Reorganized the Freestyle Python API in a hierarchical package structure.

Both C- and Python-coded API components were rearranged into logical groups.
New Python modules are packaged as follows:

freestyle - Top-level package
freestyle.types - Classes for core data structues (e.g., view map)
freestyle.chainingiterators - Pre-defined chaining iterators
freestyle.functions - Pre-defined 0D and 1D functions
freestyle.predicates - Pre-defined 0D and 1D predicates
freestyle.shaders - Pre-defined stroke shaders
freestyle.utils - Utility functions

The Python modules are installed in scripts/freestyle/modules.  Pre-defined
styles are installed in scripts/freestyle/styles.

To-do: update styles according to the new Freestyle API package structure.
This commit is contained in:
Tamito Kajiyama
2013-11-24 00:28:01 +00:00
parent df471369e2
commit 6498b96ce7
57 changed files with 2808 additions and 2663 deletions

View File

@@ -0,0 +1,21 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import Operators
from . import chainingiterators, functions, predicates, types, utils

View File

@@ -0,0 +1,716 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import (
ChainPredicateIterator,
ChainSilhouetteIterator,
)
# modules for implementing chaining iterators
from freestyle.types import (
AdjacencyIterator,
ChainingIterator,
)
from freestyle.utils import ContextFunctions as CF
import bpy
## the natural chaining iterator
## It follows the edges of same nature following the topology of
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
class pyChainSilhouetteIterator(ChainingIterator):
def __init__(self, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, True, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = it.object
it.increment()
if count != 1:
winner = None
break
return winner
## the natural chaining iterator
## It follows the edges of same nature on the same
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
## You can specify whether to chain iterate over edges that were
## already visited or not.
class pyChainSilhouetteGenericIterator(ChainingIterator):
def __init__(self, stayInSelection=True, stayInUnvisited=True):
ChainingIterator.__init__(self, stayInSelection, stayInUnvisited, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
ve = it.object
if ve.id == self.current_edge.id:
it.increment()
continue
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = ve
it.increment()
if count != 1:
winner = None
break
return winner
class pyExternalContourChainingIterator(ChainingIterator):
def __init__(self):
ChainingIterator.__init__(self, False, True, None, True)
self._isExternalContour = ExternalContourUP1D()
def init(self):
self._nEdges = 0
self._isInSelection = 1
def checkViewEdge(self, ve, orientation):
if orientation != 0:
vertex = ve.second_svertex()
else:
vertex = ve.first_svertex()
it = AdjacencyIterator(vertex,1,1)
while not it.is_end:
ave = it.object
if self._isExternalContour(ave):
return 1
it.increment()
print("pyExternlContourChainingIterator : didn't find next edge")
return 0
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if self._isExternalContour(ve):
if ve.time_stamp == CF.get_time_stamp():
winner = ve
it.increment()
self._nEdges = self._nEdges+1
if winner is None:
orient = 1
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if it.is_incoming:
orient = 0
good = self.checkViewEdge(ve,orient)
if good != 0:
winner = ve
it.increment()
return winner
## the natural chaining iterator
## with a sketchy multiple touch
class pySketchyChainSilhouetteIterator(ChainingIterator):
def __init__(self, nRounds=3,stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, False, None, True)
self._timeStamp = CF.get_time_stamp()+nRounds
self._nRounds = nRounds
def init(self):
self._timeStamp = CF.get_time_stamp()+self._nRounds
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
ve = it.object
if ve.id == self.current_edge.id:
it.increment()
continue
if (oNature & natures[i]) != 0:
if (natures[i] != oNature) != 0:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = ve
it.increment()
if count != 1:
winner = None
break
if winner is None:
winner = self.current_edge
if winner.chaining_time_stamp == self._timeStamp:
winner = None
return winner
# Chaining iterator designed for sketchy style.
# can chain several times the same ViewEdge
# in order to produce multiple strokes per ViewEdge.
class pySketchyChainingIterator(ChainingIterator):
def __init__(self, nRounds=3, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, False, None, True)
self._timeStamp = CF.get_time_stamp()+nRounds
self._nRounds = nRounds
def init(self):
self._timeStamp = CF.get_time_stamp()+self._nRounds
def traverse(self, iter):
winner = None
found = False
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if ve.id == self.current_edge.id:
found = True
it.increment()
continue
winner = ve
it.increment()
if not found:
# This is a fatal error condition: self.current_edge must be found
# among the edges seen by the AdjacencyIterator [bug #35695].
if bpy.app.debug_freestyle:
print('pySketchyChainingIterator: current edge not found')
return None
if winner is None:
winner = self.current_edge
if winner.chaining_time_stamp == self._timeStamp:
return None
return winner
## Chaining iterator that fills small occlusions
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillOcclusionsRelativeChainingIterator(ChainingIterator):
def __init__(self, percent):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if connexl > self._percent * self._length:
winner = None
return winner
## Chaining iterator that fills small occlusions
## size
## The max length of the occluded part
## expressed in pixels
class pyFillOcclusionsAbsoluteChainingIterator(ChainingIterator):
def __init__(self, length):
ChainingIterator.__init__(self, False, True, None, True)
self._length = float(length)
def init(self):
pass
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if connexl > self._length:
winner = None
return winner
## Chaining iterator that fills small occlusions
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillOcclusionsAbsoluteAndRelativeChainingIterator(ChainingIterator):
def __init__(self, percent, l):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._absLength = l
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if (connexl > self._percent * self._length) or (connexl > self._absLength):
winner = None
return winner
## Chaining iterator that fills small occlusions without caring about the
## actual selection
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillQi0AbsoluteAndRelativeChainingIterator(ChainingIterator):
def __init__(self, percent, l):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._absLength = l
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.qi != 0:
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while not _cit.is_end and _cit.object.qi != 0:
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if (connexl > self._percent * self._length) or (connexl > self._absLength):
winner = None
return winner
## the natural chaining iterator
## It follows the edges of same nature on the same
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
class pyNoIdChainSilhouetteIterator(ChainingIterator):
def __init__(self, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, True, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
feB = self.current_edge.last_fedge
feA = ve.first_fedge
vB = feB.second_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.first_fedge
feB = ve.last_fedge
vB = feB.second_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.last_fedge
feB = ve.last_fedge
vB = feB.second_svertex
vA = feA.second_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.first_fedge
feB = ve.first_fedge
vB = feB.first_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = it.object
it.increment()
if count != 1:
winner = None
break
return winner

View File

@@ -0,0 +1,197 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import (
ChainingTimeStampF1D,
Curvature2DAngleF0D,
Curvature2DAngleF1D,
CurveNatureF0D,
CurveNatureF1D,
DensityF0D,
DensityF1D,
GetCompleteViewMapDensityF1D,
GetCurvilinearAbscissaF0D,
GetDirectionalViewMapDensityF1D,
GetOccludeeF0D,
GetOccludeeF1D,
GetOccludersF0D,
GetOccludersF1D,
GetParameterF0D,
GetProjectedXF0D,
GetProjectedXF1D,
GetProjectedYF0D,
GetProjectedYF1D,
GetProjectedZF0D,
GetProjectedZF1D,
GetShapeF0D,
GetShapeF1D,
GetSteerableViewMapDensityF1D,
GetViewMapGradientNormF0D,
GetViewMapGradientNormF1D,
GetXF0D,
GetXF1D,
GetYF0D,
GetYF1D,
GetZF0D,
GetZF1D,
IncrementChainingTimeStampF1D,
LocalAverageDepthF0D,
LocalAverageDepthF1D,
MaterialF0D,
Normal2DF0D,
Normal2DF1D,
Orientation2DF1D,
Orientation3DF1D,
QuantitativeInvisibilityF0D,
QuantitativeInvisibilityF1D,
ReadCompleteViewMapPixelF0D,
ReadMapPixelF0D,
ReadSteerableViewMapPixelF0D,
ShapeIdF0D,
TimeStampF1D,
VertexOrientation2DF0D,
VertexOrientation3DF0D,
ZDiscontinuityF0D,
ZDiscontinuityF1D,
)
# modules for implementing functions
from freestyle.types import (
IntegrationType,
UnaryFunction0DDouble,
UnaryFunction0DMaterial,
UnaryFunction0DVec2f,
UnaryFunction1DDouble,
)
from freestyle.utils import ContextFunctions as CF
import math
import mathutils
## Functions for 0D elements (vertices)
#######################################
class CurveMaterialF0D(UnaryFunction0DMaterial):
# A replacement of the built-in MaterialF0D for stroke creation.
# MaterialF0D does not work with Curves and Strokes.
def __call__(self, inter):
cp = inter.object
assert(isinstance(cp, CurvePoint))
fe = cp.first_svertex.get_fedge(cp.second_svertex)
assert(fe is not None)
return fe.material if fe.is_smooth else fe.material_left
class pyInverseCurvature2DAngleF0D(UnaryFunction0DDouble):
def __call__(self, inter):
func = Curvature2DAngleF0D()
c = func(inter)
return (3.1415 - c)
class pyCurvilinearLengthF0D(UnaryFunction0DDouble):
def __call__(self, inter):
cp = inter.object
assert(isinstance(cp, CurvePoint))
return cp.t2d
## estimate anisotropy of density
class pyDensityAnisotropyF0D(UnaryFunction0DDouble):
def __init__(self,level):
UnaryFunction0DDouble.__init__(self)
self.IsoDensity = ReadCompleteViewMapPixelF0D(level)
self.d0Density = ReadSteerableViewMapPixelF0D(0, level)
self.d1Density = ReadSteerableViewMapPixelF0D(1, level)
self.d2Density = ReadSteerableViewMapPixelF0D(2, level)
self.d3Density = ReadSteerableViewMapPixelF0D(3, level)
def __call__(self, inter):
c_iso = self.IsoDensity(inter)
c_0 = self.d0Density(inter)
c_1 = self.d1Density(inter)
c_2 = self.d2Density(inter)
c_3 = self.d3Density(inter)
cMax = max(max(c_0,c_1), max(c_2,c_3))
cMin = min(min(c_0,c_1), min(c_2,c_3))
if c_iso == 0:
v = 0
else:
v = (cMax-cMin)/c_iso
return v
## Returns the gradient vector for a pixel
## l
## the level at which one wants to compute the gradient
class pyViewMapGradientVectorF0D(UnaryFunction0DVec2f):
def __init__(self, l):
UnaryFunction0DVec2f.__init__(self)
self._l = l
self._step = math.pow(2,self._l)
def __call__(self, iter):
p = iter.object.point_2d
gx = CF.read_complete_view_map_pixel(self._l, int(p.x+self._step), int(p.y)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
gy = CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y+self._step)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
return mathutils.Vector([gx, gy])
class pyViewMapGradientNormF0D(UnaryFunction0DDouble):
def __init__(self, l):
UnaryFunction0DDouble.__init__(self)
self._l = l
self._step = math.pow(2,self._l)
def __call__(self, iter):
p = iter.object.point_2d
gx = CF.read_complete_view_map_pixel(self._l, int(p.x+self._step), int(p.y)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
gy = CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y+self._step)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
grad = mathutils.Vector([gx, gy])
return grad.length
## Functions for 1D elements (curves)
#####################################
class pyGetInverseProjectedZF1D(UnaryFunction1DDouble):
def __call__(self, inter):
func = GetProjectedZF1D()
z = func(inter)
return (1.0 - z)
class pyGetSquareInverseProjectedZF1D(UnaryFunction1DDouble):
def __call__(self, inter):
func = GetProjectedZF1D()
z = func(inter)
return (1.0 - z*z)
class pyDensityAnisotropyF1D(UnaryFunction1DDouble):
def __init__(self,level, integrationType=IntegrationType.MEAN, sampling=2.0):
UnaryFunction1DDouble.__init__(self, integrationType)
self._func = pyDensityAnisotropyF0D(level)
self._integration = integrationType
self._sampling = sampling
def __call__(self, inter):
v = integrate(self._func, inter.pointsBegin(self._sampling), inter.pointsEnd(self._sampling), self._integration)
return v
class pyViewMapGradientNormF1D(UnaryFunction1DDouble):
def __init__(self,l, integrationType, sampling=2.0):
UnaryFunction1DDouble.__init__(self, integrationType)
self._func = pyViewMapGradientNormF0D(l)
self._integration = integrationType
self._sampling = sampling
def __call__(self, inter):
v = integrate(self._func, inter.pointsBegin(self._sampling), inter.pointsEnd(self._sampling), self._integration)
return v

View File

@@ -0,0 +1,516 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import (
ContourUP1D,
DensityLowerThanUP1D,
EqualToChainingTimeStampUP1D,
EqualToTimeStampUP1D,
ExternalContourUP1D,
FalseBP1D,
FalseUP0D,
FalseUP1D,
Length2DBP1D,
QuantitativeInvisibilityUP1D,
SameShapeIdBP1D,
ShapeUP1D,
TrueBP1D,
TrueUP0D,
TrueUP1D,
ViewMapGradientNormBP1D,
WithinImageBoundaryUP1D,
)
# modules for implementing predicates
from freestyle.types import (
IntegrationType,
BinaryPredicate1D,
UnaryPredicate0D,
UnaryPredicate1D,
)
from freestyle.functions import (
pyCurvilinearLengthF0D,
pyDensityAnisotropyF1D,
pyViewMapGradientNormF1D,
)
import random
## Unary predicates for 0D elements (vertices)
##############################################
class pyHigherCurvature2DAngleUP0D(UnaryPredicate0D):
def __init__(self,a):
UnaryPredicate0D.__init__(self)
self._a = a
def __call__(self, inter):
func = Curvature2DAngleF0D()
a = func(inter)
return (a > self._a)
class pyUEqualsUP0D(UnaryPredicate0D):
def __init__(self,u, w):
UnaryPredicate0D.__init__(self)
self._u = u
self._w = w
def __call__(self, inter):
func = pyCurvilinearLengthF0D()
u = func(inter)
return (u > (self._u-self._w)) and (u < (self._u+self._w))
class pyVertexNatureUP0D(UnaryPredicate0D):
def __init__(self,nature):
UnaryPredicate0D.__init__(self)
self._nature = nature
def __call__(self, inter):
v = inter.object
return (v.nature & self._nature) != 0
## check whether an Interface0DIterator
## is a TVertex and is the one that is
## hidden (inferred from the context)
class pyBackTVertexUP0D(UnaryPredicate0D):
def __init__(self):
UnaryPredicate0D.__init__(self)
self._getQI = QuantitativeInvisibilityF0D()
def __call__(self, iter):
if (iter.object.nature & Nature.T_VERTEX) == 0:
return 0
if iter.is_end:
return 0
if self._getQI(iter) != 0:
return 1
return 0
class pyParameterUP0DGoodOne(UnaryPredicate0D):
def __init__(self,pmin,pmax):
UnaryPredicate0D.__init__(self)
self._m = pmin
self._M = pmax
#self.getCurvilinearAbscissa = GetCurvilinearAbscissaF0D()
def __call__(self, inter):
#s = self.getCurvilinearAbscissa(inter)
u = inter.u
#print(u)
return ((u>=self._m) and (u<=self._M))
class pyParameterUP0D(UnaryPredicate0D):
def __init__(self,pmin,pmax):
UnaryPredicate0D.__init__(self)
self._m = pmin
self._M = pmax
#self.getCurvilinearAbscissa = GetCurvilinearAbscissaF0D()
def __call__(self, inter):
func = Curvature2DAngleF0D()
c = func(inter)
b1 = (c>0.1)
#s = self.getCurvilinearAbscissa(inter)
u = inter.u
#print(u)
b = ((u>=self._m) and (u<=self._M))
return b and b1
## Unary predicates for 1D elements (curves)
############################################
class AndUP1D(UnaryPredicate1D):
def __init__(self, pred1, pred2):
UnaryPredicate1D.__init__(self)
self.__pred1 = pred1
self.__pred2 = pred2
def __call__(self, inter):
return self.__pred1(inter) and self.__pred2(inter)
class OrUP1D(UnaryPredicate1D):
def __init__(self, pred1, pred2):
UnaryPredicate1D.__init__(self)
self.__pred1 = pred1
self.__pred2 = pred2
def __call__(self, inter):
return self.__pred1(inter) or self.__pred2(inter)
class NotUP1D(UnaryPredicate1D):
def __init__(self, pred):
UnaryPredicate1D.__init__(self)
self.__pred = pred
def __call__(self, inter):
return not self.__pred(inter)
class pyNFirstUP1D(UnaryPredicate1D):
def __init__(self, n):
UnaryPredicate1D.__init__(self)
self.__n = n
self.__count = 0
def __call__(self, inter):
self.__count = self.__count + 1
if self.__count <= self.__n:
return 1
return 0
class pyHigherLengthUP1D(UnaryPredicate1D):
def __init__(self,l):
UnaryPredicate1D.__init__(self)
self._l = l
def __call__(self, inter):
return (inter.length_2d > self._l)
class pyNatureUP1D(UnaryPredicate1D):
def __init__(self,nature):
UnaryPredicate1D.__init__(self)
self._nature = nature
self._getNature = CurveNatureF1D()
def __call__(self, inter):
if(self._getNature(inter) & self._nature):
return 1
return 0
class pyHigherNumberOfTurnsUP1D(UnaryPredicate1D):
def __init__(self,n,a):
UnaryPredicate1D.__init__(self)
self._n = n
self._a = a
def __call__(self, inter):
count = 0
func = Curvature2DAngleF0D()
it = inter.vertices_begin()
while not it.is_end:
if func(it) > self._a:
count = count+1
if count > self._n:
return 1
it.increment()
return 0
class pyDensityUP1D(UnaryPredicate1D):
def __init__(self,wsize,threshold, integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._wsize = wsize
self._threshold = threshold
self._integration = integration
self._func = DensityF1D(self._wsize, self._integration, sampling)
def __call__(self, inter):
if self._func(inter) < self._threshold:
return 1
return 0
class pyLowSteerableViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
def __call__(self, inter):
func = GetSteerableViewMapDensityF1D(self._level, self._integration)
v = func(inter)
#print(v)
if v < self._threshold:
return 1
return 0
class pyLowDirectionalViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, orientation, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._orientation = orientation
self._level = level
self._integration = integration
def __call__(self, inter):
func = GetDirectionalViewMapDensityF1D(self._orientation, self._level, self._integration)
v = func(inter)
#print(v)
if v < self._threshold:
return 1
return 0
class pyHighSteerableViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
self._func = GetSteerableViewMapDensityF1D(self._level, self._integration)
def __call__(self, inter):
v = self._func(inter)
if v > self._threshold:
return 1
return 0
class pyHighDirectionalViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, orientation, level,integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._orientation = orientation
self._level = level
self._integration = integration
self._sampling = sampling
def __call__(self, inter):
func = GetDirectionalViewMapDensityF1D(self._orientation, self._level, self._integration, self._sampling)
v = func(inter)
if v > self._threshold:
return 1
return 0
class pyHighViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
self._sampling = sampling
self._func = GetCompleteViewMapDensityF1D(self._level, self._integration, self._sampling) # 2.0 is the smpling
def __call__(self, inter):
#print("toto")
#print(func.name)
#print(inter.name)
v= self._func(inter)
if v > self._threshold:
return 1
return 0
class pyDensityFunctorUP1D(UnaryPredicate1D):
def __init__(self,wsize,threshold, functor, funcmin=0.0, funcmax=1.0, integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._wsize = wsize
self._threshold = float(threshold)
self._functor = functor
self._funcmin = float(funcmin)
self._funcmax = float(funcmax)
self._integration = integration
def __call__(self, inter):
func = DensityF1D(self._wsize, self._integration)
res = self._functor(inter)
k = (res-self._funcmin)/(self._funcmax-self._funcmin)
if func(inter) < self._threshold*k:
return 1
return 0
class pyZSmallerUP1D(UnaryPredicate1D):
def __init__(self,z, integration=IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._z = z
self._integration = integration
def __call__(self, inter):
func = GetProjectedZF1D(self._integration)
if func(inter) < self._z:
return 1
return 0
class pyIsOccludedByUP1D(UnaryPredicate1D):
def __init__(self,id):
UnaryPredicate1D.__init__(self)
self._id = id
def __call__(self, inter):
func = GetShapeF1D()
shapes = func(inter)
for s in shapes:
if(s.id == self._id):
return 0
it = inter.vertices_begin()
itlast = inter.vertices_end()
itlast.decrement()
v = it.object
vlast = itlast.object
tvertex = v.viewvertex
if type(tvertex) is TVertex:
#print("TVertex: [ ", tvertex.id.first, ",", tvertex.id.second," ]")
eit = tvertex.edges_begin()
while not eit.is_end:
ve, incoming = eit.object
if ve.id == self._id:
return 1
#print("-------", ve.id.first, "-", ve.id.second)
eit.increment()
tvertex = vlast.viewvertex
if type(tvertex) is TVertex:
#print("TVertex: [ ", tvertex.id.first, ",", tvertex.id.second," ]")
eit = tvertex.edges_begin()
while not eit.is_end:
ve, incoming = eit.object
if ve.id == self._id:
return 1
#print("-------", ve.id.first, "-", ve.id.second)
eit.increment()
return 0
class pyIsInOccludersListUP1D(UnaryPredicate1D):
def __init__(self,id):
UnaryPredicate1D.__init__(self)
self._id = id
def __call__(self, inter):
func = GetOccludersF1D()
occluders = func(inter)
for a in occluders:
if a.id == self._id:
return 1
return 0
class pyIsOccludedByItselfUP1D(UnaryPredicate1D):
def __init__(self):
UnaryPredicate1D.__init__(self)
self.__func1 = GetOccludersF1D()
self.__func2 = GetShapeF1D()
def __call__(self, inter):
lst1 = self.__func1(inter)
lst2 = self.__func2(inter)
for vs1 in lst1:
for vs2 in lst2:
if vs1.id == vs2.id:
return 1
return 0
class pyIsOccludedByIdListUP1D(UnaryPredicate1D):
def __init__(self, idlist):
UnaryPredicate1D.__init__(self)
self._idlist = idlist
self.__func1 = GetOccludersF1D()
def __call__(self, inter):
lst1 = self.__func1(inter)
for vs1 in lst1:
for _id in self._idlist:
if vs1.id == _id:
return 1
return 0
class pyShapeIdListUP1D(UnaryPredicate1D):
def __init__(self,idlist):
UnaryPredicate1D.__init__(self)
self._idlist = idlist
self._funcs = []
for _id in idlist :
self._funcs.append(ShapeUP1D(_id.first, _id.second))
def __call__(self, inter):
for func in self._funcs :
if func(inter) == 1:
return 1
return 0
## deprecated
class pyShapeIdUP1D(UnaryPredicate1D):
def __init__(self, _id):
UnaryPredicate1D.__init__(self)
self._id = _id
def __call__(self, inter):
func = GetShapeF1D()
shapes = func(inter)
for a in shapes:
if a.id == self._id:
return 1
return 0
class pyHighDensityAnisotropyUP1D(UnaryPredicate1D):
def __init__(self,threshold, level, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._l = threshold
self.func = pyDensityAnisotropyF1D(level, IntegrationType.MEAN, sampling)
def __call__(self, inter):
return (self.func(inter) > self._l)
class pyHighViewMapGradientNormUP1D(UnaryPredicate1D):
def __init__(self,threshold, l, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._GetGradient = pyViewMapGradientNormF1D(l, IntegrationType.MEAN)
def __call__(self, inter):
gn = self._GetGradient(inter)
#print(gn)
return (gn > self._threshold)
class pyDensityVariableSigmaUP1D(UnaryPredicate1D):
def __init__(self,functor, sigmaMin,sigmaMax, lmin, lmax, tmin, tmax, integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._functor = functor
self._sigmaMin = float(sigmaMin)
self._sigmaMax = float(sigmaMax)
self._lmin = float(lmin)
self._lmax = float(lmax)
self._tmin = tmin
self._tmax = tmax
self._integration = integration
self._sampling = sampling
def __call__(self, inter):
sigma = (self._sigmaMax-self._sigmaMin)/(self._lmax-self._lmin)*(self._functor(inter)-self._lmin) + self._sigmaMin
t = (self._tmax-self._tmin)/(self._lmax-self._lmin)*(self._functor(inter)-self._lmin) + self._tmin
if sigma < self._sigmaMin:
sigma = self._sigmaMin
self._func = DensityF1D(sigma, self._integration, self._sampling)
d = self._func(inter)
if d < t:
return 1
return 0
class pyClosedCurveUP1D(UnaryPredicate1D):
def __call__(self, inter):
it = inter.vertices_begin()
itlast = inter.vertices_end()
itlast.decrement()
vlast = itlast.object
v = it.object
#print(v.id.first, v.id.second)
#print(vlast.id.first, vlast.id.second)
if v.id == vlast.id:
return 1
return 0
## Binary predicates for 1D elements (curves)
#############################################
class pyZBP1D(BinaryPredicate1D):
def __call__(self, i1, i2):
func = GetZF1D()
return (func(i1) > func(i2))
class pyZDiscontinuityBP1D(BinaryPredicate1D):
def __init__(self, iType = IntegrationType.MEAN):
BinaryPredicate1D.__init__(self)
self._GetZDiscontinuity = ZDiscontinuityF1D(iType)
def __call__(self, i1, i2):
return (self._GetZDiscontinuity(i1) > self._GetZDiscontinuity(i2))
class pyLengthBP1D(BinaryPredicate1D):
def __call__(self, i1, i2):
return (i1.length_2d > i2.length_2d)
class pySilhouetteFirstBP1D(BinaryPredicate1D):
def __call__(self, inter1, inter2):
bpred = SameShapeIdBP1D()
if (bpred(inter1, inter2) != 1):
return 0
if (inter1.nature & Nature.SILHOUETTE):
return (inter2.nature & Nature.SILHOUETTE) != 0
return (inter1.nature == inter2.nature)
class pyNatureBP1D(BinaryPredicate1D):
def __call__(self, inter1, inter2):
return (inter1.nature & inter2.nature)
class pyViewMapGradientNormBP1D(BinaryPredicate1D):
def __init__(self,l, sampling=2.0):
BinaryPredicate1D.__init__(self)
self._GetGradient = pyViewMapGradientNormF1D(l, IntegrationType.MEAN)
def __call__(self, i1,i2):
#print("compare gradient")
return (self._GetGradient(i1) > self._GetGradient(i2))
class pyShuffleBP1D(BinaryPredicate1D):
def __init__(self):
BinaryPredicate1D.__init__(self)
random.seed(1)
def __call__(self, inter1, inter2):
r1 = random.uniform(0,1)
r2 = random.uniform(0,1)
return (r1<r2)

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,81 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import (
AdjacencyIterator,
BBox,
BinaryPredicate0D,
BinaryPredicate1D,
Chain,
ChainingIterator,
Curve,
CurvePoint,
CurvePointIterator,
FEdge,
FEdgeSharp,
FEdgeSmooth,
Id,
IntegrationType,
Interface0D,
Interface0DIterator,
Interface1D,
Iterator,
Material,
MediumType,
Nature,
Noise,
NonTVertex,
SShape,
SVertex,
SVertexIterator,
Stroke,
StrokeShader,
StrokeAttribute,
StrokeVertex,
StrokeVertexIterator,
TVertex,
UnaryFunction0D,
UnaryFunction0DDouble,
UnaryFunction0DEdgeNature,
UnaryFunction0DFloat,
UnaryFunction0DId,
UnaryFunction0DMaterial,
UnaryFunction0DUnsigned,
UnaryFunction0DVec2f,
UnaryFunction0DVec3f,
UnaryFunction0DVectorViewShape,
UnaryFunction0DViewShape,
UnaryFunction1D,
UnaryFunction1DDouble,
UnaryFunction1DEdgeNature,
UnaryFunction1DFloat,
UnaryFunction1DUnsigned,
UnaryFunction1DVec2f,
UnaryFunction1DVec3f,
UnaryFunction1DVectorViewShape,
UnaryFunction1DVoid,
UnaryPredicate0D,
UnaryPredicate1D,
ViewEdge,
ViewEdgeIterator,
ViewMap,
ViewShape,
ViewVertex,
orientedViewEdgeIterator,
)

View File

@@ -0,0 +1,24 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# module members
from _freestyle import (
ContextFunctions,
getCurrentScene,
integrate,
)

View File

@@ -26,6 +26,8 @@ import math
import mathutils
import time
from _freestyle import blendRamp, evaluateColorRamp, evaluateCurveMappingF
from ChainingIterators import pySketchyChainSilhouetteIterator, pySketchyChainingIterator
from freestyle import BackboneStretcherShader, BezierCurveShader, BinaryPredicate1D, ChainPredicateIterator, \
ChainSilhouetteIterator, ConstantColorShader, ContourUP1D, Curvature2DAngleF0D, ExternalContourUP1D, \

View File

@@ -1,713 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : ChainingIterators.py
# Author : Stephane Grabli
# Date : 04/08/2005
# Purpose : Chaining Iterators to be used with chaining operators
from freestyle import AdjacencyIterator, ChainingIterator, ExternalContourUP1D, Nature, TVertex
from freestyle import ContextFunctions as CF
import bpy
## the natural chaining iterator
## It follows the edges of same nature following the topology of
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
class pyChainSilhouetteIterator(ChainingIterator):
def __init__(self, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, True, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = it.object
it.increment()
if count != 1:
winner = None
break
return winner
## the natural chaining iterator
## It follows the edges of same nature on the same
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
## You can specify whether to chain iterate over edges that were
## already visited or not.
class pyChainSilhouetteGenericIterator(ChainingIterator):
def __init__(self, stayInSelection=True, stayInUnvisited=True):
ChainingIterator.__init__(self, stayInSelection, stayInUnvisited, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
ve = it.object
if ve.id == self.current_edge.id:
it.increment()
continue
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = ve
it.increment()
if count != 1:
winner = None
break
return winner
class pyExternalContourChainingIterator(ChainingIterator):
def __init__(self):
ChainingIterator.__init__(self, False, True, None, True)
self._isExternalContour = ExternalContourUP1D()
def init(self):
self._nEdges = 0
self._isInSelection = 1
def checkViewEdge(self, ve, orientation):
if orientation != 0:
vertex = ve.second_svertex()
else:
vertex = ve.first_svertex()
it = AdjacencyIterator(vertex,1,1)
while not it.is_end:
ave = it.object
if self._isExternalContour(ave):
return 1
it.increment()
print("pyExternlContourChainingIterator : didn't find next edge")
return 0
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if self._isExternalContour(ve):
if ve.time_stamp == CF.get_time_stamp():
winner = ve
it.increment()
self._nEdges = self._nEdges+1
if winner is None:
orient = 1
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if it.is_incoming:
orient = 0
good = self.checkViewEdge(ve,orient)
if good != 0:
winner = ve
it.increment()
return winner
## the natural chaining iterator
## with a sketchy multiple touch
class pySketchyChainSilhouetteIterator(ChainingIterator):
def __init__(self, nRounds=3,stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, False, None, True)
self._timeStamp = CF.get_time_stamp()+nRounds
self._nRounds = nRounds
def init(self):
self._timeStamp = CF.get_time_stamp()+self._nRounds
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
ve = it.object
if ve.id == self.current_edge.id:
it.increment()
continue
if (oNature & natures[i]) != 0:
if (natures[i] != oNature) != 0:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = ve
it.increment()
if count != 1:
winner = None
break
if winner is None:
winner = self.current_edge
if winner.chaining_time_stamp == self._timeStamp:
winner = None
return winner
# Chaining iterator designed for sketchy style.
# can chain several times the same ViewEdge
# in order to produce multiple strokes per ViewEdge.
class pySketchyChainingIterator(ChainingIterator):
def __init__(self, nRounds=3, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, False, None, True)
self._timeStamp = CF.get_time_stamp()+nRounds
self._nRounds = nRounds
def init(self):
self._timeStamp = CF.get_time_stamp()+self._nRounds
def traverse(self, iter):
winner = None
found = False
it = AdjacencyIterator(iter)
while not it.is_end:
ve = it.object
if ve.id == self.current_edge.id:
found = True
it.increment()
continue
winner = ve
it.increment()
if not found:
# This is a fatal error condition: self.current_edge must be found
# among the edges seen by the AdjacencyIterator [bug #35695].
if bpy.app.debug_freestyle:
print('pySketchyChainingIterator: current edge not found')
return None
if winner is None:
winner = self.current_edge
if winner.chaining_time_stamp == self._timeStamp:
return None
return winner
## Chaining iterator that fills small occlusions
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillOcclusionsRelativeChainingIterator(ChainingIterator):
def __init__(self, percent):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if connexl > self._percent * self._length:
winner = None
return winner
## Chaining iterator that fills small occlusions
## size
## The max length of the occluded part
## expressed in pixels
class pyFillOcclusionsAbsoluteChainingIterator(ChainingIterator):
def __init__(self, length):
ChainingIterator.__init__(self, False, True, None, True)
self._length = float(length)
def init(self):
pass
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if connexl > self._length:
winner = None
return winner
## Chaining iterator that fills small occlusions
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillOcclusionsAbsoluteAndRelativeChainingIterator(ChainingIterator):
def __init__(self, percent, l):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._absLength = l
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.time_stamp != CF.get_time_stamp():
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while _cit.is_end == 0 and _cit.object.time_stamp != CF.get_time_stamp():
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if (connexl > self._percent * self._length) or (connexl > self._absLength):
winner = None
return winner
## Chaining iterator that fills small occlusions without caring about the
## actual selection
## percent
## The max length of the occluded part
## expressed in % of the total chain length
class pyFillQi0AbsoluteAndRelativeChainingIterator(ChainingIterator):
def __init__(self, percent, l):
ChainingIterator.__init__(self, False, True, None, True)
self._length = 0
self._absLength = l
self._percent = float(percent)
def init(self):
# each time we're evaluating a chain length
# we try to do it once. Thus we reinit
# the chain length here:
self._length = 0
def traverse(self, iter):
winner = None
winnerOrientation = 0
#print(self.current_edge.id.first, self.current_edge.id.second)
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
if ve.id == mateVE.id:
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for nat in natures:
if (self.current_edge.nature & nat) != 0:
count=0
while not it.is_end:
ve = it.object
if (ve.nature & nat) != 0:
count = count+1
winner = ve
if not it.is_incoming:
winnerOrientation = 1
else:
winnerOrientation = 0
it.increment()
if count != 1:
winner = None
break
if winner is not None:
# check whether this edge was part of the selection
if winner.qi != 0:
#print("---", winner.id.first, winner.id.second)
# if not, let's check whether it's short enough with
# respect to the chain made without staying in the selection
#------------------------------------------------------------
# Did we compute the prospective chain length already ?
if self._length == 0:
#if not, let's do it
_it = pyChainSilhouetteGenericIterator(0,0)
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
_it.init()
while not _it.is_end:
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.increment()
if _it.is_begin:
break;
_it.begin = winner
_it.current_edge = winner
_it.orientation = winnerOrientation
if not _it.is_begin:
_it.decrement()
while (not _it.is_end) and (not _it.is_begin):
ve = _it.object
#print("--------", ve.id.first, ve.id.second)
self._length = self._length + ve.length_2d
_it.decrement()
# let's do the comparison:
# nw let's compute the length of this connex non selected part:
connexl = 0
_cit = pyChainSilhouetteGenericIterator(0,0)
_cit.begin = winner
_cit.current_edge = winner
_cit.orientation = winnerOrientation
_cit.init()
while not _cit.is_end and _cit.object.qi != 0:
ve = _cit.object
#print("-------- --------", ve.id.first, ve.id.second)
connexl = connexl + ve.length_2d
_cit.increment()
if (connexl > self._percent * self._length) or (connexl > self._absLength):
winner = None
return winner
## the natural chaining iterator
## It follows the edges of same nature on the same
## objects with preseance on silhouettes, then borders,
## then suggestive contours, then everything else. It doesn't chain the same ViewEdge twice
## You can specify whether to stay in the selection or not.
class pyNoIdChainSilhouetteIterator(ChainingIterator):
def __init__(self, stayInSelection=True):
ChainingIterator.__init__(self, stayInSelection, True, None, True)
def init(self):
pass
def traverse(self, iter):
winner = None
it = AdjacencyIterator(iter)
tvertex = self.next_vertex
if type(tvertex) is TVertex:
mateVE = tvertex.get_mate(self.current_edge)
while not it.is_end:
ve = it.object
feB = self.current_edge.last_fedge
feA = ve.first_fedge
vB = feB.second_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.first_fedge
feB = ve.last_fedge
vB = feB.second_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.last_fedge
feB = ve.last_fedge
vB = feB.second_svertex
vA = feA.second_svertex
if vA.id.first == vB.id.first:
winner = ve
break
feA = self.current_edge.first_fedge
feB = ve.first_fedge
vB = feB.first_svertex
vA = feA.first_svertex
if vA.id.first == vB.id.first:
winner = ve
break
it.increment()
else:
## case of NonTVertex
natures = [Nature.SILHOUETTE,Nature.BORDER,Nature.CREASE,Nature.SUGGESTIVE_CONTOUR,Nature.VALLEY,Nature.RIDGE]
for i in range(len(natures)):
currentNature = self.current_edge.nature
if (natures[i] & currentNature) != 0:
count=0
while not it.is_end:
visitNext = 0
oNature = it.object.nature
if (oNature & natures[i]) != 0:
if natures[i] != oNature:
for j in range(i):
if (natures[j] & oNature) != 0:
visitNext = 1
break
if visitNext != 0:
break
count = count+1
winner = it.object
it.increment()
if count != 1:
winner = None
break
return winner

View File

@@ -1,105 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : Functions0D.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 30/06/2005
# Purpose : Functions (functors) to be used for 0D elements
from freestyle import Curvature2DAngleF0D, CurvePoint, ReadCompleteViewMapPixelF0D, \
ReadSteerableViewMapPixelF0D, UnaryFunction0DDouble, UnaryFunction0DMaterial, \
UnaryFunction0DVec2f
from freestyle import ContextFunctions as CF
import math
import mathutils
class CurveMaterialF0D(UnaryFunction0DMaterial):
# A replacement of the built-in MaterialF0D for stroke creation.
# MaterialF0D does not work with Curves and Strokes.
def __call__(self, inter):
cp = inter.object
assert(isinstance(cp, CurvePoint))
fe = cp.first_svertex.get_fedge(cp.second_svertex)
assert(fe is not None)
return fe.material if fe.is_smooth else fe.material_left
class pyInverseCurvature2DAngleF0D(UnaryFunction0DDouble):
def __call__(self, inter):
func = Curvature2DAngleF0D()
c = func(inter)
return (3.1415 - c)
class pyCurvilinearLengthF0D(UnaryFunction0DDouble):
def __call__(self, inter):
cp = inter.object
assert(isinstance(cp, CurvePoint))
return cp.t2d
## estimate anisotropy of density
class pyDensityAnisotropyF0D(UnaryFunction0DDouble):
def __init__(self,level):
UnaryFunction0DDouble.__init__(self)
self.IsoDensity = ReadCompleteViewMapPixelF0D(level)
self.d0Density = ReadSteerableViewMapPixelF0D(0, level)
self.d1Density = ReadSteerableViewMapPixelF0D(1, level)
self.d2Density = ReadSteerableViewMapPixelF0D(2, level)
self.d3Density = ReadSteerableViewMapPixelF0D(3, level)
def __call__(self, inter):
c_iso = self.IsoDensity(inter)
c_0 = self.d0Density(inter)
c_1 = self.d1Density(inter)
c_2 = self.d2Density(inter)
c_3 = self.d3Density(inter)
cMax = max(max(c_0,c_1), max(c_2,c_3))
cMin = min(min(c_0,c_1), min(c_2,c_3))
if c_iso == 0:
v = 0
else:
v = (cMax-cMin)/c_iso
return v
## Returns the gradient vector for a pixel
## l
## the level at which one wants to compute the gradient
class pyViewMapGradientVectorF0D(UnaryFunction0DVec2f):
def __init__(self, l):
UnaryFunction0DVec2f.__init__(self)
self._l = l
self._step = math.pow(2,self._l)
def __call__(self, iter):
p = iter.object.point_2d
gx = CF.read_complete_view_map_pixel(self._l, int(p.x+self._step), int(p.y)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
gy = CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y+self._step)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
return mathutils.Vector([gx, gy])
class pyViewMapGradientNormF0D(UnaryFunction0DDouble):
def __init__(self, l):
UnaryFunction0DDouble.__init__(self)
self._l = l
self._step = math.pow(2,self._l)
def __call__(self, iter):
p = iter.object.point_2d
gx = CF.read_complete_view_map_pixel(self._l, int(p.x+self._step), int(p.y)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
gy = CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y+self._step)) - \
CF.read_complete_view_map_pixel(self._l, int(p.x), int(p.y))
grad = mathutils.Vector([gx, gy])
return grad.length

View File

@@ -1,58 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : Functions1D.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 08/04/2005
# Purpose : Functions (functors) to be used for 1D elements
from freestyle import GetProjectedZF1D, IntegrationType, UnaryFunction1DDouble, integrate
from Functions0D import pyDensityAnisotropyF0D, pyViewMapGradientNormF0D
import string
class pyGetInverseProjectedZF1D(UnaryFunction1DDouble):
def __call__(self, inter):
func = GetProjectedZF1D()
z = func(inter)
return (1.0 - z)
class pyGetSquareInverseProjectedZF1D(UnaryFunction1DDouble):
def __call__(self, inter):
func = GetProjectedZF1D()
z = func(inter)
return (1.0 - z*z)
class pyDensityAnisotropyF1D(UnaryFunction1DDouble):
def __init__(self,level, integrationType=IntegrationType.MEAN, sampling=2.0):
UnaryFunction1DDouble.__init__(self, integrationType)
self._func = pyDensityAnisotropyF0D(level)
self._integration = integrationType
self._sampling = sampling
def __call__(self, inter):
v = integrate(self._func, inter.pointsBegin(self._sampling), inter.pointsEnd(self._sampling), self._integration)
return v
class pyViewMapGradientNormF1D(UnaryFunction1DDouble):
def __init__(self,l, integrationType, sampling=2.0):
UnaryFunction1DDouble.__init__(self, integrationType)
self._func = pyViewMapGradientNormF0D(l)
self._integration = integrationType
self._sampling = sampling
def __call__(self, inter):
v = integrate(self._func, inter.pointsBegin(self._sampling), inter.pointsEnd(self._sampling), self._integration)
return v

View File

@@ -1,73 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : PredicatesB1D.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 08/04/2005
# Purpose : Binary predicates (functors) to be used for 1D elements
from freestyle import BinaryPredicate1D, GetZF1D, IntegrationType, Nature, SameShapeIdBP1D, ZDiscontinuityF1D
from Functions1D import pyViewMapGradientNormF1D
import random
class pyZBP1D(BinaryPredicate1D):
def __call__(self, i1, i2):
func = GetZF1D()
return (func(i1) > func(i2))
class pyZDiscontinuityBP1D(BinaryPredicate1D):
def __init__(self, iType = IntegrationType.MEAN):
BinaryPredicate1D.__init__(self)
self._GetZDiscontinuity = ZDiscontinuityF1D(iType)
def __call__(self, i1, i2):
return (self._GetZDiscontinuity(i1) > self._GetZDiscontinuity(i2))
class pyLengthBP1D(BinaryPredicate1D):
def __call__(self, i1, i2):
return (i1.length_2d > i2.length_2d)
class pySilhouetteFirstBP1D(BinaryPredicate1D):
def __call__(self, inter1, inter2):
bpred = SameShapeIdBP1D()
if (bpred(inter1, inter2) != 1):
return 0
if (inter1.nature & Nature.SILHOUETTE):
return (inter2.nature & Nature.SILHOUETTE) != 0
return (inter1.nature == inter2.nature)
class pyNatureBP1D(BinaryPredicate1D):
def __call__(self, inter1, inter2):
return (inter1.nature & inter2.nature)
class pyViewMapGradientNormBP1D(BinaryPredicate1D):
def __init__(self,l, sampling=2.0):
BinaryPredicate1D.__init__(self)
self._GetGradient = pyViewMapGradientNormF1D(l, IntegrationType.MEAN)
def __call__(self, i1,i2):
#print("compare gradient")
return (self._GetGradient(i1) > self._GetGradient(i2))
class pyShuffleBP1D(BinaryPredicate1D):
def __init__(self):
BinaryPredicate1D.__init__(self)
random.seed(1)
def __call__(self, inter1, inter2):
r1 = random.uniform(0,1)
r2 = random.uniform(0,1)
return (r1<r2)

View File

@@ -1,96 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : PredicatesU0D.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 08/04/2005
# Purpose : Unary predicates (functors) to be used for 0D elements
from freestyle import Curvature2DAngleF0D, Nature, QuantitativeInvisibilityF0D, UnaryPredicate0D
from Functions0D import pyCurvilinearLengthF0D
class pyHigherCurvature2DAngleUP0D(UnaryPredicate0D):
def __init__(self,a):
UnaryPredicate0D.__init__(self)
self._a = a
def __call__(self, inter):
func = Curvature2DAngleF0D()
a = func(inter)
return (a > self._a)
class pyUEqualsUP0D(UnaryPredicate0D):
def __init__(self,u, w):
UnaryPredicate0D.__init__(self)
self._u = u
self._w = w
def __call__(self, inter):
func = pyCurvilinearLengthF0D()
u = func(inter)
return (u > (self._u-self._w)) and (u < (self._u+self._w))
class pyVertexNatureUP0D(UnaryPredicate0D):
def __init__(self,nature):
UnaryPredicate0D.__init__(self)
self._nature = nature
def __call__(self, inter):
v = inter.object
return (v.nature & self._nature) != 0
## check whether an Interface0DIterator
## is a TVertex and is the one that is
## hidden (inferred from the context)
class pyBackTVertexUP0D(UnaryPredicate0D):
def __init__(self):
UnaryPredicate0D.__init__(self)
self._getQI = QuantitativeInvisibilityF0D()
def __call__(self, iter):
if (iter.object.nature & Nature.T_VERTEX) == 0:
return 0
if iter.is_end:
return 0
if self._getQI(iter) != 0:
return 1
return 0
class pyParameterUP0DGoodOne(UnaryPredicate0D):
def __init__(self,pmin,pmax):
UnaryPredicate0D.__init__(self)
self._m = pmin
self._M = pmax
#self.getCurvilinearAbscissa = GetCurvilinearAbscissaF0D()
def __call__(self, inter):
#s = self.getCurvilinearAbscissa(inter)
u = inter.u
#print(u)
return ((u>=self._m) and (u<=self._M))
class pyParameterUP0D(UnaryPredicate0D):
def __init__(self,pmin,pmax):
UnaryPredicate0D.__init__(self)
self._m = pmin
self._M = pmax
#self.getCurvilinearAbscissa = GetCurvilinearAbscissaF0D()
def __call__(self, inter):
func = Curvature2DAngleF0D()
c = func(inter)
b1 = (c>0.1)
#s = self.getCurvilinearAbscissa(inter)
u = inter.u
#print(u)
b = ((u>=self._m) and (u<=self._M))
return b and b1

View File

@@ -1,342 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : PredicatesU1D.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 08/04/2005
# Purpose : Unary predicates (functors) to be used for 1D elements
from freestyle import Curvature2DAngleF0D, CurveNatureF1D, DensityF1D, GetCompleteViewMapDensityF1D, \
GetDirectionalViewMapDensityF1D, GetOccludersF1D, GetProjectedZF1D, GetShapeF1D, GetSteerableViewMapDensityF1D, \
IntegrationType, ShapeUP1D, TVertex, UnaryPredicate1D
from Functions1D import pyDensityAnisotropyF1D, pyViewMapGradientNormF1D
class pyNFirstUP1D(UnaryPredicate1D):
def __init__(self, n):
UnaryPredicate1D.__init__(self)
self.__n = n
self.__count = 0
def __call__(self, inter):
self.__count = self.__count + 1
if self.__count <= self.__n:
return 1
return 0
class pyHigherLengthUP1D(UnaryPredicate1D):
def __init__(self,l):
UnaryPredicate1D.__init__(self)
self._l = l
def __call__(self, inter):
return (inter.length_2d > self._l)
class pyNatureUP1D(UnaryPredicate1D):
def __init__(self,nature):
UnaryPredicate1D.__init__(self)
self._nature = nature
self._getNature = CurveNatureF1D()
def __call__(self, inter):
if(self._getNature(inter) & self._nature):
return 1
return 0
class pyHigherNumberOfTurnsUP1D(UnaryPredicate1D):
def __init__(self,n,a):
UnaryPredicate1D.__init__(self)
self._n = n
self._a = a
def __call__(self, inter):
count = 0
func = Curvature2DAngleF0D()
it = inter.vertices_begin()
while not it.is_end:
if func(it) > self._a:
count = count+1
if count > self._n:
return 1
it.increment()
return 0
class pyDensityUP1D(UnaryPredicate1D):
def __init__(self,wsize,threshold, integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._wsize = wsize
self._threshold = threshold
self._integration = integration
self._func = DensityF1D(self._wsize, self._integration, sampling)
def __call__(self, inter):
if self._func(inter) < self._threshold:
return 1
return 0
class pyLowSteerableViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
def __call__(self, inter):
func = GetSteerableViewMapDensityF1D(self._level, self._integration)
v = func(inter)
#print(v)
if v < self._threshold:
return 1
return 0
class pyLowDirectionalViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, orientation, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._orientation = orientation
self._level = level
self._integration = integration
def __call__(self, inter):
func = GetDirectionalViewMapDensityF1D(self._orientation, self._level, self._integration)
v = func(inter)
#print(v)
if v < self._threshold:
return 1
return 0
class pyHighSteerableViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
self._func = GetSteerableViewMapDensityF1D(self._level, self._integration)
def __call__(self, inter):
v = self._func(inter)
if v > self._threshold:
return 1
return 0
class pyHighDirectionalViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, orientation, level,integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._orientation = orientation
self._level = level
self._integration = integration
self._sampling = sampling
def __call__(self, inter):
func = GetDirectionalViewMapDensityF1D(self._orientation, self._level, self._integration, self._sampling)
v = func(inter)
if v > self._threshold:
return 1
return 0
class pyHighViewMapDensityUP1D(UnaryPredicate1D):
def __init__(self,threshold, level,integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._level = level
self._integration = integration
self._sampling = sampling
self._func = GetCompleteViewMapDensityF1D(self._level, self._integration, self._sampling) # 2.0 is the smpling
def __call__(self, inter):
#print("toto")
#print(func.name)
#print(inter.name)
v= self._func(inter)
if v > self._threshold:
return 1
return 0
class pyDensityFunctorUP1D(UnaryPredicate1D):
def __init__(self,wsize,threshold, functor, funcmin=0.0, funcmax=1.0, integration = IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._wsize = wsize
self._threshold = float(threshold)
self._functor = functor
self._funcmin = float(funcmin)
self._funcmax = float(funcmax)
self._integration = integration
def __call__(self, inter):
func = DensityF1D(self._wsize, self._integration)
res = self._functor(inter)
k = (res-self._funcmin)/(self._funcmax-self._funcmin)
if func(inter) < self._threshold*k:
return 1
return 0
class pyZSmallerUP1D(UnaryPredicate1D):
def __init__(self,z, integration=IntegrationType.MEAN):
UnaryPredicate1D.__init__(self)
self._z = z
self._integration = integration
def __call__(self, inter):
func = GetProjectedZF1D(self._integration)
if func(inter) < self._z:
return 1
return 0
class pyIsOccludedByUP1D(UnaryPredicate1D):
def __init__(self,id):
UnaryPredicate1D.__init__(self)
self._id = id
def __call__(self, inter):
func = GetShapeF1D()
shapes = func(inter)
for s in shapes:
if(s.id == self._id):
return 0
it = inter.vertices_begin()
itlast = inter.vertices_end()
itlast.decrement()
v = it.object
vlast = itlast.object
tvertex = v.viewvertex
if type(tvertex) is TVertex:
#print("TVertex: [ ", tvertex.id.first, ",", tvertex.id.second," ]")
eit = tvertex.edges_begin()
while not eit.is_end:
ve, incoming = eit.object
if ve.id == self._id:
return 1
#print("-------", ve.id.first, "-", ve.id.second)
eit.increment()
tvertex = vlast.viewvertex
if type(tvertex) is TVertex:
#print("TVertex: [ ", tvertex.id.first, ",", tvertex.id.second," ]")
eit = tvertex.edges_begin()
while not eit.is_end:
ve, incoming = eit.object
if ve.id == self._id:
return 1
#print("-------", ve.id.first, "-", ve.id.second)
eit.increment()
return 0
class pyIsInOccludersListUP1D(UnaryPredicate1D):
def __init__(self,id):
UnaryPredicate1D.__init__(self)
self._id = id
def __call__(self, inter):
func = GetOccludersF1D()
occluders = func(inter)
for a in occluders:
if a.id == self._id:
return 1
return 0
class pyIsOccludedByItselfUP1D(UnaryPredicate1D):
def __init__(self):
UnaryPredicate1D.__init__(self)
self.__func1 = GetOccludersF1D()
self.__func2 = GetShapeF1D()
def __call__(self, inter):
lst1 = self.__func1(inter)
lst2 = self.__func2(inter)
for vs1 in lst1:
for vs2 in lst2:
if vs1.id == vs2.id:
return 1
return 0
class pyIsOccludedByIdListUP1D(UnaryPredicate1D):
def __init__(self, idlist):
UnaryPredicate1D.__init__(self)
self._idlist = idlist
self.__func1 = GetOccludersF1D()
def __call__(self, inter):
lst1 = self.__func1(inter)
for vs1 in lst1:
for _id in self._idlist:
if vs1.id == _id:
return 1
return 0
class pyShapeIdListUP1D(UnaryPredicate1D):
def __init__(self,idlist):
UnaryPredicate1D.__init__(self)
self._idlist = idlist
self._funcs = []
for _id in idlist :
self._funcs.append(ShapeUP1D(_id.first, _id.second))
def __call__(self, inter):
for func in self._funcs :
if func(inter) == 1:
return 1
return 0
## deprecated
class pyShapeIdUP1D(UnaryPredicate1D):
def __init__(self, _id):
UnaryPredicate1D.__init__(self)
self._id = _id
def __call__(self, inter):
func = GetShapeF1D()
shapes = func(inter)
for a in shapes:
if a.id == self._id:
return 1
return 0
class pyHighDensityAnisotropyUP1D(UnaryPredicate1D):
def __init__(self,threshold, level, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._l = threshold
self.func = pyDensityAnisotropyF1D(level, IntegrationType.MEAN, sampling)
def __call__(self, inter):
return (self.func(inter) > self._l)
class pyHighViewMapGradientNormUP1D(UnaryPredicate1D):
def __init__(self,threshold, l, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._threshold = threshold
self._GetGradient = pyViewMapGradientNormF1D(l, IntegrationType.MEAN)
def __call__(self, inter):
gn = self._GetGradient(inter)
#print(gn)
return (gn > self._threshold)
class pyDensityVariableSigmaUP1D(UnaryPredicate1D):
def __init__(self,functor, sigmaMin,sigmaMax, lmin, lmax, tmin, tmax, integration = IntegrationType.MEAN, sampling=2.0):
UnaryPredicate1D.__init__(self)
self._functor = functor
self._sigmaMin = float(sigmaMin)
self._sigmaMax = float(sigmaMax)
self._lmin = float(lmin)
self._lmax = float(lmax)
self._tmin = tmin
self._tmax = tmax
self._integration = integration
self._sampling = sampling
def __call__(self, inter):
sigma = (self._sigmaMax-self._sigmaMin)/(self._lmax-self._lmin)*(self._functor(inter)-self._lmin) + self._sigmaMin
t = (self._tmax-self._tmin)/(self._lmax-self._lmin)*(self._functor(inter)-self._lmin) + self._tmin
if sigma < self._sigmaMin:
sigma = self._sigmaMin
self._func = DensityF1D(sigma, self._integration, self._sampling)
d = self._func(inter)
if d < t:
return 1
return 0
class pyClosedCurveUP1D(UnaryPredicate1D):
def __call__(self, inter):
it = inter.vertices_begin()
itlast = inter.vertices_end()
itlast.decrement()
vlast = itlast.object
v = it.object
#print(v.id.first, v.id.second)
#print(vlast.id.first, vlast.id.second)
if v.id == vlast.id:
return 1
return 0

View File

@@ -1,47 +0,0 @@
# ##### BEGIN GPL LICENSE BLOCK #####
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#
# ##### END GPL LICENSE BLOCK #####
# Filename : logical_operators.py
# Authors : Fredo Durand, Stephane Grabli, Francois Sillion, Emmanuel Turquin
# Date : 08/04/2005
# Purpose : Logical unary predicates (functors) for 1D elements
from freestyle import UnaryPredicate1D
class AndUP1D(UnaryPredicate1D):
def __init__(self, pred1, pred2):
UnaryPredicate1D.__init__(self)
self.__pred1 = pred1
self.__pred2 = pred2
def __call__(self, inter):
return self.__pred1(inter) and self.__pred2(inter)
class OrUP1D(UnaryPredicate1D):
def __init__(self, pred1, pred2):
UnaryPredicate1D.__init__(self)
self.__pred1 = pred1
self.__pred2 = pred2
def __call__(self, inter):
return self.__pred1(inter) or self.__pred2(inter)
class NotUP1D(UnaryPredicate1D):
def __init__(self, pred):
UnaryPredicate1D.__init__(self)
self.__pred = pred
def __call__(self, inter):
return not self.__pred(inter)

File diff suppressed because it is too large Load Diff

View File

@@ -55,7 +55,7 @@ void Path::setRootDir(const string& iRootDir)
string(DIR_SEP.c_str()) + "variation_patterns" + string(DIR_SEP.c_str());
_BrushesPath = _ProjectDir + string(DIR_SEP.c_str()) + "data" + string(DIR_SEP.c_str()) + "textures" +
string(DIR_SEP.c_str()) + "brushes" + string(DIR_SEP.c_str());
_PythonPath = _ProjectDir + string(DIR_SEP.c_str()) + "style_modules" + string(DIR_SEP.c_str());
_PythonPath = _ProjectDir + string(DIR_SEP.c_str()) + "modules" + string(DIR_SEP.c_str());
if (getenv("PYTHONPATH")) {
_PythonPath += string(PATH_SEP.c_str()) + string(getenv("PYTHONPATH"));
}

View File

@@ -469,7 +469,7 @@ static PyMethodDef module_functions[] = {
static PyModuleDef module_definition = {
PyModuleDef_HEAD_INIT,
"freestyle",
"_freestyle",
module_docstring,
-1,
module_functions