(Bug) Run-time functions can produce wrong results

[IanB]

@Gavino, short answer is it's gone!

It sneaked into the code a while back and I meant to remove it, but forgot. It was simply returning (int32)(frame_number*audioSamplesPerFrame) which is so wrong on so many levels. So R.I.P.

[Gavino]

Originally Posted by gzarkadas

I wonder whether it would worth the effort to make the global vartable being just an alias (for not breaking existing code) of the script-level local vartable and get rid of the global keyword in exchange for a new can of worms .

@gzarkadas: Much as I would like to see changes in the way variable scopes operate in Avisynth, I think this probably is too big a can of worms to open, at least in Avisynth 2.5x.

I think any revision worth doing would be difficult to do a in a 100% backwards compatible way, and it certainly would require at least some changes to the parser and evaluating engine (something I know you have a particular aversion against ).

@IanB: I see we haven't convinced you enough to put the 'global current_frame' change into the latest release.
When does voting close?

Votes so far:
For : 3 (Gavino, gzarkadas, Wilbert)
Against: 1 (IanB)
Abstentions: several 1000's

[Wilbert]

@gzarkadas: Much as I would like to see changes in the way variable scopes operate in Avisynth, I think this probably is too big a can of worms to open, at least in Avisynth 2.5x.

I think any revision worth doing would be difficult to do a in a 100% backwards compatible way, and it certainly would require at least some changes to the parser and evaluating engine (something I know you have a particular aversion against ).

In that case, I think such a change doesn't belong in a release candidate