| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | pack: Add _read to filenames |  Runxi Yu | 2026-01-28 | 1 | -578/+0 |
| * | hash: Use a hashAlgorithmDetails struct for single source of truth•••hashAlgorithm's are assumed to be valid; methods on invalid
hashAlgorithms will panic from out-of-bounds read when it's not found in
hashAlgorithmTable and that's expected and intended.
v0.1.2 | Runxi Yu | 2026-01-27 | 1 | -2/+2 |
| * | More changes from the import URL update |  Runxi Yu | 2026-01-19 | 1 | -2/+2 |
| * | repo: Drop hashSize and use hashAlgo.size() | Runxi Yu | 2026-01-17 | 1 | -2/+2 |
| * | hash: Key by algorithm, not size | Runxi Yu | 2026-01-17 | 1 | -2/+2 |
| * | loose/pack: Remove commented-out verifyTypedObject•••It's not supposed to be used there anyway
| Runxi Yu | 2026-01-15 | 1 | -4/+0 |
| * | pack: Do not madvise•••Apparently although it's not fully sequential, the general pattern is
sort-of increasing in offsets as trees are being traversed.
| Runxi Yu | 2025-11-24 | 1 | -11/+0 |
| * | bufpool: Return bytes.Buffer, rather than a pointer to it•••It's silly to allocate a bytes.Buffer struct, however small it is,
every time Borrow is called, since the entire purpose is to reduce
allocations.
| Runxi Yu | 2025-11-23 | 1 | -32/+32 |
| * | obj: call parseObjectBody in ReadObject instead of loose/pack readers | Runxi Yu | 2025-11-22 | 1 | -10/+7 |
| * | pack: Allocate the header on the stack instead | Runxi Yu | 2025-11-22 | 1 | -1/+2 |
| * | bufpool: Return *Buffer | Runxi Yu | 2025-11-22 | 1 | -27/+27 |
| * | bufpool: Improve perf by using buckets of different size classes | Runxi Yu | 2025-11-22 | 1 | -1/+1 |
| * | pack: Don't use bytes.Reader cursor, use mmaped slice directly | Runxi Yu | 2025-11-22 | 1 | -74/+66 |
| * | Revert "Add initial support for multi pack indexes"•••This reverts commit 6f8acbf1503d2fa1ef705d35a743fc6b279942e5.
Apparently my MIDX support is broken and it breaks some repos...
it'll be added back when ready.
| Runxi Yu | 2025-11-20 | 1 | -13/+0 |
| * | Probably should name the custom packages specially | Runxi Yu | 2025-11-19 | 1 | -3/+2 |
| * | Remove some redundant code | Runxi Yu | 2025-11-19 | 1 | -54/+13 |
| * | pack: Iterative instead of recursive resolver | Runxi Yu | 2025-11-19 | 1 | -116/+160 |
| * | Initial attempt to make a compressor with less overhead•••io.Reader actually has massive overhead...
| Runxi Yu | 2025-11-19 | 1 | -25/+30 |
| * | madvise | Runxi Yu | 2025-11-18 | 1 | -0/+12 |
| * | Fix import order | Runxi Yu | 2025-11-18 | 1 | -1/+2 |
| * | Use a pooled zlib implementation•••It's trivially forked from the standard library and improves git ls-tree
-r --long speed by approximately 25%.
| Runxi Yu | 2025-11-18 | 1 | -1/+1 |
| * | Disable per-object validation during reads•••In general, git inflates the packed objects and recomputes the object
hash before accepting them into hte object database during remote
operations. But when simply inflating/parsing the objects for read from
the local object store, it does not recompute the hash.
| Runxi Yu | 2025-11-17 | 1 | -4/+4 |
| * | pack: Use a Go map with a mutex instead of a sync.Map for packfiles•••Very few writes, you don't typically see more than a dozen packfiles.
A ton of reads. Go maps are the obvious choice.
| Runxi Yu | 2025-11-17 | 1 | -0/+24 |
| * | Revert "Compute checksum when reading packfiles"•••This reverts commit 1dcb92427c23d0a8b23c0154b892243c749afa5a.
Yeah this should be part of the network protocol rather than being done
on *every read*.
| Runxi Yu | 2025-11-17 | 1 | -22/+2 |
| * | Compute checksum when reading packfiles | Runxi Yu | 2025-11-17 | 1 | -2/+22 |
| * | Fix some docs and API types | Runxi Yu | 2025-11-16 | 1 | -2/+2 |
| * | Documentation overhaul | Runxi Yu | 2025-11-16 | 1 | -1/+0 |
| * | Extract bufpool into its own package | Runxi Yu | 2025-11-16 | 1 | -45/+47 |
| * | Rename ObjType enum | Runxi Yu | 2025-11-16 | 1 | -39/+39 |
| * | Unexport Repository.HashSize | Runxi Yu | 2025-11-16 | 1 | -4/+4 |
| * | Rename ObjType -> ObjectType | Runxi Yu | 2025-11-16 | 1 | -9/+9 |
| * | hash: Make fewer helper functions need explicit hash length fields | Runxi Yu | 2025-11-16 | 1 | -4/+6 |
| * | Revert "hash: Generic hash-algorithm API"•••This reverts commit 94bfb1fa147f80e6ec39009d41fc2f853925e0a5.
Generics actually kinda suck for these purposes... once you look at it
from the user's perspective.
| Runxi Yu | 2025-11-16 | 1 | -16/+16 |
| * | hash: Generic hash-algorithm API | Runxi Yu | 2025-11-16 | 1 | -16/+16 |
| * | Support multiple hash sizes in one build | Runxi Yu | 2025-11-16 | 1 | -2/+2 |
| * | Add initial support for multi pack indexes | Runxi Yu | 2025-11-15 | 1 | -0/+13 |
| * | Unexport PackLocation | Runxi Yu | 2025-11-14 | 1 | -9/+9 |
| * | Read types and sizes without inflating entire object | Runxi Yu | 2025-11-14 | 1 | -0/+83 |
| * | Initial commit | Runxi Yu | 2025-11-13 | 1 | -0/+473 |
Runxi Yu
Runxi Yu