Hello,
I wonder how i could detect RGB image format, without loading image in a
buffer, and then parsing its header.
For the moment, i'm doing this in my classe, which is far from being
efficient:
Fl_Image * Images::load (std::string filename)
{
Fl_Image *image = is_image_already_loaded(filename);
if (!image) {
image = new Fl_JPEG_Image (filename.c_str());
if (image->w()) {
//the image is cached when loading once.
liste.insert(std::pair<std::string, Fl_Image *>(filename,image) );
} else {
delete image;
image = 0;
}
}
if (!image) {
image = new Fl_PNG_Image (filename.c_str());
if (image->w()) {
liste.insert(std::pair<std::string, Fl_Image *>(filename,image) );
} else {
delete image;
image = 0;
}
}
if (!image) {
image = new Fl_BMP_Image (filename.c_str());
if (image->w()) {
liste.insert(std::pair<std::string, Fl_Image *>(filename,image) );
} else {
CRITICAL ("unknwown format for : %s\n",filename.c_str());
delete image;
image = 0;
}
}
return image;
}
Of course, i could preload the image in a buffer, and then trying
Fl_JPEG/PNG/BMP_Image from this buffer.
I could also use a template for the three almost identical load.
As i cache the picture once loaded, i think i don't get a big speed
penalty, even if i load BMP pictures.
Thank you.
(«premature optimizing is the root of all evil», i know too :) )
_______________________________________________
fltk mailing list
[email protected]
http://lists.easysw.com/mailman/listinfo/fltk
