
                         children,
                         copy_source,
                         full_path,
                         base_name_start: u16::try_from(path.base_name_start())
                             // Could only panic for paths over 64 KiB
                             .expect("dirstate-v2 path length overflow")
                             .into(),
                         descendants_with_entry_count: node
                             .descendants_with_entry_count
                             .into(),
                         tracked_descendants_count: node
                             .tracked_descendants_count
                             .into(),
                         flags: flags.bits().into(),
                         size,
                         mtime,
                     }
                 }
                 NodeRef::OnDisk(node) => Node {
                     children,
                     copy_source,
                     full_path,
                     ..*node
                 },
             })
         }
         // … so we can write them contiguously, after writing everything else
         // they refer to.
         let start = self.current_offset();
         let len = child_nodes_len_from_usize(nodes_len);
         self.out.extend(on_disk_nodes.as_bytes());
         Ok(ChildNodes { start, len })
     }
 
     /// If the given slice of items is within `on_disk`, returns its offset
     /// from the start of `on_disk`.
     fn on_disk_offset_of<T>(&self, slice: &[T]) -> Option<Offset>
     where
         T: BytesCast,
     {
         fn address_range(slice: &[u8]) -> std::ops::RangeInclusive<usize> {
             let start = slice.as_ptr() as usize;
             let end = start + slice.len();
             start..=end
         }
         let slice_addresses = address_range(slice.as_bytes());
         let on_disk_addresses = address_range(self.dirstate_map.on_disk);
         if on_disk_addresses.contains(slice_addresses.start())
             && on_disk_addresses.contains(slice_addresses.end())
         {
             let offset = slice_addresses.start() - on_disk_addresses.start();
             Some(offset_from_usize(offset))
         } else {
             None
         }
     }
 
     fn current_offset(&mut self) -> Offset {
         let mut offset = self.out.len();
         if self.append {
             offset += self.dirstate_map.on_disk.len()
         }
         offset_from_usize(offset)
     }
 
     fn write_path(&mut self, slice: &[u8]) -> PathSlice {
         let len = path_len_from_usize(slice.len());
         // Reuse an already-written path if possible
         if self.append {
             if let Some(start) = self.on_disk_offset_of(slice) {
                 return PathSlice { start, len };
             }
         }
         let start = self.current_offset();
         self.out.extend(slice.as_bytes());
         PathSlice { start, len }
     }
 }
 
 fn offset_from_usize(x: usize) -> Offset {
     u32::try_from(x)
         // Could only panic for a dirstate file larger than 4 GiB
         .expect("dirstate-v2 offset overflow")
         .into()
 }
 
 fn child_nodes_len_from_usize(x: usize) -> Size {
     u32::try_from(x)
         // Could only panic with over 4 billion nodes
         .expect("dirstate-v2 slice length overflow")
         .into()
 }
 
 fn path_len_from_usize(x: usize) -> PathSize {
     u16::try_from(x)
         // Could only panic for paths over 64 KiB
         .expect("dirstate-v2 path length overflow")
         .into()
 }
 
 impl From<TruncatedTimestamp> for PackedTruncatedTimestamp {
     fn from(timestamp: TruncatedTimestamp) -> Self {
         Self {
             truncated_seconds: timestamp.truncated_seconds().into(),
             nanoseconds: timestamp.nanoseconds().into(),
         }
     }
 }
 
 impl TryFrom<PackedTruncatedTimestamp> for TruncatedTimestamp {
     type Error = DirstateV2ParseError;
 
     fn try_from(
         timestamp: PackedTruncatedTimestamp,
     ) -> Result<Self, Self::Error> {
         Self::from_already_truncated(
             timestamp.truncated_seconds.get(),
             timestamp.nanoseconds.get(),
             false,
         )
     }
 }
 impl PackedTruncatedTimestamp {
     fn null() -> Self {
         Self {
             truncated_seconds: 0.into(),
             nanoseconds: 0.into(),
         }
     }
 }

rust/hg-cpython/src/dirstate/item.rs

@@ -1,246 +1,247 b''
 use cpython::exc;
 use cpython::ObjectProtocol;
 use cpython::PyBytes;
 use cpython::PyErr;
 use cpython::PyNone;
 use cpython::PyObject;
 use cpython::PyResult;
 use cpython::Python;
 use cpython::PythonObject;
 use hg::dirstate::DirstateEntry;
+use hg::dirstate::DirstateV2Data;
 use hg::dirstate::TruncatedTimestamp;
 use std::cell::Cell;
 
 py_class!(pub class DirstateItem |py| {
     data entry: Cell<DirstateEntry>;
 
     def __new__(
         _cls,
         wc_tracked: bool = false,
         p1_tracked: bool = false,
         p2_info: bool = false,
         has_meaningful_data: bool = true,
         has_meaningful_mtime: bool = true,
         parentfiledata: Option<(u32, u32, Option<(u32, u32, bool)>)> = None,
         fallback_exec: Option<bool> = None,
         fallback_symlink: Option<bool> = None,
 
     ) -> PyResult<DirstateItem> {
         let mut mode_size_opt = None;
         let mut mtime_opt = None;
         if let Some((mode, size, mtime)) = parentfiledata {
             if has_meaningful_data {
                 mode_size_opt = Some((mode, size))
             }
             if has_meaningful_mtime {
                 if let Some(m) = mtime {
                     mtime_opt = Some(timestamp(py, m)?);
                 }
             }
         }
-        let entry = DirstateEntry::from_v2_data(
-            wc_tracked,
+        let entry = DirstateEntry::from_v2_data(DirstateV2Data {
+            wc_tracked: wc_tracked,
             p1_tracked,
             p2_info,
-            mode_size_opt,
-            mtime_opt,
+            mode_size: mode_size_opt,
+            mtime: mtime_opt,
             fallback_exec,
             fallback_symlink,
-        );
+        });
         DirstateItem::create_instance(py, Cell::new(entry))
     }
 
     @property
     def state(&self) -> PyResult<PyBytes> {
         let state_byte: u8 = self.entry(py).get().state().into();
         Ok(PyBytes::new(py, &[state_byte]))
     }
 
     @property
     def mode(&self) -> PyResult<i32> {
         Ok(self.entry(py).get().mode())
     }
 
     @property
     def size(&self) -> PyResult<i32> {
         Ok(self.entry(py).get().size())
     }
 
     @property
     def mtime(&self) -> PyResult<i32> {
         Ok(self.entry(py).get().mtime())
     }
 
     @property
     def has_fallback_exec(&self) -> PyResult<bool> {
         match self.entry(py).get().get_fallback_exec() {
             Some(_) => Ok(true),
             None => Ok(false),
         }
     }
 
     @property
     def fallback_exec(&self) -> PyResult<Option<bool>> {
         match self.entry(py).get().get_fallback_exec() {
             Some(exec) => Ok(Some(exec)),
             None => Ok(None),
         }
     }
 
     @fallback_exec.setter
     def set_fallback_exec(&self, value: Option<PyObject>) -> PyResult<()> {
         match value {
             None => {self.entry(py).get().set_fallback_exec(None);},
             Some(value) => {
             if value.is_none(py) {
                 self.entry(py).get().set_fallback_exec(None);
             } else {
                 self.entry(py).get().set_fallback_exec(
                     Some(value.is_true(py)?)
                 );
             }},
         }
         Ok(())
     }
 
     @property
     def has_fallback_symlink(&self) -> PyResult<bool> {
         match self.entry(py).get().get_fallback_symlink() {
             Some(_) => Ok(true),
             None => Ok(false),
         }
     }
 
     @property
     def fallback_symlink(&self) -> PyResult<Option<bool>> {
         match self.entry(py).get().get_fallback_symlink() {
             Some(symlink) => Ok(Some(symlink)),
             None => Ok(None),
         }
     }
 
     @fallback_symlink.setter
     def set_fallback_symlink(&self, value: Option<PyObject>) -> PyResult<()> {
         match value {
             None => {self.entry(py).get().set_fallback_symlink(None);},
             Some(value) => {
             if value.is_none(py) {
                 self.entry(py).get().set_fallback_symlink(None);
             } else {
                 self.entry(py).get().set_fallback_symlink(
                     Some(value.is_true(py)?)
                 );
             }},
         }
         Ok(())
     }
 
     @property
     def tracked(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().tracked())
     }
 
     @property
     def p1_tracked(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().p1_tracked())
     }
 
     @property
     def added(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().added())
     }
 
 
     @property
     def p2_info(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().p2_info())
     }
 
     @property
     def removed(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().removed())
     }
 
     @property
     def maybe_clean(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().maybe_clean())
     }
 
     @property
     def any_tracked(&self) -> PyResult<bool> {
         Ok(self.entry(py).get().any_tracked())
     }
 
     def mtime_likely_equal_to(&self, other: (u32, u32, bool))
         -> PyResult<bool> {
         if let Some(mtime) = self.entry(py).get().truncated_mtime() {
             Ok(mtime.likely_equal(timestamp(py, other)?))
         } else {
             Ok(false)
         }
     }
 
     def drop_merge_data(&self) -> PyResult<PyNone> {
         self.update(py, |entry| entry.drop_merge_data());
         Ok(PyNone)
     }
 
     def set_clean(
         &self,
         mode: u32,
         size: u32,
         mtime: (u32, u32, bool),
     ) -> PyResult<PyNone> {
         let mtime = timestamp(py, mtime)?;
         self.update(py, |entry| entry.set_clean(mode, size, mtime));
         Ok(PyNone)
     }
 
     def set_possibly_dirty(&self) -> PyResult<PyNone> {
         self.update(py, |entry| entry.set_possibly_dirty());
         Ok(PyNone)
     }
 
     def set_tracked(&self) -> PyResult<PyNone> {
         self.update(py, |entry| entry.set_tracked());
         Ok(PyNone)
     }
 
     def set_untracked(&self) -> PyResult<PyNone> {
         self.update(py, |entry| entry.set_untracked());
         Ok(PyNone)
     }
 });
 
 impl DirstateItem {
     pub fn new_as_pyobject(
         py: Python<'_>,
         entry: DirstateEntry,
     ) -> PyResult<PyObject> {
         Ok(DirstateItem::create_instance(py, Cell::new(entry))?.into_object())
     }
 
     pub fn get_entry(&self, py: Python<'_>) -> DirstateEntry {
         self.entry(py).get()
     }
 
     // TODO: Use https://doc.rust-lang.org/std/cell/struct.Cell.html#method.update instead when it’s stable
     pub fn update(&self, py: Python<'_>, f: impl FnOnce(&mut DirstateEntry)) {
         let mut entry = self.entry(py).get();
         f(&mut entry);
         self.entry(py).set(entry)
     }
 }
 
 pub(crate) fn timestamp(
     py: Python<'_>,
     (s, ns, second_ambiguous): (u32, u32, bool),
 ) -> PyResult<TruncatedTimestamp> {
     TruncatedTimestamp::from_already_truncated(s, ns, second_ambiguous)
         .map_err(|_| {
             PyErr::new::<exc::ValueError, _>(
                 py,
                 "expected mtime truncated to 31 bits",
             )
         })
 }