std.d.lexer - discussion (not the voting thread)

[Walter Bright]

On 10/4/2013 11:32 AM, Jacob Carlborg wrote:
> On 2013-10-04 20:10, Brian Schott wrote:
>
>> "lexer.d is 86% covered"
>
> Do we have a minimum coverage level?

No, but any "low hanging fruit" uncovered lines need to get test cases added, 
i.e. there needs to be some sort of justification for lines not covered.

In general, I'd say we need to be shooting for >= 95%.

If you look at phobos' win32.mak, which lists coverage percentages for the 
various phobos modules, a lot of phobos modules are very inadequately covered.