[R] "try-error" can not be test. Why?

[David Winsemius]

On Sep 8, 2010, at 11:46 PM, Philippe Grosjean wrote:

> On 08/09/10 19:25, David Winsemius wrote:
>>
>> On Sep 8, 2010, at 1:18 PM, telm8 wrote:
>>
>>>
>>> Hi,
>>>
>>> I am having some strange problem with detecting "try-error". From  
>>> what I
>>> have read so far the following statement:
>>>
>>>
>>> try( log("a") ) == "try-error"
>>>
>>>
>>> should yield TRUE, however, it yields FALSE. I can not figure out  
>>> why.
>>> Can
>>> someone help?
>>>
>>
>> > class(try( log("a"), silent=TRUE )) == "try-error"
>> [1] TRUE
>
> This is perfectly correct in this case, but while we are mentioning  
> a test on the class of an object, the better syntax is:
>
> > inherits(try(log("a)), "try-error")
>
> In a more general context, class may be defined with multiple  
> strings (R way of subclassing S3 objects). For instance, this does  
> not work:
>
> > if (class(Sys.time()) == "POSIXct") "ok" else "not ok"
>
> ... because the class of a `POSIXct' object is defined as:  
> c("POSIXt", "POSIXct"). This works:
>
> > if (inherits(Sys.time(), "POSIXct")) "ok" else "not ok"
>
> Alternate valid tests would be (but a little bit less readable):
>
> > if (any(class(Sys.time()) == "POSIXct")) "ok" else "not ok"
>
> or, by installing the "operators" package, a less conventional, but  
> cleaner code:
>
> > install.packages("operators")
> > library(operators)
> > if (Sys.time() %of% "POSIXct") "ok" else "not ok"

I have also used if ("try-error" %in% class(try(log("a"))) ) { } else  
{ }, but the inherits() form looks at the very least less clunky.

>
> Best,
>
> Philippe Grosjean
>
>>> Many thanks
>>> --
>>> View this message in context:
>>> http://r.789695.n4.nabble.com/try-error-can-not-be-test-Why-tp2531675p2531675.html
>>>
>>> Sent from the R help mailing list archive at Nabble.com.
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>
>> David Winsemius, MD
>> West Hartford, CT
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>>