planning for 4.5

[Dirk Hohndel]

On Mon, Apr 20, 2015 at 09:12:08PM -0400, Miika Turkia wrote:
> On Mon, Apr 20, 2015 at 6:04 PM, Dirk Hohndel <dirk at hohndel.org> wrote:
> > On Mon, Apr 20, 2015 at 05:45:15PM -0400, Miika Turkia wrote:
> >> On Mon, Apr 20, 2015 at 1:23 PM, Dirk Hohndel <dirk at hohndel.org> wrote:
> >> > can you please look at the current state of things
> >> > and let us know what needs to be fixed before we can cut a release?
> >>
> >> The CSV imports are mostly broken due to the changes for file format
> >> version 3. Most of the imports get silently dropped. (If the imported
> >> would state the XML to be v3, then it would work, but the location
> >> conversion would not be run.)
> >>
> >> I am not sure how this should be fixed, and frankly, at the moment
> >> don't have time to look at it. (IIRC the consensus was that the
> >> importers should state version 2 so that the location conversion would
> >> be run with the "tested" C implementation and not to try do it in
> >> XSLT.)
> >
> > I remember this. So what needs to be done for this to be functional?
> > Is this just the fact that it goes back to ask the user (and thereby
> > aborts the loading)? That's easy to work around...
> 
> Something like that. If I use the current daily build, no dive is
> imported when starting Subsurface as follows:
> 
> subsurface --import dives/TestDiveDivingLog5.08.uddf
> 
> Same import obviously works with Subsurface version without the change
> in the locations. If I open the UDDF file without the import
> parameter, then it is opened properly. Anyway, the same issue is
> present with the CSV based imports, but this sample file was the first
> I was certain works with old version and not with the new, thus it
> should be an easy test case to make sure this works...

I believe I have fixed this in the latest master. Please give it a try.

/D