Kazuki Yoshida

```sh _R_CHECK_LENGTH_1_LOGIC2_="abort,verbose"; export _R_CHECK_LENGTH_1_LOGIC2_ ; make test ``` ``` ... Unit tests for svy* user functions : svyCprod(estfun %*% Ainv, design$strata, design$cluster[[1]], design$fpc, design$nPSU, design$certainty, design$postStrata) where 2: svy.varcoef(m, design)...

Skipping the following two test files avoids the issue. ``` test-modules-svy.R test-svyCreateTableOne.R ```

Will proceed with the above temporary solution (disabling tests that call `survey`) for now.

You need to convert it to data_frame. A pull request would be appreciated.

This has been on my TODO list for a while. I would appreciate your contribution.

It is not supported. I do not think there is a well-established way to present SMD after MI. You could run tableone with each imputed dataset and average SMD.

Hi, I'm sorry for the late response. Which platform are you on? Another report of a similar issue was filed by a Windows user. It may be the same issue,...

Hi, I'm sorry for the late response. What happens if you ```M-:``` (or ```M-x eval-expression```) and then type in ```(shell)``` in the mini-buffer and hit RET? ```eval-in-repl-shell``` calls ```shell``` behind...

Is "a single unified front end and implementation hiding" what you mean? I mean currently we need to configure each specific implementation and call it directly. This surely goes against...

Thanks for your interest in the package. I wasn't aware of this. Deleting the blank line before reaching the inferior python process may be a quick fix. I'm currently busy,...