Re: Poll: Inline methods sufficient?

[Kai Großjohann]

lawrence mitchell <address@hidden> writes:

> Kai Grossjohann wrote:
>
>> Are the inline methods sufficient that Tramp offers now?  I mean the
>> ways to encode a file for transit, not all the alternatives to ssh
>> and telnet that people might think of.
>
>> Tramp tries base64 encoding via mimencode, mmencode, two Perl scripts
>> and GNU recode, and then it tries UU encoding via uuencode/uudecode.
>
> It seems like enough.  Is there a pointer in the manual for where
> one might find uuencode for Windows?  If not: <URL:
> http://www.concentric.net/~Wkiernan/UUencode/uuencode.html> seems
> to offer a version, though they are not free (as in speech
> software).  The GNU versions (of uucode) also come with cygwin
> (mingw doesn't seem to do the sharutils), but getting all of
> cygwin seems a bit excessive.

I wish there was another free implementation of uu*code for Windows.
Does anyone know?

> There's a pointer to metamail on the NTEmacs FAQ, which I would
> guess is probably enough.

Metamail needs to be on the remote host.  Emacs has base64
de/encoding built-in.

>> People running Windows with remote systems requiring UU encoding
>> might have problems because of a missing uuencode on Windows.
>
>> There are plans to support another encoding which uses "od -b" to
>> read files from the remote host and SysV-style "echo -e" to send
>> files to the remote host (including \000 for null bytes).  Is this
>> really necessary, or is what is offered now enough?
>
> Again for windows, the "echo -e" would have to be emulated
> somehow, as the echo builtin is pathetic.

The plan is to do it in Lisp inside Emacs: from Emacs, generate a
sequence of "echo -e" commands which are then executed on the remote
host.

> But I think that there are generally enough fallback inline methods.

Good :-)  Other opinions?

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)