Nuke: Collect representation files based on Write

openpype/hosts/nuke/plugins/publish/extract_render_local.py

@@ -42,12 +42,22 @@ def process(self, instance):
  self.log.info("Start frame: {}".format(first_frame))
  self.log.info("End frame: {}".format(last_frame))
 
- # write node url might contain nuke's ctl expressin
- # as [python ...]/path...
- path = node["file"].evaluate()
+ node_file = node["file"]
+ # Collecte expected filepaths for each frame
+ # - for cases that output is still image is first created set of
+ # paths which is then sorted and converted to list
+ expected_paths = list(sorted({
+ node_file.evaluate(frame)
+ for frame in range(first_frame, last_frame + 1)
+ }))
+ # Extract only filenames for representation
+ filenames = [
+ os.path.basename(filepath)
+ for filepath in expected_paths
+ ]
 
  # Ensure output directory exists.
- out_dir = os.path.dirname(path)
+ out_dir = os.path.dirname(expected_paths[0])
  if not os.path.exists(out_dir):
  os.makedirs(out_dir)
 
@@ -67,12 +77,11 @@ def process(self, instance):
  if "representations" not in instance.data:
  instance.data["representations"] = []
 
- collected_frames = os.listdir(out_dir)
- if len(collected_frames) == 1:
+ if len(filenames) == 1:
  repre = {
  'name': ext,
  'ext': ext,
- 'files': collected_frames.pop(),
+ 'files': filenames[0],
  "stagingDir": out_dir
  }
  else:
@@ -81,7 +90,7 @@ def process(self, instance):
  'ext': ext,
  'frameStart': "%0{}d".format(
  len(str(last_frame))) % first_frame,
- 'files': collected_frames,
+ 'files': filenames,
  "stagingDir": out_dir
  }
  instance.data["representations"].append(repre)
@@ -105,7 +114,7 @@ def process(self, instance):
  families.remove('still.local')
  instance.data["families"] = families
 
- collections, remainder = clique.assemble(collected_frames)
+ collections, remainder = clique.assemble(filenames)
  self.log.info('collections: {}'.format(str(collections)))
 
  if collections:

openpype/hosts/nuke/plugins/publish/precollect_writes.py

@@ -56,9 +56,21 @@ def process(self, instance):
  first_frame = int(node["first"].getValue())
  last_frame = int(node["last"].getValue())
 
- # get path
+ # Prepare expected output paths by evaluating each frame of write node
+ # - paths are first collected to set to avoid duplicated paths, then
+ # sorted and converted to list
+ node_file = node["file"]
+ expected_paths = list(sorted({
+ node_file.evaluate(frame)
+ for frame in range(first_frame, last_frame + 1)
+ }))
+ expected_filenames = [
+ os.path.basename(filepath)
+ for filepath in expected_paths
+ ]
  path = nuke.filename(node)
  output_dir = os.path.dirname(path)
+
  self.log.debug('output dir: {}'.format(output_dir))
 
  # create label
@@ -83,8 +95,11 @@ def process(self, instance):
  }
 
  try:
- collected_frames = [f for f in os.listdir(output_dir)
- if ext in f]
+ collected_frames = [
+ filename
+ for filename in os.listdir(output_dir)
+ if filename in expected_filenames
+ ]
  if collected_frames:
  collected_frames_len = len(collected_frames)
  frame_start_str = "%0{}d".format(