pvalue_fun in add_pvalue not working

[fedenichetti]

I would like to round the log rank p value to 3 decimals, but when running

test %>% 
 ggsurvfit() +
  #add_confidence_interval() +
 add_risktable(
        theme =   # increase font size of risk table title and y-axis label
      list(
        theme_risktable_default(),
        theme(plot.title = element_text(face = "bold"))
      )
  ) + 
  add_censor_mark(size = 2, alpha = 0.8) +
  add_quantile(y_value = 0.5, linetype = "dotted", color = "grey30", linewidth = 0.5) +
  scale_ggsurvfit() + #add_confidence_interval() +
  scale_color_manual(values = c("#0B3F54","#D14055")) +  
  add_pvalue(location  = "annotation", x = 10, pvalue_fun = ~ style_pvalue(.x, digits = 3))

Error in `update_add_pvalue()`:
! The `add_pvalue(pvalue_fun=)` argument must be a function.
Run `rlang::last_trace()` to see where the error occurred.

How should I solve this? Thanks!

[ddsjoberg]

Dear @fedenichetti ,

The error message says that add_pvalue(pvalue_fun=) argument must be a function, and you've passed a formula. There are some tidyverse functions that convert that formula notation to a function, but that does not occur here. Try this:

add_pvalue(pvalue_fun = function(.x) style_pvalue(.x, digits = 3))

FYI GitHub is primarily a space to report bugs and make feature requests. For assistance with the package in the future, it's best to post on stackoverflow.com. When you do post, please include a minimal reproducible example. https://reprex.tidyverse.org/

[fedenichetti]

Many thanks and sorry for the inappropriate post. Best,