Author Topic: [1.5.8.20] derived command counter is wrong  (Read 6771 times)

BlackJack

  • Jr. Member
  • **
  • Posts: 68
[1.5.8.20] derived command counter is wrong
« on: May 28, 2016, 04:02:03 PM »
The derived command counter isn't working correct.
My profiles are growing by update from 1.5.8.18. to 1.5.8.20
I've exported a profile and imported this profile again and the commands grows from 12966 to 15000
An other profile with 76 commands has 6558 derived command. After deleting the four commands with square brackets, VA tels me that I still have 6556 derived and 70 normal commands.
It looks also, that they need more time to load.
Same with 1.5.8.21.

Gary

  • Administrator
  • Hero Member
  • *****
  • Posts: 2832
Re: [1.5.8.20] derived command counter is wrong
« Reply #1 on: May 31, 2016, 09:52:29 PM »
Hi, Blackjack.

The derived command counter is made up of the following:

The actual command count (this includes multi-part commands (the commands that are separated by semicolon))
Composite commands (prefix and suffix commands)
Dynamic commands (commands made up with the square brackets)
Any commands from a linked profile
'Override listening when my spoken phrase starts with' selected (this in itself will double the derived list)

What is it that makes the count wrong?
« Last Edit: May 31, 2016, 11:58:31 PM by gmagenheimer »

Gary

  • Administrator
  • Hero Member
  • *****
  • Posts: 2832
Re: [1.5.8.20] derived command counter is wrong
« Reply #2 on: May 31, 2016, 11:47:01 PM »
I did notice that if a voice command is disabled, the counter will still show them as included.  This has been corrected.


On a side note, something I've noticed is that long sentences included as commands (in bulk) tend to bog down the speech engine load time.  The only workaround for this right now is to shorten your commands :(

Gary

  • Administrator
  • Hero Member
  • *****
  • Posts: 2832
Re: [1.5.8.20] derived command counter is wrong
« Reply #3 on: June 01, 2016, 11:39:41 AM »
Extra note on this.  I *think* I may have come up with some code to help with larger profile load times.  I probably just jinxed it by saying something o_O  This will be in the next release... shooting for today.

BlackJack

  • Jr. Member
  • **
  • Posts: 68
Re: [1.5.8.20] derived command counter is wrong
« Reply #4 on: June 01, 2016, 12:52:30 PM »
Hi, Gary.

I always try to keep my phrases as short as possible in order to hold the loading time down.
But the affected huge profile was linked with a even huger profile and I can't remember, that I linked them.
It would be illogical to do that, as leonard would say, because there would be no benefit.
I must have been sleepwalking again. But the fact that I realized the high number of commands after an update and
I've recently read  in a post, you planed to modify something around this topic, let me think something went wrong. sorry  ::)
I didn't even know, that the counter include the commands from linked profiles. It makes sens.
So, thank you for the hints. The loading time now is as fast as before.

Gary

  • Administrator
  • Hero Member
  • *****
  • Posts: 2832
Re: [1.5.8.20] derived command counter is wrong
« Reply #5 on: June 01, 2016, 06:13:28 PM »
There is a new beta out there to help with the loading stuff if you find that you've got a lot of commands ;)

BlackJack

  • Jr. Member
  • **
  • Posts: 68
Re: [1.5.8.20] derived command counter is wrong
« Reply #6 on: June 02, 2016, 03:54:49 PM »
Oh Yeah! I can see the effect. Good work!