GnuPG keypair generation takes more than an hour

[WanEye]

The window doesn say whether the process is finished or failed or times out. A status message would be convenient.

[annejan]

There used to be feedback both positive and negative, I'll look into this regression.

[rdoeffinger]

As to addressing the long key generation times, if compatibility to old gpg versions isn't a huge priority, maybe should use
--default-new-key-algo ed25519
for generating keys?
ECC keys should require much less randomness and thus be much faster to generate.

[FiloSpaTeam]

I created a branch with a QFuture usage to handle this key generation so the main process will not stuck. Idk if it can solve but using this i can link the events from Pass to QFuture to control the process

[rdoeffinger]

I am not sure if it should take this long, for me generating a key even with 3000 bit RSA takes a few seconds at most.
However I noticed that QtPass sometimes seems to not notice that the gpg process has exited due to some error and the hangs forever... So do we know for sure it was really gpg taking this long or did it maybe crash or otherwise fail and QtPass fails to handle THAT?

[FiloSpaTeam]

QtPass calls directly the command so maybe we should use a QProcess or something like that to know who stucks.
QFuture or QtConcurrent module can be useful to isolate in another Thread and handle with QObject or events the results and the progress.

[lherschi]

#640

[wxl]

This looks fixed as far as I can tell; please close.