archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Packages Activity

netrender

Browse Source 
Support for fluid files and better support for point cache (including external cache for particles)

This also fixes a couple of bugs with frame based dependencies and with file transfer.

NOTE: With external point cache and fluids, the path needs to be relative or relative to the file (starting with //) if the files are not on a shared drive. It should eventually warn if that is not the case, but doesn't right now, so be careful.
This commit is contained in:
Martin Poirier
2009-10-09 01:52:57 +00:00
parent 1b6a09847a
commit cc4dd3f04e
3 changed files with 91 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

144
release/scripts/io/netrender/client.py
View File

@@ -8,6 +8,68 @@ import netrender.slave as slave
import netrender.master as master
from netrender.utils import *
def addFluidFiles(job, path):
if os.path.exists(path):
pattern = re.compile("fluidsurface_(final|preview)_([0-9]+)\.(bobj|bvel)\.gz")
for fluid_file in sorted(os.listdir(path)):
match = pattern.match(fluid_file)
if match:
current_frame = int(match.groups()[1])
job.addFile(path + fluid_file, current_frame, current_frame)
def addPointCache(job, ob, point_cache, default_path):
if not point_cache.disk_cache:
return
name = point_cache.name
if name == "":
name = "".join(["%02X" % ord(c) for c in ob.name])
cache_path = bpy.sys.expandpath(point_cache.filepath) if point_cache.external else default_path
index = "%02i" % point_cache.index
if os.path.exists(cache_path):
pattern = re.compile(name + "_([0-9]+)_" + index + "\.bphys")
cache_files = []
for cache_file in sorted(os.listdir(cache_path)):
match = pattern.match(cache_file)
if match:
cache_frame = int(match.groups()[0])
cache_files.append((cache_frame, cache_file))
cache_files.sort()
if len(cache_files) == 1:
cache_frame, cache_file = cache_files[0]
job.addFile(cache_path + cache_file, cache_frame, cache_frame)
else:
for i in range(len(cache_files)):
current_item = cache_files[i]
next_item = cache_files[i+1] if i + 1 < len(cache_files) else None
previous_item = cache_files[i - 1] if i > 0 else None
current_frame, current_file = current_item
if not next_item and not previous_item:
job.addFile(cache_path + current_file, current_frame, current_frame)
elif next_item and not previous_item:
next_frame = next_item[0]
job.addFile(cache_path + current_file, current_frame, next_frame - 1)
elif not next_item and previous_item:
previous_frame = previous_item[0]
job.addFile(cache_path + current_file, previous_frame + 1, current_frame)
else:
next_frame = next_item[0]
previous_frame = previous_item[0]
job.addFile(cache_path + current_file, previous_frame + 1, next_frame - 1)
def clientSendJob(conn, scene, anim = False):
netsettings = scene.network_render
job = netrender.model.RenderJob()
@@ -23,6 +85,7 @@ def clientSendJob(conn, scene, anim = False):
job_name = netsettings.job_name
path, name = os.path.split(filename)
path += os.sep
if job_name == "[default]":
job_name = name
@@ -30,67 +93,38 @@ def clientSendJob(conn, scene, anim = False):
# LIBRARIES
###########################
for lib in bpy.data.libraries:
lib_path = lib.filename
if lib_path.startswith("//"):
lib_path = path + os.sep + lib_path[2:]
job.addFile(lib_path)
###########################
# POINT CACHES
###########################
root, ext = os.path.splitext(name)
cache_path = path + os.sep + "blendcache_" + root + os.sep # need an API call for that
if os.path.exists(cache_path):
caches = {}
pattern = re.compile("([a-zA-Z0-9]+)_([0-9]+)_[0-9]+\.bphys")
for cache_file in sorted(os.listdir(cache_path)):
match = pattern.match(cache_file)
if match:
cache_id = match.groups()[0]
cache_frame = int(match.groups()[1])
cache_files = caches.get(cache_id, [])
cache_files.append((cache_frame, cache_file))
caches[cache_id] = cache_files
for cache in caches.values():
cache.sort()
if len(cache) == 1:
cache_frame, cache_file = cache[0]
job.addFile(cache_path + cache_file, cache_frame, cache_frame)
else:
for i in range(len(cache)):
current_item = cache[i]
next_item = cache[i+1] if i + 1 < len(cache) else None
previous_item = cache[i - 1] if i > 0 else None
current_frame, current_file = current_item
if not next_item and not previous_item:
job.addFile(cache_path + current_file, current_frame, current_frame)
elif next_item and not previous_item:
next_frame = next_item[0]
job.addFile(cache_path + current_file, current_frame, next_frame - 1)
elif not next_item and previous_item:
previous_frame = previous_item[0]
job.addFile(cache_path + current_file, previous_frame + 1, current_frame)
else:
next_frame = next_item[0]
previous_frame = previous_item[0]
job.addFile(cache_path + current_file, previous_frame + 1, next_frame - 1)
job.addFile(bpy.sys.expandpath(lib_path))
###########################
# IMAGES
###########################
for image in bpy.data.images:
if image.source == "FILE" and not image.packed_file:
job.addFile(image.filename)
job.addFile(bpy.sys.expandpath(image.filename))
###########################
# FLUID + POINT CACHE
###########################
root, ext = os.path.splitext(name)
default_path = path + "blendcache_" + root + os.sep # need an API call for that
for object in bpy.data.objects:
for modifier in object.modifiers:
if modifier.type == 'FLUID_SIMULATION' and modifier.settings.type == "DOMAIN":
addFluidFiles(job, bpy.sys.expandpath(modifier.settings.path))
elif modifier.type == "CLOTH":
addPointCache(job, object, modifier.point_cache, default_path)
elif modifier.type == "SOFT_BODY":
addPointCache(job, object, modifier.point_cache, default_path)
elif modifier.type == "SMOKE" and modifier.smoke_type == "TYPE_DOMAIN":
addPointCache(job, object, modifier.domain_settings.point_cache_low, default_path)
if modifier.domain_settings.highres:
addPointCache(job, object, modifier.domain_settings.point_cache_high, default_path)
# particles modifier are stupid and don't contain data
# we have to go through the object property
for psys in object.particle_systems:
addPointCache(job, object, psys.point_cache, default_path)
# print(job.files)

2
release/scripts/io/netrender/model.py
View File

@@ -149,7 +149,7 @@ class RenderJob:
"id": self.id,
"type": self.type,
"name": self.name,
"files": [f for f in self.files if f[1] == -1 or not frames or (f[1] <= min_frame <= f[2] or f[1] <= max_frame <= f[2])],
"files": [f for f in self.files if f[1] == -1 or not frames or (f[1] <= max_frame and f[2] >= min_frame)],
"frames": [f.serialize() for f in self.frames if not frames or f in frames],
"chunks": self.chunks,
"priority": self.priority,

2
release/scripts/io/netrender/utils.py
View File

@@ -75,7 +75,7 @@ def prefixPath(prefix_directory, file_path, prefix_path):
if prefix_path and p.startswith(prefix_path):
directory = prefix_directory + p[len(prefix_path):]
full_path = directory + n
full_path = directory + os.sep + n
if not os.path.exists(directory):
os.mkdir(directory)
else: