Cool_Tools/bemaniutils
1
0
Fork 0
mirror of synced 2024-09-24 03:18:22 +02:00
Code Releases Activity

Implement proper affine transformations instead of just translation.

Browse Source
This commit is contained in:
Jennifer Taylor 2021-04-17 23:31:36 +00:00
parent c6e19d0dfa
commit e1c6ad429c
2 changed files with 100 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

98
bemani/format/afp/render.py
View File

@ -261,18 +261,15 @@ class AFPRenderer(VerboseOutput):
# Double check supported options.
if tag.mult_color or tag.add_color:
# TODO: Handle additive and multiplicative color.
print(f"WARNING: Unhandled color blend request Mult: {tag.mult_color} Add: {tag.add_color}!")
# Look up the affine transformation matrix and rotation/origin.
transform = tag.transform or Matrix.identity()
origin = tag.rotation_offset or Point.identity()
transform = parent_transform.multiply(tag.transform or Matrix.identity())
origin = parent_origin.add(tag.rotation_offset or Point.identity())
# TODO: Need to do actual affine transformations here.
if transform.b != 0.0 or transform.c != 0.0 or transform.a != 1.0 or transform.d != 1.0:
print("WARNING: Unhandled affine transformation request!")
if parent_transform.b != 0.0 or parent_transform.c != 0.0 or parent_transform.a != 1.0 or parent_transform.d != 1.0:
print("WARNING: Unhandled affine transformation request!")
offset = parent_transform.multiply_point(transform.multiply_point(Point.identity().subtract(origin).subtract(parent_origin)))
# Calculate the inverse so we can map canvas space back to texture space.
inverse = transform.inverse()
# Look up source shape.
if tag.source_tag_id not in self.__registered_shapes:
@ -310,18 +307,60 @@ class AFPRenderer(VerboseOutput):
if texture is not None:
# Now, render out the texture.
cutin = Point(offset.x, offset.y)
cutoff = Point.identity()
if cutin.x < 0:
cutoff.x = -cutin.x
cutin.x = 0
if cutin.y < 0:
cutoff.y = -cutin.y
cutin.y = 0
imgmap = list(img.getdata())
texmap = list(texture.getdata())
# Calculate the maximum range of update this texture can possibly reside in.
pix1 = transform.multiply_point(Point.identity().subtract(origin))
pix2 = transform.multiply_point(Point.identity().subtract(origin).add(Point(texture.width, 0)))
pix3 = transform.multiply_point(Point.identity().subtract(origin).add(Point(0, texture.height)))
pix4 = transform.multiply_point(Point.identity().subtract(origin).add(Point(texture.width, texture.height)))
# Map this to the rectangle we need to sweep in the rendering image.
minx = max(int(min(pix1.x, pix2.x, pix3.x, pix4.x)), 0)
maxx = min(int(max(pix1.x, pix2.x, pix3.x, pix4.x)) + 1, img.width)
miny = max(int(min(pix1.y, pix2.y, pix3.y, pix4.y)), 0)
maxy = min(int(max(pix1.y, pix2.y, pix3.y, pix4.y)) + 1, img.height)
for imgy in range(miny, maxy):
for imgx in range(minx, maxx):
# Determine offset
imgoff = imgx + (imgy * img.width)
# Calculate what texture pixel data goes here.
texloc = inverse.multiply_point(Point(float(imgx), float(imgy))).add(origin)
texx, texy = texloc.as_tuple()
# If we're out of bounds, don't update.
if texx < 0 or texy < 0 or texx >= texture.width or texy >= texture.height:
continue
# Blend it.
texoff = texx + (texy * texture.width)
imgmap[imgoff] = self.__blend(imgmap[imgoff], texmap[texoff])
img = Image.new("RGBA", (img.width, img.height))
img.putdata(imgmap)
img.alpha_composite(texture, cutin.as_tuple(), cutoff.as_tuple())
return img
def __blend(self, bg: Tuple[int, int, int, int], fg: Tuple[int, int, int, int]) -> Tuple[int, int, int, int]:
# Short circuit for speed.
if fg[3] == 0:
return bg
if fg[3] == 255:
return fg
# Calculate alpha blending.
fgpercent = (float(fg[3]) / 255.0)
bgpercent = 1.0 - fgpercent
return (
max(int(float(bg[0]) * bgpercent + float(fg[0]) * fgpercent), 255),
max(int(float(bg[1]) * bgpercent + float(fg[1]) * fgpercent), 255),
max(int(float(bg[2]) * bgpercent + float(fg[2]) * fgpercent), 255),
255,
)
def __render(self, swf: SWF, export_tag: Optional[str]) -> Tuple[int, List[Image.Image]]:
# If we are rendering an exported tag, we want to perform the actions of the
# rest of the SWF but not update any layers as a result.
@ -349,10 +388,10 @@ class AFPRenderer(VerboseOutput):
# Create a new image to render into.
time = spf * float(frameno)
color = swf.color or Color(0.0, 0.0, 0.0, 0.0)
curimage = Image.new("RGBA", (swf.location.width, swf.location.height), color=color.as_tuple())
self.vprint(f"Rendering Frame {frameno} ({time}s)")
# Go through all registered clips, place all needed tags.
changed = False
while any(c.dirty for c in self.__clips):
newclips: List[Clip] = []
for clip in self.__clips:
@ -361,6 +400,7 @@ class AFPRenderer(VerboseOutput):
self.vprint(f" Sprite Tag ID: {clip.tag_id}, Current Frame: {clip.frame.start_tag_offset + clip.frame.current_tag}, Num Frames: {clip.frame.num_tags}")
newclips.extend(self.__place(clip.tags[clip.frame.start_tag_offset + clip.frame.current_tag], parent_clip=clip.tag_id))
clip.frame.current_tag += 1
changed = True
if clip.dirty and clip.frame.current_tag == clip.frame.num_tags:
# We handled this clip.
@ -369,15 +409,21 @@ class AFPRenderer(VerboseOutput):
# Add any new clips that we should process next frame.
self.__clips.extend(newclips)
# Now, render out the placed objects. We sort by depth so that we can
# get the layering correct, but its important to preserve the original
# insertion order for delete requests.
for obj in sorted(self.__placed_objects, key=lambda obj: obj.depth):
if self.__visible_tag != obj.parent_clip:
continue
if changed or frameno == 0:
# Now, render out the placed objects. We sort by depth so that we can
# get the layering correct, but its important to preserve the original
# insertion order for delete requests.
curimage = Image.new("RGBA", (swf.location.width, swf.location.height), color=color.as_tuple())
for obj in sorted(self.__placed_objects, key=lambda obj: obj.depth):
if self.__visible_tag != obj.parent_clip:
continue
self.vprint(f" Rendering placed object ID {obj.object_id} from sprite {obj.parent_clip} onto Depth {obj.depth}")
curimage = self.__render_object(curimage, obj.tag, Matrix.identity(), Point.identity())
self.vprint(f" Rendering placed object ID {obj.object_id} from sprite {obj.parent_clip} onto Depth {obj.depth}")
curimage = self.__render_object(curimage, obj.tag, Matrix.identity(), Point.identity())
else:
# Nothing changed, make a copy of the previous render.
self.vprint(" Using previous frame render")
curimage = frames[-1].copy()
# Advance all the clips and frame now that we processed and rendered them.
for clip in self.__clips:

34
bemani/format/afp/types/generic.py
View File

@ -47,14 +47,14 @@ class Point:
return (int(self.x), int(self.y))
def add(self, other: "Point") -> "Point":
self.x += other.x
self.y += other.y
return self
x = self.x + other.x
y = self.y + other.y
return Point(x, y)
def subtract(self, other: "Point") -> "Point":
self.x -= other.x
self.y -= other.y
return self
x = self.x - other.x
y = self.y - other.y
return Point(x, y)
def __repr__(self) -> str:
return f"x: {round(self.x, 5)}, y: {round(self.y, 5)}"
@ -110,5 +110,27 @@ class Matrix:
y=(self.b * point.x) + (self.d * point.y) + self.ty,
)
def multiply(self, other: "Matrix") -> "Matrix":
return Matrix(
a=self.a * other.a + self.b * other.c,
b=self.a * other.b + self.b * other.d,
c=self.c * other.a + self.d * other.c,
d=self.c * other.b + self.d * other.d,
tx=self.tx * other.a + self.ty * other.c + other.tx,
ty=self.tx * other.b + self.ty * other.d + other.ty,
)
def inverse(self) -> "Matrix":
denom = (self.a * self.d - self.b * self.c)
return Matrix(
a=self.d / denom,
b=-self.b / denom,
c=-self.c / denom,
d=self.a / denom,
tx=(self.c * self.ty - self.d * self.tx) / denom,
ty=-(self.a * self.ty - self.b * self.tx) / denom,
)
def __repr__(self) -> str:
return f"a: {round(self.a, 5)}, b: {round(self.b, 5)}, c: {round(self.c, 5)}, d: {round(self.d, 5)}, tx: {round(self.tx, 5)}, ty: {round(self.ty, 5)}"