use std::{ io, path::Path, str::FromStr, sync::{atomic::AtomicBool, Arc},}; use anyhow::{anyhow, Context as AnyhowContext, Result};use bytesize::ByteSize;use git_repository::{ hash::ObjectId, object, odb, odb::{pack, pack::index}, progress, Progress,};pub use index::verify::Mode; use crate::OutputFormat; #[derive(Debug, Eq, PartialEq, Hash, Clone, Copy)]pub enum Algorithm { LessTime, LessMemory,} impl Algorithm { pub fn variants() -> &'static [&'static str] { &["less-time", "less-memory"] }} impl FromStr for Algorithm { type Err = String; fn from_str(s: &str) -> Result<Self, Self::Err> { let s_lc = s.to_ascii_lowercase(); Ok(match s_lc.as_str() { "less-memory" => Algorithm::LessMemory, "less-time" => Algorithm::LessTime, _ => return Err(format!("Invalid verification algorithm: '{}'", s)), }) }} impl From<Algorithm> for index::traverse::Algorithm { fn from(v: Algorithm) -> Self { match v { Algorithm::LessMemory => index::traverse::Algorithm::Lookup, Algorithm::LessTime => index::traverse::Algorithm::DeltaTreeLookup, } }} /// A general purpose context for many operations provided herepub struct Context<W1: io::Write, W2: io::Write> { /// If set, provide statistics to `out` in the given format pub output_statistics: Option<OutputFormat>, /// A stream to which to output operation results pub out: W1, /// A stream to which to errors pub err: W2, /// If set, don't use more than this amount of threads. /// Otherwise, usually use as many threads as there are logical cores. /// A value of 0 is interpreted as no-limit pub thread_limit: Option<usize>, pub mode: index::verify::Mode, pub algorithm: Algorithm, pub should_interrupt: Arc<AtomicBool>,} impl Default for Context<Vec<u8>, Vec<u8>> { fn default() -> Self { Context { output_statistics: None, thread_limit: None, mode: index::verify::Mode::Sha1Crc32, algorithm: Algorithm::LessMemory, out: Vec::new(), err: Vec::new(), should_interrupt: Default::default(), } }} enum EitherCache<const SIZE: usize> { Left(pack::cache::Never), Right(pack::cache::lru::StaticLinkedList<SIZE>),} impl<const SIZE: usize> pack::cache::DecodeEntry for EitherCache<SIZE> { fn put(&mut self, pack_id: u32, offset: u64, data: &[u8], kind: object::Kind, compressed_size: usize) { match self { EitherCache::Left(v) => v.put(pack_id, offset, data, kind, compressed_size), EitherCache::Right(v) => v.put(pack_id, offset, data, kind, compressed_size), } } fn get(&mut self, pack_id: u32, offset: u64, out: &mut Vec<u8>) -> Option<(object::Kind, usize)> { match self { EitherCache::Left(v) => v.get(pack_id, offset, out), EitherCache::Right(v) => v.get(pack_id, offset, out), } }} pub fn pack_or_pack_index<W1, W2>( path: impl AsRef<Path>, progress: Option<impl Progress>, Context { mut out, mut err, mode, output_statistics, thread_limit, algorithm, should_interrupt, }: Context<W1, W2>,) -> Result<(ObjectId, Option<index::traverse::Outcome>)>where W1: io::Write, W2: io::Write,{ let path = path.as_ref(); let ext = path.extension().and_then(|ext| ext.to_str()).ok_or_else(|| { anyhow!( "Cannot determine data type on path without extension '{}', expecting default extensions 'idx' and 'pack'", path.display() ) })?; let res = match ext { "pack" => { let pack = odb::pack::data::File::at(path).with_context(|| "Could not open pack file")?; pack.verify_checksum( progress::DoOrDiscard::from(progress).add_child("Sha1 of pack"), &should_interrupt, ) .map(|id| (id, None))? } "idx" => { let idx = odb::pack::index::File::at(path).with_context(|| "Could not open pack index file")?; let packfile_path = path.with_extension("pack"); let pack = odb::pack::data::File::at(&packfile_path) .map_err(|e| { writeln!( err, "Could not find matching pack file at '{}' - only index file will be verified, error was: {}", packfile_path.display(), e ) .ok(); e }) .ok(); const CACHE_SIZE: usize = 64; let cache = || -> EitherCache<CACHE_SIZE> { if matches!(algorithm, Algorithm::LessMemory) { if output_statistics.is_some() { // turn off acceleration as we need to see entire chains all the time EitherCache::Left(pack::cache::Never) } else { EitherCache::Right(pack::cache::lru::StaticLinkedList::<CACHE_SIZE>::default()) } } else { EitherCache::Left(pack::cache::Never) } }; idx.verify_integrity( pack.as_ref().map(|p| (p, mode, algorithm.into(), cache)), thread_limit, progress, should_interrupt, ) .map(|(a, b, _)| (a, b)) .with_context(|| "Verification failure")? } ext => return Err(anyhow!("Unknown extension {:?}, expecting 'idx' or 'pack'", ext)), }; if let Some(stats) = res.1.as_ref() { #[cfg_attr(not(feature = "serde1"), allow(clippy::single_match))] match output_statistics { Some(OutputFormat::Human) => drop(print_statistics(&mut out, stats)), #[cfg(feature = "serde1")] Some(OutputFormat::Json) => serde_json::to_writer_pretty(out, stats)?, _ => {} }; } Ok(res)} fn print_statistics(out: &mut impl io::Write, stats: &index::traverse::Outcome) -> io::Result<()> { writeln!(out, "objects per delta chain length")?; let mut chain_length_to_object: Vec<_> = stats.objects_per_chain_length.iter().map(|(a, b)| (*a, *b)).collect(); chain_length_to_object.sort_by_key(|e| e.0); let mut total_object_count = 0; for (chain_length, object_count) in chain_length_to_object.into_iter() { total_object_count += object_count; writeln!(out, "\t{:>2}: {}", chain_length, object_count)?; } writeln!(out, "\t->: {}", total_object_count)?; let pack::data::decode_entry::Outcome { kind: _, num_deltas, decompressed_size, compressed_size, object_size, } = stats.average; let width = 30; writeln!(out, "\naverages")?; #[rustfmt::skip] writeln!( out, "\t{:<width$} {};\n\t{:<width$} {};\n\t{:<width$} {};\n\t{:<width$} {};", "delta chain length:", num_deltas, "decompressed entry [B]:", decompressed_size, "compressed entry [B]:", compressed_size, "decompressed object size [B]:", object_size, width = width )?; writeln!(out, "\ncompression")?; #[rustfmt::skip] writeln!( out, "\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}", "compressed entries size", ByteSize(stats.total_compressed_entries_size), "decompressed entries size", ByteSize(stats.total_decompressed_entries_size), "total object size", ByteSize(stats.total_object_size), "pack size", ByteSize(stats.pack_size), width = width )?; #[rustfmt::skip] writeln!( out, "\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}", "num trees", stats.num_trees, "num blobs", stats.num_blobs, "num commits", stats.num_commits, "num tags", stats.num_tags, width = width )?; let compression_ratio = stats.total_decompressed_entries_size as f64 / stats.total_compressed_entries_size as f64; let delta_compression_ratio = stats.total_object_size as f64 / stats.total_compressed_entries_size as f64; #[rustfmt::skip] writeln!( out, "\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.3}%", "compression ratio", compression_ratio, "delta compression ratio", delta_compression_ratio, "delta gain", delta_compression_ratio / compression_ratio, "pack overhead", (1.0 - (stats.total_compressed_entries_size as f64 / stats.pack_size as f64)) * 100.0, width = width )?; Ok(())}
use std::{
io,
path::Path,
str::FromStr,
sync::{atomic::AtomicBool, Arc},
};
use anyhow::{anyhow, Context as AnyhowContext, Result};
use bytesize::ByteSize;
use git_repository::{
hash::ObjectId,
object, odb,
odb::{pack, pack::index},
progress, Progress,
};
pub use index::verify::Mode;
use crate::OutputFormat;
#[derive(Debug, Eq, PartialEq, Hash, Clone, Copy)]
pub enum Algorithm {
LessTime,
LessMemory,
}
impl Algorithm {
pub fn variants() -> &'static [&'static str] {
&["less-time", "less-memory"]
}
}
impl FromStr for Algorithm {
type Err = String;
fn from_str(s: &str) -> Result<Self, Self::Err> {
let s_lc = s.to_ascii_lowercase();
Ok(match s_lc.as_str() {
"less-memory" => Algorithm::LessMemory,
"less-time" => Algorithm::LessTime,
_ => return Err(format!("Invalid verification algorithm: '{}'", s)),
})
}
}
impl From<Algorithm> for index::traverse::Algorithm {
fn from(v: Algorithm) -> Self {
match v {
Algorithm::LessMemory => index::traverse::Algorithm::Lookup,
Algorithm::LessTime => index::traverse::Algorithm::DeltaTreeLookup,
}
}
}
/// A general purpose context for many operations provided here
pub struct Context<W1: io::Write, W2: io::Write> {
/// If set, provide statistics to `out` in the given format
pub output_statistics: Option<OutputFormat>,
/// A stream to which to output operation results
pub out: W1,
/// A stream to which to errors
pub err: W2,
/// If set, don't use more than this amount of threads.
/// Otherwise, usually use as many threads as there are logical cores.
/// A value of 0 is interpreted as no-limit
pub thread_limit: Option<usize>,
pub mode: index::verify::Mode,
pub algorithm: Algorithm,
pub should_interrupt: Arc<AtomicBool>,
}
impl Default for Context<Vec<u8>, Vec<u8>> {
fn default() -> Self {
Context {
output_statistics: None,
thread_limit: None,
mode: index::verify::Mode::Sha1Crc32,
algorithm: Algorithm::LessMemory,
out: Vec::new(),
err: Vec::new(),
should_interrupt: Default::default(),
}
}
}
enum EitherCache<const SIZE: usize> {
Left(pack::cache::Never),
Right(pack::cache::lru::StaticLinkedList<SIZE>),
}
impl<const SIZE: usize> pack::cache::DecodeEntry for EitherCache<SIZE> {
fn put(&mut self, pack_id: u32, offset: u64, data: &[u8], kind: object::Kind, compressed_size: usize) {
match self {
EitherCache::Left(v) => v.put(pack_id, offset, data, kind, compressed_size),
EitherCache::Right(v) => v.put(pack_id, offset, data, kind, compressed_size),
}
}
fn get(&mut self, pack_id: u32, offset: u64, out: &mut Vec<u8>) -> Option<(object::Kind, usize)> {
match self {
EitherCache::Left(v) => v.get(pack_id, offset, out),
EitherCache::Right(v) => v.get(pack_id, offset, out),
}
}
}
pub fn pack_or_pack_index<W1, W2>(
path: impl AsRef<Path>,
progress: Option<impl Progress>,
Context {
mut out,
mut err,
mode,
output_statistics,
thread_limit,
algorithm,
should_interrupt,
}: Context<W1, W2>,
) -> Result<(ObjectId, Option<index::traverse::Outcome>)>
where
W1: io::Write,
W2: io::Write,
{
let path = path.as_ref();
let ext = path.extension().and_then(|ext| ext.to_str()).ok_or_else(|| {
anyhow!(
"Cannot determine data type on path without extension '{}', expecting default extensions 'idx' and 'pack'",
path.display()
)
})?;
let res = match ext {
"pack" => {
let pack = odb::pack::data::File::at(path).with_context(|| "Could not open pack file")?;
pack.verify_checksum(
progress::DoOrDiscard::from(progress).add_child("Sha1 of pack"),
&should_interrupt,
)
.map(|id| (id, None))?
}
"idx" => {
let idx = odb::pack::index::File::at(path).with_context(|| "Could not open pack index file")?;
let packfile_path = path.with_extension("pack");
let pack = odb::pack::data::File::at(&packfile_path)
.map_err(|e| {
writeln!(
err,
"Could not find matching pack file at '{}' - only index file will be verified, error was: {}",
packfile_path.display(),
e
)
.ok();
e
})
.ok();
const CACHE_SIZE: usize = 64;
let cache = || -> EitherCache<CACHE_SIZE> {
if matches!(algorithm, Algorithm::LessMemory) {
if output_statistics.is_some() {
// turn off acceleration as we need to see entire chains all the time
EitherCache::Left(pack::cache::Never)
} else {
EitherCache::Right(pack::cache::lru::StaticLinkedList::<CACHE_SIZE>::default())
}
} else {
EitherCache::Left(pack::cache::Never)
}
};
idx.verify_integrity(
pack.as_ref().map(|p| (p, mode, algorithm.into(), cache)),
thread_limit,
progress,
should_interrupt,
)
.map(|(a, b, _)| (a, b))
.with_context(|| "Verification failure")?
}
ext => return Err(anyhow!("Unknown extension {:?}, expecting 'idx' or 'pack'", ext)),
};
if let Some(stats) = res.1.as_ref() {
#[cfg_attr(not(feature = "serde1"), allow(clippy::single_match))]
match output_statistics {
Some(OutputFormat::Human) => drop(print_statistics(&mut out, stats)),
#[cfg(feature = "serde1")]
Some(OutputFormat::Json) => serde_json::to_writer_pretty(out, stats)?,
_ => {}
};
}
Ok(res)
}
fn print_statistics(out: &mut impl io::Write, stats: &index::traverse::Outcome) -> io::Result<()> {
writeln!(out, "objects per delta chain length")?;
let mut chain_length_to_object: Vec<_> = stats.objects_per_chain_length.iter().map(|(a, b)| (*a, *b)).collect();
chain_length_to_object.sort_by_key(|e| e.0);
let mut total_object_count = 0;
for (chain_length, object_count) in chain_length_to_object.into_iter() {
total_object_count += object_count;
writeln!(out, "\t{:>2}: {}", chain_length, object_count)?;
}
writeln!(out, "\t->: {}", total_object_count)?;
let pack::data::decode_entry::Outcome {
kind: _,
num_deltas,
decompressed_size,
compressed_size,
object_size,
} = stats.average;
let width = 30;
writeln!(out, "\naverages")?;
#[rustfmt::skip]
writeln!(
out,
"\t{:<width$} {};\n\t{:<width$} {};\n\t{:<width$} {};\n\t{:<width$} {};",
"delta chain length:", num_deltas,
"decompressed entry [B]:", decompressed_size,
"compressed entry [B]:", compressed_size,
"decompressed object size [B]:", object_size,
width = width
)?;
writeln!(out, "\ncompression")?;
#[rustfmt::skip]
writeln!(
out, "\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}",
"compressed entries size", ByteSize(stats.total_compressed_entries_size),
"decompressed entries size", ByteSize(stats.total_decompressed_entries_size),
"total object size", ByteSize(stats.total_object_size),
"pack size", ByteSize(stats.pack_size),
width = width
)?;
#[rustfmt::skip]
writeln!(
out,
"\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}\n\t{:<width$}: {}",
"num trees", stats.num_trees,
"num blobs", stats.num_blobs,
"num commits", stats.num_commits,
"num tags", stats.num_tags,
width = width
)?;
let compression_ratio = stats.total_decompressed_entries_size as f64 / stats.total_compressed_entries_size as f64;
let delta_compression_ratio = stats.total_object_size as f64 / stats.total_compressed_entries_size as f64;
#[rustfmt::skip]
writeln!(
out,
"\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.2}\n\t{:<width$}: {:.3}%",
"compression ratio", compression_ratio,
"delta compression ratio", delta_compression_ratio,
"delta gain", delta_compression_ratio / compression_ratio,
"pack overhead", (1.0 - (stats.total_compressed_entries_size as f64 / stats.pack_size as f64)) * 100.0,
width = width
)?;
Ok(())
}