[R] else problem

[David Winsemius]

On Jun 24, 2011, at 7:47 AM, Petr PIKAL wrote:

> Hi
>>
>> Thank you for all your help! I did not know to use "" when  
>> searching for
>
>> help, as ?mean, etc, had always worked for me in the past.
>> It makes perfect sense why 'else' was causing me the trouble the  
>> way I
> was
>> using it. I think it was working in my other code, despite the same
>> format, because it was part of a function and thus would have been
>> executed completely within the function?
>
> Yes and no
>
> if you made your function like that
>
> fff<-function(x)
> if (x>0) print("A")
> else print("B")
>
> you would get immediate error
>
>> else {
> Error: unexpected 'else' in "else"
>
> So you probably enclosed whole function into brackets like
>
> fff<-function(x) {
> if (x>0) print("A")
> else print("B")
> }

If you successfully get the code into a function there will be no  
error. I suspect your demonstration of the error was at the console.  
If you use Bert's "brace-enclosure" method to delay evaluation until  
the full function is entered you get a working function that has no  
braces in its body (as I showed earlier):

 > {fntest <- function() if (FALSE)  print("T")
+                         else  print("F") }
 >  fntest
function() if (FALSE)  print("T")
                         else  print("F")
 > fntest()
[1] "F"

-- 
David.

>
> and in that case you did not experienced this else problem not  
> because it
> is a function but because you used those brackets.
>
> Regards
> Petr
>
>
>
>> Cheers,
>> Kara
>> ________________________________________
>> From: Bert Gunter [gunter.berton at gene.com]
>> Sent: June 23, 2011 8:27 AM
>> To: David Winsemius
>> Cc: Kara Przeczek; r-help at r-project.org
>> Subject: Re: [R] else problem
>>
>> Perhaps some additional clarification...  (???)
>>
>>>> if (length(dat2f$year)%%2==0) {
>>>> md <-dat2f[, list(med_year = max(year[which(abs(tot_km3y -
>>>> median(tot_km3y)) == min(abs(tot_km3y - median(tot_km3y)))) ]),
> med_TotQ =
>>>> median(tot_km3y))]  }
>>>> else {
>>>
>>> If this line is executed at a console session it will fail because  
>>> the
>>> interpreter does not keep a copy of the last condition. If you moved
> the
>>> closing curley-brace to just befor the 'else', you should get the
> behavior
>>> you expect (if your other code is correct):
>>
>> Well, um.. not sure if this is what you meant, but what is happening
>> at the console is that when you type <return>, the interpreter checks
>> for a syntactically complete statement. If it finds what has been
>> given to it **thus far** is, it tries to execute it (if not, it gives
>> a continuation character and waits for more input) and, as you said,
>> then starts anew to interpret the next line(s) entered, "forgetting"
>> all previous. The problem above is that the" if()" statement up to  
>> the
>> close bracket, "}" is syntactically complete, and so the "else{" that
>> follows makes no sense as the beginnig of a new line to  be
>> interpreted.
>>
>> The simplest and universal solution to this is to simply enclose the
>> whole conditional in" { }":
>>
>> {if(length ...
>> ...
>> else {...}
>> }
>>
>> This forces the interpreter to wait for the last "}" before it will
>> interpret and execute.
>>
>> Hoping this clarifies rather than obfuscates.
>>
>> -- Bert
>>
>>
>>
>>>
>>> Try instead:
>>>
>>> if (length(dat2f$year)%%2==0) {
>>> md <-dat2f[, list(med_year = max(year[which(abs(tot_km3y -
>>> median(tot_km3y)) == min(abs(tot_km3y - median(tot_km3y)))) ]),
> med_TotQ =
>>> median(tot_km3y))]
>>> } else {
>>> md <-dat2f[, list(med_year = year[which(tot_km3y ==
> median(tot_km3y))],
>>> med_TotQ = median(tot_km3y))]
>>> }
>>>
>>>
>>>> md <-dat2f[, list(med_year = year[which(tot_km3y ==
> median(tot_km3y))],
>>>> med_TotQ = median(tot_km3y))]
>>>> }
>>>>
>>>> Each individual piece works perfectly on its own, but together I  
>>>> get
> the
>>>> following error:
>>>>
>>>>> if (length(dat2f$year)%%2==0) {
>>>>
>>>> +   md <-dat2f[, list(med_year = max(year[which(abs(tot_km3y -
>>>> median(tot_km3y)) == min(abs(tot_km3y - median(tot_km3y)))) ]),
> med_TotQ =
>>>> median(tot_km3y))] }
>>>>>
>>>>> else {
>>>>
>>>> Error: unexpected 'else' in "else"
>>>>>
>>>>> md <-dat2f[, list(med_year = year[which(tot_km3y ==
> median(tot_km3y))],
>>>>> med_TotQ = median(tot_km3y))]
>>>>> }
>>>>
>>>> Error: unexpected '}' in "  }"
>>>>>
>>>>
>>>> When I tried to look up "else" I got this error:
>>>>
>>>>> ?else
>>>>
>>>> Error: unexpected 'else' in "?else"
>>>
>>> Try instead:
>>>
>>> ?"else"
>>>
>>> --
>>> David
>>>>
>>>> I have used exactly the same set up with if...else in other code  
>>>> and
> it
>>>> worked fine then. I tried to run it again, and I got the same error
> as
>>>> above. What is the problem? I hope it isn't something simple and
> silly!
>>>>
>>>> I realize that I can use the first line:
>>>>
>>>> md <- dat2f[, list(med_year = max(year[which(abs(tot_km3y -
>>>> median(tot_km3y)) == min(abs(tot_km3y - median(tot_km3y)))) ]),
> med_TotQ =
>>>> median(tot_km3y))]
>>>>
>>>> for all data sets and it will give me the median for both odd and
>>>> even-length data sets, but it is now about the principle; why won't
> the
>>>> if...else work?
>>>>
>>>> Thank you very much for your time!
>>>>
>>>> Kara
>>>
>>> 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.
>>>
>>
>>
>>
>> --
>> "Men by nature long to get on to the ultimate truths, and will often
>> be impatient with elementary studies or fight shy of them. If it were
>> possible to reach the ultimate truths without the elementary studies
>> usually prefixed to them, these would not be preparatory studies but
>> superfluous diversions."
>>
>> -- Maimonides (1135-1204)
>>
>> Bert Gunter
>> Genentech Nonclinical Biostatistics
>>
>> ______________________________________________
>> 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.
>
> ______________________________________________
> 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