Re: [R] 'autocompletion' of named elements of a list

2011-03-11 Thread Peter Ehlers
On 2011-03-11 01:07, Ivan Calandra wrote: Hi, From ?"$", you can see that using [[ instead would do what you're looking for. You should read and try to understand the whole help file. The reason is that for [[ the default is exact=TRUE, wheareas for $ the only possible value is exact=FALSE, wh

Re: [R] 'autocompletion' of named elements of a list

2011-03-11 Thread peter dalgaard
On Mar 11, 2011, at 09:55 , Håvard Rue wrote: > > since NULL elements are removed from the list, and that A$a auto-expand > to A$aa, my error appeared. > > To me, this seems not a feature I want in order to have a robust > program. > Partial matching of arguments and list item names is

Re: [R] 'autocompletion' of named elements of a list

2011-03-11 Thread Ivan Calandra
Hi, From ?"$", you can see that using [[ instead would do what you're looking for. You should read and try to understand the whole help file. The reason is that for [[ the default is exact=TRUE, wheareas for $ the only possible value is exact=FALSE, which means partial matching if possible.

[R] 'autocompletion' of named elements of a list

2011-03-11 Thread Håvard Rue
I had an obscure bug that boiled down to this ``feature'' in R, Read 3921 items > A = list(aa = 1) > A $aa [1] 1 > if (A$a) print("a is there") [1] "a is there" The test appear to check is A$a is TRUE, but what happen is that it auto-complete (silently), and expand to 'A$aa'. The problem was