Skip to content

ls() pattern question

6 messages · Andrew Simmons, Kai Yang, Duncan Murdoch +1 more

Kai Yang
# 
Hello List,
I have many data frames in environment.? I need to keep 3 data frames only, con DB2 and ora.?
I write the script to do this.?
rm(ls(pattern != c("(con|DB2|ora)")))


but it give me an error message:


Error in rm(ls(pattern != c("(con|DB2|ora)"))) :?
? ... must contain names or character strings

I think the pattern option doesn't support != ? and is it possible to fix this?
Thank you,
Kai
Andrew Simmons
# 
Hello,


First, `ls` does not support `!=` for pattern, but it's actually throwing a
different error. For `rm`, the objects provided into `...` are substituted
(not evaluated), so you should really do something like

rm(list = ls(pattern = ...))

As for all except "con", "DB2", and "ora", I would try something like

setdiff(ls(), c("con", "DB2", "ora"))

and then add `rm` to that like

rm(list = setdiff(ls(), c("con", "DB2", "ora")))

On Wed, Jul 14, 2021 at 7:41 PM Kai Yang via R-help <r-help at r-project.org>
wrote:

  
  


  


      

    

    

      
      

        


  

        

      Bert Gunter
    

    

      
      #
    

  


  

      
Actually fun( param != something..) is syntactically incorrect in the first
place for any function!

ls sees "pat != whatever"  as the "name" argument of ls() and can't make
any sense of it, of course.

Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )

        
On Wed, Jul 14, 2021 at 5:01 PM Andrew Simmons <akwsimmo at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Kai Yang
    

    

      
      #
    

  


  

      
Thanks Andrew. it works well. --- Kai

        
On Wednesday, July 14, 2021, 05:22:01 PM PDT, Bert Gunter <bgunter.4567 at gmail.com> wrote:

        
Actually fun( param != something..) is syntactically incorrect in the first place for any function! 

ls sees "pat != whatever"? as the "name" argument of ls() and can't make any sense of it, of course. 
 
Bert Gunter

"The trouble with having an open mind is that people keep coming along and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )

        
On Wed, Jul 14, 2021 at 5:01 PM Andrew Simmons <akwsimmo at gmail.com> wrote:

        
Hello,


First, `ls` does not support `!=` for pattern, but it's actually throwing a
different error. For `rm`, the objects provided into `...` are substituted
(not evaluated), so you should really do something like

rm(list = ls(pattern = ...))

As for all except "con", "DB2", and "ora", I would try something like

setdiff(ls(), c("con", "DB2", "ora"))

and then add `rm` to that like

rm(list = setdiff(ls(), c("con", "DB2", "ora")))

On Wed, Jul 14, 2021 at 7:41 PM Kai Yang via R-help <r-help at r-project.org>
wrote:

            

      
      
      
    

        

      
      
    

        
? ? ? ? [[alternative HTML version deleted]]

______________________________________________
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.

  
  


  


      

    

    

      
      

        


  

        

      Duncan Murdoch
    

    

      
      #
    

  


  

      
On 14/07/2021 8:21 p.m., Bert Gunter wrote:

            

      
      
      
    

        
You have to be careful with absolute statements:

   > f <- function(pattern) cat("It's legal!")
   > f(pattern != something..)
   It's legal!

"pattern != something.." is a legal expression, which would return a 
logical vector.  My function didn't ever evaluate it, so the fact that 
variables named "pattern" and "something.." didn't exist in the global 
env didn't make any difference.  It's legal syntax, just wrong.

Duncan Murdoch

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Bert Gunter
    

    

      
      #
    

  


  

      
Yes,. Thanks for the clarification. I tried to indicate this with my
comment about it being used as the "name" argument of ls().
What I should have said is that it is syntactically wrong as a way to
specify an argument of any function -- "myarg != something" does not
specify anything about the argument "myarg" of a function not equalling
something --  unless, as you pointed out, it positionally matches to
"myarg" and evaluates to something. But that was certainly not the OP's
intent, and I thought it was worth pointing out that he appeared to be
confused about the syntax of argument specification in general.


Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Thu, Jul 15, 2021 at 2:09 AM Duncan Murdoch <murdoch.duncan at gmail.com>
wrote: