lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

Probably an error ...
Sorry for that.

Read More...

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

No problem Hy, I think you're right. Also, the header of my file is probably wrong too. But, I'm happy because it helps to fix wrong code.
Last remark, maybe it would be better to apply same fix for yoffset? In order to avoid junk row?

Many thanks,
Cheers,

Read More...

lock thanked Hy Murveit in topic Bug in fitsdata.cpp 3 months ago

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

@hy:

OK, my tests shows (am I wrong?) that:

<code>
case DC1394_COLOR_FILTER_GRBG:
debayerParams.filter = DC1394_COLOR_FILTER_GBRG
</code>

Would be the fine transformation.

Read More...

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

I think you switched bayer pattern, I would write:

case DC1394_COLOR_FILTER_RGGB:
            debayerParams.filter = DC1394_COLOR_FILTER_GBRG;
            break;
        case DC1394_COLOR_FILTER_GBRG:
            debayerParams.filter = DC1394_COLOR_FILTER_RGGB;
            break;
        case DC1394_COLOR_FILTER_GRBG:
            debayerParams.filter = DC1394_COLOR_FILTER_BGGR;
            break;
        case DC1394_COLOR_FILTER_BGGR:
            debayerParams.filter = DC1394_COLOR_FILTER_GRBG;
            break;
        }


Read More...

lock thanked Hy Murveit in topic Bug in fitsdata.cpp 3 months ago

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

Does 0 and 1 the only values that can take these fields?

Read More...

lock thanked Hy Murveit in topic Bug in fitsdata.cpp 3 months ago

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

Some crash in the fitsviewer when opening a FITS image containing a field XBAYROFF or YBAYROFF different from 0.

Read More...

lock replied to the topic 'Bug in fitsdata.cpp' in the forum. 3 months ago

Any idea about this?

Read More...

lock created a new topic ' Bug in fitsdata.cpp' in the forum. 3 months ago

When reading a FITS file with X or Y offsets, crash could occur.
With valgrind I have the following error:

==627021== Thread 17 Thread (pooled):
==627021== Invalid read of size 2
==627021==    at 0xBE7A0C: dc1394_bayer_NearestNeighbor_uint16 (bayer.c:978)
==627021==    by 0xBF2176: dc1394_bayer_decoding_16bit (bayer.c:2544)
==627021==    by 0x401A47: FITSData::debayer_16bit() (fitsdata.cpp:3461)
==627021==    by 0x4013E5: FITSData::debayer() (fitsdata.cpp:3347)
==627021==    by 0x3F9676: FITSData::privateLoad(void*, unsigned long, bool) (fitsdata.cpp:347)
==627021==    by 0x45E7B6: QtConcurrent::StoredMemberFunctionPointerCall3<bool, FITSData, void*, decltype(nullptr), unsigned long, int, bool, bool>::runFunctor() (qtconcurrentstoredfunctioncall.h:1275)
==627021==    by 0x222C7A: QtConcurrent::RunFunctionTask<bool>::run() (qtconcurrentrunbase.h:108)
==627021==    by 0x6BB3D31: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.12.5)
==627021==    by 0x6BB08B1: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.12.5)
==627021==    by 0x75C6FB6: start_thread (pthread_create.c:486)
==627021==    by 0x7C621AE: clone (clone.S:95)
==627021==  Address 0x37689840 is 0 bytes after a block of size 32,778,240 alloc'd
==627021==    at 0x483750F: operator new[](unsigned long) (vg_replace_malloc.c:433)
==627021==    by 0x3F942D: FITSData::privateLoad(void*, unsigned long, bool) (fitsdata.cpp:325)
==627021==    by 0x45E7B6: QtConcurrent::StoredMemberFunctionPointerCall3<bool, FITSData, void*, decltype(nullptr), unsigned long, int, bool, bool>::runFunctor() (qtconcurrentstoredfunctioncall.h:1275)
==627021==    by 0x222C7A: QtConcurrent::RunFunctionTask<bool>::run() (qtconcurrentrunbase.h:108)
==627021==    by 0x6BB3D31: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.12.5)
==627021==    by 0x6BB08B1: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.12.5)
==627021==    by 0x75C6FB6: start_thread (pthread_create.c:486)
==627021==    by 0x7C621AE: clone (clone.S:95)
==627021== 


Read More...

lock created a new topic ' Add of a new field in FITS keyword' in the forum. 9 months ago

Hello, could it be possible to add a field in the FITS keyword. This field could be named ORIENT (for orientation) and be filled with:
UP-BOTTOM
or
BOTTOM-UP

Indeed, some software do not use the same read orientation, but the orientation is important for the debayer step.
Using this keyword could fix this issue if processing software start to read that.

I will add this in Siril if you're ok with that.

Read More...

lock replied to the topic 'Mirrored Fits : Fits header missing order information ?' in the forum. 2 years ago

Could it be possible to add a keyword, even non-standard, to write the type of data ?
bottom-top, ... or others ?
I believe that some softwares use this kind of keywords.
In this case I would add something in Siril.

Read More...

lock replied to the topic 'Cannot connect Canon DSLR' in the forum. 3 years ago

OK. Some news.

LANG=C gphoto2 --camera « Canon EOS 1100D » --port usb: --capture-image-and-download
will give an error
Saving file as capt0000.cr2
Deleting file /capt0000.cr2 on the camera
*** Error (-1: 'Unspecified error') ***

but capt0000.cr2 is well saved in /home..... So gphoto seems working.

Read More...

lock replied to the topic 'Cannot connect Canon DSLR' in the forum. 3 years ago

I tried gphoto with a sdcard ...
"funny" things, I have another error message without the mirror noise this time:

*** Error ***              
An error occurred in the io-library ('Could not claim the USB device'): Could not claim interface 0 (Device or resource busy). Make sure no other program (gvfs-gphoto2-volume-monitor) or kernel module (such as sdc2xx, stv680, spca50x) is using the device and you have read/write access to the device.
ERROR: Could not capture image.
ERROR: Could not capture.
*** Error (-53: 'Could not claim the USB device') ***


Read More...