rhg: Expose FilelogEntry that wraps RevlogEntry

This can be later extended to access metadata such as `uncompressed_len` without
necessarily resolving deltas.

Differential Revision: https://phab.mercurial-scm.org/D11961

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

@@ -1,89 +1,117
 use crate::errors::HgError;
 use crate::repo::Repo;
 use crate::revlog::path_encode::path_encode;
+use crate::revlog::revlog::RevlogEntry;
 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::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
+    /// The given node ID is that of the file as found in a filelog, not of a
     /// changeset.
     pub fn data_for_node(
         &self,
         file_node: impl Into<NodePrefix>,
     ) -> Result<FilelogRevisionData, RevlogError> {
         let file_rev = self.revlog.rev_from_node(file_node.into())?;
         self.data_for_rev(file_rev)
     }
 
-    /// The given revision is that of the file as found in a manifest, not of a
+    /// The given revision is that of the file as found in a filelog, not of a
     /// changeset.
     pub fn data_for_rev(
         &self,
         file_rev: Revision,
     ) -> Result<FilelogRevisionData, RevlogError> {
         let data: Vec<u8> = self.revlog.get_rev_data(file_rev)?.into_owned();
         Ok(FilelogRevisionData(data.into()))
     }
+
+    /// The given node ID is that of the file as found in a filelog, not of a
+    /// changeset.
+    pub fn entry_for_node(
+        &self,
+        file_node: impl Into<NodePrefix>,
+    ) -> Result<FilelogEntry, RevlogError> {
+        let file_rev = self.revlog.rev_from_node(file_node.into())?;
+        self.entry_for_rev(file_rev)
+    }
+
+    /// The given revision is that of the file as found in a filelog, not of a
+    /// changeset.
+    pub fn entry_for_rev(
+        &self,
+        file_rev: Revision,
+    ) -> Result<FilelogEntry, RevlogError> {
+        Ok(FilelogEntry(self.revlog.get_entry(file_rev)?))
+    }
 }
 
 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<'a>(RevlogEntry<'a>);
+
+impl FilelogEntry<'_> {
+    pub fn data(&self) -> Result<FilelogRevisionData, HgError> {
+        Ok(FilelogRevisionData(self.0.data()?.into_owned()))
+    }
+}
+
 /// The data for one revision in a filelog, uncompressed and delta-resolved.
 pub struct FilelogRevisionData(Vec<u8>);
 
 impl FilelogRevisionData {
     /// 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 file_data(&self) -> Result<&[u8], HgError> {
         let (_metadata, data) = self.split()?;
         Ok(data)
     }
 
     /// Consume the entry, and convert it into data, discarding any metadata,
     /// if present.
     pub fn into_file_data(self) -> Result<Vec<u8>, HgError> {
         if let (Some(_metadata), data) = self.split()? {
             Ok(data.to_owned())
         } else {
             Ok(self.0)
         }
     }
 }

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

@@ -1,431 +1,434
 use std::borrow::Cow;
 use std::io::Read;
 use std::ops::Deref;
 use std::path::Path;
 
 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(),
         }
     }
 }
 
+fn corrupted() -> HgError {
+    HgError::corrupted("corrupted revlog")
+}
+
 impl RevlogError {
     fn corrupted() -> Self {
-        RevlogError::Other(HgError::corrupted("corrupted revlog"))
+        RevlogError::Other(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, HgError> {
         let index_path = index_path.as_ref();
         let index = {
             match repo.store_vfs().mmap_open_opt(&index_path)? {
                 None => Index::new(Box::new(vec![])),
                 Some(index_mmap) => {
                     let index = Index::new(Box::new(index_mmap))?;
                     Ok(index)
                 }
             }
         }?;
 
         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 = if index.is_inline() {
             None
         } else {
             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> {
         if rev == NULL_REVISION {
             return Some(&NULL_NODE);
         }
         Some(self.index.get_entry(rev)?.hash())
     }
 
     /// Return the revision number for the given node ID, if it exists in this
     /// revlog
     #[timed]
     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<Cow<[u8]>, RevlogError> {
         if rev == NULL_REVISION {
             return Ok(Cow::Borrowed(&[]));
         };
-        self.get_entry(rev)?.data()
+        Ok(self.get_entry(rev)?.data()?)
     }
 
     /// 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> {
+    ) -> Result<Vec<u8>, HgError> {
         let snapshot = snapshot.data_chunk()?;
         let deltas = deltas
             .iter()
             .rev()
             .map(RevlogEntry::data_chunk)
-            .collect::<Result<Vec<Cow<'_, [u8]>>, RevlogError>>()?;
+            .collect::<Result<Vec<_>, _>>()?;
         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> {
+    pub 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 {
             revlog: self,
             rev,
             bytes: data,
             compressed_len: index_entry.compressed_len(),
             uncompressed_len: index_entry.uncompressed_len(),
             base_rev_or_base_of_delta_chain: if index_entry
                 .base_revision_or_base_of_delta_chain()
                 == rev
             {
                 None
             } else {
                 Some(index_entry.base_revision_or_base_of_delta_chain())
             },
         };
         Ok(entry)
     }
 
     /// when resolving internal references within revlog, any errors
     /// should be reported as corruption, instead of e.g. "invalid revision"
     fn get_entry_internal(
         &self,
         rev: Revision,
-    ) -> Result<RevlogEntry, RevlogError> {
-        return self.get_entry(rev).map_err(|_| RevlogError::corrupted());
+    ) -> Result<RevlogEntry, HgError> {
+        return self.get_entry(rev).map_err(|_| corrupted());
     }
 }
 
 /// The revlog entry's bytes and the necessary informations to extract
 /// the entry's data.
 #[derive(Clone)]
 pub struct RevlogEntry<'a> {
     revlog: &'a Revlog,
     rev: Revision,
     bytes: &'a [u8],
     compressed_len: usize,
     uncompressed_len: usize,
     base_rev_or_base_of_delta_chain: Option<Revision>,
 }
 
 impl<'a> RevlogEntry<'a> {
     pub fn revision(&self) -> Revision {
         self.rev
     }
 
     /// The data for this entry, after resolving deltas if any.
-    pub fn data(&self) -> Result<Cow<'a, [u8]>, RevlogError> {
+    pub fn data(&self) -> Result<Cow<'a, [u8]>, HgError> {
         let mut entry = self.clone();
         let mut delta_chain = vec![];
 
         // The meaning of `base_rev_or_base_of_delta_chain` depends on
         // generaldelta. See the doc on `ENTRY_DELTA_BASE` in
         // `mercurial/revlogutils/constants.py` and the code in
         // [_chaininfo] and in [index_deltachain].
         let uses_generaldelta = self.revlog.index.uses_generaldelta();
         while let Some(base_rev) = entry.base_rev_or_base_of_delta_chain {
             let base_rev = if uses_generaldelta {
                 base_rev
             } else {
                 entry.rev - 1
             };
             delta_chain.push(entry);
             entry = self.revlog.get_entry_internal(base_rev)?;
         }
 
         // TODO do not look twice in the index
         let index_entry = self
             .revlog
             .index
             .get_entry(self.rev)
-            .ok_or(RevlogError::InvalidRevision)?;
+            .ok_or_else(corrupted)?;
 
         let data = if delta_chain.is_empty() {
             entry.data_chunk()?
         } else {
             Revlog::build_data_from_deltas(entry, &delta_chain)?.into()
         };
 
         if self.revlog.check_hash(
             index_entry.p1(),
             index_entry.p2(),
             index_entry.hash().as_bytes(),
             &data,
         ) {
             Ok(data)
         } else {
-            Err(RevlogError::corrupted())
+            Err(corrupted())
         }
     }
 
     /// Extract the data contained in the entry.
     /// This may be a delta. (See `is_delta`.)
-    fn data_chunk(&self) -> Result<Cow<'a, [u8]>, RevlogError> {
+    fn data_chunk(&self) -> Result<Cow<'a, [u8]>, HgError> {
         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()),
+            _format_type => Err(corrupted()),
         }
     }
 
-    fn uncompressed_zlib_data(&self) -> Result<Vec<u8>, RevlogError> {
+    fn uncompressed_zlib_data(&self) -> Result<Vec<u8>, HgError> {
         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())?;
+            decoder.read_to_end(&mut buf).map_err(|_| corrupted())?;
             Ok(buf)
         } else {
             let mut buf = vec![0; self.uncompressed_len];
-            decoder
-                .read_exact(&mut buf)
-                .map_err(|_| RevlogError::corrupted())?;
+            decoder.read_exact(&mut buf).map_err(|_| corrupted())?;
             Ok(buf)
         }
     }
 
-    fn uncompressed_zstd_data(&self) -> Result<Vec<u8>, RevlogError> {
+    fn uncompressed_zstd_data(&self) -> Result<Vec<u8>, HgError> {
         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())?;
+                .map_err(|_| 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())?;
+                .map_err(|_| corrupted())?;
             if len != self.uncompressed_len {
-                Err(RevlogError::corrupted())
+                Err(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_or_base_of_delta_chain.is_some()
     }
 }
 
 /// 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()
 }

rust/rhg/src/commands/status.rs

@@ -1,538 +1,539
 // status.rs
 //
 // Copyright 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.
 
 use crate::error::CommandError;
 use crate::ui::Ui;
 use crate::utils::path_utils::RelativizePaths;
 use clap::{Arg, SubCommand};
 use format_bytes::format_bytes;
 use hg;
 use hg::config::Config;
 use hg::dirstate::has_exec_bit;
 use hg::dirstate::status::StatusPath;
 use hg::dirstate::TruncatedTimestamp;
 use hg::dirstate::RANGE_MASK_31BIT;
 use hg::errors::{HgError, IoResultExt};
 use hg::lock::LockError;
 use hg::manifest::Manifest;
 use hg::matchers::AlwaysMatcher;
 use hg::repo::Repo;
 use hg::utils::files::get_bytes_from_os_string;
 use hg::utils::files::get_bytes_from_path;
 use hg::utils::files::get_path_from_bytes;
 use hg::utils::hg_path::{hg_path_to_path_buf, HgPath};
 use hg::StatusOptions;
 use log::info;
 use std::io;
 use std::path::PathBuf;
 
 pub const HELP_TEXT: &str = "
 Show changed files in the working directory
 
 This is a pure Rust version of `hg status`.
 
 Some options might be missing, check the list below.
 ";
 
 pub fn args() -> clap::App<'static, 'static> {
     SubCommand::with_name("status")
         .alias("st")
         .about(HELP_TEXT)
         .arg(
             Arg::with_name("all")
                 .help("show status of all files")
                 .short("-A")
                 .long("--all"),
         )
         .arg(
             Arg::with_name("modified")
                 .help("show only modified files")
                 .short("-m")
                 .long("--modified"),
         )
         .arg(
             Arg::with_name("added")
                 .help("show only added files")
                 .short("-a")
                 .long("--added"),
         )
         .arg(
             Arg::with_name("removed")
                 .help("show only removed files")
                 .short("-r")
                 .long("--removed"),
         )
         .arg(
             Arg::with_name("clean")
                 .help("show only clean files")
                 .short("-c")
                 .long("--clean"),
         )
         .arg(
             Arg::with_name("deleted")
                 .help("show only deleted files")
                 .short("-d")
                 .long("--deleted"),
         )
         .arg(
             Arg::with_name("unknown")
                 .help("show only unknown (not tracked) files")
                 .short("-u")
                 .long("--unknown"),
         )
         .arg(
             Arg::with_name("ignored")
                 .help("show only ignored files")
                 .short("-i")
                 .long("--ignored"),
         )
         .arg(
             Arg::with_name("copies")
                 .help("show source of copied files (DEFAULT: ui.statuscopies)")
                 .short("-C")
                 .long("--copies"),
         )
         .arg(
             Arg::with_name("no-status")
                 .help("hide status prefix")
                 .short("-n")
                 .long("--no-status"),
         )
 }
 
 /// Pure data type allowing the caller to specify file states to display
 #[derive(Copy, Clone, Debug)]
 pub struct DisplayStates {
     pub modified: bool,
     pub added: bool,
     pub removed: bool,
     pub clean: bool,
     pub deleted: bool,
     pub unknown: bool,
     pub ignored: bool,
 }
 
 pub const DEFAULT_DISPLAY_STATES: DisplayStates = DisplayStates {
     modified: true,
     added: true,
     removed: true,
     clean: false,
     deleted: true,
     unknown: true,
     ignored: false,
 };
 
 pub const ALL_DISPLAY_STATES: DisplayStates = DisplayStates {
     modified: true,
     added: true,
     removed: true,
     clean: true,
     deleted: true,
     unknown: true,
     ignored: true,
 };
 
 impl DisplayStates {
     pub fn is_empty(&self) -> bool {
         !(self.modified
             || self.added
             || self.removed
             || self.clean
             || self.deleted
             || self.unknown
             || self.ignored)
     }
 }
 
 pub fn run(invocation: &crate::CliInvocation) -> Result<(), CommandError> {
     let status_enabled_default = false;
     let status_enabled = invocation.config.get_option(b"rhg", b"status")?;
     if !status_enabled.unwrap_or(status_enabled_default) {
         return Err(CommandError::unsupported(
             "status is experimental in rhg (enable it with 'rhg.status = true' \
             or enable fallback with 'rhg.on-unsupported = fallback')"
         ));
     }
 
     // TODO: lift these limitations
     if invocation.config.get_bool(b"ui", b"tweakdefaults")? {
         return Err(CommandError::unsupported(
             "ui.tweakdefaults is not yet supported with rhg status",
         ));
     }
     if invocation.config.get_bool(b"ui", b"statuscopies")? {
         return Err(CommandError::unsupported(
             "ui.statuscopies is not yet supported with rhg status",
         ));
     }
     if invocation
         .config
         .get(b"commands", b"status.terse")
         .is_some()
     {
         return Err(CommandError::unsupported(
             "status.terse is not yet supported with rhg status",
         ));
     }
 
     let ui = invocation.ui;
     let config = invocation.config;
     let args = invocation.subcommand_args;
 
     let verbose = !ui.plain()
         && !args.is_present("print0")
         && (config.get_bool(b"ui", b"verbose")?
             || config.get_bool(b"commands", b"status.verbose")?);
     if verbose {
         return Err(CommandError::unsupported(
             "verbose status is not supported yet",
         ));
     }
 
     let all = args.is_present("all");
     let display_states = if all {
         // TODO when implementing `--quiet`: it excludes clean files
         // from `--all`
         ALL_DISPLAY_STATES
     } else {
         let requested = DisplayStates {
             modified: args.is_present("modified"),
             added: args.is_present("added"),
             removed: args.is_present("removed"),
             clean: args.is_present("clean"),
             deleted: args.is_present("deleted"),
             unknown: args.is_present("unknown"),
             ignored: args.is_present("ignored"),
         };
         if requested.is_empty() {
             DEFAULT_DISPLAY_STATES
         } else {
             requested
         }
     };
     let no_status = args.is_present("no-status");
     let list_copies = all
         || args.is_present("copies")
         || config.get_bool(b"ui", b"statuscopies")?;
 
     let repo = invocation.repo?;
 
     if repo.has_sparse() || repo.has_narrow() {
         return Err(CommandError::unsupported(
             "rhg status is not supported for sparse checkouts or narrow clones yet"
         ));
     }
 
     let mut dmap = repo.dirstate_map_mut()?;
 
     let options = StatusOptions {
         // we're currently supporting file systems with exec flags only
         // anyway
         check_exec: true,
         list_clean: display_states.clean,
         list_unknown: display_states.unknown,
         list_ignored: display_states.ignored,
         list_copies,
         collect_traversed_dirs: false,
     };
     let (mut ds_status, pattern_warnings) = dmap.status(
         &AlwaysMatcher,
         repo.working_directory_path().to_owned(),
         ignore_files(repo, config),
         options,
     )?;
     for warning in pattern_warnings {
         match warning {
             hg::PatternFileWarning::InvalidSyntax(path, syntax) => ui
                 .write_stderr(&format_bytes!(
                     b"{}: ignoring invalid syntax '{}'\n",
                     get_bytes_from_path(path),
                     &*syntax
                 ))?,
             hg::PatternFileWarning::NoSuchFile(path) => {
                 let path = if let Ok(relative) =
                     path.strip_prefix(repo.working_directory_path())
                 {
                     relative
                 } else {
                     &*path
                 };
                 ui.write_stderr(&format_bytes!(
                     b"skipping unreadable pattern file '{}': \
                       No such file or directory\n",
                     get_bytes_from_path(path),
                 ))?
             }
         }
     }
 
     for (path, error) in ds_status.bad {
         let error = match error {
             hg::BadMatch::OsError(code) => {
                 std::io::Error::from_raw_os_error(code).to_string()
             }
             hg::BadMatch::BadType(ty) => {
                 format!("unsupported file type (type is {})", ty)
             }
         };
         ui.write_stderr(&format_bytes!(
             b"{}: {}\n",
             path.as_bytes(),
             error.as_bytes()
         ))?
     }
     if !ds_status.unsure.is_empty() {
         info!(
             "Files to be rechecked by retrieval from filelog: {:?}",
             ds_status.unsure.iter().map(|s| &s.path).collect::<Vec<_>>()
         );
     }
     let mut fixup = Vec::new();
     if !ds_status.unsure.is_empty()
         && (display_states.modified || display_states.clean)
     {
         let p1 = repo.dirstate_parents()?.p1;
         let manifest = repo.manifest_for_node(p1).map_err(|e| {
             CommandError::from((e, &*format!("{:x}", p1.short())))
         })?;
         for to_check in ds_status.unsure {
             if unsure_is_modified(repo, &manifest, &to_check.path)? {
                 if display_states.modified {
                     ds_status.modified.push(to_check);
                 }
             } else {
                 if display_states.clean {
                     ds_status.clean.push(to_check.clone());
                 }
                 fixup.push(to_check.path.into_owned())
             }
         }
     }
     let relative_paths = (!ui.plain())
         && config
             .get_option(b"commands", b"status.relative")?
             .unwrap_or(config.get_bool(b"ui", b"relative-paths")?);
     let output = DisplayStatusPaths {
         ui,
         no_status,
         relativize: if relative_paths {
             Some(RelativizePaths::new(repo)?)
         } else {
             None
         },
     };
     if display_states.modified {
         output.display(b"M", ds_status.modified)?;
     }
     if display_states.added {
         output.display(b"A", ds_status.added)?;
     }
     if display_states.removed {
         output.display(b"R", ds_status.removed)?;
     }
     if display_states.deleted {
         output.display(b"!", ds_status.deleted)?;
     }
     if display_states.unknown {
         output.display(b"?", ds_status.unknown)?;
     }
     if display_states.ignored {
         output.display(b"I", ds_status.ignored)?;
     }
     if display_states.clean {
         output.display(b"C", ds_status.clean)?;
     }
 
     let mut dirstate_write_needed = ds_status.dirty;
     let filesystem_time_at_status_start =
         ds_status.filesystem_time_at_status_start;
 
     if (fixup.is_empty() || filesystem_time_at_status_start.is_none())
         && !dirstate_write_needed
     {
         // Nothing to update
         return Ok(());
     }
 
     // Update the dirstate on disk if we can
     let with_lock_result =
         repo.try_with_wlock_no_wait(|| -> Result<(), CommandError> {
             if let Some(mtime_boundary) = filesystem_time_at_status_start {
                 for hg_path in fixup {
                     use std::os::unix::fs::MetadataExt;
                     let fs_path = hg_path_to_path_buf(&hg_path)
                         .expect("HgPath conversion");
                     // Specifically do not reuse `fs_metadata` from
                     // `unsure_is_clean` which was needed before reading
                     // contents. Here we access metadata again after reading
                     // content, in case it changed in the meantime.
                     let fs_metadata = repo
                         .working_directory_vfs()
                         .symlink_metadata(&fs_path)?;
                     if let Some(mtime) =
                         TruncatedTimestamp::for_reliable_mtime_of(
                             &fs_metadata,
                             &mtime_boundary,
                         )
                         .when_reading_file(&fs_path)?
                     {
                         let mode = fs_metadata.mode();
                         let size = fs_metadata.len() as u32 & RANGE_MASK_31BIT;
                         let mut entry = dmap
                             .get(&hg_path)?
                             .expect("ambiguous file not in dirstate");
                         entry.set_clean(mode, size, mtime);
                         dmap.add_file(&hg_path, entry)?;
                         dirstate_write_needed = true
                     }
                 }
             }
             drop(dmap); // Avoid "already mutably borrowed" RefCell panics
             if dirstate_write_needed {
                 repo.write_dirstate()?
             }
             Ok(())
         });
     match with_lock_result {
         Ok(closure_result) => closure_result?,
         Err(LockError::AlreadyHeld) => {
             // Not updating the dirstate is not ideal but not critical:
             // don’t keep our caller waiting until some other Mercurial
             // process releases the lock.
         }
         Err(LockError::Other(HgError::IoError { error, .. }))
             if error.kind() == io::ErrorKind::PermissionDenied =>
         {
             // `hg status` on a read-only repository is fine
         }
         Err(LockError::Other(error)) => {
             // Report other I/O errors
             Err(error)?
         }
     }
     Ok(())
 }
 
 fn ignore_files(repo: &Repo, config: &Config) -> Vec<PathBuf> {
     let mut ignore_files = Vec::new();
     let repo_ignore = repo.working_directory_vfs().join(".hgignore");
     if repo_ignore.exists() {
         ignore_files.push(repo_ignore)
     }
     for (key, value) in config.iter_section(b"ui") {
         if key == b"ignore" || key.starts_with(b"ignore.") {
             let path = get_path_from_bytes(value);
             // TODO: expand "~/" and environment variable here, like Python
             // does with `os.path.expanduser` and `os.path.expandvars`
 
             let joined = repo.working_directory_path().join(path);
             ignore_files.push(joined);
         }
     }
     ignore_files
 }
 
 struct DisplayStatusPaths<'a> {
     ui: &'a Ui,
     no_status: bool,
     relativize: Option<RelativizePaths>,
 }
 
 impl DisplayStatusPaths<'_> {
     // Probably more elegant to use a Deref or Borrow trait rather than
     // harcode HgPathBuf, but probably not really useful at this point
     fn display(
         &self,
         status_prefix: &[u8],
         mut paths: Vec<StatusPath<'_>>,
     ) -> Result<(), CommandError> {
         paths.sort_unstable();
         for StatusPath { path, copy_source } in paths {
             let relative;
             let path = if let Some(relativize) = &self.relativize {
                 relative = relativize.relativize(&path);
                 &*relative
             } else {
                 path.as_bytes()
             };
             // TODO optim, probably lots of unneeded copies here, especially
             // if out stream is buffered
             if self.no_status {
                 self.ui.write_stdout(&format_bytes!(b"{}\n", path))?
             } else {
                 self.ui.write_stdout(&format_bytes!(
                     b"{} {}\n",
                     status_prefix,
                     path
                 ))?
             }
             if let Some(source) = copy_source {
                 self.ui.write_stdout(&format_bytes!(
                     b"  {}\n",
                     source.as_bytes()
                 ))?
             }
         }
         Ok(())
     }
 }
 
 /// Check if a file is modified by comparing actual repo store and file system.
 ///
 /// This meant to be used for those that the dirstate cannot resolve, due
 /// to time resolution limits.
 fn unsure_is_modified(
     repo: &Repo,
     manifest: &Manifest,
     hg_path: &HgPath,
 ) -> Result<bool, HgError> {
     let vfs = repo.working_directory_vfs();
     let fs_path = hg_path_to_path_buf(hg_path).expect("HgPath conversion");
     let fs_metadata = vfs.symlink_metadata(&fs_path)?;
     let is_symlink = fs_metadata.file_type().is_symlink();
     // TODO: Also account for `FALLBACK_SYMLINK` and `FALLBACK_EXEC` from the
     // dirstate
     let fs_flags = if is_symlink {
         Some(b'l')
     } else if has_exec_bit(&fs_metadata) {
         Some(b'x')
     } else {
         None
     };
 
     let entry = manifest
         .find_by_path(hg_path)?
         .expect("ambgious file not in p1");
     if entry.flags != fs_flags {
         return Ok(true);
     }
     let filelog = repo.filelog(hg_path)?;
     let fs_len = fs_metadata.len();
+    let filelog_entry =
+        filelog.entry_for_node(entry.node_id()?).map_err(|_| {
+            HgError::corrupted("filelog missing node from manifest")
+        })?;
     // TODO: check `fs_len` here like below, but based on
     // `RevlogEntry::uncompressed_len` without decompressing the full filelog
     // contents where possible. This is only valid if the revlog data does not
     // contain metadata. See how Python’s `revlog.rawsize` calls
     // `storageutil.filerevisioncopied`.
     // (Maybe also check for content-modifying flags? See `revlog.size`.)
-    let filelog_entry =
-        filelog.data_for_node(entry.node_id()?).map_err(|_| {
-            HgError::corrupted("filelog missing node from manifest")
-        })?;
-    let contents_in_p1 = filelog_entry.file_data()?;
+    let filelog_data = filelog_entry.data()?;
+    let contents_in_p1 = filelog_data.file_data()?;
     if contents_in_p1.len() as u64 != fs_len {
         // No need to read the file contents:
         // it cannot be equal if it has a different length.
         return Ok(true);
     }
 
     let fs_contents = if is_symlink {
         get_bytes_from_os_string(vfs.read_link(fs_path)?.into_os_string())
     } else {
         vfs.read(fs_path)?
     };
     Ok(contents_in_p1 != &*fs_contents)
 }