Conversation
It is present in all lossless samples available on RPU (ILCE-1, ILCE-7M4, ILCE-7RM5)
M/S files are not CFAs but demosaiced YCbCr (PhotometricInterpretation=6)
kmilos
force-pushed
the
kmilos/sony_ljpeg_tweaks
branch
from
7330b4e to
dce5b00
Compare
| const TiffEntry* size_entry = raw->hasEntry(TiffTag::SONYRAWIMAGESIZE) | ||
| ? raw->getEntry(TiffTag::SONYRAWIMAGESIZE) | ||
| : raw->getEntry(TiffTag::DEFAULTCROPSIZE); | ||
| const TiffEntry* size_entry = raw->getEntry(TiffTag::SONYRAWIMAGESIZE); |
There was a problem hiding this comment.
so no fallback is necessary, we're always sure to have this EXIF marker available?
There was a problem hiding this comment.
we're always sure to have this EXIF marker available?
Seems to be the case so far, as mentioned.
In the worst case, I'd rather return the uncropped image w/ black borders than the DNG crop, I feel it's more consistent with the stated behaviour:
RawSpeed does NOT…
crop the image to the same sizes as manufactures, but supplies biggest possible images.
There was a problem hiding this comment.
FWIW, getEntry() throws if you query for an entry that will not be found.
|
Makes sense, thank you! |
|
Do we feel cheeky enough to drop this on dt? ;) I suspect they'll freeze and tag 4.4 very soon... @TurboGit |
|
Even though |
|
@LebedevRI : It is planed for June 14th, I can move it one day on 15th if necessary. |
|
Ah, that's plenty of time then. |
4 participants
Follow-up to #484