Strange problem caused by extra space characters

[Paul R. Jackson]

I would say its just a slight oversight. A similar problem occurs if you set
correct response to say 'aA' i.e. meaning that the correct response is
capital or lower case 'A'. In this situation eprime will never get a correct
response because eprime is waiting for the 'aA' key!!

These sorts of things are all part of the joys of computer programming! ;)

Paul

~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Paul R. Jackson
 Experimental Programmer

 School of Psychology
 University of Queensland
 E:paulj at psy.uq.edu.au
 P:3365-6713
~~~~~~~~~~~~~~~~~~~~~~~~~~~


> -----Original Message-----
> From: eprime at mail.talkbank.org
> [mailto:eprime at mail.talkbank.org] On Behalf Of Harvey Shulman
> Sent: Friday, 30 July 2004 4:01 AM
> To: eprime at mail.talkbank.org
> Subject: Strange problem caused by extra space characters
>
> I wonder if this is a well known problem to everyone but me ?
>  We created a list structure that included our stimulus words
> and an attribute named corr_resp that contained the required
> keypress for each
> stimulus.   The two values used for corr_resp are {UPARROW}
> and {DOWNARROW}.
>
> We then specified in the textdisplay that presented the
> stimulus that the allowable responses were
> {UPARROW}{DOWNARROW} and the correctresponse was [corr_resp].
>
> This worked just fine until we added some new stimuli and
> then the program would hit a run-time error with the message that the
> correctresponse was not in the allowable set.   This could be
> fixed by
> changing the allowable set to {ANY} but why was it happening ?
>
> The answer turned out to be that in adding new stimuli to our
> list we had sometimes entered the value of corr_resp as
> {UPARROW}__  where __ was just one or two (invisible) space
> characters after the bracket.
> Erasing them fixed the problem.
>
> I find this a very strange 'feature' in EPRIME.  The trailing
> spaces are being interpreted as part of the code for a
> correct response.  Is there a benefit to this feature or is
> it just an oversight in parsing the correct response field in
> the text display ?
>
> --
> Dr. Harvey G. Shulman
> Department of Psychology
> Ohio State University
> Columbus, OH 43210
> fax 614 688-3984  ph 614 292-2759
>