rust: Rename the `Revlog::get_node_rev` method to `rev_from_node`

This better describes the input and outputs of this method.

Also rewrite the doc-comment, which seemed to have been left from copy-paste
of another method.

Differential Revision: https://phab.mercurial-scm.org/D11415

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

@@ -1,67 +1,67 b''
 use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::Revision;
 use crate::revlog::{Node, NodePrefix};
 
 /// 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, HgError> {
         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)?;
+        let rev = self.revlog.rev_from_node(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 })
     }
 
     pub fn node_from_rev(&self, rev: Revision) -> Option<&Node> {
         self.revlog.node_from_rev(rev)
     }
 }
 
 /// `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<Node, HgError> {
         Node::from_hex_for_repo(
             self.lines()
                 .next()
                 .ok_or_else(|| HgError::corrupted("empty changelog entry"))?,
         )
     }
 }

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

@@ -1,79 +1,79 b''
 use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::path_encode::path_encode;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefix;
 use crate::revlog::Revision;
 use crate::utils::files::get_path_from_bytes;
 use crate::utils::hg_path::HgPath;
 use crate::utils::SliceExt;
 use std::borrow::Cow;
 use std::path::PathBuf;
 
 /// A specialized `Revlog` to work with file data logs.
 pub struct Filelog {
     /// The generic `revlog` format.
     revlog: Revlog,
 }
 
 impl Filelog {
     pub fn open(repo: &Repo, file_path: &HgPath) -> Result<Self, HgError> {
         let index_path = store_path(file_path, b".i");
         let data_path = store_path(file_path, b".d");
         let revlog = Revlog::open(repo, index_path, Some(&data_path))?;
         Ok(Self { revlog })
     }
 
     /// The given node ID is that of the file as found in a manifest, not of a
     /// changeset.
     pub fn get_node(
         &self,
         file_node: impl Into<NodePrefix>,
     ) -> Result<FilelogEntry, RevlogError> {
-        let file_rev = self.revlog.get_node_rev(file_node.into())?;
+        let file_rev = self.revlog.rev_from_node(file_node.into())?;
         self.get_rev(file_rev)
     }
 
     /// The given revision is that of the file as found in a manifest, not of a
     /// changeset.
     pub fn get_rev(
         &self,
         file_rev: Revision,
     ) -> Result<FilelogEntry, RevlogError> {
         let data = self.revlog.get_rev_data(file_rev)?;
         Ok(FilelogEntry(data.into()))
     }
 }
 
 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()
 }
 
 pub struct FilelogEntry<'filelog>(Cow<'filelog, [u8]>);
 
 impl<'filelog> FilelogEntry<'filelog> {
     /// Split into metadata and data
     pub fn split(&self) -> Result<(Option<&[u8]>, &[u8]), HgError> {
         const DELIMITER: &[u8; 2] = &[b'\x01', b'\n'];
 
         if let Some(rest) = self.0.drop_prefix(DELIMITER) {
             if let Some((metadata, data)) = rest.split_2_by_slice(DELIMITER) {
                 Ok((Some(metadata), data))
             } else {
                 Err(HgError::corrupted(
                     "Missing metadata end delimiter in filelog entry",
                 ))
             }
         } else {
             Ok((None, &self.0))
         }
     }
 
     /// Returns the file contents at this revision, stripped of any metadata
     pub fn data(&self) -> Result<&[u8], HgError> {
         let (_metadata, data) = self.split()?;
         Ok(data)
     }
 }

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

@@ -1,84 +1,84 b''
 use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::Revision;
 use crate::revlog::{Node, NodePrefix};
 use crate::utils::hg_path::HgPath;
 
 /// A specialized `Revlog` to work with `manifest` data format.
 pub struct Manifestlog {
     /// The generic `revlog` format.
     revlog: Revlog,
 }
 
 impl Manifestlog {
     /// Open the `manifest` of a repository given by its root.
     pub fn open(repo: &Repo) -> Result<Self, HgError> {
         let revlog = Revlog::open(repo, "00manifest.i", None)?;
         Ok(Self { revlog })
     }
 
     /// Return the `ManifestEntry` of a given node id.
     pub fn get_node(&self, node: NodePrefix) -> Result<Manifest, RevlogError> {
-        let rev = self.revlog.get_node_rev(node)?;
+        let rev = self.revlog.rev_from_node(node)?;
         self.get_rev(rev)
     }
 
     /// Return the `ManifestEntry` of a given node revision.
     pub fn get_rev(&self, rev: Revision) -> Result<Manifest, RevlogError> {
         let bytes = self.revlog.get_rev_data(rev)?;
         Ok(Manifest { bytes })
     }
 }
 
 /// `Manifestlog` entry which knows how to interpret the `manifest` data bytes.
 #[derive(Debug)]
 pub struct Manifest {
     bytes: Vec<u8>,
 }
 
 impl Manifest {
     /// Return an iterator over the lines of the entry.
     pub fn lines(&self) -> impl Iterator<Item = &[u8]> {
         self.bytes
             .split(|b| b == &b'\n')
             .filter(|line| !line.is_empty())
     }
 
     /// Return an iterator over the files of the entry.
     pub fn files(&self) -> impl Iterator<Item = &HgPath> {
         self.lines().filter(|line| !line.is_empty()).map(|line| {
             let pos = line
                 .iter()
                 .position(|x| x == &b'\0')
                 .expect("manifest line should contain \\0");
             HgPath::new(&line[..pos])
         })
     }
 
     /// Return an iterator over the files of the entry.
     pub fn files_with_nodes(&self) -> impl Iterator<Item = (&HgPath, &[u8])> {
         self.lines().filter(|line| !line.is_empty()).map(|line| {
             let pos = line
                 .iter()
                 .position(|x| x == &b'\0')
                 .expect("manifest line should contain \\0");
             let hash_start = pos + 1;
             let hash_end = hash_start + 40;
             (HgPath::new(&line[..pos]), &line[hash_start..hash_end])
         })
     }
 
     /// If the given path is in this manifest, return its filelog node ID
     pub fn find_file(&self, path: &HgPath) -> Result<Option<Node>, HgError> {
         // TODO: use binary search instead of linear scan. This may involve
         // building (and caching) an index of the byte indicex of each manifest
         // line.
         for (manifest_path, node) in self.files_with_nodes() {
             if manifest_path == path {
                 return Ok(Some(Node::from_hex_for_repo(node)?));
             }
         }
         Ok(None)
     }
 }

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

@@ -1,406 +1,406 b''
 use std::borrow::Cow;
 use std::io::Read;
 use std::ops::Deref;
 use std::path::Path;
 
 use byteorder::{BigEndian, ByteOrder};
 use flate2::read::ZlibDecoder;
 use micro_timer::timed;
 use sha1::{Digest, Sha1};
 use zstd;
 
 use super::index::Index;
 use super::node::{NodePrefix, NODE_BYTES_LENGTH, NULL_NODE};
 use super::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;
 use crate::{Node, NULL_REVISION};
 
 #[derive(derive_more::From)]
 pub enum RevlogError {
     InvalidRevision,
     /// Working directory is not supported
     WDirUnsupported,
     /// Found more than one entry whose ID match the requested prefix
     AmbiguousPrefix,
     #[from]
     Other(HgError),
 }
 
 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, HgError> {
         let index_path = index_path.as_ref();
         let index_mmap = repo.store_vfs().mmap_open(&index_path)?;
 
         let version = get_version(&index_mmap);
         if version != 1 {
             // A proper new version should have had a repo/store requirement.
             return Err(HgError::corrupted("corrupted revlog"));
         }
 
         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)?;
                 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()
     }
 
     /// Returns the node ID for the given revision number, if it exists in this revlog
     pub fn node_from_rev(&self, rev: Revision) -> Option<&Node> {
         Some(self.index.get_entry(rev)?.hash())
     }
 
-    /// Return the full data associated to a node.
+    /// Return the revision number for the given node ID, if it exists in this revlog
     #[timed]
-    pub fn get_node_rev(
+    pub fn rev_from_node(
         &self,
         node: NodePrefix,
     ) -> Result<Revision, RevlogError> {
         if node.is_prefix_of(&NULL_NODE) {
             return Ok(NULL_REVISION);
         }
 
         if let Some(nodemap) = &self.nodemap {
             return nodemap
                 .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(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)
                 .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())
         }
     }
 
     /// 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()) == 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()?)),
             // 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)
                 .map_err(|_| RevlogError::corrupted())?;
             Ok(buf)
         } else {
             let mut buf = vec![0; self.uncompressed_len];
             decoder
                 .read_exact(&mut buf)
                 .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)
                 .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)
                 .map_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],
 ) -> [u8; NODE_BYTES_LENGTH] {
     let mut hasher = Sha1::new();
     let (a, b) = (p1_hash, p2_hash);
     if a > b {
         hasher.update(b);
         hasher.update(a);
     } else {
         hasher.update(a);
         hasher.update(b);
     }
     hasher.update(data);
     *hasher.finalize().as_ref()
 }
 
 #[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/hg-core/src/revset.rs

@@ -1,67 +1,67 b''
 //! The revset query language
 //!
 //! <https://www.mercurial-scm.org/repo/hg/help/revsets>
 
 use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::revlog::{Revlog, RevlogError};
 use crate::revlog::NodePrefix;
 use crate::revlog::{Revision, NULL_REVISION, WORKING_DIRECTORY_HEX};
 use crate::Node;
 
 /// 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 = repo.changelog()?;
 
     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(
         HgError::unsupported(format!("cannot parse revset '{}'", input))
             .into(),
     )
 }
 
 /// 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> {
     // The Python equivalent of this is part of `revsymbol` in
     // `mercurial/scmutil.py`
 
     if let Ok(integer) = input.parse::<i32>() {
         if integer.to_string() == input
             && integer >= 0
             && revlog.has_rev(integer)
         {
             return Ok(integer);
         }
     }
     if let Ok(prefix) = NodePrefix::from_hex(input) {
         if prefix.is_prefix_of(&Node::from_hex(WORKING_DIRECTORY_HEX).unwrap())
         {
             return Err(RevlogError::WDirUnsupported);
         }
-        return revlog.get_node_rev(prefix);
+        return revlog.rev_from_node(prefix);
     }
     Err(RevlogError::InvalidRevision)
 }