Update text_summary() to support fixed designs from gs_design_ahr()

[jdblischak]

When a fixed design is selected in the app, the following error appears below the summary table:

Error: $ operator is invalid for atomic vectors

This is because text_summary() is expecting a function when determining the upper bound.

gsDesign2/R/text_summary.R

Line 192 in 5103d9c

sfu <- get_sf(x$input$upar$sf)

A fixed design has no upper bound, so I fixed this by conditioning on n_analysis > 1.

Now the text summary for a fixed design works as below:

One-sided group sequential design with 1 analyses, time-to-event outcome with sample size 1124 and 865 events, percent (1-sided) Type I error. Enrollment and total study durations are assumed to be 24 and 30 months, respectively. With hazard ratio of 1 during the first 3 months and 0.7 thereafter, the power is 90 percent.

library(gsDesign)
library(gsDesign2)

# Enrollment
enroll_rate <- define_enroll_rate(duration = c(2, 2, 2, 18), rate = c(1, 2, 3, 4), stratum = "All")

# Failure rates
duration <- diff(c(0, 3, 103))
control_rate <- log(2)/c(6, 6)
fail_rate <- define_fail_rate(duration, fail_rate = control_rate, dropout_rate = c(0.001, 0.001), hr = c(1, 0.7), stratum = "All")

x <- gs_design_ahr(
  enroll_rate = enroll_rate,
  fail_rate = fail_rate,
  alpha = 0.025,
  beta = 0.1,
  analysis_time = 30,
  ratio = 1,
  binding = FALSE,
  upper = "gs_b",
  upar = qnorm(1 - 0.025),
  lower = "gs_b",
  lpar = -Inf,
  h1_spending = TRUE,
  test_lower = FALSE,
  info_scale = "h0_info"
) |>
  to_integer()

text_summary(x)
## [1] "One-sided group sequential design with 1 analyses, time-to-event outcome with sample size 1124 and 865 events,  percent (1-sided) Type I error. Enrollment and total study durations are assumed to be 24 and 30 months, respectively. With hazard ratio of 1 during the first 3 months and 0.7 thereafter, the power is 90 percent."

[yihui]

Looks good to me!