Rok Mandeljc

> I quite often accidentally clobber my spec files by editing them then blindly running the most recent `pyinstaller ...` command in my terminal history. Yeah, same here... > I've...

This performance hit is caused by 91fb851cb384fb45986cd1e80d2c63a7c22415b8; every time `pefile.PE` is closed (either explicitly via `close` method, or due to use of context, or if the loading fails with exception),...

@erocarrera In case it helps with confirmation/verification: ```python # perftest.py import sys import time import numpy.core import pefile import numpy as np mode = sys.argv[1] if len(sys.argv) > 1 else...

Considering the underlying error is `MemoryError` during PYZ extraction, I would say either this is a low-memory issue, or you have somehow ended up with (partially?) corrupted executable and its...

> Do you think moving CDEmu to that would be possible? Theoretically, it might be; but I imagine will also require redesign of the clients and their functionality (for example,...

> And [@rokm](https://github.com/rokm): if cdemu is in "maintenance phase", the hardest part is probably maintaining that vbha-module kernel module, which constantly breaks since it's out-of-tree and the Linux kernel API...

It is likely that cdemu's (rather primitive delay emulation) is not accurate enough to fool the protection shemes in question. If you want to investigate this, you should probably start...

> > perhaps the part that "short-cuts" the delay for long rotations is problematic; etc. > > I assumed this was the problem, but after experimenting with more advanced handling...

> > Nope, I don't have an exact model (I don't think I ever really set out to set one up, either). https://en.wikipedia.org/wiki/CD-ROM has high-level overview of transfer rates and...

> Do you have any information about how PC disc drives choose to read in CLV vs CAV, and whether you have any control over that? I think that's implementation...