`forestplot()` fails or drops boxes when estimates are negative

[rempsyc]

Description

When using forestplot() (version 3.1.7), the function behaves inconsistently when estimates are negative.

Specifically:

1. When coefficients are made negative but CIs remain positive, the box is not drawn (only CI lines appear).
2. When both coefficients and CIs are negative, the function throws an internal error:

Warning in min(cwidth[cwidth > 0], na.rm = TRUE):
  no non-missing arguments to min; returning Inf

Error in if (any(textHeight * (nrow(estimates) + 0.5) * 1.5 < boxsize)) :
  missing value where TRUE/FALSE needed

packageVersion("forestplot")
#> [1] '3.1.7'

library(forestplot)
#> Loading required package: grid
#> Loading required package: checkmate
#> Loading required package: abind

# A basic example, create some fake data
row_names <- list(list("test = 1", expression(test >= 2)))
test_data <- data.frame(
  coef = c(1.59, 1.24),
  low = c(1.4, 0.78),
  high = c(1.8, 1.55)
)
forestplot(test_data,
           labeltext = row_names,
           mean = coef,
           lower = low,
           upper = high)

# Make coefficient negative
test_data$coef <- test_data$coef * -1

forestplot(test_data,
           labeltext = row_names,
           mean = coef,
           lower = low,
           upper = high)

# Make CI negative
test_data$low <- test_data$low * -1
test_data$high <- test_data$high * -1

forestplot(test_data,
           labeltext = row_names,
           mean = coef,
           lower = low,
           upper = high)

#> Warning in min(cwidth[cwidth > 0], na.rm = TRUE): no non-missing arguments to
#> min; returning Inf
#> Error in `if (any(textHeight * (nrow(estimates) + 0.5) * 1.5 < boxsize)) ...`:
#> ! missing value where TRUE/FALSE needed

^{Created on 2026-02-24 with reprex v2.1.1}