OIIO #143
OIIO #143
Conversation
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
…ibrary Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
|
Sorry for non-technical comment.
|
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
| @@ -566,7 +529,7 @@ bool ImageConverter::configure( | |||
| OIIO::TypeDesc( OIIO::TypeDesc::FLOAT, 4 ), | |||
| user_mul ); | |||
|
|
|||
| if ( _read_raw ) | |||
| // if ( _read_raw ) | |||
There was a problem hiding this comment.
U sure not left accidentally?
(sorry, I don't know what's happening here at all, but this change of logic flow from outside looks like potential unintentional change)
Or if itentional, then deleting it is a better option?
There was a problem hiding this comment.
It is intentional, yes. A temporary fix until I can find a better solution. The problem is we are a bit limited in what we can do when dealing with libraw via OIIO. We don't have the same level of control as in the older version where we could access libraw directly.
|
|
||
| # setup our library dependencies (i.e. ilmbase) | ||
| include( "${RAWTOACES_CMAKE_DIR}/RAWTOACESLibraryDepends.cmake" ) | ||
| # include( "${RAWTOACES_CMAKE_DIR}/RAWTOACESLibraryDepends.cmake" ) |
There was a problem hiding this comment.
another commented out line not to forget about (unless, it's a special syntax in c++, which I don't know. sry if so)
There was a problem hiding this comment.
There is no such file in the repo, whoever wrote it has forgotten to push the file, I assume.
| // here. This creates a bit of overhead, we may want to decouple | ||
| // the selection of illuminant and WB calculation in the core | ||
| // library. But this is still faster than running the matrix solver | ||
| // everytime. |
There was a problem hiding this comment.
Reminder to deal with this TODO before merging. Don't know how you guys do it. At my place we either deal with them right there or create and issue and link to it. If it's ok to have TODO left for future generation - ok then.
There was a problem hiding this comment.
The PR is still marked as draft :)
There was a problem hiding this comment.
Thanks for taking a look by the way. I will be pushing more changes soon, there are still things missing/broken.
There was a problem hiding this comment.
@antond-weta sorry. I will back off =) I would be annoyed. Let me know if you need beta tester (I am a mac user) for changes.
There was a problem hiding this comment.
No worries! Just saying that this PR is not going to be merged in any time soon, there are still many things to do.
The code in this branch is mostly functional. There are some bits still missing, like white balancing by averaging all pixels or a box. In any other mode you should get the result identical to the master.
Feel free to use/test, any feedback/suggestions would be appreciated.
antond-weta
force-pushed
the
OIIO
branch
5 times, most recently
from
b36f004 to
a87817a
Compare
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
Signed-off-by: Anton Dukhovnikov <[email protected]>
This is a proposed replacement of the 'rawtoaces_util' library and 'rawtoaces' command line tool.
Main changes:
The makefile contains 2 new targets rawtoaces_util2 and rawtoaces2. The old targets are still in place.