This camera produces a RAW file that has a REGB layout which means there are 4 colors. This results in a non-square rgb_cam. However, the display code in raw-identify (v 0.19.6) appears to be incorrect. The code is as below:
for (int i = 0; i < P1.colors; i++)
printf("%6.4f\t%6.4f\t%6.4f\n", C.rgb_cam[i][0], C.rgb_cam[i][1], C.rgb_cam[i][2]);
I'm getting pink clouds when trying to convert a RAW to a TIF/JPEG with LibRaw 0.20.0. I've run into this a while ago in fact, when using the RawPy wrapper library for Python.
Hello everyone.
First thank you for your big work on this lib I used for my software.
Since the last version I have issues with some camera (especially canon). This is probably due to a big change you've done.
As stated in Changelog:
* Bayer images: ensure that even margins have the same COLOR() for both the active sensor area and the full sensor area.
I've been sent some CR3 files that were reported as causing my application to freeze.
The problem appears to be that open_file() never returns and just sits there eating CPU.
These same files also cause Windows Explorer to freeze for a number of seconds and then Explorer closes and restarts (maybe they're using LibRaw internally to the raw file codec).
I've placed a zip file containing the bad files on Dropbox:
I am trying to convert CR3 images to 16 bit fits files used RAW2FITS, which uses libraw. I have the newest libraw installed, which supports cr3.
When converting to 8 bit images, everything looks fine. However, when setting the output_bps to 16 everything is garbled, at the libraw level. First of all, pixel values are all still 8 bit (max values of 255) and their coordinates are wrong also.
They fail when calling unpack() with the error "Corrupted data or unexpected EOF". Call to open_file works. Hence raw-identify is able to parse metadata from the file.
I found some samples of Fuji X-T4 raws that LibRaw seems to have problems handling.
There seems to be a few things going wrong with this one. The version that parse_fuji_compressed_header() finds is version 0 which is not supported as the check is for version != 1. This image does appear to be compressed though (half the file size of a normal X-T4). Not sure if LibRaw will support a version 0 compression algorithm but just something I noticed, also all the other conditions are satisfied as true for the compression check.
I am using mostly libraw to convert IIQ to tiff and I did notice that some EXIF information such as the GPS coordinates and some various metadata related to the camera itself were missing after the conversion. It there any way to avoid this?
Recent comments