First, thanks for creating this version of DCRaw. I have been using the original DCRAW as part of a utility I have developed to make it easy to run the NIK Collection standalone. One of the added features is the ability to input RAW files. I needed a way to support Fuji X compressed files and this is probably the best and easiest way to do that. It also seems to be quicker than the original DCRaw.

Before I run
LibRaw::raw2image() or LibRaw::dcraw_process()

LibRaw::imgdata.sizes contain proper dimensions but after running one or the other the values are 0. Both functions return LIBRAW_SUCCESS (0)

I have no idea how to debug this. All images tested are cr2 and open just fine in other programs including "dcraw_emu.exe".

everything I did is below.

I have a .CR2 image taken by a Canon 80D camera. I'm using LibRaw to convert it to a .TIF. The .TIF which is created is darker than the original .CR2 file (when viewed using Canon's Digital Photo Professional or Irfan View)

I read the contents of the .CR2 file to a buffer, then pass the buffer to this method to convert and save to .TIF:

Hello,

If you compile LibRaw with RawSpeed and extract a tiff for a NX300 file you will see that the developed image is different as compared to the developed image with LibRaw without RawSpeed. By different I mean very different, blues become orange, yellows become blue etc. You can see this problem in the NX300 sample in www.rawsamples.ch.

By using the example programs I can get the image I want by using the following
dcraw_emu -w -T

I want to be able to to this in a program so that I can do it in memory and not have to write it to the disk. To do this I wanted to replicate the draw_emu command line first, then change it to everything happening in memory.

But the code I have doesn't seem to use the camera white balance or output to TIFF.

int ret;
LibRaw libRaw;
#define OUT libRaw.imgdata.params

putenv("TZ=UTC");