[R] Problem with geterrmessage()

[Martin Maechler]

>>>>> Dennis Fisher <fisher at plessthan.com>
>>>>>     on Thu, 22 Feb 2018 13:01:37 -0800 writes:

    > Luke
    > Thanks — I revised the code to:
    > ERRORMESSAGE <- try(source(USERSCRIPTFILE, local=T), silent=T) 

    > print(ERRORMESSAGE) now returns:
    > $value
    > [1] 0

    > $visible
    > [1] FALSE

    > Not clear what to make of that.
    > Dennis

The help page   ?try   has

contained for a long time

     ‘try’ is implemented using ‘tryCatch’; for programming, instead of
     ‘try(expr, silent = TRUE)’, something like ‘tryCatch(expr, error =
     function(e) e)’ (or other simple error handler functions) may be
     more efficient and flexible.

and you do use 'silent=T' (which is "unsafe" (*) !)

I'd strongly advocate you use the 2nd proposition by given by
Luke Tierney (cited below):
 Learn to use tryCatch() instead of try() and then such things
 can be done considerably less obscurely. 

Best,
Martin Maechler, ETH Zurich

--
*) Using 'T' instead of 'TRUE'  (of 'F' instead of 'FALSE' *is* unsafe):
   a previous  'T <- 0'  will change what you really wanted.
   TRUE and FALSE are  "constants" in R, whereas  T and F are variables


    > Dennis Fisher MD
    > P < (The "P Less Than" Company)
    > Phone / Fax: 1-866-PLessThan (1-866-753-7784)
    > www.PLessThan.com




    >> On Feb 22, 2018, at 12:45 PM, luke-tierney at uiowa.edu wrote:
    >> 
    >> Only the default error handler puts the error message in a buffer
    >> where it can be retrieved with geterrmessage. try() replaces the
    >> default error handler. Either look at the value returned by try() or
    >> use tryCatch with conditionMessage.
    >> 
    >> Best,
    >> 
    >> luke
    >> 
    >> On Thu, 22 Feb 2018, Dennis Fisher wrote:
    >> 
    >>> R 3.4.3
    >>> OS X
    >>> 
    >>> Colleagues
    >>> 
    >>> I have a 20K line script in which I encounter an unexpected problem.
    >>> 
    >>> If the script detects presence of a particular file USERCODE.txt, it executes:
    >>> source(“USERCODE.txt”)
    >>> If that file is not present, the script executes without a problem.
    >>> 
    >>> There might be syntax errors in USERCODE.txt; therefore, the code above is embedded in a try command:
    >>> try(source(“USERCODE.txt", local=T), silent=T)
    >>> followed by:
    >>> ERRORMESSAGE <- geterrmessage()
    >>> 
    >>> For unclear reasons, an earlier command is yielding an error message:
    >>> unused argument (\"\\n\")
    >>> Despite identifying the exact source of that error, I can’t fix it (and it is of no consequence).
    >>> 
    >>> Ideally, I would like to clear out the pre-existing error message immediately before the “try” command (or perhaps at that particular location where it is being created) — but I can’t figure out how to do so.
    >>> 
    >>> Any suggestions would be welcome.
    >>> 
    >>> Dennis
    >>> 
    >>> Dennis Fisher MD
    >>> P < (The "P Less Than" Company)
    >>> Phone / Fax: 1-866-PLessThan (1-866-753-7784)
    >>> www.PLessThan.com
    >>> 
    >>> ______________________________________________
    >>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
    >>> 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.
    >>> 
    >> 
    >> -- 
    >> Luke Tierney
    >> Ralph E. Wareham Professor of Mathematical Sciences
    >> University of Iowa                  Phone:             319-335-3386
    >> Department of Statistics and        Fax:               319-335-3017
    >> Actuarial Science
    >> 241 Schaeffer Hall                  email:   luke-tierney at uiowa.edu
    >> Iowa City, IA 52242                 WWW:  http://www.stat.uiowa.edu

    > ______________________________________________
    > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
    > 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.