Code

History

Commits on git-commitgraph-v0.4.33535+
  1. reduce memory usage for index considerablySebastian Thiel committed 6 years agoaa802be
  2. and now it works!Sebastian Thiel committed 6 years agof14e10e
  3. use new traversal in index writing, but it doesn't work yetSebastian Thiel committed 6 years ago0dd5570
  4. refactorSebastian Thiel committed 6 years ago4ff69c6
  5. refactorSebastian Thiel committed 6 years ago6cbb7cc
  6. generalized tree traversal can theoretically workSebastian Thiel committed 6 years ago64158e0
  7. make traversal part of the tree for greater ease of useSebastian Thiel committed 6 years ago6629e30
  8. prepare flexible traversal on decompressed objectsSebastian Thiel committed 6 years ago7707ea6
  9. refactorSebastian Thiel committed 6 years agodeea36c
  10. refactorSebastian Thiel committed 6 years ago83a0102
  11. refactorSebastian Thiel committed 6 years agob77d148
  12. generalize tree iterationSebastian Thiel committed 6 years agofdc06de
  13. Also index using the new tree impl during verify (prepare replacement)Sebastian Thiel committed 6 years ago92039b0
  14. refactorSebastian Thiel committed 6 years agoe3ff6af
  15. support for building a tree from offsetsSebastian Thiel committed 6 years ago95858bc
  16. refactorSebastian Thiel committed 6 years ago8cfe025
  17. refactorSebastian Thiel committed 6 years agobb9e518
  18. count sorting into the progress, 7.5 mio entries takes a momentSebastian Thiel committed 6 years ago2fc4cd8
  19. Now it's probably as fast as it gets without miniz-oxide adjustmentsSebastian Thiel committed 6 years ago4cb091b
  20. Use bigger buffers when reading from disk.Sebastian Thiel committed 6 years agoe76e4eb
  21. Only keep decompressed bytes of base objects…Sebastian Thiel committed 6 years agob39ad89
  22. remove memory mode entirely (and some complexity with it)Sebastian Thiel committed 6 years ago8812e91
  23. turns out you never want to keep deltas in memorySebastian Thiel committed 6 years ago657aa2c
  24. Remove support for keeping compressed memory to reduce the index sizeSebastian Thiel committed 6 years ago1e2ec7e
  25. don't cause re-allocs of the compression bufferSebastian Thiel committed 6 years ago2bb6fd2
  26. document miniz-oxide performance opportunity instead of trying to fix itSebastian Thiel committed 6 years ago31934f6
  27. Revert "FAIL: try to use a customized version of just pieces of Miniz-oxide"Sebastian Thiel committed 6 years agoea0fdb3
  28. FAIL: try to use a customized version of just pieces of Miniz-oxideSebastian Thiel committed 6 years ago9945eba
  29. dependency upgrade + updateSebastian Thiel committed 6 years agoc6692c6
  30. refactorSebastian Thiel committed 6 years ago133e3ba
  31. Let go of another handbreak - decompression is much faster nowSebastian Thiel committed 6 years agoae9dc16
  32. thanks clippySebastian Thiel committed 6 years ago393067d
  33. Use call to produce the resolver, allowing to delay opening a file mapping…Sebastian Thiel committed 6 years agodd30e8d
  34. Fix actual memory violation (thanks to unsafe code)Sebastian Thiel committed 6 years agoc44c5e1
  35. thanks clippySebastian Thiel committed 6 years ago1083a0b