On 2016-09-09, Kornel Benko wrote:
> Am Donnerstag, 8. September 2016 um 16:35:10, schrieb Guenter Milde 
> <mi...@users.sf.net>

Dear Kornel,


>> ... actually the labeling mechanism is constructed so that inversion
>> takes precedence and prevents tagging with "unreliable".

>> Maybe this can be changed in an overhoul of the test setup...

> Yes, there is some discussion needed.

Maybe we can start with some simplification...

* Can we rename "suspiciousTests" to "invertedTests", please?
  
  The "docstring" at the top of the file says:

     # Regular expressions for tests that are known to fail
     #
     # Patterns will be automatically enclosed with '^' and '$'.
     # Matching tests get the label "inverted"
     # or (if also matching a pattern in "suspendedTests") "suspended".
     # They get also the test-feature 'inverted', i.e.
     # they are reported as failing if the export works without error.


* Do you still need the "suspendeTests"? What for?

  I know that it was me who suggested this label. 
  
  However, the sublabel mechanism supersedes this by giving more
  detailled info about problems and reasons for failure.


Thanks

Günter

Reply via email to