Bug 335102

Summary:	Fill contiguous area doesn't fill correctly with limit to current layer active
Product:	[Applications] krita	Reporter:	Bollebib <kwadraatnope>
Component:	Tools	Assignee:	Krita Bugs <krita-bugs-null>
Status:	RESOLVED DUPLICATE
Severity:	normal	CC:	dimula73, halla
Priority:	NOR
Version:	git master (please specify the git hash!)
Target Milestone:	---
Platform:	unspecified
OS:	Linux
Latest Commit:		Version Fixed In:
Sentry Crash Report:

Description Bollebib 2014-05-20 21:14:16 UTC

22:56:34 - Bollebib1: limit to current layer ,for flood fill(fill a continuous section tool) doesn't work,it fills whole layer
23:03:56 - frinring_f is now known as frinring.
23:08:25 - dmitryK: Bollebib1: oh, it might be either Fast Mode or some weirdness with trying to fill another layer. If you are sure Fast Mode is disabled, please report :)
23:08:40 - Bollebib1: yes I never use fast mode


when using flood fill with 'limit to active layer" off it works as it should
with the option on it fills the whole image or selection

Comment 1 Dmitry Kazakov 2014-05-21 12:08:47 UTC

For me it works ok if I set threshold to lower values :(

Comment 2 Dmitry Kazakov 2014-06-20 12:20:08 UTC

Hm... works here. Probably, you could make a video for this bug? Since I cannot reproduce it here.

Please reset the state of the bug to UNCONFIRMED back when you are done.

Comment 3 Bollebib 2014-06-22 14:22:56 UTC

https://www.youtube.com/watch?v=z8fkA6A1gBc&feature=youtu.be


this shows the issue I have

Comment 4 Bollebib 2014-06-22 14:23:30 UTC

apparantly fully opaque layers have no problems,but transparant ones do

Comment 5 Dmitry Kazakov 2014-06-22 19:18:04 UTC

Hi, Bollebib!

Could you please send me the file you used in the test? I cannot reproduce it :(

Comment 6 Dmitry Kazakov 2014-06-22 19:20:51 UTC

Ok, reproduced. It happens on 16bit per channel-integer images only! :(

Comment 7 Halla Rempt 2014-08-23 12:52:44 UTC

It's also a regression -- it worked fine in 2.8.

Comment 8 Halla Rempt 2015-02-03 19:12:53 UTC

Ah, and that's the same issue as in  343365. Fixed in 

commit d6b43848e24a833304e4670d3212a1ce4cfc34ff
Author: Dmitry Kazakov <dimula73@gmail.com>
Date:   Mon Feb 2 14:24:03 2015 +0300

    Fixed flood fill in 16 bit mode
    
    Well, 'reinterpret_cast<quint32*>(pixelPtr)' in a templated code was
    obviously wrong :(
    
    BUG:343365

We really could have fixed this six months earlier...

*** This bug has been marked as a duplicate of bug 343365 ***