search

artpixls / ART

The ART raw image processor
Other
1 stars 0 forks source link

Problems with TIF files metadata #147

Open agriggio opened 4 years ago

agriggio commented 4 years ago

Original report by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Hi Alberto,

there seems to be some issues with the tif files meta-data. At least in my case, sequator and Afinity Photo complain about different things when I open tif files exported from ART (see the screenshots).

All the best,

Daniel

agriggio commented 4 years ago

Original comment by agriggio (Bitbucket: agriggio, GitHub: agriggio).

Hi Daniel,

can you provide an arp (and an input file if this happens only with some files)? [and is it really necessary to ask for this all the time? – sorry, this is not directed at you in particular, it’s just that it seems like a trend recently that people report bugs – for which I’m very grateful – without providing any info on how to reproduce. I don’t mind working for free, this is a hobby after all, but chasing bugs is not fun…]

‌

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Hi Alberto,

‌

sure, I’ll attach the files. I noticed this while following my ‘regular’ night photography workflow which involves:

‌

It happens with all the tif files exported from ART in this way. I did not tried to generate tif files from other raw formats yet, but will try also this,

‌

‌

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Test files attached.

agriggio commented 4 years ago

Original comment by agriggio (Bitbucket: agriggio, GitHub: agriggio).

Thanks for the files. For the Affinity warning, did you try using a V2 profile instead of V4? It might be that Affinity doesn’t understand V4 profiles?

Regarding sequator, I can reproduce the warning, but if I look with exiv2 and/or exiftool I don’t see any null byte in the Copyright tag…

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

You are right with sequator, I also cannot see anything wrong there…

‌

About Affinity, it gives the same warning with RT V2 profile, but it works fine with RT_Rec2020. So I guess the problem is not in ART but in the other software(s).

‌

I guess this ticket can be closed now …

agriggio commented 4 years ago

Original comment by agriggio (Bitbucket: agriggio, GitHub: agriggio).

Thanks for checking. It would still be good to understand what confuses Affinity though…

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Ok, how can I investigate this? Can I compare something between the file that raises the warning and the one that does not?

agriggio commented 4 years ago

Original comment by agriggio (Bitbucket: agriggio, GitHub: agriggio).

sorry, I think I was unclear… I didn’t mean that I’m expecting you to investigate, I was just thinking of leaving this open… however, you are more than welcome to help of course! :slight_smile:

one way would be to try using other srgb profiles and see if they also trigger the warning. If not, then we should try to understand what is the difference between them, and proceed from there…

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

No problem, I’ll gladly look into it. I’ll let you know as soon as I have done something.

agriggio commented 4 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

So, out of curiosity, I opened a RAF file, set the processing profile to neutral and exported it as jpg and as tif without changing anything else.

Affinity opens the jpg just fine with no complains, but for the tif it complains that the profile is missing. This can of course come from how Affinity treats the different file types and not from how ART attaches the profile to the file.

‌

agriggio commented 3 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Hi Alberto,

I tried to open the tif file also with gimp 2.99. I get a similar warning as from Sequator (se the attached file).

I guess I can get some logs or some detailed information from gimp if you think it is useful.

agriggio commented 3 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

agriggio commented 3 years ago

Original comment by agriggio (Bitbucket: agriggio, GitHub: agriggio).

Hi Daniel,

I’ve just pushed a tentative fix for the issue. Can you try again with the latest master?

agriggio commented 3 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

Hi Alberto,

I tried with the master this evening and did not noticed any change. Both Gimp and Sequator complain as before.

agriggio commented 3 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

I also tried to actually put something in the field and that works fine, the warning is no longer shown.

After this, I deleted the content and saved again with empty copyright. The warning message appeared again.

I am not sure this things worths to spend any more time on it. After all, the files are usable in both Sequator and Gimp even if the warnings appear.

I suppose there are more interesting things to do in ART than this :wink:

‌

agriggio commented 3 years ago

Original comment by Daniel Catalina (Bitbucket: [Daniel Catalina](https://bitbucket.org/Daniel Catalina), ).

I've exported a tiff also from darktable. Sequator and gimp complain, but this time is because the Copytight tag does not end with a null byte.