Re: Output format for p-values

[Ben Pfaff]

On Tue, Apr 29, 2014 at 07:08:33PM +0200, John Darrington wrote:
> On Thu, Apr 24, 2014 at 08:58:20PM -0700, Ben Pfaff wrote:
>      On Tue, Apr 22, 2014 at 08:40:22AM +0200, John Darrington wrote:
>      > Perhaps we could change tab_double and tab_value to take a "reported 
> value class", say
>      > 
>      > enum reported_value_class {
>      >  WEIGHT,
>      >  INTEGER,
>      >  P_VALUE,
>      >  OTHER
>      > };
>      > 
>      > and then have a map which associates each class with a function to 
> determine how it should
>      > be displayed.  We would then need to provide a means for the user to 
> change the mapping 
>      > as he desires.
>      > 
>      > Such a solution should be regarded as intermediate, until we get a 
> better abstraction of
>      > our data output, but might be solve the immediate problem without too 
> much effort.
>      
>      I think that's a fine way to start.  If you have time and inclination
>      to work on it, then it makes sense to me.
> 
> Here's a first patch.  It adds the report_class parameter to tab_double and 
> updates all callers.
> It then provides a map from classes to format.  I've set the format for 
> p-values to 5.3
> I have also updated all the affected tests.  I have also removed tab_fixed 
> since it had few callers
> and did not provide much benefit over tab_double.
> 
> The limitation at the moment is that there is no way to override the 5.3 
> choice if you want more than
> three decimal places.  That needs to come later.

This patch looks like a nice improvement to the current state of
affairs.