terminal_width fix causes test failures with subunit

Bug #458741 reported by Matt Nordhoff
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Bazaar
Incomplete
Undecided
Unassigned

Bug Description

The "terminal_width is now very large for non-tty" change (bzr.dev r4764 and r4766) cause 2 test failures when subunit is enabled.

It's because bzrlib.tests.blackbox.test_status.BranchStatus.test_pending and bzrlib.tests.blackbox.test_status.CheckoutStatus.test_pending both have wide output and expect it to be truncated, but it's not anymore.

It works fine without subunit. I guess it has access to the real terminal width?

Revision history for this message
Matt Nordhoff (mnordhoff) wrote :
Revision history for this message
Vincent Ladeuil (vila) wrote :

I can't reproduce that, what do you mean with 'subunit enabled' ? Passing --subunit to selftest ?

Also, check bug #353370

Changed in bzr:
status: New → Incomplete
Revision history for this message
Matt Nordhoff (mnordhoff) wrote : Re: [Bug 458741] Re: terminal_width fix causes test failures with subunit

Vincent Ladeuil wrote:
> I can't reproduce that, what do you mean with 'subunit enabled' ?
> Passing --subunit to selftest ?

Yes. Well, I tested it by implicitly activating it with --parallel.

"bzr selftest --parallel subprocess", to be specific. Maybe that's the
only condition where it happens.

> Also, check bug #353370

I know. I felt like filing new bugs instead of adding comments to the
existing one.
--

Revision history for this message
Vincent Ladeuil (vila) wrote :

So, 'subunit enabled' meant:
  bzr selftest --parallel=subprocess

This is fixed by the latest proposal to fix bug #353370, marking as duplicate.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.