Within groups randomization and sampling without replacement

I've looked through all the DirectRT posts and haven't been able to find an answer to my, ostensibly, easy problem.

I am presenting participants with 5 good words (e.g., peace) and 5 bad words (e.g., nasty) for a total of 20 trials (each good and bad word is presented twice). What I would like to accomplish is presenting the good and bad words randomly and for all of the good and bad words to be presented once before they are repeated a second time. I have used the same within groups randomization (wgr) for all 20 trials and then included $good.txt and $bad.txt in the stim column. This does not produce the intended sampling without replacement; I’ve tried a couple other variations, but nothing has worked when wgr is used (it is more critical to have the good and bad words presented in random order).

Have you tried assigning the first set of 10 (5 good and 5 bad) a WGR value of 1 and the second set a WGR value of 2?

Just tried assigning first 10 (5 good and 5 bad) to WGR 1 and second 10 to WGR 2. It cut down on the number of times the same stimulus was presented back-to-back, but it did not cause all of the first 5 good words to be presented before a good word was repeated.

Can you attach your input file and stimlist(s) (if any)? I'd be happy to take a look.