Thanks for the suggestion Jonathan, but unfortunately that didn’t work. I’m 
about ready to just write a NoIop to inject a controllable metadata value where 
I need it, but I would love to have an idea of what voodoo is going on to get 
that written hash.

-Nathan



From: Jonathan Egstad 
Sent: Tuesday, June 14, 2011 12:51 PM
To: Nuke plug-in development discussion 
Subject: Re: [Nuke-dev] Write::getHashOfInputs() vs. Op::hash().getHash()

Did you try getting the hashes for the input Ops at different views and hashing 
them together?  Try Op::node_input() as it takes an OutputContext argument.
The order in which it's done would be important too.

-jonathan

On Jun 14, 2011, at 11:52 AM, Nathan Rusch wrote:


  I’m trying to make use of the “nuke/node_hash” metadata value in written EXR 
sequences. In my plugin, I’m fetching the value from the metadata stream of 
input1() and comparing it to the hash returned by input0().hash().getHash().

  This all works fine for mono comps; if I write out a comp at a certain point 
and then read it back in, the hash my plugin reports for the tree matches the 
hash embedded in the written EXR’s metadata.

  However, as soon as multiple views are introduced, everything breaks. Looking 
over the exrWriter source, it’s using Write::getHashOfInputs() to fetch the 
hash that gets embedded in the metadata. This doesn’t ever match the 
Op::hash().getHash() return value in stereo situations (which doesn’t vary 
per-view unless a split knob is changed for that view, even when I append 
outputContext().view() to my plugin’s hash).

  So my question is, what is Write::getHashOfInputs() doing differently to 
fetch its hash? Is there anything I can append to my plugin’s hash to get it to 
match the getHashOfInputs() return? Or are my only options to A) invent my own 
metadata hash that I can control or B) compile my own modified exrWriter that 
embeds what I want? This has me righteously stumped, so I would really 
appreciate any ideas.

  Thanks,


  -Nathan


  _______________________________________________
  Nuke-dev mailing list
  [email protected], http://forums.thefoundry.co.uk/
  http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev





--------------------------------------------------------------------------------
_______________________________________________
Nuke-dev mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev
_______________________________________________
Nuke-dev mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev

Reply via email to