gix_pack/data/
mod.rs

1//! a pack data file
2use std::path::Path;
3
4/// The offset to an entry into the pack data file, relative to its beginning.
5pub type Offset = u64;
6
7/// An identifier to uniquely identify all packs loaded within a known context or namespace.
8pub type Id = u32;
9
10use memmap2::Mmap;
11
12/// An representing an full- or delta-object within a pack
13#[derive(PartialEq, Eq, Debug, Hash, Ord, PartialOrd, Clone)]
14#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
15pub struct Entry {
16    /// The entry's header
17    pub header: entry::Header,
18    /// The decompressed size of the entry in bytes.
19    ///
20    /// Note that for non-delta entries this will be the size of the object itself.
21    pub decompressed_size: u64,
22    /// absolute offset to compressed object data in the pack, just behind the entry's header
23    pub data_offset: Offset,
24}
25
26mod file;
27pub use file::{decode, verify, Header};
28///
29pub mod header;
30
31///
32pub mod init {
33    pub use super::header::decode::Error;
34}
35
36///
37pub mod entry;
38
39///
40#[cfg(feature = "streaming-input")]
41pub mod input;
42
43/// Utilities to encode pack data entries and write them to a `Write` implementation to resemble a pack data file.
44#[cfg(feature = "generate")]
45pub mod output;
46
47/// A slice into a pack file denoting a pack entry.
48///
49/// An entry can be decoded into an object.
50pub type EntryRange = std::ops::Range<Offset>;
51
52/// Supported versions of a pack data file
53#[derive(Default, PartialEq, Eq, Debug, Hash, Ord, PartialOrd, Clone, Copy)]
54#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
55#[allow(missing_docs)]
56pub enum Version {
57    #[default]
58    V2,
59    V3,
60}
61
62/// A pack data file
63pub struct File {
64    data: Mmap,
65    path: std::path::PathBuf,
66    /// A value to represent this pack uniquely when used with cache lookup, or a way to identify this pack by its location on disk.
67    /// The same location on disk should yield the same id.
68    ///
69    /// These must be unique per pack and must be stable, that is they don't change if the pack doesn't change.
70    /// If the same id is assigned (or reassigned) to different packs, pack creation or cache access will fail in hard-to-debug ways.
71    ///
72    /// This value is controlled by the owning object store, which can use it in whichever way it wants as long as the above constraints are met.
73    pub id: Id,
74    version: Version,
75    num_objects: u32,
76    /// The size of the hash contained within. This is entirely determined by the caller, and repositories have to know which hash to use
77    /// based on their configuration.
78    hash_len: usize,
79    object_hash: gix_hash::Kind,
80}
81
82/// Information about the pack data file itself
83impl File {
84    /// The pack data version of this file
85    pub fn version(&self) -> Version {
86        self.version
87    }
88    /// The number of objects stored in this pack data file
89    pub fn num_objects(&self) -> u32 {
90        self.num_objects
91    }
92    /// The length of all mapped data, including the pack header and the pack trailer
93    pub fn data_len(&self) -> usize {
94        self.data.len()
95    }
96    /// The kind of hash we use internally.
97    pub fn object_hash(&self) -> gix_hash::Kind {
98        self.object_hash
99    }
100    /// The position of the byte one past the last pack entry, or in other terms, the first byte of the trailing hash.
101    pub fn pack_end(&self) -> usize {
102        self.data.len() - self.hash_len
103    }
104
105    /// The path to the pack data file on disk
106    pub fn path(&self) -> &Path {
107        &self.path
108    }
109
110    /// Returns the pack data at the given slice if its range is contained in the mapped pack data
111    pub fn entry_slice(&self, slice: EntryRange) -> Option<&[u8]> {
112        let entry_end: usize = slice.end.try_into().expect("end of pack fits into usize");
113        let entry_start = slice.start as usize;
114        self.data.get(entry_start..entry_end)
115    }
116
117    /// Returns the CRC32 of the pack data indicated by `pack_offset` and the `size` of the mapped data.
118    ///
119    /// _Note:_ finding the right size is only possible by decompressing
120    /// the pack entry beforehand, or by using the (to be sorted) offsets stored in an index file.
121    ///
122    /// # Panics
123    ///
124    /// If `pack_offset` or `size` are pointing to a range outside of the mapped pack data.
125    pub fn entry_crc32(&self, pack_offset: Offset, size: usize) -> u32 {
126        let pack_offset: usize = pack_offset.try_into().expect("pack_size fits into usize");
127        gix_features::hash::crc32(&self.data[pack_offset..pack_offset + size])
128    }
129}
130
131pub(crate) mod delta;
gix_pack/data/mod.rs

gix_pack/data/
mod.rs
