Add info as the output of sim_gs_n()

[LittleBeannie]

See issue #259 .

[keaven]

Looks good

[jdblischak]

The title of this PR is confusing. It looks like you added the columns info and info0 only to the test function wlr(). This has broken the sim_gs_n() feature of using different tests per cut. What is the timeline for adding info and info0 columns to rmst() and milestone()? Can we create an Issue for this?

[LittleBeannie]

Adding info and info0 to rmst and milestone will be a very future plan. I am wondering if sim_gs_n could bind_rows results from different tests. For example,

This bind_rows will also help with adding maxcombo (instead of reporting Z score, we report p-values) to sim_gs_n.

[jdblischak]

Good idea. We can filling missing columns with NAs. {dplyr} is not a dependnecy of {simtrial}, but {data.table} is, and it has an equivalent function.

library("dplyr")
library("tibble")
df1 <- tibble(x = 1:2, y = letters[1:2])
df2 <- tibble(x = 4:5, z = 1:2)
r1 <- bind_rows(df1, df2)

rbind(df1, df2)
## Error in match.names(clabs, names(xi)) :
##   names do not match previous names

library("data.table")
r2 <- data.table::rbindlist(list(df1, df2), use.names = TRUE, fill = TRUE)

identical(as.data.frame(r1), as.data.frame(r2))
## [1] TRUE

[LittleBeannie]

Shall we create a issue for this?

[jdblischak]

Shall we create a issue for this?

I already have RStudio open, so I'm going to attempt to do it right now