top_n crashes

[burkovae]

Recently I was experimenting with top_n and this (wrong) application of top_n does fry all R sessions:

data.frame(x = 1:10) %>% top_n(10, "x")

 *** caught segfault ***
address 0x20000010, cause 'memory not mapped'

Traceback:
 1: .Call("dplyr_filter_impl", PACKAGE = "dplyr", df, dots)
 2: filter_impl(.data, dots)                                                                                                                                                                     
 3: filter_.tbl_df(tbl_df(.data), .dots = dots)                                                                                                                                                  
 4: filter_(tbl_df(.data), .dots = dots)                                                                                                                                                         
 5: as.data.frame(filter_(tbl_df(.data), .dots = dots))                                                                                                                                          
 6: filter_.data.frame(.data, .dots = lazyeval::lazy_dots(...))                                                                                                                                  
 7: filter_(.data, .dots = lazyeval::lazy_dots(...))                                                                                                                                             
 8: filter(x, min_rank(desc("x")) <= 10)                                                                                                                                                         
 9: eval(expr, envir, enclos)                                                                                                                                                                    
10: eval(call)                                                                                                                                                                                   
11: top_n(., 10, "x")                                                                                                                                                                            
12: function_list[[k]](value)                                                                                                                                                                    
13: withVisible(function_list[[k]](value))                                                                                                                                                       
14: freduce(value, `_function_list`)                                                                                                                                                             
15: `_fseq`(`_lhs`)                                                                                                                                                                              
16: eval(expr, envir, enclos)
17: eval(quote(`_fseq`(`_lhs`)), env, env)
18: withVisible(eval(quote(`_fseq`(`_lhs`)), env, env))
19: data.frame(x = 1:10) %>% top_n(10, "x")

[krlmlr]

Thanks, confirmed.