How to write truncated distribution in flist?

[ctfysh]

In Stan code, there are truncated distributions, like

y ~ normal(0, 1) T[-0.5, 2.1];

So how can this be put in the flist for ulam?

[rmcelreath]

This can be done with custom(), but it's not really a supported feature.

[ctfysh]

I have tried to do this with custom(), it is useful for vectors (like observations), but not for scalars, like the following sample in Stan reference manual.

y ~ normal(0, 1);
if (y < -0.5 || y > 2.1) {
  target += negative_infinity();
} else {
  target += -log_diff_exp(normal_lcdf(2.1 | 0, 1),
                          normal_lcdf(-0.5 | 0, 1));
}

I don't know if I miss someting, or it's not supported. Thanks!