On pon, 2016-01-18 at 22:31 +0100, viq wrote: > On Sun, 2016-01-17 at 13:50 +0100, Adam Wolk wrote: > > On Sat, 09 Jan 2016 23:36:25 +0100 > > viq <vic...@gmail.com> wrote: > > > > > $ cat duply/pkg/DESCR > > > Duply is a shell front end for the duplicity backup tool. > > > It greatly simplifies its usage by implementing backup > > > job profiles, batch commands and more. Who says secure > > > backups on non-trusted spaces are no child's play. > > > > > > WWW: http://duply.net > > > > > > I've been using it on other systems, now time to have it on > > > OpenBSD > > > as > > > well ;) Works fine for me, though I really tested only with > > > duplicity > > > 0.7.x > > > > > > > I'm not a duplicity user (yet?) but decided to at least test your > > port > > since you already pinged 2 times and seems no one noticed yet :) > > Thank you :) > > > > Should I patch > > > #!/usr/bin/env bash > > > to use absolute path, or is it OK to leave as is? > > > > I think /usr/bin/env bash is completely fine. That's a proper > > portable > > way to find an interpreter. > > > > Why is the port using the lang/python module? I don't see any build > > systems or .egg files being produced here. Just being dependant on > > sysutils/duplicity should get you the interpreter used by > > duplicity. > > The port itself is a bash program calling a python program - not a > > python program. > > > > In my opinion you can drop both: > > - MODULES = lang/python > > - SUBST_VARS += MODPY_VERSION > > Good catch. An earlier version was doing some python version > detection, > so I was putting python${MODPY_VERSION} in there, but the way it's > currently there is no need for this. Removed.
Bump. I'm using this on a bunch of my machines to back up to hubic. > > I tried the master site you mentioned in the Makefile, indeed it > > doesn't > > work with the spaces. > > > > Tested the port on amd64 snapshot from Jan 15th. It builds and > > installs > > cleanly. > > Thank you. -- viq <vic...@gmail.com>
