def getScreenshot(p3dApp):
p3dApp.taskMgr.step()
p3dApp.taskMgr.step()
pnmss = PNMImage()
p3dApp.win.getScreenshot(pnmss)
resulting_ss = StringStream()
pnmss.write(resulting_ss, "screenshot.png")
screenshot_buffer = resulting_ss.getData()
pilimage = Image.open(StringIO(screenshot_buffer))
pilimage.load()
#pnmimage will sometimes output as palette mode for 8-bit png so convert
pilimage = pilimage.convert('RGBA')
return pilimage
def getScreenshot(p3dApp):
p3dApp.taskMgr.step()
p3dApp.taskMgr.step()
pnmss = PNMImage()
p3dApp.win.getScreenshot(pnmss)
resulting_ss = StringStream()
pnmss.write(resulting_ss, "screenshot.png")
screenshot_buffer = resulting_ss.getData()
pilimage = Image.open(StringIO(screenshot_buffer))
pilimage.load()
#pnmimage will sometimes output as palette mode for 8-bit png so convert
pilimage = pilimage.convert('RGBA')
return pilimage
def pilFromData(image_data):
try:
im = Image.open(StringIO(image_data))
im.load()
except IOError:
#PIL couldn't open, so try to read with panda3d which supports DDS:
im = None
tex = textureFromData(image_data)
if tex is not None:
outdata = tex.getRamImageAs('RGB').getData()
try:
im = Image.fromstring('RGB', (tex.getXSize(), tex.getYSize()), outdata)
im.load()
except IOError:
#Any problem with panda3d might generate an invalid image buffer, so don't convert this
im = None
return im
def pilFromData(image_data):
try:
im = Image.open(StringIO(image_data))
im.load()
except IOError:
#PIL couldn't open, so try to read with panda3d which supports DDS:
im = None
tex = textureFromData(image_data)
if tex is not None:
outdata = tex.getRamImageAs('RGB').getData()
try:
im = Image.fromstring('RGB', (tex.getXSize(), tex.getYSize()), outdata)
im.load()
except IOError:
#Any problem with panda3d might generate an invalid image buffer, so don't convert this
im = None
return im
def getMipMaps(mesh):
mipmaps = {}
for effect in mesh.effects:
for prop in effect.supported:
propval = getattr(effect, prop)
if isinstance(propval, collada.material.Map):
image_name = propval.sampler.surface.image.path
image_data = propval.sampler.surface.image.data
try:
im = Image.open(StringIO(image_data))
im.load()
except IOError:
from panda3d.core import Texture
from panda3d.core import StringStream
from panda3d.core import PNMImage
#PIL failed, so lets try DDS reader with panda3d
t = Texture(image_name)
success = t.readDds(StringStream(image_data))
if success == 0:
raise FilterException("Failed to read image file %s" %
image_name)
#convert DDS to PNG
outdata = t.getRamImageAs('RGBA').getData()
try:
im = Image.fromstring('RGBA',
(t.getXSize(), t.getYSize()),
outdata)
im.load()
except IOError:
raise FilterException("Failed to read image file %s" %
image_name)
#Keep JPG in same format since JPG->PNG is pretty bad
if im.format == 'JPEG':
output_format = 'JPEG'
output_extension = 'jpg'
output_options = {'quality': 95, 'optimize': True}
else:
output_format = 'PNG'
output_extension = 'png'
output_options = {'optimize': True}
#store a copy to the original image so we can resize from it directly each time
orig_im = im
width, height = im.size
#round down to power of 2
width = int(math.pow(2, int(math.log(width, 2))))
height = int(math.pow(2, int(math.log(height, 2))))
pil_images = []
while True:
im = orig_im.resize((width, height), Image.ANTIALIAS)
pil_images.insert(0, im)
if width == 1 and height == 1:
break
width = max(width / 2, 1)
height = max(height / 2, 1)
tar_buf = StringIO()
tar = tarfile.TarFile(fileobj=tar_buf, mode='w')
cur_offset = 0
byte_ranges = []
for i, pil_img in enumerate(pil_images):
buf = StringIO()
pil_img.save(buf, output_format, **output_options)
file_len = buf.tell()
cur_name = '%dx%d.%s' % (pil_img.size[0], pil_img.size[1],
output_extension)
tar_info = tarfile.TarInfo(name=cur_name)
tar_info.size = file_len
buf.seek(0)
tar.addfile(tarinfo=tar_info, fileobj=buf)
#tar files have a 512 byte header
cur_offset += 512
file_start = cur_offset
byte_ranges.append({
'offset': file_start,
'length': file_len,
'width': pil_img.size[0],
'height': pil_img.size[1]
})
#file lengths are rounded up to nearest 512 multiple
file_len = 512 * ((file_len + 512 - 1) / 512)
cur_offset += file_len
tar.close()
mipmaps[propval.sampler.surface.image.path] = (
tar_buf.getvalue(), byte_ranges)
return mipmaps
def getMipMaps(mesh):
mipmaps = {}
for effect in mesh.effects:
for prop in effect.supported:
propval = getattr(effect, prop)
if isinstance(propval, collada.material.Map):
image_name = propval.sampler.surface.image.path
image_data = propval.sampler.surface.image.data
try:
im = Image.open(StringIO(image_data))
im.load()
except IOError:
from panda3d.core import Texture
from panda3d.core import StringStream
from panda3d.core import PNMImage
#PIL failed, so lets try DDS reader with panda3d
t = Texture(image_name)
success = t.readDds(StringStream(image_data))
if success == 0:
raise FilterException("Failed to read image file %s" % image_name)
#convert DDS to PNG
outdata = t.getRamImageAs('RGBA').getData()
try:
im = Image.fromstring('RGBA', (t.getXSize(), t.getYSize()), outdata)
im.load()
except IOError:
raise FilterException("Failed to read image file %s" % image_name)
#Keep JPG in same format since JPG->PNG is pretty bad
if im.format == 'JPEG':
output_format = 'JPEG'
output_extension = 'jpg'
output_options = {'quality': 95, 'optimize':True}
else:
output_format = 'PNG'
output_extension = 'png'
output_options = {'optimize':True}
#store a copy to the original image so we can resize from it directly each time
orig_im = im
width, height = im.size
#round down to power of 2
width = int(math.pow(2, int(math.log(width, 2))))
height = int(math.pow(2, int(math.log(height, 2))))
pil_images = []
while True:
im = orig_im.resize((width, height), Image.ANTIALIAS)
pil_images.insert(0, im)
if width == 1 and height == 1:
break
width = max(width / 2, 1)
height = max(height / 2, 1)
tar_buf = StringIO()
tar = tarfile.TarFile(fileobj=tar_buf, mode='w')
cur_offset = 0
byte_ranges = []
for i, pil_img in enumerate(pil_images):
buf = StringIO()
pil_img.save(buf, output_format, **output_options)
file_len = buf.tell()
cur_name = '%dx%d.%s' % (pil_img.size[0], pil_img.size[1], output_extension)
tar_info = tarfile.TarInfo(name=cur_name)
tar_info.size=file_len
buf.seek(0)
tar.addfile(tarinfo=tar_info, fileobj=buf)
#tar files have a 512 byte header
cur_offset += 512
file_start = cur_offset
byte_ranges.append({'offset':file_start,
'length':file_len,
'width':pil_img.size[0],
'height':pil_img.size[1]})
#file lengths are rounded up to nearest 512 multiple
file_len = 512 * ((file_len + 512 - 1) / 512)
cur_offset += file_len
tar.close()
mipmaps[propval.sampler.surface.image.path] = (tar_buf.getvalue(), byte_ranges)
return mipmaps
