Re: Controlling test coverage

> 
> I’ve stumbled upon the following two entries…
> 
>   <tests change="fn-codepoints-to-string"
>          count="2"
>          description="Function fn:codepoints-to-string becomes variadic (TODO???)"/>

The tests being counted are K4-codepoints-to-string-90 and K4-codepoints-to-string-91. 

They were originally written in anticipation of issue #161, which has never been implemented, and in their current form they are just testing that the feature isn't implemented, awaiting a decision on the issue.
> 
> Which test cases are counted here? Is this entry related to the proposal from Reece?
> 
> https://github.com/qt4cg/qtspecs/issues/161
> 
>   <tests change="fn-concat"
>          count="14"
>          description="Function fn:concat allows sequence-valued arguments (TODO???)"/>
> 
> Maybe this one refers to…
> 
> https://github.com/qt4cg/qtspecs/issues/701
> 

Yes. In fact the actual changes.xml file includes the PR number:

<change id="fn-concat" PR="702">Function fn:concat allows sequence-valued arguments (TODO???)</change>

it just hasn't been copied into the test case output.

As I say, there's a lot of opportunity for cleaning up this data,.

Michael Kay