Skip to content

Commit

Permalink
Notes
Browse files Browse the repository at this point in the history
  • Loading branch information
FenPhoenix committed Jan 2, 2025
1 parent 7d5ca97 commit 8b4c945
Show file tree
Hide file tree
Showing 2 changed files with 6 additions and 0 deletions.
3 changes: 3 additions & 0 deletions FMScanner/Scanner.cs
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,9 @@
@BLOCKS: Could SharpCompress (full) allow us to stream 7z entries to memory?
Even though it's slower than native 7z.exe, if we have to extract a lot less, then maybe we'd still come out ahead.
We could scan .mis and .gam files in the usual way, decompressing in chunks etc.
UPDATE 2025-01-01: Tested this, and surprisingly we gain very little to nothing. SharpCompress is much slower
at decompressing than native 7z.exe, enough so that it erases most of our time gained. It's probably for the
best, as it made the code even more horrendously complicated than it already is.
@BLOCKS: Non-solid 7z FMs work fine, but our solid-aware paths might be doing more work than necessary in that
case. TBP non-solid scans very slightly slower than loader-friendly solid (like ~220ms vs ~190ms warm).
Expand Down
3 changes: 3 additions & 0 deletions SharpCompress/Archives/SevenZip/SevenZipArchive.cs
Original file line number Diff line number Diff line change
@@ -1,6 +1,9 @@
/*
@SharpCompress: There's plenty of performance optimization we could still do here.
Lots of byte[] allocations, stream recreations, etc. However, we're already pretty fast, so it's not urgent.
@SharpCompress: Note that this custom version is broken for actually extracting files, and I haven't been able to
figure out where the breakage is. But it's fine for just getting entries, which is all we do with it.
*/

using System.IO;
Expand Down

0 comments on commit 8b4c945

Please sign in to comment.