I am happy to announce that the newer, faster and better version of rclone web-gui is now released in beta, it's still in its early stages, so I would really appreciate if you could test it out and give feedback so together, we can make it even better.
Highlights:
New and redesigned UI.
Support to control mounts from the GUI (requires atleast rclone v1.52.1.0-DEV ).
Sorting of files and folders in Explorer with name/ size/ date modified etc.
for the open new remote, there should be a per-populated drop down box that lists all the remotes,
or an option to press the down arrow and then display a list of all remotes.
Yes, please feel free to post here,
if you have some specific bug, I would really appreciate it if you create a new issue here: https://github.com/rclone/rclone-webui-react/issues, as it becomes easier to track.
what is needed is a dual pane interface. like midnight commander.
press F5 to copy between panes.
Press F6 to move between panes.
also, using a web browser as the user interface means i cannot run your software locally.
i have a lot of window server 2019 hyper-v edition server , which is free.
has no gui, no windows explorer, no desktop, no ability to run a web browser.
however, the rclonebrowser does run on it well.
sure, i can tweak a firewall, connect via the ssh server but it is nice to be able to run software locally.
I am planning to release the webui to npm. So you should be able to use node to run the web gui without the rclone.
One thing you can do right now to connect to it, is to download the latest build for webui and then run it using a local server. I often use the python server for testing, by navigating to the extracted webui files, and running,
python -m "http.server"
This creates a local server which serves the webui.
One thing you need to allow while running the server on Windows server is to run with the --rc-allow-origin=local_address If you use the command above, you will use --rc-allow-origin=http://localhost:8000.
An even better alternative, if you don't mind using a remotely hosted web gui (on GitHub Pages).
You need to specify the user and password because, you will have to enter it while logging in.
If you are not comfortable with specifying username and password here, you can use a htpasswd file.
This is definitely needed. My feedback is that the function keys aren't as required. To me we can move things between panes by dragging them but that isn't intuitive. Additionally, on mobile browsers you can't drag and drop with touch. It would be nice to have simple arrows between the two pane that you can copy or move.
Also the ability to multi-select (check boxes?) to select those items before using the arrows buttons (and delete buttons) would be a big navigational improvement.
I'd like to collaborate on building a scheduler which would work with the GUI. Is that interesting to you? Is it something that would work with your GSoC 2020 plugin system? If this is interesting, I'll up in an idea with details.
Hi @Stephen_Klosky. That's an interesting idea. We should discuss more over it. We can either try to implement it as a plugin, or we can integrate it into the main rclone-webui project itself. Looking forward to collaborating.
I checked out the template. It will take me a bit of time to learn the framework (react, rclone api, web ui config). The concept looks great. I have a couple questions.
How does an end user select/enable a plugin?
How is the plugin deployed at runtime?
The end user can enable the plugin using the web gui. The gui makes a call to the rclone backend, and then the backend downloads the plugin via the github repository specified in the package.json of the plugin. The compilation is done using a Github action.
Here are some examples:
The current development is going on in the plugins branch of rclone/rclone and rclone/rclone-webui-react.
the plugin infrastructure basically uses iframe to load the things. So for the plugin development, you can use any js frontend framework.