#3416: Make maximumBy strict on the contents of its list
---------------------------------+------------------------------------------
    Reporter:  bdonlan           |        Owner:                  
        Type:  proposal          |       Status:  reopened        
    Priority:  normal            |    Milestone:                  
   Component:  libraries/base    |      Version:  6.10.4          
    Severity:  normal            |   Resolution:                  
    Keywords:                    |   Difficulty:  Unknown         
    Testcase:                    |           Os:  Unknown/Multiple
Architecture:  Unknown/Multiple  |  
---------------------------------+------------------------------------------
Changes (by bdonlan):

  * status:  closed => reopened
  * type:  bug => proposal
  * resolution:  wontfix =>
  * summary:  maximumBy uses too much stack => Make maximumBy strict on the
              contents of its list

Comment:

 Is being able to use maximumBy on a constant comparison function
 considered to be the common case here? Is that even _useful_?
 It's true that the Haskell Report defines 'maximum' as being non-strict on
 its arguments. This is arguably a bug in the Report, however.

 Making maximumBy strict on the list elements does indeed change the
 behavior of maximumBy, but on most reasonable programs, this should have
 no visible effect. It is hard to imagine many cases in which a lazy
 version of maximumBy is necessary and desirable, and if it is, it's simple
 enough to create one.

 Note also that this issue has confused newbies to haskell before:
 http://stackoverflow.com/questions/1238883/running-a-compiled-haskell-
 program-getting-errors

 Why shouldn't we change the behavior of maximumBy (and similar functions
 such as minimumBy etc) to be strict on its arguments?

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/3416#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

Reply via email to