Function git_pack::multi_index::chunk::index_names::write
source · Expand description
Write all paths in order to out, including padding.
Examples found in repository?
src/multi_index/write.rs (line 176)
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
pub fn write_from_index_paths<P>(
mut index_paths: Vec<PathBuf>,
out: impl std::io::Write,
mut progress: P,
should_interrupt: &AtomicBool,
Options { object_hash }: Options,
) -> Result<Outcome<P>, Error>
where
P: Progress,
{
let out = git_features::hash::Write::new(out, object_hash);
let (index_paths_sorted, index_filenames_sorted) = {
index_paths.sort();
let file_names = index_paths
.iter()
.map(|p| PathBuf::from(p.file_name().expect("file name present")))
.collect::<Vec<_>>();
(index_paths, file_names)
};
let entries = {
let mut entries = Vec::new();
let start = Instant::now();
let mut progress = progress.add_child_with_id("Collecting entries", *b"MPCE"); /* Multiindex from Paths Collecting Entries */
progress.init(Some(index_paths_sorted.len()), git_features::progress::count("indices"));
// This could be parallelized… but it's probably not worth it unless you have 500mio objects.
for (index_id, index) in index_paths_sorted.iter().enumerate() {
let mtime = index
.metadata()
.and_then(|m| m.modified())
.unwrap_or(SystemTime::UNIX_EPOCH);
let index = crate::index::File::at(index, object_hash)?;
entries.reserve(index.num_objects() as usize);
entries.extend(index.iter().map(|e| Entry {
id: e.oid,
pack_index: index_id as u32,
pack_offset: e.pack_offset,
index_mtime: mtime,
}));
progress.inc();
if should_interrupt.load(Ordering::Relaxed) {
return Err(Error::Interrupted);
}
}
progress.show_throughput(start);
let start = Instant::now();
progress.set_name("Deduplicate");
progress.init(Some(entries.len()), git_features::progress::count("entries"));
entries.sort_by(|l, r| {
l.id.cmp(&r.id)
.then_with(|| l.index_mtime.cmp(&r.index_mtime).reverse())
.then_with(|| l.pack_index.cmp(&r.pack_index))
});
entries.dedup_by_key(|e| e.id);
progress.inc_by(entries.len());
progress.show_throughput(start);
if should_interrupt.load(Ordering::Relaxed) {
return Err(Error::Interrupted);
}
entries
};
let mut cf = git_chunk::file::Index::for_writing();
cf.plan_chunk(
multi_index::chunk::index_names::ID,
multi_index::chunk::index_names::storage_size(&index_filenames_sorted),
);
cf.plan_chunk(multi_index::chunk::fanout::ID, multi_index::chunk::fanout::SIZE as u64);
cf.plan_chunk(
multi_index::chunk::lookup::ID,
multi_index::chunk::lookup::storage_size(entries.len(), object_hash),
);
cf.plan_chunk(
multi_index::chunk::offsets::ID,
multi_index::chunk::offsets::storage_size(entries.len()),
);
let num_large_offsets = multi_index::chunk::large_offsets::num_large_offsets(&entries);
if let Some(num_large_offsets) = num_large_offsets {
cf.plan_chunk(
multi_index::chunk::large_offsets::ID,
multi_index::chunk::large_offsets::storage_size(num_large_offsets),
);
}
let mut write_progress = progress.add_child_with_id("Writing multi-index", *b"MPBW"); /* Multiindex Bytes Written */
let write_start = Instant::now();
write_progress.init(
Some(cf.planned_storage_size() as usize + Self::HEADER_LEN),
git_features::progress::bytes(),
);
let mut out = git_features::progress::Write {
inner: out,
progress: write_progress,
};
let bytes_written = Self::write_header(
&mut out,
cf.num_chunks().try_into().expect("BUG: wrote more than 256 chunks"),
index_paths_sorted.len() as u32,
object_hash,
)?;
{
progress.set_name("Writing chunks");
progress.init(Some(cf.num_chunks()), git_features::progress::count("chunks"));
let mut chunk_write = cf.into_write(&mut out, bytes_written)?;
while let Some(chunk_to_write) = chunk_write.next_chunk() {
match chunk_to_write {
multi_index::chunk::index_names::ID => {
multi_index::chunk::index_names::write(&index_filenames_sorted, &mut chunk_write)?
}
multi_index::chunk::fanout::ID => multi_index::chunk::fanout::write(&entries, &mut chunk_write)?,
multi_index::chunk::lookup::ID => multi_index::chunk::lookup::write(&entries, &mut chunk_write)?,
multi_index::chunk::offsets::ID => {
multi_index::chunk::offsets::write(&entries, num_large_offsets.is_some(), &mut chunk_write)?
}
multi_index::chunk::large_offsets::ID => multi_index::chunk::large_offsets::write(
&entries,
num_large_offsets.expect("available if planned"),
&mut chunk_write,
)?,
unknown => unreachable!("BUG: forgot to implement chunk {:?}", std::str::from_utf8(&unknown)),
}
progress.inc();
if should_interrupt.load(Ordering::Relaxed) {
return Err(Error::Interrupted);
}
}
}
// write trailing checksum
let multi_index_checksum: git_hash::ObjectId = out.inner.hash.digest().into();
out.inner.inner.write_all(multi_index_checksum.as_slice())?;
out.progress.show_throughput(write_start);
Ok(Outcome {
multi_index_checksum,
progress,
})
}