rust: use HgError in RevlogError and Vfs

Differential Revision: https://phab.mercurial-scm.org/D9897

rust/hg-core/src/errors.rs

@@ -1,111 +1,114 b''
 use std::fmt;
 
 /// Common error cases that can happen in many different APIs
 #[derive(Debug)]
 pub enum HgError {
     IoError {
         error: std::io::Error,
         context: IoErrorContext,
     },
 
     /// A file under `.hg/` normally only written by Mercurial
     ///
     /// The given string is a short explanation for users, not intended to be
     /// machine-readable.
     CorruptedRepository(String),
 
     /// The respository or requested operation involves a feature not
     /// supported by the Rust implementation. Falling back to the Python
     /// implementation may or may not work.
     ///
     /// The given string is a short explanation for users, not intended to be
     /// machine-readable.
     UnsupportedFeature(String),
 }
 
 /// Details about where an I/O error happened
 #[derive(Debug, derive_more::From)]
 pub enum IoErrorContext {
     /// A filesystem operation returned `std::io::Error`
     #[from]
     File(std::path::PathBuf),
     /// `std::env::current_dir` returned `std::io::Error`
     CurrentDir,
 }
 
 impl HgError {
     pub fn corrupted(explanation: impl Into<String>) -> Self {
+        // TODO: capture a backtrace here and keep it in the error value
+        // to aid debugging?
+        // https://doc.rust-lang.org/std/backtrace/struct.Backtrace.html
         HgError::CorruptedRepository(explanation.into())
     }
 }
 
 // TODO: use `DisplayBytes` instead to show non-Unicode filenames losslessly?
 impl fmt::Display for HgError {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
             HgError::IoError { error, context } => {
                 write!(f, "{}: {}", error, context)
             }
             HgError::CorruptedRepository(explanation) => {
                 write!(f, "corrupted repository: {}", explanation)
             }
             HgError::UnsupportedFeature(explanation) => {
                 write!(f, "unsupported feature: {}", explanation)
             }
         }
     }
 }
 
 // TODO: use `DisplayBytes` instead to show non-Unicode filenames losslessly?
 impl fmt::Display for IoErrorContext {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
             IoErrorContext::File(path) => path.display().fmt(f),
             IoErrorContext::CurrentDir => f.write_str("current directory"),
         }
     }
 }
 
 pub trait IoResultExt<T> {
     /// Annotate a possible I/O error as related to a file at the given path.
     ///
     /// This allows printing something like “File not found: example.txt”
     /// instead of just “File not found”.
     ///
     /// Converts a `Result` with `std::io::Error` into one with `HgError`.
     fn for_file(self, path: &std::path::Path) -> Result<T, HgError>;
 }
 
 impl<T> IoResultExt<T> for std::io::Result<T> {
     fn for_file(self, path: &std::path::Path) -> Result<T, HgError> {
         self.map_err(|error| HgError::IoError {
             error,
             context: IoErrorContext::File(path.to_owned()),
         })
     }
 }
 
 pub trait HgResultExt<T> {
     /// Handle missing files separately from other I/O error cases.
     ///
     /// Wraps the `Ok` type in an `Option`:
     ///
     /// * `Ok(x)` becomes `Ok(Some(x))`
     /// * An I/O "not found" error becomes `Ok(None)`
     /// * Other errors are unchanged
     fn io_not_found_as_none(self) -> Result<Option<T>, HgError>;
 }
 
 impl<T> HgResultExt<T> for Result<T, HgError> {
     fn io_not_found_as_none(self) -> Result<Option<T>, HgError> {
         match self {
             Ok(x) => Ok(Some(x)),
             Err(HgError::IoError { error, .. })
                 if error.kind() == std::io::ErrorKind::NotFound =>
             {
                 Ok(None)
             }
             Err(other_error) => Err(other_error),
         }
     }
 }

rust/hg-core/src/lib.rs

@@ -1,115 +1,116 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 mod errors;
 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 config;
 pub mod operations;
 pub mod revset;
 pub mod utils;
 
 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(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, derive_more::From)]
 pub enum DirstateError {
     Map(DirstateMapError),
     Common(errors::HgError),
 }
 
 #[derive(Debug, derive_more::From)]
 pub enum PatternError {
     #[from]
     Path(HgPathError),
     UnsupportedSyntax(String),
     UnsupportedSyntaxInFile(String, String, usize),
     TooLong(usize),
     #[from]
     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)
             }
         }
     }
 }

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

@@ -1,76 +1,75 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::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::utils::files::get_path_from_bytes;
 use crate::utils::hg_path::{HgPath, HgPathBuf};
 
 const METADATA_DELIMITER: [u8; 2] = [b'\x01', b'\n'];
 
 /// 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(
     repo: &Repo,
     revset: &str,
     files: &[HgPathBuf],
 ) -> Result<Vec<u8>, RevlogError> {
     let rev = crate::revset::resolve_single(revset, repo)?;
     let changelog = Changelog::open(repo)?;
     let manifest = Manifest::open(repo)?;
     let changelog_entry = changelog.get_rev(rev)?;
-    let manifest_node = Node::from_hex(&changelog_entry.manifest_node()?)
-        .map_err(|_| RevlogError::Corrupted)?;
+    let manifest_node =
+        Node::from_hex_for_repo(&changelog_entry.manifest_node()?)?;
     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(manifest_file, b".i");
                 let data_path = store_path(manifest_file, b".d");
 
                 let file_log =
                     Revlog::open(repo, &index_path, Some(&data_path))?;
-                let file_node = Node::from_hex(node_bytes)
-                    .map_err(|_| RevlogError::Corrupted)?;
+                let file_node = Node::from_hex_for_repo(node_bytes)?;
                 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(hg_path: &HgPath, suffix: &[u8]) -> PathBuf {
     let encoded_bytes =
         path_encode(&[b"data/", hg_path.as_bytes(), suffix].concat());
     get_path_from_bytes(&encoded_bytes).into()
 }

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

@@ -1,72 +1,67 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::errors::{HgError, IoResultExt};
+use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::changelog::Changelog;
 use crate::revlog::manifest::{Manifest, ManifestEntry};
 use crate::revlog::node::Node;
 use crate::revlog::revlog::RevlogError;
 use crate::utils::hg_path::HgPath;
 use crate::EntryState;
 use rayon::prelude::*;
 
 /// 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(repo: &Repo) -> Result<Self, HgError> {
-        let content = repo
-            .hg_vfs()
-            .read("dirstate")
-            // TODO: this will be more accurate when we use `HgError` in
-            // `Vfs::read`.
-            .for_file("dirstate".as_ref())?;
+        let content = repo.hg_vfs().read("dirstate")?;
         Ok(Self { content })
     }
 
     pub fn tracked_files(&self) -> Result<Vec<&HgPath>, HgError> {
         let (_, entries, _) = parse_dirstate(&self.content)?;
         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)
     }
 }
 
 /// List files under Mercurial control at a given revision.
 pub fn list_rev_tracked_files(
     repo: &Repo,
     revset: &str,
 ) -> Result<FilesForRev, RevlogError> {
     let rev = crate::revset::resolve_single(revset, repo)?;
     let changelog = Changelog::open(repo)?;
     let manifest = Manifest::open(repo)?;
     let changelog_entry = changelog.get_rev(rev)?;
-    let manifest_node = Node::from_hex(&changelog_entry.manifest_node()?)
-        .map_err(|_| RevlogError::Corrupted)?;
+    let manifest_node =
+        Node::from_hex_for_repo(&changelog_entry.manifest_node()?)?;
     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/repo.rs

@@ -1,91 +1,86 b''
-use crate::errors::HgError;
+use crate::errors::{HgError, IoResultExt};
 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<(), HgError> {
         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))
+    ) -> Result<Vec<u8>, HgError> {
+        let path = self.base.join(relative_path);
+        std::fs::read(&path).for_file(&path)
     }
 
     pub(crate) fn mmap_open(
         &self,
         relative_path: impl AsRef<Path>,
-    ) -> std::io::Result<Mmap> {
-        let file = self.open(relative_path)?;
+    ) -> Result<Mmap, HgError> {
+        let path = self.base.join(relative_path);
+        let file = std::fs::File::open(&path).for_file(&path)?;
         // TODO: what are the safety requirements here?
-        let mmap = unsafe { MmapOptions::new().map(&file) }?;
+        let mmap = unsafe { MmapOptions::new().map(&file) }.for_file(&path)?;
         Ok(mmap)
     }
 }

rust/hg-core/src/requirements.rs

@@ -1,70 +1,67 b''
-use crate::errors::{HgError, HgResultExt, IoResultExt};
+use crate::errors::{HgError, HgResultExt};
 use crate::repo::Repo;
 
 fn parse(bytes: &[u8]) -> Result<Vec<String>, HgError> {
     // 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(HgError::corrupted("parse error in 'requires' file"))
             }
         })
         .collect()
 }
 
 pub fn load(repo: &Repo) -> Result<Vec<String>, HgError> {
-    if let Some(bytes) = repo
-        .hg_vfs()
-        .read("requires")
-        .for_file("requires".as_ref())
-        .io_not_found_as_none()?
+    if let Some(bytes) =
+        repo.hg_vfs().read("requires").io_not_found_as_none()?
     {
         parse(&bytes)
     } else {
         // 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.
         Ok(Vec::new())
     }
 }
 
 pub fn check(repo: &Repo) -> Result<(), HgError> {
     for feature in load(repo)? {
         if !SUPPORTED.contains(&&*feature) {
             // TODO: collect and all unknown features and include them in the
             // error message?
             return Err(HgError::UnsupportedFeature(format!(
                 "repository requires feature unknown to this Mercurial: {}",
                 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

@@ -1,58 +1,61 b''
+use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefix;
 use crate::revlog::Revision;
 
 /// A specialized `Revlog` to work with `changelog` data format.
 pub struct Changelog {
     /// The generic `revlog` format.
     pub(crate) revlog: Revlog,
 }
 
 impl Changelog {
     /// Open the `changelog` of a repository given by its root.
     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: NodePrefix,
     ) -> 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)
+        self.lines()
+            .next()
+            .ok_or_else(|| HgError::corrupted("empty changelog entry").into())
     }
 }

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

@@ -1,402 +1,404 b''
 use std::convert::TryInto;
 use std::ops::Deref;
 
 use byteorder::{BigEndian, ByteOrder};
 
+use crate::errors::HgError;
 use crate::revlog::node::Node;
 use crate::revlog::revlog::RevlogError;
 use crate::revlog::{Revision, NULL_REVISION};
 
 pub const INDEX_ENTRY_SIZE: usize = 64;
 
 /// A Revlog index
 pub struct Index {
     bytes: Box<dyn Deref<Target = [u8]> + Send>,
     /// Offsets of starts of index blocks.
     /// Only needed when the index is interleaved with data.
     offsets: Option<Vec<usize>>,
 }
 
 impl Index {
     /// Create an index from bytes.
     /// Calculate the start of each entry when is_inline is true.
     pub fn new(
         bytes: Box<dyn Deref<Target = [u8]> + Send>,
     ) -> Result<Self, RevlogError> {
         if is_inline(&bytes) {
             let mut offset: usize = 0;
             let mut offsets = Vec::new();
 
             while offset + INDEX_ENTRY_SIZE <= bytes.len() {
                 offsets.push(offset);
                 let end = offset + INDEX_ENTRY_SIZE;
                 let entry = IndexEntry {
                     bytes: &bytes[offset..end],
                     offset_override: None,
                 };
 
                 offset += INDEX_ENTRY_SIZE + entry.compressed_len();
             }
 
             if offset == bytes.len() {
                 Ok(Self {
                     bytes,
                     offsets: Some(offsets),
                 })
             } else {
-                Err(RevlogError::Corrupted)
+                Err(HgError::corrupted("unexpected inline revlog length")
+                    .into())
             }
         } else {
             Ok(Self {
                 bytes,
                 offsets: None,
             })
         }
     }
 
     /// Value of the inline flag.
     pub fn is_inline(&self) -> bool {
         is_inline(&self.bytes)
     }
 
     /// Return a slice of bytes if `revlog` is inline. Panic if not.
     pub fn data(&self, start: usize, end: usize) -> &[u8] {
         if !self.is_inline() {
             panic!("tried to access data in the index of a revlog that is not inline");
         }
         &self.bytes[start..end]
     }
 
     /// Return number of entries of the revlog index.
     pub fn len(&self) -> usize {
         if let Some(offsets) = &self.offsets {
             offsets.len()
         } else {
             self.bytes.len() / INDEX_ENTRY_SIZE
         }
     }
 
     /// Returns `true` if the `Index` has zero `entries`.
     pub fn is_empty(&self) -> bool {
         self.len() == 0
     }
 
     /// Return the index entry corresponding to the given revision if it
     /// exists.
     pub fn get_entry(&self, rev: Revision) -> Option<IndexEntry> {
         if rev == NULL_REVISION {
             return None;
         }
         if let Some(offsets) = &self.offsets {
             self.get_entry_inline(rev, offsets)
         } else {
             self.get_entry_separated(rev)
         }
     }
 
     fn get_entry_inline(
         &self,
         rev: Revision,
         offsets: &[usize],
     ) -> Option<IndexEntry> {
         let start = *offsets.get(rev as usize)?;
         let end = start.checked_add(INDEX_ENTRY_SIZE)?;
         let bytes = &self.bytes[start..end];
 
         // See IndexEntry for an explanation of this override.
         let offset_override = Some(end);
 
         Some(IndexEntry {
             bytes,
             offset_override,
         })
     }
 
     fn get_entry_separated(&self, rev: Revision) -> Option<IndexEntry> {
         let max_rev = self.bytes.len() / INDEX_ENTRY_SIZE;
         if rev as usize >= max_rev {
             return None;
         }
         let start = rev as usize * INDEX_ENTRY_SIZE;
         let end = start + INDEX_ENTRY_SIZE;
         let bytes = &self.bytes[start..end];
 
         // Override the offset of the first revision as its bytes are used
         // for the index's metadata (saving space because it is always 0)
         let offset_override = if rev == 0 { Some(0) } else { None };
 
         Some(IndexEntry {
             bytes,
             offset_override,
         })
     }
 }
 
 impl super::RevlogIndex for Index {
     fn len(&self) -> usize {
         self.len()
     }
 
     fn node(&self, rev: Revision) -> Option<&Node> {
         self.get_entry(rev).map(|entry| entry.hash())
     }
 }
 
 #[derive(Debug)]
 pub struct IndexEntry<'a> {
     bytes: &'a [u8],
     /// Allows to override the offset value of the entry.
     ///
     /// For interleaved index and data, the offset stored in the index
     /// corresponds to the separated data offset.
     /// It has to be overridden with the actual offset in the interleaved
     /// index which is just after the index block.
     ///
     /// For separated index and data, the offset stored in the first index
     /// entry is mixed with the index headers.
     /// It has to be overridden with 0.
     offset_override: Option<usize>,
 }
 
 impl<'a> IndexEntry<'a> {
     /// Return the offset of the data.
     pub fn offset(&self) -> usize {
         if let Some(offset_override) = self.offset_override {
             offset_override
         } else {
             let mut bytes = [0; 8];
             bytes[2..8].copy_from_slice(&self.bytes[0..=5]);
             BigEndian::read_u64(&bytes[..]) as usize
         }
     }
 
     /// Return the compressed length of the data.
     pub fn compressed_len(&self) -> usize {
         BigEndian::read_u32(&self.bytes[8..=11]) as usize
     }
 
     /// Return the uncompressed length of the data.
     pub fn uncompressed_len(&self) -> usize {
         BigEndian::read_u32(&self.bytes[12..=15]) as usize
     }
 
     /// Return the revision upon which the data has been derived.
     pub fn base_revision(&self) -> Revision {
         // TODO Maybe return an Option when base_revision == rev?
         //      Requires to add rev to IndexEntry
 
         BigEndian::read_i32(&self.bytes[16..])
     }
 
     pub fn p1(&self) -> Revision {
         BigEndian::read_i32(&self.bytes[24..])
     }
 
     pub fn p2(&self) -> Revision {
         BigEndian::read_i32(&self.bytes[28..])
     }
 
     /// Return the hash of revision's full text.
     ///
     /// Currently, SHA-1 is used and only the first 20 bytes of this field
     /// are used.
     pub fn hash(&self) -> &'a Node {
         (&self.bytes[32..52]).try_into().unwrap()
     }
 }
 
 /// Value of the inline flag.
 pub fn is_inline(index_bytes: &[u8]) -> bool {
     match &index_bytes[0..=1] {
         [0, 0] | [0, 2] => false,
         _ => true,
     }
 }
 
 #[cfg(test)]
 mod tests {
     use super::*;
 
     #[cfg(test)]
     #[derive(Debug, Copy, Clone)]
     pub struct IndexEntryBuilder {
         is_first: bool,
         is_inline: bool,
         is_general_delta: bool,
         version: u16,
         offset: usize,
         compressed_len: usize,
         uncompressed_len: usize,
         base_revision: Revision,
     }
 
     #[cfg(test)]
     impl IndexEntryBuilder {
         pub fn new() -> Self {
             Self {
                 is_first: false,
                 is_inline: false,
                 is_general_delta: true,
                 version: 2,
                 offset: 0,
                 compressed_len: 0,
                 uncompressed_len: 0,
                 base_revision: 0,
             }
         }
 
         pub fn is_first(&mut self, value: bool) -> &mut Self {
             self.is_first = value;
             self
         }
 
         pub fn with_inline(&mut self, value: bool) -> &mut Self {
             self.is_inline = value;
             self
         }
 
         pub fn with_general_delta(&mut self, value: bool) -> &mut Self {
             self.is_general_delta = value;
             self
         }
 
         pub fn with_version(&mut self, value: u16) -> &mut Self {
             self.version = value;
             self
         }
 
         pub fn with_offset(&mut self, value: usize) -> &mut Self {
             self.offset = value;
             self
         }
 
         pub fn with_compressed_len(&mut self, value: usize) -> &mut Self {
             self.compressed_len = value;
             self
         }
 
         pub fn with_uncompressed_len(&mut self, value: usize) -> &mut Self {
             self.uncompressed_len = value;
             self
         }
 
         pub fn with_base_revision(&mut self, value: Revision) -> &mut Self {
             self.base_revision = value;
             self
         }
 
         pub fn build(&self) -> Vec<u8> {
             let mut bytes = Vec::with_capacity(INDEX_ENTRY_SIZE);
             if self.is_first {
                 bytes.extend(&match (self.is_general_delta, self.is_inline) {
                     (false, false) => [0u8, 0],
                     (false, true) => [0u8, 1],
                     (true, false) => [0u8, 2],
                     (true, true) => [0u8, 3],
                 });
                 bytes.extend(&self.version.to_be_bytes());
                 // Remaining offset bytes.
                 bytes.extend(&[0u8; 2]);
             } else {
                 // Offset is only 6 bytes will usize is 8.
                 bytes.extend(&self.offset.to_be_bytes()[2..]);
             }
             bytes.extend(&[0u8; 2]); // Revision flags.
             bytes.extend(&self.compressed_len.to_be_bytes()[4..]);
             bytes.extend(&self.uncompressed_len.to_be_bytes()[4..]);
             bytes.extend(&self.base_revision.to_be_bytes());
             bytes
         }
     }
 
     #[test]
     fn is_not_inline_when_no_inline_flag_test() {
         let bytes = IndexEntryBuilder::new()
             .is_first(true)
             .with_general_delta(false)
             .with_inline(false)
             .build();
 
         assert_eq!(is_inline(&bytes), false)
     }
 
     #[test]
     fn is_inline_when_inline_flag_test() {
         let bytes = IndexEntryBuilder::new()
             .is_first(true)
             .with_general_delta(false)
             .with_inline(true)
             .build();
 
         assert_eq!(is_inline(&bytes), true)
     }
 
     #[test]
     fn is_inline_when_inline_and_generaldelta_flags_test() {
         let bytes = IndexEntryBuilder::new()
             .is_first(true)
             .with_general_delta(true)
             .with_inline(true)
             .build();
 
         assert_eq!(is_inline(&bytes), true)
     }
 
     #[test]
     fn test_offset() {
         let bytes = IndexEntryBuilder::new().with_offset(1).build();
         let entry = IndexEntry {
             bytes: &bytes,
             offset_override: None,
         };
 
         assert_eq!(entry.offset(), 1)
     }
 
     #[test]
     fn test_with_overridden_offset() {
         let bytes = IndexEntryBuilder::new().with_offset(1).build();
         let entry = IndexEntry {
             bytes: &bytes,
             offset_override: Some(2),
         };
 
         assert_eq!(entry.offset(), 2)
     }
 
     #[test]
     fn test_compressed_len() {
         let bytes = IndexEntryBuilder::new().with_compressed_len(1).build();
         let entry = IndexEntry {
             bytes: &bytes,
             offset_override: None,
         };
 
         assert_eq!(entry.compressed_len(), 1)
     }
 
     #[test]
     fn test_uncompressed_len() {
         let bytes = IndexEntryBuilder::new().with_uncompressed_len(1).build();
         let entry = IndexEntry {
             bytes: &bytes,
             offset_override: None,
         };
 
         assert_eq!(entry.uncompressed_len(), 1)
     }
 
     #[test]
     fn test_base_revision() {
         let bytes = IndexEntryBuilder::new().with_base_revision(1).build();
         let entry = IndexEntry {
             bytes: &bytes,
             offset_override: None,
         };
 
         assert_eq!(entry.base_revision(), 1)
     }
 }
 
 #[cfg(test)]
 pub use tests::IndexEntryBuilder;

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

@@ -1,384 +1,398 b''
 // Copyright 2019-2020 Georges Racinet <[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.
 
 //! Definitions and utilities for Revision nodes
 //!
 //! In Mercurial code base, it is customary to call "a node" the binary SHA
 //! of a revision.
 
+use crate::errors::HgError;
 use bytes_cast::BytesCast;
 use std::convert::{TryFrom, TryInto};
 use std::fmt;
 
 /// The length in bytes of a `Node`
 ///
 /// This constant is meant to ease refactors of this module, and
 /// are private so that calling code does not expect all nodes have
 /// the same size, should we support several formats concurrently in
 /// the future.
 pub const NODE_BYTES_LENGTH: usize = 20;
 
 /// Id of the null node.
 ///
 /// Used to indicate the absence of node.
 pub const NULL_NODE_ID: [u8; NODE_BYTES_LENGTH] = [0u8; NODE_BYTES_LENGTH];
 
 /// The length in bytes of a `Node`
 ///
 /// see also `NODES_BYTES_LENGTH` about it being private.
 const NODE_NYBBLES_LENGTH: usize = 2 * NODE_BYTES_LENGTH;
 
 /// Private alias for readability and to ease future change
 type NodeData = [u8; NODE_BYTES_LENGTH];
 
 /// Binary revision SHA
 ///
 /// ## Future changes of hash size
 ///
 /// To accomodate future changes of hash size, Rust callers
 /// should use the conversion methods at the boundaries (FFI, actual
 /// computation of hashes and I/O) only, and only if required.
 ///
 /// All other callers outside of unit tests should just handle `Node` values
 /// and never make any assumption on the actual length, using [`nybbles_len`]
 /// if they need a loop boundary.
 ///
 /// All methods that create a `Node` either take a type that enforces
 /// the size or return an error at runtime.
 ///
 /// [`nybbles_len`]: #method.nybbles_len
 #[derive(Copy, Clone, Debug, PartialEq, BytesCast, derive_more::From)]
 #[repr(transparent)]
 pub struct Node {
     data: NodeData,
 }
 
 /// The node value for NULL_REVISION
 pub const NULL_NODE: Node = Node {
     data: [0; NODE_BYTES_LENGTH],
 };
 
 /// Return an error if the slice has an unexpected length
 impl<'a> TryFrom<&'a [u8]> for &'a Node {
     type Error = ();
 
     #[inline]
     fn try_from(bytes: &'a [u8]) -> Result<Self, Self::Error> {
         match Node::from_bytes(bytes) {
             Ok((node, rest)) if rest.is_empty() => Ok(node),
             _ => Err(()),
         }
     }
 }
 
 /// Return an error if the slice has an unexpected length
 impl TryFrom<&'_ [u8]> for Node {
     type Error = std::array::TryFromSliceError;
 
     #[inline]
     fn try_from(bytes: &'_ [u8]) -> Result<Self, Self::Error> {
         let data = bytes.try_into()?;
         Ok(Self { data })
     }
 }
 
 impl fmt::LowerHex for Node {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         for &byte in &self.data {
             write!(f, "{:02x}", byte)?
         }
         Ok(())
     }
 }
 
 #[derive(Debug)]
 pub struct FromHexError;
 
 /// Low level utility function, also for prefixes
 fn get_nybble(s: &[u8], i: usize) -> u8 {
     if i % 2 == 0 {
         s[i / 2] >> 4
     } else {
         s[i / 2] & 0x0f
     }
 }
 
 impl Node {
     /// Retrieve the `i`th half-byte of the binary data.
     ///
     /// This is also the `i`th hexadecimal digit in numeric form,
     /// also called a [nybble](https://en.wikipedia.org/wiki/Nibble).
     pub fn get_nybble(&self, i: usize) -> u8 {
         get_nybble(&self.data, i)
     }
 
     /// Length of the data, in nybbles
     pub fn nybbles_len(&self) -> usize {
         // public exposure as an instance method only, so that we can
         // easily support several sizes of hashes if needed in the future.
         NODE_NYBBLES_LENGTH
     }
 
     /// Convert from hexadecimal string representation
     ///
     /// Exact length is required.
     ///
     /// To be used in FFI and I/O only, in order to facilitate future
     /// changes of hash format.
     pub fn from_hex(hex: impl AsRef<[u8]>) -> Result<Node, FromHexError> {
         let prefix = NodePrefix::from_hex(hex)?;
         if prefix.nybbles_len() == NODE_NYBBLES_LENGTH {
             Ok(Self { data: prefix.data })
         } else {
             Err(FromHexError)
         }
     }
 
+    /// `from_hex`, but for input from an internal file of the repository such
+    /// as a changelog or manifest entry.
+    ///
+    /// An error is treated as repository corruption.
+    pub fn from_hex_for_repo(hex: impl AsRef<[u8]>) -> Result<Node, HgError> {
+        Self::from_hex(hex.as_ref()).map_err(|FromHexError| {
+            HgError::CorruptedRepository(format!(
+                "Expected a full hexadecimal node ID, found {}",
+                String::from_utf8_lossy(hex.as_ref())
+            ))
+        })
+    }
+
     /// Provide access to binary data
     ///
     /// This is needed by FFI layers, for instance to return expected
     /// binary values to Python.
     pub fn as_bytes(&self) -> &[u8] {
         &self.data
     }
 }
 
 /// The beginning of a binary revision SHA.
 ///
 /// Since it can potentially come from an hexadecimal representation with
 /// odd length, it needs to carry around whether the last 4 bits are relevant
 /// or not.
 #[derive(Debug, PartialEq, Copy, Clone)]
 pub struct NodePrefix {
     /// In `1..=NODE_NYBBLES_LENGTH`
     nybbles_len: u8,
     /// The first `4 * length_in_nybbles` bits are used (considering bits
     /// within a bytes in big-endian: most significant first), the rest
     /// are zero.
     data: NodeData,
 }
 
 impl NodePrefix {
     /// Convert from hexadecimal string representation
     ///
     /// Similarly to `hex::decode`, can be used with Unicode string types
     /// (`String`, `&str`) as well as bytes.
     ///
     /// To be used in FFI and I/O only, in order to facilitate future
     /// changes of hash format.
     pub fn from_hex(hex: impl AsRef<[u8]>) -> Result<Self, FromHexError> {
         let hex = hex.as_ref();
         let len = hex.len();
         if len > NODE_NYBBLES_LENGTH || len == 0 {
             return Err(FromHexError);
         }
 
         let mut data = [0; NODE_BYTES_LENGTH];
         let mut nybbles_len = 0;
         for &ascii_byte in hex {
             let nybble = match char::from(ascii_byte).to_digit(16) {
                 Some(digit) => digit as u8,
                 None => return Err(FromHexError),
             };
             // Fill in the upper half of a byte first, then the lower half.
             let shift = if nybbles_len % 2 == 0 { 4 } else { 0 };
             data[nybbles_len as usize / 2] |= nybble << shift;
             nybbles_len += 1;
         }
         Ok(Self { data, nybbles_len })
     }
 
     pub fn nybbles_len(&self) -> usize {
         self.nybbles_len as _
     }
 
     pub fn is_prefix_of(&self, node: &Node) -> bool {
         let full_bytes = self.nybbles_len() / 2;
         if self.data[..full_bytes] != node.data[..full_bytes] {
             return false;
         }
         if self.nybbles_len() % 2 == 0 {
             return true;
         }
         let last = self.nybbles_len() - 1;
         self.get_nybble(last) == node.get_nybble(last)
     }
 
     /// Retrieve the `i`th half-byte from the prefix.
     ///
     /// This is also the `i`th hexadecimal digit in numeric form,
     /// also called a [nybble](https://en.wikipedia.org/wiki/Nibble).
     pub fn get_nybble(&self, i: usize) -> u8 {
         assert!(i < self.nybbles_len());
         get_nybble(&self.data, i)
     }
 
     fn iter_nybbles(&self) -> impl Iterator<Item = u8> + '_ {
         (0..self.nybbles_len()).map(move |i| get_nybble(&self.data, i))
     }
 
     /// Return the index first nybble that's different from `node`
     ///
     /// If the return value is `None` that means that `self` is
     /// a prefix of `node`, but the current method is a bit slower
     /// than `is_prefix_of`.
     ///
     /// Returned index is as in `get_nybble`, i.e., starting at 0.
     pub fn first_different_nybble(&self, node: &Node) -> Option<usize> {
         self.iter_nybbles()
             .zip(NodePrefix::from(*node).iter_nybbles())
             .position(|(a, b)| a != b)
     }
 }
 
 impl fmt::LowerHex for NodePrefix {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         let full_bytes = self.nybbles_len() / 2;
         for &byte in &self.data[..full_bytes] {
             write!(f, "{:02x}", byte)?
         }
         if self.nybbles_len() % 2 == 1 {
             let last = self.nybbles_len() - 1;
             write!(f, "{:x}", self.get_nybble(last))?
         }
         Ok(())
     }
 }
 
 /// A shortcut for full `Node` references
 impl From<&'_ Node> for NodePrefix {
     fn from(node: &'_ Node) -> Self {
         NodePrefix {
             nybbles_len: node.nybbles_len() as _,
             data: node.data,
         }
     }
 }
 
 /// A shortcut for full `Node` references
 impl From<Node> for NodePrefix {
     fn from(node: Node) -> Self {
         NodePrefix {
             nybbles_len: node.nybbles_len() as _,
             data: node.data,
         }
     }
 }
 
 impl PartialEq<Node> for NodePrefix {
     fn eq(&self, other: &Node) -> bool {
         Self::from(*other) == *self
     }
 }
 
 #[cfg(test)]
 mod tests {
     use super::*;
 
     const SAMPLE_NODE_HEX: &str = "0123456789abcdeffedcba9876543210deadbeef";
     const SAMPLE_NODE: Node = Node {
         data: [
             0x01, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba,
             0x98, 0x76, 0x54, 0x32, 0x10, 0xde, 0xad, 0xbe, 0xef,
         ],
     };
 
     /// Pad an hexadecimal string to reach `NODE_NYBBLES_LENGTH`
     /// The padding is made with zeros.
     pub fn hex_pad_right(hex: &str) -> String {
         let mut res = hex.to_string();
         while res.len() < NODE_NYBBLES_LENGTH {
             res.push('0');
         }
         res
     }
 
     #[test]
     fn test_node_from_hex() {
         let not_hex = "012... oops";
         let too_short = "0123";
         let too_long = format!("{}0", SAMPLE_NODE_HEX);
         assert_eq!(Node::from_hex(SAMPLE_NODE_HEX).unwrap(), SAMPLE_NODE);
         assert!(Node::from_hex(not_hex).is_err());
         assert!(Node::from_hex(too_short).is_err());
         assert!(Node::from_hex(&too_long).is_err());
     }
 
     #[test]
     fn test_node_encode_hex() {
         assert_eq!(format!("{:x}", SAMPLE_NODE), SAMPLE_NODE_HEX);
     }
 
     #[test]
     fn test_prefix_from_to_hex() -> Result<(), FromHexError> {
         assert_eq!(format!("{:x}", NodePrefix::from_hex("0e1")?), "0e1");
         assert_eq!(format!("{:x}", NodePrefix::from_hex("0e1a")?), "0e1a");
         assert_eq!(
             format!("{:x}", NodePrefix::from_hex(SAMPLE_NODE_HEX)?),
             SAMPLE_NODE_HEX
         );
         Ok(())
     }
 
     #[test]
     fn test_prefix_from_hex_errors() {
         assert!(NodePrefix::from_hex("testgr").is_err());
         let mut long = format!("{:x}", NULL_NODE);
         long.push('c');
         assert!(NodePrefix::from_hex(&long).is_err())
     }
 
     #[test]
     fn test_is_prefix_of() -> Result<(), FromHexError> {
         let mut node_data = [0; NODE_BYTES_LENGTH];
         node_data[0] = 0x12;
         node_data[1] = 0xca;
         let node = Node::from(node_data);
         assert!(NodePrefix::from_hex("12")?.is_prefix_of(&node));
         assert!(!NodePrefix::from_hex("1a")?.is_prefix_of(&node));
         assert!(NodePrefix::from_hex("12c")?.is_prefix_of(&node));
         assert!(!NodePrefix::from_hex("12d")?.is_prefix_of(&node));
         Ok(())
     }
 
     #[test]
     fn test_get_nybble() -> Result<(), FromHexError> {
         let prefix = NodePrefix::from_hex("dead6789cafe")?;
         assert_eq!(prefix.get_nybble(0), 13);
         assert_eq!(prefix.get_nybble(7), 9);
         Ok(())
     }
 
     #[test]
     fn test_first_different_nybble_even_prefix() {
         let prefix = NodePrefix::from_hex("12ca").unwrap();
         let mut node = Node::from([0; NODE_BYTES_LENGTH]);
         assert_eq!(prefix.first_different_nybble(&node), Some(0));
         node.data[0] = 0x13;
         assert_eq!(prefix.first_different_nybble(&node), Some(1));
         node.data[0] = 0x12;
         assert_eq!(prefix.first_different_nybble(&node), Some(2));
         node.data[1] = 0xca;
         // now it is a prefix
         assert_eq!(prefix.first_different_nybble(&node), None);
     }
 
     #[test]
     fn test_first_different_nybble_odd_prefix() {
         let prefix = NodePrefix::from_hex("12c").unwrap();
         let mut node = Node::from([0; NODE_BYTES_LENGTH]);
         assert_eq!(prefix.first_different_nybble(&node), Some(0));
         node.data[0] = 0x13;
         assert_eq!(prefix.first_different_nybble(&node), Some(1));
         node.data[0] = 0x12;
         assert_eq!(prefix.first_different_nybble(&node), Some(2));
         node.data[1] = 0xca;
         // now it is a prefix
         assert_eq!(prefix.first_different_nybble(&node), None);
     }
 }
 
 #[cfg(test)]
 pub use tests::hex_pad_right;

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

@@ -1,105 +1,110 b''
+use crate::errors::{HgError, HgResultExt};
 use bytes_cast::{unaligned, BytesCast};
 use memmap::Mmap;
 use std::path::{Path, PathBuf};
 
 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
 }
 
 #[derive(BytesCast)]
 #[repr(C)]
 struct DocketHeader {
     uid_size: u8,
     _tip_rev: unaligned::U64Be,
     data_length: unaligned::U64Be,
     _data_unused: unaligned::U64Be,
     tip_node_size: unaligned::U64Be,
 }
 
 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 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 docket_bytes = if let Some(bytes) =
+            repo.store_vfs().read(&docket_path).io_not_found_as_none()?
+        {
+            bytes
+        } else {
+            return Ok(None);
         };
 
         let input = if let Some((&ONDISK_VERSION, rest)) =
             docket_bytes.split_first()
         {
             rest
         } else {
             return Ok(None);
         };
 
-        let (header, rest) = DocketHeader::from_bytes(input)?;
+        /// Treat any error as a parse error
+        fn parse<T, E>(result: Result<T, E>) -> Result<T, RevlogError> {
+            result.map_err(|_| {
+                HgError::corrupted("nodemap docket parse error").into()
+            })
+        }
+
+        let (header, rest) = parse(DocketHeader::from_bytes(input))?;
         let uid_size = header.uid_size as usize;
         // TODO: do we care about overflow for 4 GB+ nodemap files on 32-bit
         // systems?
         let tip_node_size = header.tip_node_size.get() as usize;
         let data_length = header.data_length.get() as usize;
-        let (uid, rest) = u8::slice_from_bytes(rest, uid_size)?;
-        let (_tip_node, _rest) = u8::slice_from_bytes(rest, tip_node_size)?;
-        let uid =
-            std::str::from_utf8(uid).map_err(|_| RevlogError::Corrupted)?;
+        let (uid, rest) = parse(u8::slice_from_bytes(rest, uid_size))?;
+        let (_tip_node, _rest) =
+            parse(u8::slice_from_bytes(rest, tip_node_size))?;
+        let uid = parse(std::str::from_utf8(uid))?;
         let docket = NodeMapDocket { data_length };
 
         let data_path = rawdata_path(&docket_path, uid);
-        // TODO: use `std::fs::read` here when the `persistent-nodemap.mmap`
+        // TODO: use `vfs.read()` here when the `persistent-nodemap.mmap`
         // config is false?
-        match repo.store_vfs().mmap_open(&data_path) {
-            Ok(mmap) => {
-                if mmap.len() >= data_length {
-                    Ok(Some((docket, mmap)))
-                } else {
-                    Err(RevlogError::Corrupted)
-                }
+        if let Some(mmap) = repo
+            .store_vfs()
+            .mmap_open(&data_path)
+            .io_not_found_as_none()?
+        {
+            if mmap.len() >= data_length {
+                Ok(Some((docket, mmap)))
+            } else {
+                Err(HgError::corrupted("persistent nodemap too short").into())
             }
-            Err(error) => {
-                if error.kind() == std::io::ErrorKind::NotFound {
-                    Ok(None)
-                } else {
-                    Err(RevlogError::IoError(error))
-                }
-            }
+        } else {
+            Ok(None)
         }
     }
 }
 
 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

@@ -1,387 +1,393 b''
 use std::borrow::Cow;
 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 micro_timer::timed;
 use zstd;
 
 use super::index::Index;
 use super::node::{NodePrefix, NODE_BYTES_LENGTH, NULL_NODE};
 use super::nodemap;
-use super::nodemap::NodeMap;
+use super::nodemap::{NodeMap, NodeMapError};
 use super::nodemap_docket::NodeMapDocket;
 use super::patch;
+use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::Revision;
 
+#[derive(derive_more::From)]
 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),
+    #[from]
+    Other(HgError),
 }
 
-impl From<bytes_cast::FromBytesError> for RevlogError {
-    fn from(_: bytes_cast::FromBytesError) -> Self {
-        RevlogError::Corrupted
+impl From<NodeMapError> for RevlogError {
+    fn from(error: NodeMapError) -> Self {
+        match error {
+            NodeMapError::MultipleResults => RevlogError::AmbiguousPrefix,
+            NodeMapError::RevisionNotInIndex(_) => RevlogError::corrupted(),
+        }
+    }
+}
+
+impl RevlogError {
+    fn corrupted() -> Self {
+        RevlogError::Other(HgError::corrupted("corrupted revlog"))
     }
 }
 
 /// 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(
         repo: &Repo,
         index_path: impl AsRef<Path>,
         data_path: Option<&Path>,
     ) -> Result<Self, RevlogError> {
         let index_path = index_path.as_ref();
-        let index_mmap = repo
-            .store_vfs()
-            .mmap_open(&index_path)
-            .map_err(RevlogError::IoError)?;
+        let index_mmap = repo.store_vfs().mmap_open(&index_path)?;
 
         let version = get_version(&index_mmap);
         if version != 1 {
-            return Err(RevlogError::UnsuportedVersion(version));
+            // A proper new version should have had a repo/store requirement.
+            return Err(RevlogError::corrupted());
         }
 
         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 = repo
-                    .store_vfs()
-                    .mmap_open(data_path)
-                    .map_err(RevlogError::IoError)?;
+                let data_mmap = repo.store_vfs().mmap_open(data_path)?;
                 Some(Box::new(data_mmap))
             };
 
         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: NodePrefix,
     ) -> 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)?
+                .find_bin(&self.index, node)?
                 .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)?;
+                self.index.get_entry(rev).ok_or(HgError::corrupted(
+                    "revlog references a revision not in the index",
+                ))?;
             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)
     }
 
     /// Returns whether the given revision exists in this revlog.
     pub fn has_rev(&self, rev: Revision) -> bool {
         self.index.get_entry(rev).is_some()
     }
 
     /// 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))?;
+            entry = self
+                .get_entry(base_rev)
+                .map_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)
+            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)),
+            // A proper new format should have had a repo/store requirement.
+            _format_type => Err(RevlogError::corrupted()),
         }
     }
 
     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))?;
+                .map_err(|_| RevlogError::corrupted())?;
             Ok(buf)
         } else {
             let mut buf = vec![0; self.uncompressed_len];
             decoder
                 .read_exact(&mut buf)
-                .or(Err(RevlogError::Corrupted))?;
+                .map_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))?;
+                .map_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))?;
+                .map_err(|_| RevlogError::corrupted())?;
             if len != self.uncompressed_len {
-                Err(RevlogError::Corrupted)
+                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/error.rs

@@ -1,146 +1,123 b''
 use crate::exitcode;
 use crate::ui::utf8_to_local;
 use crate::ui::UiError;
 use format_bytes::format_bytes;
 use hg::errors::HgError;
 use hg::operations::FindRootError;
 use hg::revlog::revlog::RevlogError;
 use hg::utils::files::get_bytes_from_path;
 use std::convert::From;
 use std::path::PathBuf;
 
 /// The kind of command error
 #[derive(Debug, derive_more::From)]
 pub enum CommandError {
     /// The root of the repository cannot be found
     RootNotFound(PathBuf),
     /// The current directory cannot be found
     CurrentDirNotFound(std::io::Error),
     /// The standard output stream cannot be written to
     StdoutError,
     /// The standard error stream cannot be written to
     StderrError,
     /// The command aborted
     Abort(Option<Vec<u8>>),
     /// A mercurial capability as not been implemented.
     Unimplemented,
     /// Common cases
     #[from]
     Other(HgError),
 }
 
 impl CommandError {
     pub fn get_exit_code(&self) -> exitcode::ExitCode {
         match self {
             CommandError::RootNotFound(_) => exitcode::ABORT,
             CommandError::CurrentDirNotFound(_) => exitcode::ABORT,
             CommandError::StdoutError => exitcode::ABORT,
             CommandError::StderrError => exitcode::ABORT,
             CommandError::Abort(_) => exitcode::ABORT,
             CommandError::Unimplemented => exitcode::UNIMPLEMENTED_COMMAND,
             CommandError::Other(HgError::UnsupportedFeature(_)) => {
                 exitcode::UNIMPLEMENTED_COMMAND
             }
             CommandError::Other(_) => exitcode::ABORT,
         }
     }
 
     /// Return the message corresponding to the error if any
     pub fn get_error_message_bytes(&self) -> Option<Vec<u8>> {
         match self {
             CommandError::RootNotFound(path) => {
                 let bytes = get_bytes_from_path(path);
                 Some(format_bytes!(
                     b"abort: no repository found in '{}' (.hg not found)!\n",
                     bytes.as_slice()
                 ))
             }
             CommandError::CurrentDirNotFound(e) => Some(format_bytes!(
                 b"abort: error getting current working directory: {}\n",
                 e.to_string().as_bytes(),
             )),
             CommandError::Abort(message) => message.to_owned(),
 
             CommandError::StdoutError
             | CommandError::StderrError
             | CommandError::Unimplemented
             | CommandError::Other(HgError::UnsupportedFeature(_)) => None,
 
             CommandError::Other(e) => {
                 Some(format_bytes!(b"{}\n", e.to_string().as_bytes()))
             }
         }
     }
 
     /// Exist the process with the corresponding exit code.
     pub fn exit(&self) {
         std::process::exit(self.get_exit_code())
     }
 }
 
 impl From<UiError> for CommandError {
     fn from(error: UiError) -> Self {
         match error {
             UiError::StdoutError(_) => CommandError::StdoutError,
             UiError::StderrError(_) => CommandError::StderrError,
         }
     }
 }
 
 impl From<FindRootError> for CommandError {
     fn from(err: FindRootError) -> Self {
         match err {
             FindRootError::RootNotFound(path) => {
                 CommandError::RootNotFound(path)
             }
             FindRootError::GetCurrentDirError(e) => {
                 CommandError::CurrentDirNotFound(e)
             }
         }
     }
 }
 
 impl From<(RevlogError, &str)> for CommandError {
     fn from((err, rev): (RevlogError, &str)) -> CommandError {
         match err {
-            RevlogError::IoError(err) => CommandError::Abort(Some(
-                utf8_to_local(&format!("abort: {}\n", err)).into(),
-            )),
             RevlogError::InvalidRevision => CommandError::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: invalid revision identifier {}\n",
                     rev
                 ))
                 .into(),
             )),
             RevlogError::AmbiguousPrefix => CommandError::Abort(Some(
                 utf8_to_local(&format!(
                     "abort: ambiguous revision identifier {}\n",
                     rev
                 ))
                 .into(),
             )),
-            RevlogError::UnsuportedVersion(version) => {
-                CommandError::Abort(Some(
-                    utf8_to_local(&format!(
-                        "abort: unsupported revlog version {}\n",
-                        version
-                    ))
-                    .into(),
-                ))
-            }
-            RevlogError::Corrupted => {
-                CommandError::Abort(Some("abort: corrupted revlog\n".into()))
-            }
-            RevlogError::UnknowDataFormat(format) => {
-                CommandError::Abort(Some(
-                    utf8_to_local(&format!(
-                        "abort: unknow revlog dataformat {:?}\n",
-                        format
-                    ))
-                    .into(),
-                ))
-            }
+            RevlogError::Other(err) => CommandError::Other(err),
         }
     }
 }