[R] How to replace NAs in a vector of factors?

[Rolf Turner]

On 22/07/2009, at 1:21 PM, jim holtman wrote:

> Notice that three items are returned where you thought one was:
>
> [1] FALSE  TRUE  TRUE  TRUE FALSE
>> db1$olditems[db1$olditems=='']  #wait, only one item is returned?
> [1]
> Levels:  nuts soup
>> db1[db1$olditems=='',]  #somehow this works!
>   olditems prices
> 2            3.25
> 3            4.42
> 4            2.25
>> paste('[',db1$olditems[db1$olditems==''],']') # put some  
>> characters around return values
> [1] "[  ]" "[  ]" "[  ]"
>>
>
>
> The '[1]' was just an indication that this is the first value returned
> and the other two were blanks so you did not see them.  Also "" is
> just that; a blank and not NA.

	Strictly speaking --- or maybe even not so strictly --- a ***null  
character***,
	not a blank.  "" != " "

		cheers,

			Rolf Turner

	<snip>

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}