Hi @tom
Something you can do on your side (as you are on Linux):
- Exit Krita if opened
- Edit file
~/.local/share/krita/pykrita/bulicommander/bc/bcfile.py - Uncomment the last line of file to have this:
Debug.setEnabled(True) - Open a terminal and execute from command line (it’s import to get output information!)
- Open BuliCommander
You should get something like this in terminal output:DEBUG: =========================================== DEBUG: BCFileList.execute DEBUG: ........................................... DEBUG: Search in paths: [<BCFileListPath('/home/grum/Temporaire/TempKrita/testMisc/testmd5', False)>] DEBUG: Found 500 of 500 files in 0.002569913864135742s DEBUG: - Files: 500 DEBUG: - Directories: 0 DEBUG: Filter 500 files to 500 files in 0.3373551368713379s DEBUG: BCFileList.execute.99-global: 0.4158 DEBUG: BCFileList.execute.01-search: 0.0026 DEBUG: BCFileList.execute.02-filter: 0.3374 DEBUG: BCFileList.execute.03-result: 0.0369 DEBUG: BCFileList.execute.04-buildStats: 0.0387 DEBUG: BCFileList.execute.05-sort: 0.0000 DEBUG: =========================================== DEBUG: =========================================== DEBUG: BCFileList.execute DEBUG: ........................................... DEBUG: Search in paths: [<BCFileListPath('/home/grum/Data/Temp/testmd5', False)>] DEBUG: Found 500 of 500 files in 0.002535104751586914s DEBUG: - Files: 500 DEBUG: - Directories: 0 DEBUG: Filter 500 files to 500 files in 0.3148353099822998s DEBUG: BCFileList.execute.99-global: 0.3918 DEBUG: BCFileList.execute.01-search: 0.0025 DEBUG: BCFileList.execute.02-filter: 0.3148 DEBUG: BCFileList.execute.03-result: 0.0341 DEBUG: BCFileList.execute.04-buildStats: 0.0400 DEBUG: BCFileList.execute.05-sort: 0.0000 DEBUG: ===========================================
Go up from directory and go in again, just to got a comparison startup/not startup.
I think problem is not in this part of plugin (searching & analyzing files) but your stats would confirm me this ; I’m trying to determinate where to put stopwatches especially on startup to get a more precise idea of what happen
Also, doing some tests I saw that since I’ve implemented metadata cache, even if it’s quite faster than before cache implementation, if I reduce multithreading it’s now faster than before
![]()
I’ll do some more tests for that to determinate when activate/deactivate multithreading…
Grum999