Bug #9692
Test suite's --retry-find option breaks wait_for_gnome_window
100%
Description
It seems that the specified timeout is not respected, and just a few seconds after sikuli started waiting for the picture, it fails and asks for the image to be updated. I’ve seen that in:
@screen.wait_for_gnome_window('PersistenceWizardWindow.png', 40)
Subtasks
History
#1 Updated by intrigeri 2015-07-06 09:21:06
- Priority changed from Normal to High
- Target version changed from Tails_2.0 to Tails_1.8
- Parent task set to
Feature #7563
(Copying metadata from the parent ticket.)
#2 Updated by intrigeri 2015-08-02 09:05:27
- Due date set to 2016-01-15
#3 Updated by intrigeri 2015-08-02 09:05:46
- blocks #8668 added
#4 Updated by intrigeri 2015-08-26 05:55:12
- Deliverable for set to 268
#5 Updated by anonym 2015-09-03 01:12:11
Honestly, when I added --retry-find
it was supposed to be a cheap fix for solving 90% of the problem (i.e. something that makes updating the images easier). Fixing the last 10% would take probably 10x the initial implementation effort, and make the Sikuli monkey-patching we do crazy. Here the issue is that we implement our own waiting mechanism with a looped (via try_for
) find()
(so we have the same issue for at least waitAny()
as well). One way to fix this would be to temporarily disable the findfailed_hook
for find()
… some how. This will become incredibly ugly.
I’m not sure I want to fix this.
#6 Updated by intrigeri 2015-09-15 11:36:57
- Status changed from Confirmed to Rejected
> I’m not sure I want to fix this.
Agreed, the “solutions” seem to be worse than the problem.
#7 Updated by intrigeri 2015-09-15 11:37:32
- Assignee deleted (
anonym) - Target version deleted (
Tails_1.8) - Deliverable for deleted (
268)
#8 Updated by intrigeri 2015-11-24 11:00:20
- blocked by deleted (
#8668)
#9 Updated by intrigeri 2015-11-24 16:22:15
- % Done changed from 0 to 100