I am looking for some help with orchestrating downloads from Dropbox to a local filesystem. I have essentially built a file-transfer application which will allow our users to upload large files from Dropbox to our system.
Up to this point I have been using the javascript Chooser successfully-- the users login to Dropbox, select their files, and the temporary "direct" links are sent server-side, where we download with something like wget (wget -q -O - <temporary link>) . However, this method has the limitation of the 4-hour expiration. In the case of especially large transfers, I would like to avoid this limitation as I do not want to limit the size of the files that my users can transfer.
Now, I know I can have the user approve my application, perform oauth2 flow to get the code/token, and then use the SDK to query a user's Dropbox server-side. Presumably, that would give my application unlimited time to perform any downloads. We have been using the SDK to upload files from our system to Dropbox, and that's been great. HOWEVER, I like the javascript chooser and would rather not write my own "file browser" UI.
If I have the oauth2 token AND the temporary link, is there a way I can "convert" that to a path which I can use which does not have the time constraint? I'd love to be able to continue using the Chooser, and reverse-engineer the path in the user's Dropbox (to which I will have full access). Is this remotely possible?
