builder-generated scripts should not /initialise/ components with thisTrial['blah'] about psychopy HOT 4 CLOSED

I didn't explain my concern about this very well - the downside is the code for working out how to write the script will get rather more confusing (even though it should be reasonably easy to write first time) and maybe it isn't worth having confusing internal code for a non-critical feature like this.

from psychopy.

I am not completely sure if its related, but it might be useful in this situation (which came up for me and also someone else at the workshop): if you set a component's value (via a code snippet in a field in a builder window) to $value at the start of a routine, you get an error at initialization time (because the variable is unknown at that point). the non-obvious work around I used was to set value=None at the start of the experiment via a code component (so that variable "value" would be defined, even if its value was meaningless). so it sounds like only expecting things to be defined at the start of the routine would be good?

from psychopy.

Yes, this is the same issue, and I've also been seeing it more and more as an issue. You got the correct workaround too, but it's kind of ugly. I've promoted the issue to 3 ticks.

from psychopy.

Done (commit fe253ea)

from psychopy.