Re: Rclone?

[Matt M]

I have no problem signing those, Michael.  Thanks!

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Saturday, November 17, 2018 10:41 AM, Michael Albinus <address@hidden> wrote:

> Matt M address@hidden writes:
>
> > Hi, Michael,
>
> Hi Matt,
>
> > The mount seems flaky to me, at least when copying to it using
> > Finder on Macos (copying from it is fine). So, I tend to use the
> > direct commands for move, copy, copyto, delete, etc., from eshell.
>
> No problem, we can change it any time. For the beginning I believe using
> a mount would give the first results soon. If it isn't sufficient, or
> performance is too slow (looks to me like this), we could change later on.
>
> > I was thinking of writing the integration myself, but I'm relatively
> > new to elisp and don't know the tramp integration points at all. So,
> > I was hoping this was already in the works! ;)
>
> No, as said. However, I'm very glad for your interest to
> contribute. It's always great if people chime in into Tramp development.
>
> > If, however, you can point me to a document which might explain the
> > minimum functions I need to implement for a new integration, that
> > would help a lot!
>
> I recommend to study tramp-adb.el. It has a similar scope, accessing
> Android devices via the "adb" program. Here you will see waht's needed
> to get a Tramp integration. And you might read the Elisp manual about
> file name handlers. Call (info "(elisp) Magic File Names")
> This gives you also a list of file name primitives which need an own
> implementation.
>
> And of course, you can ask all kind of questions. I will try to be
> responsive, and hopefully other people will also answer.
>
> > But, if you are interested in this yourself, I should probably not
> > bother, as I would be too slow and the code poorly written!
>
> As usual, there's all too much on my todo list. I've started a new file
> tramp-rclone.el as proof of concept, and some functionality is already
> working. Likely, I will stop after this weekend, and push everything to
> the Tramp repository. You could continue with this, or start your own
> integration if you're not satified with that.
>
> If you intend to make it part of Tramp/Emacs, there is also a formal
> prerequisite. All contributors to Emacs shall sign legal papers, that
> the work is licensed to FSF. Are you willing to sign such legal papers?
>
> Best regards, Michael.