problem waiting - spellchecker

[MisterX]

Woah cool the DarFunction!

Some Interactivity seems to be showing handy thanks to the DarFunction in
this waiting-room case though!

More than i saw before... the seconds ticked but...

But even though i tried it still worked once and then
no more... It just hanged. The "message" wouldn't budge.
Same keyword... The hilite also i noticed was only working
once hilited and then no more (nu behavior since new word parser
but same result - i said before that it took 3 or 4 clicks 
before it locked.)

So it's, as i was afraid, a scripter's error! Arghh

I started commenting out all the scripts in the button that's supposed to be
hit that was unneeded. And i left the essentials. 

Suddenly i get an error!
I was adding 1 to a wrong array item... 
A code line from mambo #2 - the xtalk word parser!

RevError suddenly spewed in total geometric disarray (seen this before for
the same reasons somewhere too?)

Type add: destination has a bad format (numeric?)
Object Learn
Line add 1 to xosdictionary["currentword"]
Hint button id 1030 of group id 1080 of card id 1002 of stack
"F:/TAOO/Palettes/SpellNO2.rev"

Why wasn't that error showing up before??? I sure am glad i didn't enter a
bugzilla for the wait problem! It's the debugger not following the path
deeper into the problem... Something Bugzillaed months ago ...

Xcellent deduction thanks to DAR! My man! You saved me a rewrite ;) The next
rev made dictionary is gonna rock with your name in the credits and the
honorous free license emeritus circle in TAOO! 

SpellNO2 learned a whole paragraph in one pass thanks to you!!! Cool! Now, i
got to start watching for badly spelled words and how it suggests "the real
stuff"... And work out Xceptions... 

Thanks again Dar!!!!

revcheers
Xav
http://monsieurx.com 


> -----Original Message-----
> From: Dave Cragg [mailto:dcragg at lacscentre.co.uk] 
> Sent: Thursday, September 01, 2005 19:28
> To: x at monsieurx.com; How to use Revolution
> Subject: Re: problem waiting - spellchecker
> 
> On 1 Sep 2005, at 17:20, MisterX wrote:
> >
> >
> > repeat...
> >   ...
> >   put "wait" into xosdictionary["current"]
> >   --put the waitdepth -- always 1
> >   --if the waitdepth < 2 then
> >     wait until xosdictionary["current"] = "continue" with messages
> >   -- else skip... poor idea...
> >   --end if
> >   ...
> > end repeat
> > then, any button in the spellchecker GUI will do something like 
> > ignore, fix, etc... and then
> >   ...
> >   put "continue" into xosdictionary["current"] end mouseup
> >
> > The reason the wait is in the first place is to wait for the user 
> > input (or abort). Since it's blocked, the only way this could work 
> > (it's in a loop that parses each word) would be if i added a double 
> > condition with a property reset and that gets messy...
> >
> > This requires major code change (not again, 2 already)...
> >
> > Why can't this wait statement work each time it is called?
> > If it works 2 or 3 times why not 4 or more?
> >
> > What can affect this?
> >
> > Im removing now all outside influences for the next test run...
> >
> > The question remains, why does it lock up?
> 
> What do you mean by lock up?
> 
> If you mean it's hanging at the "wait" statement, then the 
> first thought would be that the condition isn't being met. 
> One way to explore this would be to use a function as the 
> condition, and spit out some loggable data from within the 
> function. Perhaps something like this:
> 
> repeat...
>    ...
>    put "wait" into xosdictionary["current"]
>    --put the waitdepth -- always 1
>    --if the waitdepth < 2 then
>      wait until myCondition() = "continue"  with messages
>    -- else skip... poor idea...
>    --end if
>    ...
> end repeat
> 
> function myCondition
>    put xosdictionary["current"] into tRetVal
>    put the milliseconds && ":" && tRetVal into field "log"
>    return tRetVal
> end myCondition
> 
> In this way, you can see whether the condition is being 
> continually checked or not, and monitor the value that should 
> be changing. At least it should give you an idea of where 
> things are going wrong.
> 
> Cheers
> Dave
>