livie
/
phoenix-firestorm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

EXP-3049 : Implement partial loading using the calcDataSize(discard_level) computation found in the viewer.

meow-7.2.2
Merov Linden 2012-03-15 21:59:37 -07:00
parent fe7a4d0183
commit 4f19eb3951
1 changed files with 37 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
indra/integration_tests/llimage_libtest/llimage_libtest.cpp
View File

@ -55,10 +55,10 @@ static const char USAGE[] = "\n"
" List of image files to create (assumes same order as for input files)\n"
" OR 3 letters file type extension to convert each input file into.\n"
" -load, --load_size <n>\n"
" Portion of the input image to load, in bytes."
" Will load the whole image if 0 or if the size requested is more than the file size."
" This parameter will be applied to any input image type but really makes sense only"
" for j2c images."
" Portion of the input file to load, in bytes."
" If (load == 0), it will load the whole file."
" If (load == -1), it will load the size relevant to reach the requested discard level (see -d)."
" Only valid for j2c images. Default is 0 (load whole file).\n"
" -r, --region <x0, y0, x1, y1>\n"
" Crop region applied to the input files in pixels.\n"
" Only used for j2c images. Default is no region cropping.\n"
@ -121,10 +121,40 @@ LLPointer<LLImageRaw> load_image(const std::string &src_filename, int discard_le
{
LLPointer<LLImageFormatted> image = create_image(src_filename);
// This just loads the image file stream into a buffer. No decoding done.
if (!image->load(src_filename, load_size))
// We support partial loading only for j2c images
if (image->getCodec() == IMG_CODEC_J2C)
{
return NULL;
// Load the header
if (!image->load(src_filename, 600))
{
return NULL;
}
S32 h = ((LLImageJ2C*)(image.get()))->calcHeaderSize();
S32 d = (load_size > 0 ? ((LLImageJ2C*)(image.get()))->calcDiscardLevelBytes(load_size) : 0);
S8 r = ((LLImageJ2C*)(image.get()))->getRawDiscardLevel();
std::cout << "Merov debug : header = " << h << ", load_size = " << load_size << ", discard level = " << d << ", raw discard level = " << r << std::endl;
for (d = 0; d < MAX_DISCARD_LEVEL; d++)
{
S32 data_size = ((LLImageJ2C*)(image.get()))->calcDataSize(d);
std::cout << "Merov debug : discard_level = " << d << ", data_size = " << data_size << std::endl;
}
if (load_size < 0)
{
load_size = (discard_level != -1 ? ((LLImageJ2C*)(image.get()))->calcDataSize(discard_level) : 0);
}
// Load the requested byte range
if (!image->load(src_filename, load_size))
{
return NULL;
}
}
else
{
// This just loads the image file stream into a buffer. No decoding done.
if (!image->load(src_filename))
{
return NULL;
}
}
if( (image->getComponents() != 3) && (image->getComponents() != 4) )