
To get even more heretical, a quick grep of the compiler source tree does
not show
SrcSpanOneLine / SrcSpanMultiLine / SrcSpanPoint
being used anywhere but in SrcLoc.lhs.
The use in SrcLoc is in the original constructor, the deconstructors
pulling out lines and columns, and combineSrcSpans.
Perhaprs our SrcSpan is more complicated than it needs to be?
On Mon, Oct 6, 2014 at 3:37 PM, Matthias Fischmann
Ah, you want to drop constructor field names. Then I withdraw my reservations.
thanks for the clarification, m.
On Mon, Oct 06, 2014 at 03:27:49PM +0200, Alan & Kim Zimmerman wrote:
Date: Mon, 6 Oct 2014 15:27:49 +0200 From: Alan & Kim Zimmerman
To: Matthias Fischmann Cc: "ghc-devs@haskell.org" Subject: Re: Show instance for SrcSpan Not at all, just show as e.g.
SrcSpan (RealSrcSpan (SrcSpanOneLine "./foo.hs" 4 1 6))
We just avoid showing the srcSpanFile / srcSpanLine / srcSpanSCol / srcSpanECol noise
On Mon, Oct 6, 2014 at 3:16 PM, Matthias Fischmann
wrote: On Mon, Oct 06, 2014 at 02:59:03PM +0200, Alan & Kim Zimmerman wrote:
Date: Mon, 6 Oct 2014 14:59:03 +0200 From: Alan & Kim Zimmerman
To: "ghc-devs@haskell.org" Subject: Show instance for SrcSpan Is there any reason I can't put in a diff request to replace the derived Show instance for SrcSpan with a handcrafted one that does not exhausively list the constructors, making it more readable?
Alan
I like the notion that `Read . Show` should always work. Dropping parts of a datatype would break that rule.
matthias _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs