[1/2] shell32/tests: Trace the last ShellExecute command whenever a corresponding test fails.

[Alexandre Julliard]

Francois Gouget <fgouget at codeweavers.com> writes:

> On Fri, 22 Jan 2016, Alexandre Julliard wrote:
>> Why is that better than always printing it in the failing calls?
>
> Because usually there is more than one failing call: there's the ok() 
> after the shell_execute() and then multiple okChildXxx() calls to verify 
> the data received by the child and there is often multiple failures. 
> There is no reason for each of them to repeat the information about what 
> ShellExecute() call failed. By comparing shell_call and last_shell_call 
> okShell ensures that information is printed once before the series of 
> failing tests.

OTOH, it means that in silent mode you get no indication at all. Given
that the checks are not supposed to fail, I'm not sure that improving
the output for multiple failures is that important.

-- 
Alexandre Julliard
julliard at winehq.org