Re: [Denemo-devel] Release 1.0.8

[Richard Shann]

On Fri, 2013-10-04 at 13:13 -0500, Jeremiah Benham wrote:
> Do you want me to make a 1.1 branch now?

Not yet: I would like to finish off the three things I mentioned (I have
already started on buttons to switch movements) first.


>  I am a bit confused. So I will make a 1.0.9 tarball so that the
> release will be labelled 1.0.9 for the mingw binary. 
There is no special need to create a 1.0.9 tarball, the source code is
available which is all GPL requires, I think we should just build a
mingw binary and put it in denemo.org/downloads labelled as 1.0.9 and
put a link to it from the Download page, explaining it is better than
1.0.8 which is not available for windows.

> 
> 
> Did the librubberband stuff work correctly with the mingw binary?

I didn't get a chance to test this yet. I noticed the problem with
palettes (due to it being a GTK2 build) and set to fixing that. I'll
check this next.

Richard


>  
> 
> 
> Jeremiah
> 
> 
> On Fri, Oct 4, 2013 at 11:53 AM, Richard Shann
> <address@hidden> wrote:
>         Success! The windows binary
>         http://www.denemo.org/~jjbenham/gub/uploads/denemo-0.0.0-0.mingw.exe
>         now executes correctly.
>         This is 1.0.9 that we have built, and it had a bug for GTK2
>         builds which
>         I have fixed.
>         How should we proceed now? We have already uploaded the 1.0.8
>         tarball,
>         so we shouldn't change that. But we can put a 1.0.9 windows
>         binary in
>         denemo.org/downloads and move on to releasing Denemo ver. 1.1
>         now.
>         The windows binary should be re-built (because of the GTK2 bug
>         just
>         fixed), but apart from that we can move on to preparing a 1.1
>         release.
>         
>         The main thing I think this needs is a sensible "General"
>         palette, the
>         one currently appearing immediately above the music. This
>         should have
>         all the buttons naive users expect. One thing that would be
>         nice, but I
>         haven't looked into it, would be to have icons instead of text
>         labels on
>         the buttons in the palettes.
>         
>         Could people with experience of other score-writing programs
>         suggest
>         buttons that are missing from the current "General" palette?
>         
>         Oh, and a third thing that would be good would be buttons
>         allow you to
>         switch to any movement in one click (and that would also let
>         you see how
>         many movements you have easily).
>         
>         Suggestions as to what else it would be desirable to add to
>         ver 1.1 are
>         welcome ...
>         
>         Richard
>         
>         
>         
>         
>         On Fri, 2013-10-04 at 09:10 -0500, Jeremiah Benham wrote:
>         > Ok. I have updated the binary.
>         >
>         > Jeremiah
>         >
>         > On Oct 4, 2013 3:12 AM, "Richard Shann"
>         <address@hidden>
>         > wrote:
>         >         At last I have tracked down the bug causing the
>         crash on
>         >         windows:
>         >
>         >         41ec393e949fe115d98d533b006d2a4f6d6405e6  utils
>         functions
>         >         author Éloi
>         >         Rivard <address@hidden> Thu, 20 Jun 2013 12:32:24
>         +0000
>         >         (14:32 +0200)
>         >
>         >         This commit was freeing the trailing arguments to
>         >         append_to_path(),
>         >         regardless of the caller's intentions. I think it is
>         better
>         >         for callers
>         >         to keep responsibility for their own data and not to
>         separate
>         >         allocation
>         >         from freeing if at all possible. In this case
>         (initialization)
>         >         freeing
>         >         is not needed anyway.
>         >
>         >         Jeremiah - one final build please! Unless I have
>         slipped in my
>         >         typing
>         >         the program should once again compile and work.
>         >
>         >         Richard
>         >
>         >
>         
>         
>         
> 
>