rust: introduce Repo and Vfs types for filesystem abstraction

This is similar to the corresponding Python classes.

Repo represents a repository and knows the path to the `.hg` directory,
the `store` directory, and the working directory.
Separating these will enable supporting the share extension.

A Vfs is created from a Repo for one of these three directories.
It has filesystem access APIs that take a relative std::path::Path
as a parameter.

Differential Revision: https://phab.mercurial-scm.org/D9596

rust/hg-core/src/repo.rs

@@ -0,0 +1,92 b''
+use crate::operations::{find_root, FindRootError};
+use crate::requirements;
+use memmap::{Mmap, MmapOptions};
+use std::path::{Path, PathBuf};
+
+/// A repository on disk
+pub struct Repo {
+    working_directory: PathBuf,
+    dot_hg: PathBuf,
+    store: PathBuf,
+}
+
+/// Filesystem access abstraction for the contents of a given "base" diretory
+#[derive(Clone, Copy)]
+pub(crate) struct Vfs<'a> {
+    base: &'a Path,
+}
+
+impl Repo {
+    /// Returns `None` if the given path doesn’t look like a repository
+    /// (doesn’t contain a `.hg` sub-directory).
+    pub fn for_path(root: impl Into<PathBuf>) -> Self {
+        let working_directory = root.into();
+        let dot_hg = working_directory.join(".hg");
+        Self {
+            store: dot_hg.join("store"),
+            dot_hg,
+            working_directory,
+        }
+    }
+
+    pub fn find() -> Result<Self, FindRootError> {
+        find_root().map(Self::for_path)
+    }
+
+    pub fn check_requirements(
+        &self,
+    ) -> Result<(), requirements::RequirementsError> {
+        requirements::check(self)
+    }
+
+    pub fn working_directory_path(&self) -> &Path {
+        &self.working_directory
+    }
+
+    /// For accessing repository files (in `.hg`), except for the store
+    /// (`.hg/store`).
+    pub(crate) fn hg_vfs(&self) -> Vfs<'_> {
+        Vfs { base: &self.dot_hg }
+    }
+
+    /// For accessing repository store files (in `.hg/store`)
+    pub(crate) fn store_vfs(&self) -> Vfs<'_> {
+        Vfs { base: &self.store }
+    }
+
+    /// For accessing the working copy
+
+    // The undescore prefix silences the "never used" warning. Remove before
+    // using.
+    pub(crate) fn _working_directory_vfs(&self) -> Vfs<'_> {
+        Vfs {
+            base: &self.working_directory,
+        }
+    }
+}
+
+impl Vfs<'_> {
+    pub(crate) fn read(
+        &self,
+        relative_path: impl AsRef<Path>,
+    ) -> std::io::Result<Vec<u8>> {
+        std::fs::read(self.base.join(relative_path))
+    }
+
+    pub(crate) fn open(
+        &self,
+        relative_path: impl AsRef<Path>,
+    ) -> std::io::Result<std::fs::File> {
+        std::fs::File::open(self.base.join(relative_path))
+    }
+
+    pub(crate) fn mmap_open(
+        &self,
+        relative_path: impl AsRef<Path>,
+    ) -> std::io::Result<Mmap> {
+        let file = self.open(relative_path)?;
+        // TODO: what are the safety requirements here?
+        let mmap = unsafe { MmapOptions::new().map(&file) }?;
+        Ok(mmap)
+    }
+}

rust/hg-core/src/lib.rs

@@ -1,194 +1,195 b''
 // Copyright 2018-2020 Georges Racinet <[e-mail]>
 //           and Mercurial contributors
 //
 // This software may be used and distributed according to the terms of the
 // GNU General Public License version 2 or any later version.
 mod ancestors;
 pub mod dagops;
 pub use ancestors::{AncestorsIterator, LazyAncestors, MissingAncestors};
 mod dirstate;
 pub mod discovery;
 pub mod requirements;
 pub mod testing; // unconditionally built, for use from integration tests
 pub use dirstate::{
     dirs_multiset::{DirsMultiset, DirsMultisetIter},
     dirstate_map::DirstateMap,
     parsers::{pack_dirstate, parse_dirstate, PARENT_SIZE},
     status::{
         status, BadMatch, BadType, DirstateStatus, StatusError, StatusOptions,
     },
     CopyMap, CopyMapIter, DirstateEntry, DirstateParents, EntryState,
     StateMap, StateMapIter,
 };
 pub mod copy_tracing;
 mod filepatterns;
 pub mod matchers;
+pub mod repo;
 pub mod revlog;
 pub use revlog::*;
 pub mod operations;
 pub mod utils;
 
 // Remove this to see (potential) non-artificial compile failures. MacOS
 // *should* compile, but fail to compile tests for example as of 2020-03-06
 #[cfg(not(target_os = "linux"))]
 compile_error!(
     "`hg-core` has only been tested on Linux and will most \
      likely not behave correctly on other platforms."
 );
 
 use crate::utils::hg_path::{HgPathBuf, HgPathError};
 pub use filepatterns::{
     parse_pattern_syntax, read_pattern_file, IgnorePattern,
     PatternFileWarning, PatternSyntax,
 };
 use std::collections::HashMap;
 use twox_hash::RandomXxHashBuilder64;
 
 /// This is a contract between the `micro-timer` crate and us, to expose
 /// the `log` crate as `crate::log`.
 use log;
 
 pub type LineNumber = usize;
 
 /// Rust's default hasher is too slow because it tries to prevent collision
 /// attacks. We are not concerned about those: if an ill-minded person has
 /// write access to your repository, you have other issues.
 pub type FastHashMap<K, V> = HashMap<K, V, RandomXxHashBuilder64>;
 
 #[derive(Clone, Debug, PartialEq)]
 pub enum DirstateParseError {
     TooLittleData,
     Overflow,
     // TODO refactor to use bytes instead of String
     CorruptedEntry(String),
     Damaged,
 }
 
 impl From<std::io::Error> for DirstateParseError {
     fn from(e: std::io::Error) -> Self {
         DirstateParseError::CorruptedEntry(e.to_string())
     }
 }
 
 impl ToString for DirstateParseError {
     fn to_string(&self) -> String {
         use crate::DirstateParseError::*;
         match self {
             TooLittleData => "Too little data for dirstate.".to_string(),
             Overflow => "Overflow in dirstate.".to_string(),
             CorruptedEntry(e) => format!("Corrupted entry: {:?}.", e),
             Damaged => "Dirstate appears to be damaged.".to_string(),
         }
     }
 }
 
 #[derive(Debug, PartialEq)]
 pub enum DirstatePackError {
     CorruptedEntry(String),
     CorruptedParent,
     BadSize(usize, usize),
 }
 
 impl From<std::io::Error> for DirstatePackError {
     fn from(e: std::io::Error) -> Self {
         DirstatePackError::CorruptedEntry(e.to_string())
     }
 }
 #[derive(Debug, PartialEq)]
 pub enum DirstateMapError {
     PathNotFound(HgPathBuf),
     EmptyPath,
     InvalidPath(HgPathError),
 }
 
 impl ToString for DirstateMapError {
     fn to_string(&self) -> String {
         match self {
             DirstateMapError::PathNotFound(_) => {
                 "expected a value, found none".to_string()
             }
             DirstateMapError::EmptyPath => "Overflow in dirstate.".to_string(),
             DirstateMapError::InvalidPath(e) => e.to_string(),
         }
     }
 }
 
 #[derive(Debug)]
 pub enum DirstateError {
     Parse(DirstateParseError),
     Pack(DirstatePackError),
     Map(DirstateMapError),
     IO(std::io::Error),
 }
 
 impl From<DirstateParseError> for DirstateError {
     fn from(e: DirstateParseError) -> Self {
         DirstateError::Parse(e)
     }
 }
 
 impl From<DirstatePackError> for DirstateError {
     fn from(e: DirstatePackError) -> Self {
         DirstateError::Pack(e)
     }
 }
 
 #[derive(Debug)]
 pub enum PatternError {
     Path(HgPathError),
     UnsupportedSyntax(String),
     UnsupportedSyntaxInFile(String, String, usize),
     TooLong(usize),
     IO(std::io::Error),
     /// Needed a pattern that can be turned into a regex but got one that
     /// can't. This should only happen through programmer error.
     NonRegexPattern(IgnorePattern),
 }
 
 impl ToString for PatternError {
     fn to_string(&self) -> String {
         match self {
             PatternError::UnsupportedSyntax(syntax) => {
                 format!("Unsupported syntax {}", syntax)
             }
             PatternError::UnsupportedSyntaxInFile(syntax, file_path, line) => {
                 format!(
                     "{}:{}: unsupported syntax {}",
                     file_path, line, syntax
                 )
             }
             PatternError::TooLong(size) => {
                 format!("matcher pattern is too long ({} bytes)", size)
             }
             PatternError::IO(e) => e.to_string(),
             PatternError::Path(e) => e.to_string(),
             PatternError::NonRegexPattern(pattern) => {
                 format!("'{:?}' cannot be turned into a regex", pattern)
             }
         }
     }
 }
 
 impl From<DirstateMapError> for DirstateError {
     fn from(e: DirstateMapError) -> Self {
         DirstateError::Map(e)
     }
 }
 
 impl From<std::io::Error> for DirstateError {
     fn from(e: std::io::Error) -> Self {
         DirstateError::IO(e)
     }
 }
 
 impl From<std::io::Error> for PatternError {
     fn from(e: std::io::Error) -> Self {
         PatternError::IO(e)
     }
 }
 
 impl From<HgPathError> for PatternError {
     fn from(e: HgPathError) -> Self {
         PatternError::Path(e)
     }
 }

rust/hg-core/src/operations/cat.rs

@@ -1,139 +1,135 b''
 // list_tracked_files.rs
 //
 // Copyright 2020 Antoine Cezar <[e-mail]>
 //
 // This software may be used and distributed according to the terms of the
 // GNU General Public License version 2 or any later version.
 
 use std::convert::From;
-use std::path::{Path, PathBuf};
+use std::path::PathBuf;
 
+use crate::repo::Repo;
 use crate::revlog::changelog::Changelog;
 use crate::revlog::manifest::Manifest;
 use crate::revlog::path_encode::path_encode;
 use crate::revlog::revlog::Revlog;
 use crate::revlog::revlog::RevlogError;
 use crate::revlog::Node;
 use crate::revlog::NodePrefix;
 use crate::revlog::Revision;
 use crate::utils::files::get_path_from_bytes;
 use crate::utils::hg_path::{HgPath, HgPathBuf};
 
 const METADATA_DELIMITER: [u8; 2] = [b'\x01', b'\n'];
 
 /// Kind of error encountered by `CatRev`
 #[derive(Debug)]
 pub enum CatRevErrorKind {
     /// Error when reading a `revlog` file.
     IoError(std::io::Error),
     /// The revision has not been found.
     InvalidRevision,
     /// Found more than one revision whose ID match the requested prefix
     AmbiguousPrefix,
     /// A `revlog` file is corrupted.
     CorruptedRevlog,
     /// The `revlog` format version is not supported.
     UnsuportedRevlogVersion(u16),
     /// The `revlog` data format is not supported.
     UnknowRevlogDataFormat(u8),
 }
 
 /// A `CatRev` error
 #[derive(Debug)]
 pub struct CatRevError {
     /// Kind of error encountered by `CatRev`
     pub kind: CatRevErrorKind,
 }
 
 impl From<CatRevErrorKind> for CatRevError {
     fn from(kind: CatRevErrorKind) -> Self {
         CatRevError { kind }
     }
 }
 
 impl From<RevlogError> for CatRevError {
     fn from(err: RevlogError) -> Self {
         match err {
             RevlogError::IoError(err) => CatRevErrorKind::IoError(err),
             RevlogError::UnsuportedVersion(version) => {
                 CatRevErrorKind::UnsuportedRevlogVersion(version)
             }
             RevlogError::InvalidRevision => CatRevErrorKind::InvalidRevision,
             RevlogError::AmbiguousPrefix => CatRevErrorKind::AmbiguousPrefix,
             RevlogError::Corrupted => CatRevErrorKind::CorruptedRevlog,
             RevlogError::UnknowDataFormat(format) => {
                 CatRevErrorKind::UnknowRevlogDataFormat(format)
             }
         }
         .into()
     }
 }
 
 /// List files under Mercurial control at a given revision.
 ///
 /// * `root`: Repository root
 /// * `rev`: The revision to cat the files from.
 /// * `files`: The files to output.
 pub fn cat(
-    root: &Path,
+    repo: &Repo,
     rev: &str,
     files: &[HgPathBuf],
 ) -> Result<Vec<u8>, CatRevError> {
-    let changelog = Changelog::open(&root)?;
-    let manifest = Manifest::open(&root)?;
+    let changelog = Changelog::open(repo)?;
+    let manifest = Manifest::open(repo)?;
 
     let changelog_entry = match rev.parse::<Revision>() {
         Ok(rev) => changelog.get_rev(rev)?,
         _ => {
             let changelog_node = NodePrefix::from_hex(&rev)
                 .map_err(|_| CatRevErrorKind::InvalidRevision)?;
             changelog.get_node(changelog_node.borrow())?
         }
     };
     let manifest_node = Node::from_hex(&changelog_entry.manifest_node()?)
         .map_err(|_| CatRevErrorKind::CorruptedRevlog)?;
 
     let manifest_entry = manifest.get_node((&manifest_node).into())?;
     let mut bytes = vec![];
 
     for (manifest_file, node_bytes) in manifest_entry.files_with_nodes() {
         for cat_file in files.iter() {
             if cat_file.as_bytes() == manifest_file.as_bytes() {
-                let index_path = store_path(root, manifest_file, b".i");
-                let data_path = store_path(root, manifest_file, b".d");
+                let index_path = store_path(manifest_file, b".i");
+                let data_path = store_path(manifest_file, b".d");
 
-                let file_log = Revlog::open(&index_path, Some(&data_path))?;
+                let file_log =
+                    Revlog::open(repo, &index_path, Some(&data_path))?;
                 let file_node = Node::from_hex(node_bytes)
                     .map_err(|_| CatRevErrorKind::CorruptedRevlog)?;
                 let file_rev = file_log.get_node_rev((&file_node).into())?;
                 let data = file_log.get_rev_data(file_rev)?;
                 if data.starts_with(&METADATA_DELIMITER) {
                     let end_delimiter_position = data
                         [METADATA_DELIMITER.len()..]
                         .windows(METADATA_DELIMITER.len())
                         .position(|bytes| bytes == METADATA_DELIMITER);
                     if let Some(position) = end_delimiter_position {
                         let offset = METADATA_DELIMITER.len() * 2;
                         bytes.extend(data[position + offset..].iter());
                     }
                 } else {
                     bytes.extend(data);
                 }
             }
         }
     }
 
     Ok(bytes)
 }
 
-fn store_path(root: &Path, hg_path: &HgPath, suffix: &[u8]) -> PathBuf {
+fn store_path(hg_path: &HgPath, suffix: &[u8]) -> PathBuf {
     let encoded_bytes =
         path_encode(&[b"data/", hg_path.as_bytes(), suffix].concat());
-    [
-        root,
-        &Path::new(".hg/store/"),
-        get_path_from_bytes(&encoded_bytes),
-    ]
-    .iter()
-    .collect()
+    get_path_from_bytes(&encoded_bytes).into()
 }

rust/hg-core/src/operations/debugdata.rs

@@ -1,103 +1,102 b''
 // debugdata.rs
 //
 // Copyright 2020 Antoine Cezar <[e-mail]>
 //
 // This software may be used and distributed according to the terms of the
 // GNU General Public License version 2 or any later version.
 
-use std::path::Path;
-
+use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefix;
 use crate::revlog::Revision;
 
 /// Kind of data to debug
 #[derive(Debug, Copy, Clone)]
 pub enum DebugDataKind {
     Changelog,
     Manifest,
 }
 
 /// Kind of error encountered by DebugData
 #[derive(Debug)]
 pub enum DebugDataErrorKind {
     /// Error when reading a `revlog` file.
     IoError(std::io::Error),
     /// The revision has not been found.
     InvalidRevision,
     /// Found more than one revision whose ID match the requested prefix
     AmbiguousPrefix,
     /// A `revlog` file is corrupted.
     CorruptedRevlog,
     /// The `revlog` format version is not supported.
     UnsuportedRevlogVersion(u16),
     /// The `revlog` data format is not supported.
     UnknowRevlogDataFormat(u8),
 }
 
 /// A DebugData error
 #[derive(Debug)]
 pub struct DebugDataError {
     /// Kind of error encountered by DebugData
     pub kind: DebugDataErrorKind,
 }
 
 impl From<DebugDataErrorKind> for DebugDataError {
     fn from(kind: DebugDataErrorKind) -> Self {
         DebugDataError { kind }
     }
 }
 
 impl From<std::io::Error> for DebugDataError {
     fn from(err: std::io::Error) -> Self {
         let kind = DebugDataErrorKind::IoError(err);
         DebugDataError { kind }
     }
 }
 
 impl From<RevlogError> for DebugDataError {
     fn from(err: RevlogError) -> Self {
         match err {
             RevlogError::IoError(err) => DebugDataErrorKind::IoError(err),
             RevlogError::UnsuportedVersion(version) => {
                 DebugDataErrorKind::UnsuportedRevlogVersion(version)
             }
             RevlogError::InvalidRevision => {
                 DebugDataErrorKind::InvalidRevision
             }
             RevlogError::AmbiguousPrefix => {
                 DebugDataErrorKind::AmbiguousPrefix
             }
             RevlogError::Corrupted => DebugDataErrorKind::CorruptedRevlog,
             RevlogError::UnknowDataFormat(format) => {
                 DebugDataErrorKind::UnknowRevlogDataFormat(format)
             }
         }
         .into()
     }
 }
 
 /// Dump the contents data of a revision.
 pub fn debug_data(
-    root: &Path,
+    repo: &Repo,
     rev: &str,
     kind: DebugDataKind,
 ) -> Result<Vec<u8>, DebugDataError> {
     let index_file = match kind {
-        DebugDataKind::Changelog => root.join(".hg/store/00changelog.i"),
-        DebugDataKind::Manifest => root.join(".hg/store/00manifest.i"),
+        DebugDataKind::Changelog => "00changelog.i",
+        DebugDataKind::Manifest => "00manifest.i",
     };
-    let revlog = Revlog::open(&index_file, None)?;
+    let revlog = Revlog::open(repo, index_file, None)?;
 
     let data = match rev.parse::<Revision>() {
         Ok(rev) => revlog.get_rev_data(rev)?,
         _ => {
             let node = NodePrefix::from_hex(&rev)
                 .map_err(|_| DebugDataErrorKind::InvalidRevision)?;
             let rev = revlog.get_node_rev(node.borrow())?;
             revlog.get_rev_data(rev)?
         }
     };
 
     Ok(data)
 }

rust/hg-core/src/operations/list_tracked_files.rs

@@ -1,167 +1,165 b''
 // list_tracked_files.rs
 //
 // Copyright 2020 Antoine Cezar <[e-mail]>
 //
 // This software may be used and distributed according to the terms of the
 // GNU General Public License version 2 or any later version.
 
 use crate::dirstate::parsers::parse_dirstate;
+use crate::repo::Repo;
 use crate::revlog::changelog::Changelog;
 use crate::revlog::manifest::{Manifest, ManifestEntry};
 use crate::revlog::node::{Node, NodePrefix};
 use crate::revlog::revlog::RevlogError;
 use crate::revlog::Revision;
 use crate::utils::hg_path::HgPath;
 use crate::{DirstateParseError, EntryState};
 use rayon::prelude::*;
 use std::convert::From;
-use std::fs;
-use std::path::Path;
 
 /// Kind of error encountered by `ListDirstateTrackedFiles`
 #[derive(Debug)]
 pub enum ListDirstateTrackedFilesErrorKind {
     /// Error when reading the `dirstate` file
     IoError(std::io::Error),
     /// Error when parsing the `dirstate` file
     ParseError(DirstateParseError),
 }
 
 /// A `ListDirstateTrackedFiles` error
 #[derive(Debug)]
 pub struct ListDirstateTrackedFilesError {
     /// Kind of error encountered by `ListDirstateTrackedFiles`
     pub kind: ListDirstateTrackedFilesErrorKind,
 }
 
 impl From<ListDirstateTrackedFilesErrorKind>
     for ListDirstateTrackedFilesError
 {
     fn from(kind: ListDirstateTrackedFilesErrorKind) -> Self {
         ListDirstateTrackedFilesError { kind }
     }
 }
 
 impl From<std::io::Error> for ListDirstateTrackedFilesError {
     fn from(err: std::io::Error) -> Self {
         let kind = ListDirstateTrackedFilesErrorKind::IoError(err);
         ListDirstateTrackedFilesError { kind }
     }
 }
 
 /// List files under Mercurial control in the working directory
 /// by reading the dirstate
 pub struct Dirstate {
     /// The `dirstate` content.
     content: Vec<u8>,
 }
 
 impl Dirstate {
-    pub fn new(root: &Path) -> Result<Self, ListDirstateTrackedFilesError> {
-        let dirstate = root.join(".hg/dirstate");
-        let content = fs::read(&dirstate)?;
+    pub fn new(repo: &Repo) -> Result<Self, ListDirstateTrackedFilesError> {
+        let content = repo.hg_vfs().read("dirstate")?;
         Ok(Self { content })
     }
 
     pub fn tracked_files(
         &self,
     ) -> Result<Vec<&HgPath>, ListDirstateTrackedFilesError> {
         let (_, entries, _) = parse_dirstate(&self.content)
             .map_err(ListDirstateTrackedFilesErrorKind::ParseError)?;
         let mut files: Vec<&HgPath> = entries
             .into_iter()
             .filter_map(|(path, entry)| match entry.state {
                 EntryState::Removed => None,
                 _ => Some(path),
             })
             .collect();
         files.par_sort_unstable();
         Ok(files)
     }
 }
 
 /// Kind of error encountered by `ListRevTrackedFiles`
 #[derive(Debug)]
 pub enum ListRevTrackedFilesErrorKind {
     /// Error when reading a `revlog` file.
     IoError(std::io::Error),
     /// The revision has not been found.
     InvalidRevision,
     /// Found more than one revision whose ID match the requested prefix
     AmbiguousPrefix,
     /// A `revlog` file is corrupted.
     CorruptedRevlog,
     /// The `revlog` format version is not supported.
     UnsuportedRevlogVersion(u16),
     /// The `revlog` data format is not supported.
     UnknowRevlogDataFormat(u8),
 }
 
 /// A `ListRevTrackedFiles` error
 #[derive(Debug)]
 pub struct ListRevTrackedFilesError {
     /// Kind of error encountered by `ListRevTrackedFiles`
     pub kind: ListRevTrackedFilesErrorKind,
 }
 
 impl From<ListRevTrackedFilesErrorKind> for ListRevTrackedFilesError {
     fn from(kind: ListRevTrackedFilesErrorKind) -> Self {
         ListRevTrackedFilesError { kind }
     }
 }
 
 impl From<RevlogError> for ListRevTrackedFilesError {
     fn from(err: RevlogError) -> Self {
         match err {
             RevlogError::IoError(err) => {
                 ListRevTrackedFilesErrorKind::IoError(err)
             }
             RevlogError::UnsuportedVersion(version) => {
                 ListRevTrackedFilesErrorKind::UnsuportedRevlogVersion(version)
             }
             RevlogError::InvalidRevision => {
                 ListRevTrackedFilesErrorKind::InvalidRevision
             }
             RevlogError::AmbiguousPrefix => {
                 ListRevTrackedFilesErrorKind::AmbiguousPrefix
             }
             RevlogError::Corrupted => {
                 ListRevTrackedFilesErrorKind::CorruptedRevlog
             }
             RevlogError::UnknowDataFormat(format) => {
                 ListRevTrackedFilesErrorKind::UnknowRevlogDataFormat(format)
             }
         }
         .into()
     }
 }
 
 /// List files under Mercurial control at a given revision.
 pub fn list_rev_tracked_files(
-    root: &Path,
+    repo: &Repo,
     rev: &str,
 ) -> Result<FilesForRev, ListRevTrackedFilesError> {
-    let changelog = Changelog::open(root)?;
-    let manifest = Manifest::open(root)?;
+    let changelog = Changelog::open(repo)?;
+    let manifest = Manifest::open(repo)?;
 
     let changelog_entry = match rev.parse::<Revision>() {
         Ok(rev) => changelog.get_rev(rev)?,
         _ => {
             let changelog_node = NodePrefix::from_hex(&rev)
                 .or(Err(ListRevTrackedFilesErrorKind::InvalidRevision))?;
             changelog.get_node(changelog_node.borrow())?
         }
     };
     let manifest_node = Node::from_hex(&changelog_entry.manifest_node()?)
         .or(Err(ListRevTrackedFilesErrorKind::CorruptedRevlog))?;
     let manifest_entry = manifest.get_node((&manifest_node).into())?;
     Ok(FilesForRev(manifest_entry))
 }
 
 pub struct FilesForRev(ManifestEntry);
 
 impl FilesForRev {
     pub fn iter(&self) -> impl Iterator<Item = &HgPath> {
         self.0.files()
     }
 }

rust/hg-core/src/requirements.rs

@@ -1,76 +1,76 b''
+use crate::repo::Repo;
 use std::io;
-use std::path::Path;
 
 #[derive(Debug)]
 pub enum RequirementsError {
     // TODO: include a path?
     Io(io::Error),
     /// The `requires` file is corrupted
     Corrupted,
     /// The repository requires a feature that we don't support
     Unsupported {
         feature: String,
     },
 }
 
 fn parse(bytes: &[u8]) -> Result<Vec<String>, ()> {
     // The Python code reading this file uses `str.splitlines`
     // which looks for a number of line separators (even including a couple of
     // non-ASCII ones), but Python code writing it always uses `\n`.
     let lines = bytes.split(|&byte| byte == b'\n');
 
     lines
         .filter(|line| !line.is_empty())
         .map(|line| {
             // Python uses Unicode `str.isalnum` but feature names are all
             // ASCII
             if line[0].is_ascii_alphanumeric() && line.is_ascii() {
                 Ok(String::from_utf8(line.into()).unwrap())
             } else {
                 Err(())
             }
         })
         .collect()
 }
 
-pub fn load(repo_root: &Path) -> Result<Vec<String>, RequirementsError> {
-    match std::fs::read(repo_root.join(".hg").join("requires")) {
+pub fn load(repo: &Repo) -> Result<Vec<String>, RequirementsError> {
+    match repo.hg_vfs().read("requires") {
         Ok(bytes) => parse(&bytes).map_err(|()| RequirementsError::Corrupted),
 
         // Treat a missing file the same as an empty file.
         // From `mercurial/localrepo.py`:
         // > requires file contains a newline-delimited list of
         // > features/capabilities the opener (us) must have in order to use
         // > the repository. This file was introduced in Mercurial 0.9.2,
         // > which means very old repositories may not have one. We assume
         // > a missing file translates to no requirements.
         Err(error) if error.kind() == std::io::ErrorKind::NotFound => {
             Ok(Vec::new())
         }
 
         Err(error) => Err(RequirementsError::Io(error))?,
     }
 }
 
-pub fn check(repo_root: &Path) -> Result<(), RequirementsError> {
-    for feature in load(repo_root)? {
+pub fn check(repo: &Repo) -> Result<(), RequirementsError> {
+    for feature in load(repo)? {
         if !SUPPORTED.contains(&&*feature) {
             return Err(RequirementsError::Unsupported { feature });
         }
     }
     Ok(())
 }
 
 // TODO: set this to actually-supported features
 const SUPPORTED: &[&str] = &[
     "dotencode",
     "fncache",
     "generaldelta",
     "revlogv1",
     "sparserevlog",
     "store",
     // As of this writing everything rhg does is read-only.
     // When it starts writing to the repository, it’ll need to either keep the
     // persistent nodemap up to date or remove this entry:
     "persistent-nodemap",
 ];

rust/hg-core/src/revlog/changelog.rs

@@ -1,59 +1,58 b''
+use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefixRef;
 use crate::revlog::Revision;
-use std::path::Path;
 
 /// A specialized `Revlog` to work with `changelog` data format.
 pub struct Changelog {
     /// The generic `revlog` format.
     revlog: Revlog,
 }
 
 impl Changelog {
     /// Open the `changelog` of a repository given by its root.
-    pub fn open(root: &Path) -> Result<Self, RevlogError> {
-        let index_file = root.join(".hg/store/00changelog.i");
-        let revlog = Revlog::open(&index_file, None)?;
+    pub fn open(repo: &Repo) -> Result<Self, RevlogError> {
+        let revlog = Revlog::open(repo, "00changelog.i", None)?;
         Ok(Self { revlog })
     }
 
     /// Return the `ChangelogEntry` a given node id.
     pub fn get_node(
         &self,
         node: NodePrefixRef,
     ) -> Result<ChangelogEntry, RevlogError> {
         let rev = self.revlog.get_node_rev(node)?;
         self.get_rev(rev)
     }
 
     /// Return the `ChangelogEntry` of a given node revision.
     pub fn get_rev(
         &self,
         rev: Revision,
     ) -> Result<ChangelogEntry, RevlogError> {
         let bytes = self.revlog.get_rev_data(rev)?;
         Ok(ChangelogEntry { bytes })
     }
 }
 
 /// `Changelog` entry which knows how to interpret the `changelog` data bytes.
 #[derive(Debug)]
 pub struct ChangelogEntry {
     /// The data bytes of the `changelog` entry.
     bytes: Vec<u8>,
 }
 
 impl ChangelogEntry {
     /// Return an iterator over the lines of the entry.
     pub fn lines(&self) -> impl Iterator<Item = &[u8]> {
         self.bytes
             .split(|b| b == &b'\n')
             .filter(|line| !line.is_empty())
     }
 
     /// Return the node id of the `manifest` referenced by this `changelog`
     /// entry.
     pub fn manifest_node(&self) -> Result<&[u8], RevlogError> {
         self.lines().next().ok_or(RevlogError::Corrupted)
     }
 }

rust/hg-core/src/revlog/manifest.rs

@@ -1,77 +1,76 b''
+use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefixRef;
 use crate::revlog::Revision;
 use crate::utils::hg_path::HgPath;
-use std::path::Path;
 
 /// A specialized `Revlog` to work with `manifest` data format.
 pub struct Manifest {
     /// The generic `revlog` format.
     revlog: Revlog,
 }
 
 impl Manifest {
     /// Open the `manifest` of a repository given by its root.
-    pub fn open(root: &Path) -> Result<Self, RevlogError> {
-        let index_file = root.join(".hg/store/00manifest.i");
-        let revlog = Revlog::open(&index_file, None)?;
+    pub fn open(repo: &Repo) -> Result<Self, RevlogError> {
+        let revlog = Revlog::open(repo, "00manifest.i", None)?;
         Ok(Self { revlog })
     }
 
     /// Return the `ManifestEntry` of a given node id.
     pub fn get_node(
         &self,
         node: NodePrefixRef,
     ) -> Result<ManifestEntry, RevlogError> {
         let rev = self.revlog.get_node_rev(node)?;
         self.get_rev(rev)
     }
 
     /// Return the `ManifestEntry` of a given node revision.
     pub fn get_rev(
         &self,
         rev: Revision,
     ) -> Result<ManifestEntry, RevlogError> {
         let bytes = self.revlog.get_rev_data(rev)?;
         Ok(ManifestEntry { bytes })
     }
 }
 
 /// `Manifest` entry which knows how to interpret the `manifest` data bytes.
 #[derive(Debug)]
 pub struct ManifestEntry {
     bytes: Vec<u8>,
 }
 
 impl ManifestEntry {
     /// Return an iterator over the lines of the entry.
     pub fn lines(&self) -> impl Iterator<Item = &[u8]> {
         self.bytes
             .split(|b| b == &b'\n')
             .filter(|line| !line.is_empty())
     }
 
     /// Return an iterator over the files of the entry.
     pub fn files(&self) -> impl Iterator<Item = &HgPath> {
         self.lines().filter(|line| !line.is_empty()).map(|line| {
             let pos = line
                 .iter()
                 .position(|x| x == &b'\0')
                 .expect("manifest line should contain \\0");
             HgPath::new(&line[..pos])
         })
     }
 
     /// Return an iterator over the files of the entry.
     pub fn files_with_nodes(&self) -> impl Iterator<Item = (&HgPath, &[u8])> {
         self.lines().filter(|line| !line.is_empty()).map(|line| {
             let pos = line
                 .iter()
                 .position(|x| x == &b'\0')
                 .expect("manifest line should contain \\0");
             let hash_start = pos + 1;
             let hash_end = hash_start + 40;
             (HgPath::new(&line[..pos]), &line[hash_start..hash_end])
         })
     }
 }

rust/hg-core/src/revlog/nodemap_docket.rs

@@ -1,119 +1,121 b''
 use memmap::Mmap;
 use std::convert::TryInto;
 use std::path::{Path, PathBuf};
 
-use super::revlog::{mmap_open, RevlogError};
+use super::revlog::RevlogError;
+use crate::repo::Repo;
 use crate::utils::strip_suffix;
 
 const ONDISK_VERSION: u8 = 1;
 
 pub(super) struct NodeMapDocket {
     pub data_length: usize,
     // TODO: keep here more of the data from `parse()` when we need it
 }
 
 impl NodeMapDocket {
     /// Return `Ok(None)` when the caller should proceed without a persistent
     /// nodemap:
     ///
     /// * This revlog does not have a `.n` docket file (it is not generated for
     ///   small revlogs), or
     /// * The docket has an unsupported version number (repositories created by
     ///   later hg, maybe that should be a requirement instead?), or
     /// * The docket file points to a missing (likely deleted) data file (this
     ///   can happen in a rare race condition).
     pub fn read_from_file(
+        repo: &Repo,
         index_path: &Path,
     ) -> Result<Option<(Self, Mmap)>, RevlogError> {
         let docket_path = index_path.with_extension("n");
-        let docket_bytes = match std::fs::read(&docket_path) {
+        let docket_bytes = match repo.store_vfs().read(&docket_path) {
             Err(e) if e.kind() == std::io::ErrorKind::NotFound => {
                 return Ok(None)
             }
             Err(e) => return Err(RevlogError::IoError(e)),
             Ok(bytes) => bytes,
         };
 
         let mut input = if let Some((&ONDISK_VERSION, rest)) =
             docket_bytes.split_first()
         {
             rest
         } else {
             return Ok(None);
         };
         let input = &mut input;
 
         let uid_size = read_u8(input)? as usize;
         let _tip_rev = read_be_u64(input)?;
         // TODO: do we care about overflow for 4 GB+ nodemap files on 32-bit
         // systems?
         let data_length = read_be_u64(input)? as usize;
         let _data_unused = read_be_u64(input)?;
         let tip_node_size = read_be_u64(input)? as usize;
         let uid = read_bytes(input, uid_size)?;
         let _tip_node = read_bytes(input, tip_node_size)?;
 
         let uid =
             std::str::from_utf8(uid).map_err(|_| RevlogError::Corrupted)?;
         let docket = NodeMapDocket { data_length };
 
         let data_path = rawdata_path(&docket_path, uid);
         // TODO: use `std::fs::read` here when the `persistent-nodemap.mmap`
         // config is false?
-        match mmap_open(&data_path) {
+        match repo.store_vfs().mmap_open(&data_path) {
             Ok(mmap) => {
                 if mmap.len() >= data_length {
                     Ok(Some((docket, mmap)))
                 } else {
                     Err(RevlogError::Corrupted)
                 }
             }
             Err(error) => {
                 if error.kind() == std::io::ErrorKind::NotFound {
                     Ok(None)
                 } else {
                     Err(RevlogError::IoError(error))
                 }
             }
         }
     }
 }
 
 fn read_bytes<'a>(
     input: &mut &'a [u8],
     count: usize,
 ) -> Result<&'a [u8], RevlogError> {
     if let Some(start) = input.get(..count) {
         *input = &input[count..];
         Ok(start)
     } else {
         Err(RevlogError::Corrupted)
     }
 }
 
 fn read_u8<'a>(input: &mut &[u8]) -> Result<u8, RevlogError> {
     Ok(read_bytes(input, 1)?[0])
 }
 
 fn read_be_u64<'a>(input: &mut &[u8]) -> Result<u64, RevlogError> {
     let array = read_bytes(input, std::mem::size_of::<u64>())?
         .try_into()
         .unwrap();
     Ok(u64::from_be_bytes(array))
 }
 
 fn rawdata_path(docket_path: &Path, uid: &str) -> PathBuf {
     let docket_name = docket_path
         .file_name()
         .expect("expected a base name")
         .to_str()
         .expect("expected an ASCII file name in the store");
     let prefix = strip_suffix(docket_name, ".n.a")
         .or_else(|| strip_suffix(docket_name, ".n"))
         .expect("expected docket path in .n or .n.a");
     let name = format!("{}-{}.nd", prefix, uid);
     docket_path
         .parent()
         .expect("expected a non-root path")
         .join(name)
 }

rust/hg-core/src/revlog/revlog.rs

@@ -1,377 +1,376 b''
 use std::borrow::Cow;
-use std::fs::File;
 use std::io::Read;
 use std::ops::Deref;
 use std::path::Path;
 
 use byteorder::{BigEndian, ByteOrder};
 use crypto::digest::Digest;
 use crypto::sha1::Sha1;
 use flate2::read::ZlibDecoder;
-use memmap::{Mmap, MmapOptions};
 use micro_timer::timed;
 use zstd;
 
 use super::index::Index;
 use super::node::{NodePrefixRef, NODE_BYTES_LENGTH, NULL_NODE};
 use super::nodemap;
 use super::nodemap::NodeMap;
 use super::nodemap_docket::NodeMapDocket;
 use super::patch;
+use crate::repo::Repo;
 use crate::revlog::Revision;
 
 pub enum RevlogError {
     IoError(std::io::Error),
     UnsuportedVersion(u16),
     InvalidRevision,
     /// Found more than one entry whose ID match the requested prefix
     AmbiguousPrefix,
     Corrupted,
     UnknowDataFormat(u8),
 }
 
-pub(super) fn mmap_open(path: &Path) -> Result<Mmap, std::io::Error> {
-    let file = File::open(path)?;
-    let mmap = unsafe { MmapOptions::new().map(&file) }?;
-    Ok(mmap)
-}
-
 /// Read only implementation of revlog.
 pub struct Revlog {
     /// When index and data are not interleaved: bytes of the revlog index.
     /// When index and data are interleaved: bytes of the revlog index and
     /// data.
     index: Index,
     /// When index and data are not interleaved: bytes of the revlog data
     data_bytes: Option<Box<dyn Deref<Target = [u8]> + Send>>,
     /// When present on disk: the persistent nodemap for this revlog
     nodemap: Option<nodemap::NodeTree>,
 }
 
 impl Revlog {
     /// Open a revlog index file.
     ///
     /// It will also open the associated data file if index and data are not
     /// interleaved.
     #[timed]
     pub fn open(
-        index_path: &Path,
+        repo: &Repo,
+        index_path: impl AsRef<Path>,
         data_path: Option<&Path>,
     ) -> Result<Self, RevlogError> {
-        let index_mmap =
-            mmap_open(&index_path).map_err(RevlogError::IoError)?;
+        let index_path = index_path.as_ref();
+        let index_mmap = repo
+            .store_vfs()
+            .mmap_open(&index_path)
+            .map_err(RevlogError::IoError)?;
 
         let version = get_version(&index_mmap);
         if version != 1 {
             return Err(RevlogError::UnsuportedVersion(version));
         }
 
         let index = Index::new(Box::new(index_mmap))?;
 
         let default_data_path = index_path.with_extension("d");
 
         // type annotation required
         // won't recognize Mmap as Deref<Target = [u8]>
         let data_bytes: Option<Box<dyn Deref<Target = [u8]> + Send>> =
             if index.is_inline() {
                 None
             } else {
                 let data_path = data_path.unwrap_or(&default_data_path);
-                let data_mmap =
-                    mmap_open(data_path).map_err(RevlogError::IoError)?;
+                let data_mmap = repo
+                    .store_vfs()
+                    .mmap_open(data_path)
+                    .map_err(RevlogError::IoError)?;
                 Some(Box::new(data_mmap))
             };
 
-        let nodemap = NodeMapDocket::read_from_file(index_path)?.map(
+        let nodemap = NodeMapDocket::read_from_file(repo, index_path)?.map(
             |(docket, data)| {
                 nodemap::NodeTree::load_bytes(
                     Box::new(data),
                     docket.data_length,
                 )
             },
         );
 
         Ok(Revlog {
             index,
             data_bytes,
             nodemap,
         })
     }
 
     /// Return number of entries of the `Revlog`.
     pub fn len(&self) -> usize {
         self.index.len()
     }
 
     /// Returns `true` if the `Revlog` has zero `entries`.
     pub fn is_empty(&self) -> bool {
         self.index.is_empty()
     }
 
     /// Return the full data associated to a node.
     #[timed]
     pub fn get_node_rev(
         &self,
         node: NodePrefixRef,
     ) -> Result<Revision, RevlogError> {
         if let Some(nodemap) = &self.nodemap {
             return nodemap
                 .find_bin(&self.index, node)
                 // TODO: propagate details of this error:
                 .map_err(|_| RevlogError::Corrupted)?
                 .ok_or(RevlogError::InvalidRevision);
         }
 
         // Fallback to linear scan when a persistent nodemap is not present.
         // This happens when the persistent-nodemap experimental feature is not
         // enabled, or for small revlogs.
         //
         // TODO: consider building a non-persistent nodemap in memory to
         // optimize these cases.
         let mut found_by_prefix = None;
         for rev in (0..self.len() as Revision).rev() {
             let index_entry =
                 self.index.get_entry(rev).ok_or(RevlogError::Corrupted)?;
             if node == *index_entry.hash() {
                 return Ok(rev);
             }
             if node.is_prefix_of(index_entry.hash()) {
                 if found_by_prefix.is_some() {
                     return Err(RevlogError::AmbiguousPrefix);
                 }
                 found_by_prefix = Some(rev)
             }
         }
         found_by_prefix.ok_or(RevlogError::InvalidRevision)
     }
 
     /// Return the full data associated to a revision.
     ///
     /// All entries required to build the final data out of deltas will be
     /// retrieved as needed, and the deltas will be applied to the inital
     /// snapshot to rebuild the final data.
     #[timed]
     pub fn get_rev_data(&self, rev: Revision) -> Result<Vec<u8>, RevlogError> {
         // Todo return -> Cow
         let mut entry = self.get_entry(rev)?;
         let mut delta_chain = vec![];
         while let Some(base_rev) = entry.base_rev {
             delta_chain.push(entry);
             entry =
                 self.get_entry(base_rev).or(Err(RevlogError::Corrupted))?;
         }
 
         // TODO do not look twice in the index
         let index_entry = self
             .index
             .get_entry(rev)
             .ok_or(RevlogError::InvalidRevision)?;
 
         let data: Vec<u8> = if delta_chain.is_empty() {
             entry.data()?.into()
         } else {
             Revlog::build_data_from_deltas(entry, &delta_chain)?
         };
 
         if self.check_hash(
             index_entry.p1(),
             index_entry.p2(),
             index_entry.hash().as_bytes(),
             &data,
         ) {
             Ok(data)
         } else {
             Err(RevlogError::Corrupted)
         }
     }
 
     /// Check the hash of some given data against the recorded hash.
     pub fn check_hash(
         &self,
         p1: Revision,
         p2: Revision,
         expected: &[u8],
         data: &[u8],
     ) -> bool {
         let e1 = self.index.get_entry(p1);
         let h1 = match e1 {
             Some(ref entry) => entry.hash(),
             None => &NULL_NODE,
         };
         let e2 = self.index.get_entry(p2);
         let h2 = match e2 {
             Some(ref entry) => entry.hash(),
             None => &NULL_NODE,
         };
 
         hash(data, h1.as_bytes(), h2.as_bytes()).as_slice() == expected
     }
 
     /// Build the full data of a revision out its snapshot
     /// and its deltas.
     #[timed]
     fn build_data_from_deltas(
         snapshot: RevlogEntry,
         deltas: &[RevlogEntry],
     ) -> Result<Vec<u8>, RevlogError> {
         let snapshot = snapshot.data()?;
         let deltas = deltas
             .iter()
             .rev()
             .map(RevlogEntry::data)
             .collect::<Result<Vec<Cow<'_, [u8]>>, RevlogError>>()?;
         let patches: Vec<_> =
             deltas.iter().map(|d| patch::PatchList::new(d)).collect();
         let patch = patch::fold_patch_lists(&patches);
         Ok(patch.apply(&snapshot))
     }
 
     /// Return the revlog data.
     fn data(&self) -> &[u8] {
         match self.data_bytes {
             Some(ref data_bytes) => &data_bytes,
             None => panic!(
                 "forgot to load the data or trying to access inline data"
             ),
         }
     }
 
     /// Get an entry of the revlog.
     fn get_entry(&self, rev: Revision) -> Result<RevlogEntry, RevlogError> {
         let index_entry = self
             .index
             .get_entry(rev)
             .ok_or(RevlogError::InvalidRevision)?;
         let start = index_entry.offset();
         let end = start + index_entry.compressed_len();
         let data = if self.index.is_inline() {
             self.index.data(start, end)
         } else {
             &self.data()[start..end]
         };
         let entry = RevlogEntry {
             rev,
             bytes: data,
             compressed_len: index_entry.compressed_len(),
             uncompressed_len: index_entry.uncompressed_len(),
             base_rev: if index_entry.base_revision() == rev {
                 None
             } else {
                 Some(index_entry.base_revision())
             },
         };
         Ok(entry)
     }
 }
 
 /// The revlog entry's bytes and the necessary informations to extract
 /// the entry's data.
 #[derive(Debug)]
 pub struct RevlogEntry<'a> {
     rev: Revision,
     bytes: &'a [u8],
     compressed_len: usize,
     uncompressed_len: usize,
     base_rev: Option<Revision>,
 }
 
 impl<'a> RevlogEntry<'a> {
     /// Extract the data contained in the entry.
     pub fn data(&self) -> Result<Cow<'_, [u8]>, RevlogError> {
         if self.bytes.is_empty() {
             return Ok(Cow::Borrowed(&[]));
         }
         match self.bytes[0] {
             // Revision data is the entirety of the entry, including this
             // header.
             b'\0' => Ok(Cow::Borrowed(self.bytes)),
             // Raw revision data follows.
             b'u' => Ok(Cow::Borrowed(&self.bytes[1..])),
             // zlib (RFC 1950) data.
             b'x' => Ok(Cow::Owned(self.uncompressed_zlib_data()?)),
             // zstd data.
             b'\x28' => Ok(Cow::Owned(self.uncompressed_zstd_data()?)),
             format_type => Err(RevlogError::UnknowDataFormat(format_type)),
         }
     }
 
     fn uncompressed_zlib_data(&self) -> Result<Vec<u8>, RevlogError> {
         let mut decoder = ZlibDecoder::new(self.bytes);
         if self.is_delta() {
             let mut buf = Vec::with_capacity(self.compressed_len);
             decoder
                 .read_to_end(&mut buf)
                 .or(Err(RevlogError::Corrupted))?;
             Ok(buf)
         } else {
             let mut buf = vec![0; self.uncompressed_len];
             decoder
                 .read_exact(&mut buf)
                 .or(Err(RevlogError::Corrupted))?;
             Ok(buf)
         }
     }
 
     fn uncompressed_zstd_data(&self) -> Result<Vec<u8>, RevlogError> {
         if self.is_delta() {
             let mut buf = Vec::with_capacity(self.compressed_len);
             zstd::stream::copy_decode(self.bytes, &mut buf)
                 .or(Err(RevlogError::Corrupted))?;
             Ok(buf)
         } else {
             let mut buf = vec![0; self.uncompressed_len];
             let len = zstd::block::decompress_to_buffer(self.bytes, &mut buf)
                 .or(Err(RevlogError::Corrupted))?;
             if len != self.uncompressed_len {
                 Err(RevlogError::Corrupted)
             } else {
                 Ok(buf)
             }
         }
     }
 
     /// Tell if the entry is a snapshot or a delta
     /// (influences on decompression).
     fn is_delta(&self) -> bool {
         self.base_rev.is_some()
     }
 }
 
 /// Format version of the revlog.
 pub fn get_version(index_bytes: &[u8]) -> u16 {
     BigEndian::read_u16(&index_bytes[2..=3])
 }
 
 /// Calculate the hash of a revision given its data and its parents.
 fn hash(data: &[u8], p1_hash: &[u8], p2_hash: &[u8]) -> Vec<u8> {
     let mut hasher = Sha1::new();
     let (a, b) = (p1_hash, p2_hash);
     if a > b {
         hasher.input(b);
         hasher.input(a);
     } else {
         hasher.input(a);
         hasher.input(b);
     }
     hasher.input(data);
     let mut hash = vec![0; NODE_BYTES_LENGTH];
     hasher.result(&mut hash);
     hash
 }
 
 #[cfg(test)]
 mod tests {
     use super::*;
 
     use super::super::index::IndexEntryBuilder;
 
     #[test]
     fn version_test() {
         let bytes = IndexEntryBuilder::new()
             .is_first(true)
             .with_version(1)
             .build();
 
         assert_eq!(get_version(&bytes), 1)
     }
 }

rust/rhg/src/commands/cat.rs

@@ -1,106 +1,105 b''
 use crate::commands::Command;
 use crate::error::{CommandError, CommandErrorKind};
 use crate::ui::utf8_to_local;
 use crate::ui::Ui;
-use hg::operations::find_root;
 use hg::operations::{cat, CatRevError, CatRevErrorKind};
-use hg::requirements;
+use hg::repo::Repo;
 use hg::utils::hg_path::HgPathBuf;
 use micro_timer::timed;
 use std::convert::TryFrom;
 
 pub const HELP_TEXT: &str = "
 Output the current or given revision of files
 ";
 
 pub struct CatCommand<'a> {
     rev: Option<&'a str>,
     files: Vec<&'a str>,
 }
 
 impl<'a> CatCommand<'a> {
     pub fn new(rev: Option<&'a str>, files: Vec<&'a str>) -> Self {
         Self { rev, files }
     }
 
     fn display(&self, ui: &Ui, data: &[u8]) -> Result<(), CommandError> {
         ui.write_stdout(data)?;
         Ok(())
     }
 }
 
 impl<'a> Command for CatCommand<'a> {
     #[timed]
     fn run(&self, ui: &Ui) -> Result<(), CommandError> {
-        let root = find_root()?;
-        requirements::check(&root)?;
+        let repo = Repo::find()?;
+        repo.check_requirements()?;
         let cwd = std::env::current_dir()
             .or_else(|e| Err(CommandErrorKind::CurrentDirNotFound(e)))?;
 
         let mut files = vec![];
         for file in self.files.iter() {
             let normalized = cwd.join(&file);
             let stripped = normalized
-                .strip_prefix(&root)
+                .strip_prefix(&repo.working_directory_path())
                 .or(Err(CommandErrorKind::Abort(None)))?;
             let hg_file = HgPathBuf::try_from(stripped.to_path_buf())
                 .or(Err(CommandErrorKind::Abort(None)))?;
             files.push(hg_file);
         }
 
         match self.rev {
             Some(rev) => {
-                let data = cat(&root, rev, &files)
+                let data = cat(&repo, rev, &files)
                     .map_err(|e| map_rev_error(rev, e))?;
                 self.display(ui, &data)
             }
             None => Err(CommandErrorKind::Unimplemented.into()),
         }
     }
 }
 
 /// Convert `CatRevErrorKind` to `CommandError`
 fn map_rev_error(rev: &str, err: CatRevError) -> CommandError {
     CommandError {
         kind: match err.kind {
             CatRevErrorKind::IoError(err) => CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!("abort: {}\n", err)).into(),
             )),
             CatRevErrorKind::InvalidRevision => CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: invalid revision identifier {}\n",
                     rev
                 ))
                 .into(),
             )),
             CatRevErrorKind::AmbiguousPrefix => CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: ambiguous revision identifier {}\n",
                     rev
                 ))
                 .into(),
             )),
             CatRevErrorKind::UnsuportedRevlogVersion(version) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: unsupported revlog version {}\n",
                         version
                     ))
                     .into(),
                 ))
             }
             CatRevErrorKind::CorruptedRevlog => CommandErrorKind::Abort(Some(
                 "abort: corrupted revlog\n".into(),
             )),
             CatRevErrorKind::UnknowRevlogDataFormat(format) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: unknow revlog dataformat {:?}\n",
                         format
                     ))
                     .into(),
                 ))
             }
         },
     }
 }

rust/rhg/src/commands/debugdata.rs

@@ -1,91 +1,91 b''
 use crate::commands::Command;
 use crate::error::{CommandError, CommandErrorKind};
 use crate::ui::utf8_to_local;
 use crate::ui::Ui;
-use hg::operations::find_root;
 use hg::operations::{
     debug_data, DebugDataError, DebugDataErrorKind, DebugDataKind,
 };
+use hg::repo::Repo;
 use micro_timer::timed;
 
 pub const HELP_TEXT: &str = "
 Dump the contents of a data file revision
 ";
 
 pub struct DebugDataCommand<'a> {
     rev: &'a str,
     kind: DebugDataKind,
 }
 
 impl<'a> DebugDataCommand<'a> {
     pub fn new(rev: &'a str, kind: DebugDataKind) -> Self {
         DebugDataCommand { rev, kind }
     }
 }
 
 impl<'a> Command for DebugDataCommand<'a> {
     #[timed]
     fn run(&self, ui: &Ui) -> Result<(), CommandError> {
-        let root = find_root()?;
-        let data = debug_data(&root, self.rev, self.kind)
+        let repo = Repo::find()?;
+        let data = debug_data(&repo, self.rev, self.kind)
             .map_err(|e| to_command_error(self.rev, e))?;
 
         let mut stdout = ui.stdout_buffer();
         stdout.write_all(&data)?;
         stdout.flush()?;
 
         Ok(())
     }
 }
 
 /// Convert operation errors to command errors
 fn to_command_error(rev: &str, err: DebugDataError) -> CommandError {
     match err.kind {
         DebugDataErrorKind::IoError(err) => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!("abort: {}\n", err)).into(),
             )),
         },
         DebugDataErrorKind::InvalidRevision => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: invalid revision identifier{}\n",
                     rev
                 ))
                 .into(),
             )),
         },
         DebugDataErrorKind::AmbiguousPrefix => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: ambiguous revision identifier{}\n",
                     rev
                 ))
                 .into(),
             )),
         },
         DebugDataErrorKind::UnsuportedRevlogVersion(version) => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: unsupported revlog version {}\n",
                     version
                 ))
                 .into(),
             )),
         },
         DebugDataErrorKind::CorruptedRevlog => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 "abort: corrupted revlog\n".into(),
             )),
         },
         DebugDataErrorKind::UnknowRevlogDataFormat(format) => CommandError {
             kind: CommandErrorKind::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: unknow revlog dataformat {:?}\n",
                     format
                 ))
                 .into(),
             )),
         },
     }
 }

rust/rhg/src/commands/debugrequirements.rs

@@ -1,30 +1,30 b''
 use crate::commands::Command;
 use crate::error::CommandError;
 use crate::ui::Ui;
-use hg::operations::find_root;
+use hg::repo::Repo;
 use hg::requirements;
 
 pub const HELP_TEXT: &str = "
 Print the current repo requirements.
 ";
 
 pub struct DebugRequirementsCommand {}
 
 impl DebugRequirementsCommand {
     pub fn new() -> Self {
         DebugRequirementsCommand {}
     }
 }
 
 impl Command for DebugRequirementsCommand {
     fn run(&self, ui: &Ui) -> Result<(), CommandError> {
-        let root = find_root()?;
+        let repo = Repo::find()?;
         let mut output = String::new();
-        for req in requirements::load(&root)? {
+        for req in requirements::load(&repo)? {
             output.push_str(&req);
             output.push('\n');
         }
         ui.write_stdout(output.as_bytes())?;
         Ok(())
     }
 }

rust/rhg/src/commands/files.rs

@@ -1,144 +1,142 b''
 use crate::commands::Command;
 use crate::error::{CommandError, CommandErrorKind};
 use crate::ui::utf8_to_local;
 use crate::ui::Ui;
-use hg::operations::find_root;
 use hg::operations::{
     list_rev_tracked_files, ListRevTrackedFilesError,
     ListRevTrackedFilesErrorKind,
 };
 use hg::operations::{
     Dirstate, ListDirstateTrackedFilesError, ListDirstateTrackedFilesErrorKind,
 };
-use hg::requirements;
+use hg::repo::Repo;
 use hg::utils::files::{get_bytes_from_path, relativize_path};
 use hg::utils::hg_path::{HgPath, HgPathBuf};
-use std::path::Path;
 
 pub const HELP_TEXT: &str = "
 List tracked files.
 
 Returns 0 on success.
 ";
 
 pub struct FilesCommand<'a> {
     rev: Option<&'a str>,
 }
 
 impl<'a> FilesCommand<'a> {
     pub fn new(rev: Option<&'a str>) -> Self {
         FilesCommand { rev }
     }
 
     fn display_files(
         &self,
         ui: &Ui,
-        root: &Path,
+        repo: &Repo,
         files: impl IntoIterator<Item = &'a HgPath>,
     ) -> Result<(), CommandError> {
         let cwd = std::env::current_dir()
             .or_else(|e| Err(CommandErrorKind::CurrentDirNotFound(e)))?;
         let rooted_cwd = cwd
-            .strip_prefix(root)
+            .strip_prefix(repo.working_directory_path())
             .expect("cwd was already checked within the repository");
         let rooted_cwd = HgPathBuf::from(get_bytes_from_path(rooted_cwd));
 
         let mut stdout = ui.stdout_buffer();
 
         for file in files {
             stdout.write_all(relativize_path(file, &rooted_cwd).as_ref())?;
             stdout.write_all(b"\n")?;
         }
         stdout.flush()?;
         Ok(())
     }
 }
 
 impl<'a> Command for FilesCommand<'a> {
     fn run(&self, ui: &Ui) -> Result<(), CommandError> {
-        let root = find_root()?;
-        requirements::check(&root)?;
+        let repo = Repo::find()?;
+        repo.check_requirements()?;
         if let Some(rev) = self.rev {
-            let files = list_rev_tracked_files(&root, rev)
+            let files = list_rev_tracked_files(&repo, rev)
                 .map_err(|e| map_rev_error(rev, e))?;
-            self.display_files(ui, &root, files.iter())
+            self.display_files(ui, &repo, files.iter())
         } else {
-            let distate = Dirstate::new(&root).map_err(map_dirstate_error)?;
+            let distate = Dirstate::new(&repo).map_err(map_dirstate_error)?;
             let files = distate.tracked_files().map_err(map_dirstate_error)?;
-            self.display_files(ui, &root, files)
+            self.display_files(ui, &repo, files)
         }
     }
 }
 
 /// Convert `ListRevTrackedFilesErrorKind` to `CommandError`
 fn map_rev_error(rev: &str, err: ListRevTrackedFilesError) -> CommandError {
     CommandError {
         kind: match err.kind {
             ListRevTrackedFilesErrorKind::IoError(err) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!("abort: {}\n", err)).into(),
                 ))
             }
             ListRevTrackedFilesErrorKind::InvalidRevision => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: invalid revision identifier {}\n",
                         rev
                     ))
                     .into(),
                 ))
             }
             ListRevTrackedFilesErrorKind::AmbiguousPrefix => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: ambiguous revision identifier {}\n",
                         rev
                     ))
                     .into(),
                 ))
             }
             ListRevTrackedFilesErrorKind::UnsuportedRevlogVersion(version) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: unsupported revlog version {}\n",
                         version
                     ))
                     .into(),
                 ))
             }
             ListRevTrackedFilesErrorKind::CorruptedRevlog => {
                 CommandErrorKind::Abort(Some(
                     "abort: corrupted revlog\n".into(),
                 ))
             }
             ListRevTrackedFilesErrorKind::UnknowRevlogDataFormat(format) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!(
                         "abort: unknow revlog dataformat {:?}\n",
                         format
                     ))
                     .into(),
                 ))
             }
         },
     }
 }
 
 /// Convert `ListDirstateTrackedFilesError` to `CommandError`
 fn map_dirstate_error(err: ListDirstateTrackedFilesError) -> CommandError {
     CommandError {
         kind: match err.kind {
             ListDirstateTrackedFilesErrorKind::IoError(err) => {
                 CommandErrorKind::Abort(Some(
                     utf8_to_local(&format!("abort: {}\n", err)).into(),
                 ))
             }
             ListDirstateTrackedFilesErrorKind::ParseError(_) => {
                 CommandErrorKind::Abort(Some(
                     // TODO find a better error message
                     b"abort: parse error\n".to_vec(),
                 ))
             }
         },
     }
 }

rust/rhg/src/commands/root.rs

@@ -1,32 +1,29 b''
 use crate::commands::Command;
 use crate::error::CommandError;
 use crate::ui::Ui;
 use format_bytes::format_bytes;
-use hg::operations::find_root;
+use hg::repo::Repo;
 use hg::utils::files::get_bytes_from_path;
 
 pub const HELP_TEXT: &str = "
 Print the root directory of the current repository.
 
 Returns 0 on success.
 ";
 
 pub struct RootCommand {}
 
 impl RootCommand {
     pub fn new() -> Self {
         RootCommand {}
     }
 }
 
 impl Command for RootCommand {
     fn run(&self, ui: &Ui) -> Result<(), CommandError> {
-        let path_buf = find_root()?;
-
-        let bytes = get_bytes_from_path(path_buf);
-
+        let repo = Repo::find()?;
+        let bytes = get_bytes_from_path(repo.working_directory_path());
         ui.write_stdout(&format_bytes!(b"{}\n", bytes.as_slice()))?;
-
         Ok(())
     }
 }