[R-sig-ME] predict.merMod with random slopes - revisiting an older issue?

[Robert Zimbardo]

Hi all

I am writing with regard to an issue that came up here before and that
I thought was resolved; here are the links:

https://github.com/lme4/lme4/issues/153
https://stat.ethz.ch/pipermail/r-sig-mixed-models/2013q4/021284.html

Specifically, I have a very simple model

m <- lmer(y~x+(x|i)

Predictions without random effects, no problem:
predict(m, newdata=data.frame(i="3", x=0.2), re.form=NA) # works perfectly

But trying to get predictions with the random effects returns an error
know from the above links
predict(m, newdata=data.frame(i="3", x=0.2), re.form=NULL) # returns
"Error: sum(nb) == q is not TRUE"

I'm using RRO 8.0.1.beta, RStudio 0.99.408, lme4 1.1-7 on Win7 x64.

Any ideas?