jyapayne/glymur
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Speed up of about 33%, maybe? #210

Browse source 
Added a real fix for #209, seeing about a 22% speed up there.
This commit is contained in:
John Evans 2014-04-02 11:28:00 -04:00
commit eb1df90fb9
1 changed files with 14 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

32
glymur/jp2box.py
View file

@ -1882,21 +1882,14 @@ class PaletteBox(Jp2kBox):
fptr.write(write_buffer) fptr.write(write_buffer)
bps = self.bits_per_component bps = self.bits_per_component
if any(b != bps[0] for b in bps): if all(b == bps[0] for b in bps):
# All components are the same. Writing is straightforward. # All components are the same. Writing is straightforward.
if self.bits_per_component[0] <= 8: if self.bits_per_component[0] <= 8:
code = 'B' write_buffer = np.getbuffer(self.palette.astype(np.uint8))
dtype = np.uint8
elif self.bits_per_component[0] <= 16: elif self.bits_per_component[0] <= 16:
code = 'H' write_buffer = np.getbuffer(self.palette.astype(np.uint16))
dtype = np.uint16
elif self.bits_per_component[0] <= 32: elif self.bits_per_component[0] <= 32:
code = 'I' write_buffer = np.getbuffer(self.palette.astype(np.uint32))
dtype = np.uint32
nelts = self.palette.shape[0] * self.palette.shape[1]
fmt = '>{0}{1}'.format(nelts, code)
write_buffer = struct.pack(fmt,
self.palette.astype(dtype).flatten())
fptr.write(write_buffer) fptr.write(write_buffer)
else: else:
# Not all the components are the same. More general, but much rarer # Not all the components are the same. More general, but much rarer
@ -1937,19 +1930,22 @@ class PaletteBox(Jp2kBox):
bps = [((x & 0x7f) + 1) for x in bps_signed] bps = [((x & 0x7f) + 1) for x in bps_signed]
signed = [((x & 0x80) > 1) for x in bps_signed] signed = [((x & 0x80) > 1) for x in bps_signed]
if any(b != bps_signed[0] for b in bps_signed): if all(b == bps_signed[0] for b in bps_signed):
# Ok the palette has the same datatype for all columns. We should # Ok the palette has the same datatype for all columns. We should
# be able to efficiently read it. # be able to efficiently read it.
if bps <= 8: if bps[0] <= 8:
nbytes_per_row = num_columns
dtype = np.uint8 dtype = np.uint8
elif bps <= 16: elif bps[0] <= 16:
nbytes_per_row = 2 * num_columns
dtype = np.uint16 dtype = np.uint16
elif bps <= 32: elif bps[0] <= 32:
nbytes_per_row = 3 * num_columns
dtype = np.uint32 dtype = np.uint32
read_buffer = fptr.read(num_entries * np.sum(bps) / 8) read_buffer = fptr.read(num_entries * nbytes_per_row)
palette = np.frombuffer(read_buffer, dtype) palette = np.frombuffer(read_buffer, dtype=dtype)
palette.reshape((num_entries, num_columns)) palette = np.reshape(palette, (num_entries, num_columns))
else: else:
# General case where the columns may not be the same width. # General case where the columns may not be the same width.