Both of those add_layer lines are problem-causing..
Incidentally, the redguard1 part is harmless, the problem is the
modifications to builtin rgb/rgba layers, done by "rgba.edgeblur",
"rgb.red", "rgb.r", "rgba.beta" etc..
The redguard1 layer would be removed by the "suppress bad channels"
option added in Nuke 6.3, and forced on in 7.. except it doesn't
consider the rgb/rgba layers as bad, so keeps the channels in it.
Bug report I made a while ago:
On 23/02/13 06:03, The Foundry Support wrote:
> Hello Ben,
>
> I've reproduced this and logged a bug as:
>
> Bug 34525 - suppress bad channels not working properly
>
> Engineering has been notified.
>
> Regards,
>
> - Dan Knittle
>
> The Foundry Support
>
>
>
> Ben Dickson<[email protected]> wrote:
>
>> In both Nuke 6 with the "suppress bad channels" option enabled, and Nuke
>> 7.0v2:
>>
>> 1. I copy in this bunch of nodes, which include a bad add_layer command
>> like so:
>>
>> set cut_paste_input [stack 0]
>> version 6.3 v8
>> push $cut_paste_input
>> add_layer {rgb rgb.r rgb.g rgb.b rgb.a}
>> NoOp {
>> }
>>
>> 2. Then I create any node with a channel selector - say a Blur node.
>>
>> 3. I set the Blur channel to "rgb"
>>
>>
>> Result is the 3-channel 'rgb' layer now has a forth checkbox, set to
>> 'rgb.r', which is nonsense. These channels propagate to other scripts
>> via copy-and-paste, "File>import script", nuke.scriptRead etc, which
>> makes it difficult to get rid of.
>>
>>
>>
>>
>> The problem is, some unknown series of steps results in the script being
>> corrupted. Specifically, the parsing of the node tree will stop with one
>> of the following errors:
>>
>> Shuffle4.out: rgb.r is not a layer or channel name
>>
>> Can't select rgba.alpha; it conflicts with other selected channels
>>
>> Anything beyond this point in the .nk file will not be read, resulting
>> in a truncated script.
>>
>> The nk file can be fixed manually with a text editor - however this is
>> tedious, error prone and should not be necessary.
>>
>> We have a script which attempts to remove the bad "add_layers" lines,
>> which might be responsible for one of the errors, but I have not yet
>> worked out the exact steps.
>>
>>
>> Again, the "suppress bad channels" option (enabled by default in Nuke 7)
>> does not help in this case.
>>
>>
>>
>> In short:
>>
>> You cannot edit a builtin 'rgb' layer in the UI. However, the
>> 'add_layer' command will happily add them, and this can cause scripts to
>> become 'virally' broken (via copy-and-paste etc)
>>
>> Is this a known bug? Is it something that will be fixed in an upcoming
>> 7.0 Nuke version?
>> --
>> ben dickson
>> 2D TD | [email protected]
>> rising sun pictures | www.rsp.com.au
On 11/09/13 18:12, John RA Benson wrote:
ack! caution!
don't paste that into a script before getting rid of this line!
add_layer {rgba *redguard1.glow* rgba.edgeblur}
jrab
On 09/10/2013 10:51 PM, Holger Hummel|Celluloid VFX wrote:
hi Paul,
check the code in the 'statusmessage' python script knob of the group
i pasted below.
is this what you're after?
cheers,
Holger
Group {
name Group1
selected true
xpos -96
ypos -675
disable true
}
Input {
inputs 0
name Input1
selected true
xpos 0
ypos 46
}
add_layer {rgb rgb.red rgb.green rgb.blue}
add_layer {rgba redguard1.glow rgba.edgeblur}
Grade {
name Grade1
selected true
xpos 0
ypos 140
addUserKnob {20 User}
addUserKnob {22 statusmessage l " status message " T "status =
nuke.thisGroup()\['disable'].getValue()\n\nnuke.message(\"disable knob
status of parent group: \" + str(status))\n" +STARTLINE}
}
Output {
name Output1
selected true
xpos 0
ypos 253
}
end_group
Paul Hudson wrote:
Hi,
If I check the disable knob of a node inside a disabled Group, it is
not True. This makes sense, because the knob itself isn't set.
However, is there a method to check if a node has 'inherited' a
disabled state from its parent(s)?
Thanks,
Paul
------------------------------------------------------------------------
_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
--
ben dickson
2D TD | [email protected]
rising sun pictures | www.rsp.com.au
_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
