filter editor : lot of bugs when duplicating

Bug #1122495 reported by v1nce
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Inkscape
Confirmed
Medium
Unassigned

Bug Description

I can't find a bug report for this but there are numerous bugs in the filter editor for years.

1) turbulence
===========
create a new filter.
add turbulence
duplicate.
set base frequencies (for last stage)
=> inkscape is lost. it sets the values for both stages.

add a new turbulence
set base frequencies (for last stage) => the new values are applied onto the two last stages.

set base frequency for first stage => inkscape applies it for the two first stages
Same goes for the seed.

I thought it was an (duplicated) id problem. But toggling link works well (if you toggle link at stage n it doesn't affect the others)

Settings octave is the weirdest. The value impacts two stages plus the slider is incorrectly redrawn in one of them (it is slightly offseted)

2) feOffset
=========
create a new filter.
add offset
duplicate
set x (|y) values for last stage
bug: the first stage get the value too.

3) Fill
====
create a new filter.
add fill
duplicate
set color => ok.
set opacity => ko. opacity is set on both stages.

...
same for displacement map...
...
I can't list all of them but as soon as you duplicate a fx, there are high chances at least one parameter goes mad.

Those bugs were here for years in stable release. It looks like some of them were fixed in dev version =>
I can't crash feColormatrix using the dev version even though I'm 100% sure it was buggy with some recent release (I'll have to reinstall a stable one to check)

my version of inkscape : 0.48 12114 34 precise 1
ubuntu 12.04

Tags: filters-svg ui
su_v (suv-lp)
tags: added: filters-svg ui
Revision history for this message
jazzynico (jazzynico) wrote :

Just to be sure: by duplicate, do you mean duplicate the primitive? If so, the bug is likely a duplicate of Bug #1111459 "Filters editor jumps to the first primitive".
When you set a parameter on the second (duplicate) primitive, Inkscape sets the value as expected, but also jumps to the first primitive and keep the mouse click status active, and thus sets the same value in the first primitive.

Changed in inkscape:
importance: Undecided → Medium
status: New → Incomplete
Revision history for this message
v1nce (vincent-pennec) wrote :

> Just to be sure: by duplicate, do you mean duplicate the primitive?

Yes. But it fails with adding too.And by stage I meant primitive

> When you set a parameter on the second (duplicate) primitive, Inkscape sets the value as expected, but also jumps to the first primitive and keep the mouse click status active, and thus sets the same value in the first primitive.

No. It doesn't jump to the first primitive but stay onto the currently selected.
(Unless you mean it jumps hiddenly to first primitive set the value then come back to current primitive)

if you click the first primitive, (some) values match the ones you'd just set in second primitive.

create new filter.
add turbulence x10times
onto the last
set frequency = 0.5, octave = 4, seed=300
click the n-2 primitive
=> values of n-2 are freq = 0.5, octave = 1, seed = 300 (sometimes seed =0)
click the n-1
=> values of n-1are freq = 0.5, octave = 1, seed = 0
set seed = 500 on n-1
values are now
n : seed = 300
n-1: seed = 500
n-2: seed = 500
frq = 0.5 on all n-x other stages

Revision history for this message
v1nce (vincent-pennec) wrote :

tested with 0.48.3.1.1-ubuntu1:

no Bug #1111459
but this one is present.

Revision history for this message
su_v (suv-lp) wrote :

Earlier reported problems related to changing parameters of filter primitives:
Bug #194408 “Filters don't always use parameters as displayed in dialog.”
Bug #216945 “Filter Effects - editing text control problem”
Bug #462748 “Filter editor dialog display size incorrect”
Bug #386627 “Data display remanence in filters Color matrix dialog”
Bug #1081973 “filter parameters changing”

Revision history for this message
jazzynico (jazzynico) wrote :

Confirmed on Debian testing, Inkscape trunk revision 12241.

Closely related to some (if not all) of the reports aforementioned.

Changed in inkscape:
status: Incomplete → Confirmed
Revision history for this message
v1nce (vincent-pennec) wrote :

0.91 ubuntu.
Still present.
havoc [=crash] occurs rapidly when you start to reorder primitives inside a filter or when duplicating or renaming filter.

e.g. a scale property (of displacement map) was duplicated into feturbulence and fecomposite or a basefrequency into a fecomposite.

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.