[Users] Finally, I had to switch to Thunderbird. CM is too slow to be usable!

Steve Litt slitt at troubleshooters.com
Thu Oct 31 21:02:17 CET 2013


On Thu, 31 Oct 2013 17:44:04 +0100
wwp <subscript at free.fr> wrote:

> Hello,
> 
> 
> On Thu, 31 Oct 2013 17:33:27 +0100 Andrej Kacian <andrej at kacian.sk>
> wrote:
> 
> > On Wed, 30 Oct 2013 22:17:00 -0400
> > Steve Litt <slitt at troubleshooters.com> wrote:
> > 
> > > > Short of getting some amount of multithreaded behavior into
> > > > Claws (an admittedly large and disruptive task no doubt), I
> > > > wonder if it would  
> > > 
> > > Am I on drugs, or is multithreading not the only way to do
> > > simultaneous processes? I have a shellscript that allows me to
> > > type in info for the next vinyl record, while this vinyl record
> > > plays and then converts to .ogg and normalizes. All I did was
> > > exec a process. 
> > 
> > Yes, but multi-process code is even more complex than
> > multithreading, what with all the IPC involved. :)
> 
> Right, what matters is not the way you de-synchronize two processes,
> but the way you synchronize them back, whatever it's about handling
> concurrent access to resources or process communication. That's what
> probably doesn't happen at all in SteveT's example (script that
> manages vynil-to-ogg flow).

OK, how about this:

Put a flag on the code to grab email so it can be turned off. Then make
a copy of the email grabbing routine, and put it in its own executable.
People like the original poster and I can just run the email grab from
the command line, when *we* decide to, and it won't impact our ability
to use Claws while grabbing the mail.

If you're worried about someone doing a "delete all" while this is in
effect, there could be a flag somewhere (put it in a file for all I
care --- reading a file takes no time compared to anything else) saying
"this folder is being written to." Seriously, there has to be a simple
way.

Thanks,

SteveT

Steve Litt                *  http://www.troubleshooters.com/
Troubleshooting Training  *  Human Performance



More information about the Users mailing list