HDA creation with rops

[Magnus Pettersson]

Hey i have a small issue. Is there a way of creating and rendering a digital asset in /out/ (rop) context that contain an alfred and a mantra node while keeping the asset locked?

Right now I have one alfred and one mantra node inside my HDA. When pressing the Render button that automatically gets created with the asset, both my mantra and alfred gets started and i get one job rendering locally and one started over alfred to our renderfarm, which is not what i want, i want to ether render the mantra node, or send it to alfred. Functioning like when having the two nodes normally in /out/ context. Plus alfred needs the asset to be unlocked to work otherwise i get a permission error.

When i add the "render" buttons from mantra and alfred to my HDA and press for example alfreds render button in the asset, alfred starts up but it gets permission errors since its inside a locked asset, unlocking my HDA and testing again works fine. My render button for the mantra node is the same, if the asset is locked the render starts but when the frame range is finished rendering the progress window hangs up on 100% with the info "evaluating python" and only way to get rid of it is to kill it (and restart houdini if i ever want to se a progress window again). If i unlock the asset again it works fine. Sometimes nothing happens when i press the mantra render button on my asset, then i have to dive into the asset and out again and press the button and now it works fine.

So far it seems best for me to use the built in render button and unfortunately maybe i have to drop the idea of having the alfred node inside my asset if i dont want it unlocked or maybe someone knows a way around this? It seems to be too much work to create my own output driver and hook up with soho for just using already created output drivers (alfred and mantra) since all my HDA does is using python to streamlining rendering for our pipeline with correct folders, naming and publishing features and such. Would be nice to get alfred to work too with the asset since it makes for one less human error (like rendering an alfred node that points to wrong mantra node that ends up in renders getting overwritten)

for the curious the permisson error i get with the locked asset is as follows:

Traceback (most recent call last):
  File "/net/sw/opt/sidefx/hfs11.0.629/houdini/soho/python2.5/AlfredRun.py", line 308, in <module>
    processLine(line)
  File "/net/sw/opt/sidefx/hfs11.0.629/houdini/soho/python2.5/AlfredRun.py", line 300, in processLine
    cback(args)
  File "/net/sw/opt/sidefx/hfs11.0.629/houdini/soho/python2.5/AlfredRun.py", line 273, in renderCommand
    processRender(sys.stdout, driver, tmpdir, frame)
  File "/net/sw/opt/sidefx/hfs11.0.629/houdini/soho/python2.5/AlfredRun.py", line 185, in processRender
    trange_parm.set('off')
  File "/net/sw/opt/sidefx/hfs11.0.629/houdini/python2.5libs/hou.py", line 20903, in set
    return _hou.Parm_set(*args)
hou.PermissionError: Failed to modify node or parameter because of a permission error.  Possible causes include locked assets, takes, product permissions or user specified permissions