Import from url


Hi all,
From time to time, we need to import huge amont of files into Seafile that are already available from a url (http(s), ftp,…).
Is there any sense to have an import > from url option in the web interface that could do the job without having to download the file on a client before uploading it ? Would it work with folder structure (FTP or Webdav) ?


I like this idea!


+1 Great idea


How about using seafile client and sync the library from the server to a folder. Then simply use wget or rsync or something to copy the files form your “url” to the sync folder and let the seafile client take care of the upload/sync to the server?
You could also use cron or some other task scheduler on the client to run the wget periodically.


Hi @andi-blafasl,
Honestly, i don’t think it is as simplly as you think from an end user point of view.
A button in the Web UI may be simpler to use. I don’t presume it is simple to implement…


Installing seafile client and sync a library with a folder (let’s asume path c:\html) => quiet easy, even for non admin users :wink:
Download wget from here using the setup will get you wget for windows, with all it’s dependencies. => should be very easy for an administrator
Set up a scheduled task with the following commandline wget -r (which of course have to be a site with links to your files you want to download) => should be very easy for an administrator that is capable of installing a seafile server?

The scheduled task make it even easier for future updates! No need to log in to seafile web interface and click a shiny button manualy :confounded:

PS: if you are familiar enough with linux to set up a seafile server, you could also set up the seafile client and wget on a very small linux vm that is running headless and doing the job.


BTW, if you are the server admin of the webserver where your files are comming from, you could also install a seafile client on this system and let it sync the “webfolder” automatically to your library inside seafile. This way around you do not need another system that is doing the sync :wink:


HI @andi-blafasl,
Thank you for the detailed explanation, but when i think of users in this case, i think about end users who have no specific skill at all and just use the web interface to, for instance, import into Seafile a file that is located in Google drive ore anything else.
But as the server admin, i’ll keep in mind all your tips ! :slight_smile:


Hi all,

Any opinion from seafile staff about this idea ?




This is something I’m missing that would be very nice to have as an upload option. Some way for the server to fetch some resource from a link e.g. Dropbox or Google drive.


And what about script running on Server2 which upload files over WebDav?


Hi, i was thinking about an end user feature, not a script.


FYI:This RFE is is on the github project: