small issue: prettify digits problem in CI upper about paper

[hofnerb]

First of all sorry that I didn't reply earlier. I didn't receive any notification.

Unfortunatley, I cannot check your example as there is no (reproducible) code. Could you please provide a minimal working example which displays the behavior? Thx.

[likelet]

       # Load data set ovarian
        data(ovarian, package = "survival")

        ## fit a Cox model
        mod5 <- coxph(Surv(futime, fustat) ~ age, data=ovarian)
        summary(mod5)
        ## Make pretty summary
        prettify(summary(mod5),digit=3)

          coef    Hazard Ratio CI (lower) CI (upper) se(coef)    z Pr(>|z|)    
1 age 0.162         1.18         1.07        **1.3**   0.0497 3.25    0.001  **

you can find CI(upper) has less digits

[hofnerb]

There is some strange behavior of format, which is (at least part of) the underlying problem.

Using

format(exp(confint(mod5)), digits = 3)

returns the correct results:

##     2.5 %  97.5 %
## age "1.07" "1.30"

Yet, if one runs

debugonce("prettifyPValue")
prettify(summary(mod5), digits=3)
format(object, digits = 3)

one gets

       coef Hazard Ratio CI (lower) CI (upper) se(coef)    z Pr(>|z|)
1 age 0.162         1.18       1.07        1.3   0.0497 3.25  0.00116

i.e. the 0 is again removed in CI (upper).

The reason is that format.data.frame applies format by columns. I need to check how to fix this.

[hofnerb]

So in short, the real issue is that trailing zeros are droped despite not requested to do this:

format(1.299, digits = 3, drop0trailing = FALSE)

The option is simply ignored.