-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Limited metadata info #477
Comments
Hey @mamadoumakalou, good to hear that you like vimiv and great to see other people caring about metadata. This shows us that, even though rather troublesome, trying to improve the metadata system is really worth it. I would assume that by using pyexiv2 you should be able to extract all information exiv2 can extract. But having encountered already lots of troubles with pyexiv2, I would not be surprised if that is not the case. Anyways, have you tried Also, according to the exiv2 wiki, there should be a key I hope that helps. If not, let me know |
Thanks for the quick reply, @jcjgraf. Good questions, and I appreciate your awareness that pyexiv2/exiv2 is a work in progress. Yes, I've thoroughly searched and tested :metadata-list-keys to no avail re color temperature for my camera make (Fuji). It's helpful to know that it seems to work for Canon. I hadn't found the exiv2 wiki, so that is helpful as well. On the wiki, I see that exiv2 doesn't use/find/read the tag for color temp. However, I know that Fuji does provide this metadata. I can easily see it using:
The output for a particular image is:
The -H flag for exiftool includes the hex code for that tag, and I see in the exiv2 wiki for Fuji that this tag (0x1005) is not used. This is the root of my particular problem. The question, to my mind, is why doen't exiv2 use this tag given that Fuji makes it available. Perhaps the next step is for me to open this issue with exiv2, which I may do. But I would also like to open the possibility here that vimiv might become more robust if users could, as I mentioned in the original issue, output any info they need to a widget that overlays the image, much like the metadata widget does. I have seen this work well on other programs, namely, feh. Perhaps there's a reason to avoid this? If so, that's a development choice that is completely understandable. But, it would be a solution or workaround for those experience the limitations of exiv2, or for that matter other tools. Again, much thanks for your time and consideration. |
Upstream exiv2 is very kind and helpful, they can certainly help you out. Vimiv's metadata system is far from perfect. So, any feedback is very welcome. I have also considered moving the metadata system out to plugin-space, so that everyone can extract image metadata in whatever way that works for them (#467). You can run any shell command from within vimiv (e.g. |
I like the idea of moving metadata to plugin space. I hadn't seen #467, thank you. I have had issues with running shell commands in vimiv. For example, if I run the same code you mention above, I share all this in case it is useful. Now, regarding #478, thank you for working on this option. I am new to git, and as such have only used binary packages thus far. I used yay to install the git package. Is that an okay way to get the latest updates? It seems to have installed vimiv-qt-git-r2268.4a1a0d18-1 correctly. I maybe be confused about where to "append"
I tried a lot of other things too, including quoting differently, but these seem representative. I would not be surprised if this is due to my lack of experience with install git packages, or even how to properly quote commands. I apologize if this is the case. Please let me know what else I can test, etc. Also, as I am new to github, please let me know if I should split this email and add the content related to #478 to that issue, instead of here. |
You are right, my proposed command is indeed not working. It appears that the pipe is not interpreted as a pipe but rather as a normal character (or as a filename in this case). I should have tested my commands before posting them here; sorry for that 😬 I need to have a look at the code and see what I can do
When using the git package from the AUR, you are running the vimiv version which is currently in the master branch. But you want to run the version which is in my PRs (Pull Request) branch. So what you want to do instead:
This pulls the latest version of vimiv from github, switches to the right branch, and runs vimiv from there. Then you should be able to run something like
If you have direct feedback for #478, you can directly comment in there. But since this issue is already about many other things that just metadata, we can also keep the discussion in here Anyways, I will have a look at he code in the next few days (hopefully) and see what I can do concerning pipes. |
Thanks, @jcjgraf, for walking me through switching from the master to your branch. Should have thought of that. I was able to test and see your option working! I used The ability to use pipes and scroll output would be nice, but there are ways to get around these issues at least in many cases. One key feature would be that the output does something similar to the current metadata output, where it takes up minimal screen space, and, (possibly, though this is more involved and less necessary) stays on screen and re-runs the command. |
Thanks for the feedback.
I can see its potential. But I think that is something which is better implemented as a plugin. |
Sorry if this is too obscure, but it's helpful for me to see color temp when reviewing photo metadata. This doesn't appear to be an option. I know the camera provides it as I can extract this info using the exiftool command, but pyexiv2 doesn't seem to provide access to this tag, or at least, that's not apparent using :metadata. Is there a way to add access to this tag? If not, how about a way to display output from a linux command on the photo, much in the same way the metadata info displays (as a widget?, not sure if this is the term being used). Many thanks, I've used many photo-viewing tools and this is by far the best.
The text was updated successfully, but these errors were encountered: