rhg: centralize parsing of `--rev` CLI arguments

This new module will be the place to implement more of the revset language
when we do so.

Differential Revision: https://phab.mercurial-scm.org/D9873

rust/hg-core/src/revset.rs

@@ -0,0 +1,53 b''
+//! The revset query language
+//!
+//! <https://www.mercurial-scm.org/repo/hg/help/revsets>
+
+use crate::repo::Repo;
+use crate::revlog::changelog::Changelog;
+use crate::revlog::revlog::{Revlog, RevlogError};
+use crate::revlog::NodePrefix;
+use crate::revlog::{Revision, NULL_REVISION};
+
+/// Resolve a query string into a single revision.
+///
+/// Only some of the revset language is implemented yet.
+pub fn resolve_single(
+    input: &str,
+    repo: &Repo,
+) -> Result<Revision, RevlogError> {
+    let changelog = Changelog::open(repo)?;
+
+    match resolve_rev_number_or_hex_prefix(input, &changelog.revlog) {
+        Err(RevlogError::InvalidRevision) => {} // Try other syntax
+        result => return result,
+    }
+
+    if input == "null" {
+        return Ok(NULL_REVISION);
+    }
+
+    // TODO: support for the rest of the language here.
+
+    Err(RevlogError::InvalidRevision)
+}
+
+/// Resolve the small subset of the language suitable for revlogs other than
+/// the changelog, such as in `hg debugdata --manifest` CLI argument.
+///
+/// * A non-negative decimal integer for a revision number, or
+/// * An hexadecimal string, for the unique node ID that starts with this
+///   prefix
+pub fn resolve_rev_number_or_hex_prefix(
+    input: &str,
+    revlog: &Revlog,
+) -> Result<Revision, RevlogError> {
+    if let Ok(integer) = input.parse::<i32>() {
+        if integer >= 0 && revlog.has_rev(integer) {
+            return Ok(integer);
+        }
+    }
+    if let Ok(prefix) = NodePrefix::from_hex(input) {
+        return revlog.get_node_rev(prefix);
+    }
+    Err(RevlogError::InvalidRevision)
+}

rust/hg-core/src/lib.rs

@@ -1,188 +1,189 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 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(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

@@ -1,135 +1,125 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::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(
     repo: &Repo,
-    rev: &str,
+    revset: &str,
     files: &[HgPathBuf],
 ) -> Result<Vec<u8>, CatRevError> {
+    let rev = crate::revset::resolve_single(revset, repo)?;
     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)?
-        }
-    };
+    let changelog_entry = changelog.get_rev(rev)?;
     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(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(|_| 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(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/debugdata.rs

@@ -1,102 +1,92 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 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(
     repo: &Repo,
-    rev: &str,
+    revset: &str,
     kind: DebugDataKind,
 ) -> Result<Vec<u8>, DebugDataError> {
     let index_file = match kind {
         DebugDataKind::Changelog => "00changelog.i",
         DebugDataKind::Manifest => "00manifest.i",
     };
     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)?;
-            revlog.get_rev_data(rev)?
-        }
-    };
-
+    let rev =
+        crate::revset::resolve_rev_number_or_hex_prefix(revset, &revlog)?;
+    let data = revlog.get_rev_data(rev)?;
     Ok(data)
 }

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

@@ -1,165 +1,157 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::node::Node;
 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;
 
 /// 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(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(
     repo: &Repo,
-    rev: &str,
+    revset: &str,
 ) -> Result<FilesForRev, ListRevTrackedFilesError> {
+    let rev = crate::revset::resolve_single(revset, repo)?;
     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)?
-        }
-    };
+    let changelog_entry = changelog.get_rev(rev)?;
     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/revlog/changelog.rs

@@ -1,58 +1,58 b''
 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.
-    revlog: Revlog,
+    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)
     }
 }

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

@@ -1,382 +1,387 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_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),
 }
 
 impl From<bytes_cast::FromBytesError> for RevlogError {
     fn from(_: bytes_cast::FromBytesError) -> Self {
         RevlogError::Corrupted
     }
 }
 
 /// 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 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 = repo
                     .store_vfs()
                     .mmap_open(data_path)
                     .map_err(RevlogError::IoError)?;
                 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)?
                 .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)
     }
 
+    /// 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))?;
         }
 
         // 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)
     }
 }