arsenetar/dupeguru
1
0
Fork 0
mirror of https://github.com/arsenetar/dupeguru.git synced 2026-01-22 14:41:39 +00:00
Code Releases Wiki Activity

serialize/deserialize colors to/from bytes instead of strings

Browse Source 
it's a tiny bit faster and saves a bit of memory
This commit is contained in:
Dobatymo
2022-09-27 17:34:57 +08:00
parent 1f1dfa88dc
commit f1153c85c0
9 changed files with 47 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
core/pe/matchblock.py
View File

@@ -27,7 +27,7 @@ from core.pe.block import avgdiff, DifferentBlockCountError, NoBlocksError
# to files in other chunks. So chunkifying doesn't save us any actual comparison, but the advantage
# is that instead of reading blocks from disk number_of_files**2 times, we read it
# number_of_files*number_of_chunks times.
# Determining the right chunk size is tricky, bceause if it's too big, too many blocks will be in
# Determining the right chunk size is tricky, because if it's too big, too many blocks will be in
# memory at the same time and we might end up with memory trashing, which is awfully slow. So,
# because our *real* bottleneck is CPU, the chunk size must simply be enough so that the CPU isn't
# starved by Disk IOs.