Josh Andler <scislac@...> writes:
> Hey All!
> Are there any larger features people are planning on still for this
> cycle that aren't well known? If you want to get something in that is
> unknown, please speak up soon because we want to help you test it. :)
> I know of the following items which we're looking to get in:
> *All of the current GSoC projects. :) *fingers crossed*
> *What will also probably be more work towards GTK3 compatibility.
> On the refactoring front I can report good news that Abhishek Sharma
> has gotten all proper libcroco stuff submitted upstream. If lucky, we
> may be just a few distribution releases away from being able to take
> libcroco out of trunk!
> Additionally, Alex Valavanis has brought our in-trunk copy of libgdl
> up-to-par with the 2.X (whatever is "current") branch of libgdl. I'm
> unsure if patch submissions have been made in that direction for the
> couple of custom features that we have in our copy. Either way,
> hopefully we'll be able to get our copy out of trunk around the same
> Campbell Barton has been plugging away on cmake related stuff. I'm not
> sure of the status of it, especially cross-platform-wise.
> Obviously, Krzysztof Konsinski's GSoC work from last year has been
> merged and all I can say is... please stress test this the best you
> can. Pretend you don't know how to use Inkscape and break it like a
> newb. ;)
> Are there any big refactoring goals we're still missing out on that
> people think we should shoot for? (please be reasonable)
> I'm planning on looking to look into a couple Windows things, but will
> save that for a different email.
> All of that aside, here is a VERY loose and preliminary proposed release plan:
> Mid-August for Chill
> Beginning of September for Frost (Bug Hunt begins)
> End of September for Feature Freeze
> End of October for Hard Freeze
> Sometime in November for Branch & Release
> I'd like feedback on the bug hunt goal. I think we've actually gotten
> a lot accomplished for a release just with the cairo renderer alone.
> Newly created bugs aside (as in, if you introduce bugs during this
> cycle, they don't count toward the total), 250 points seems like a
> reasonable goal. Do note, we will be releasing an uber-stable 0.48.3
> more likely than not, so if we release "0.49" to a wider audience to
> get more testing, it's not so bad.
> For those new to the process, our bug hunt will begin at Feature
> Freeze. Traditionally, we specify a point target for us to reach, and
> award us points based on the severity of the bug: 3, 6, 9, or 12
> points for low, medium, high, and critical.
> Lastly... version numbering... we have way too many opinions about
> this. I do want to reiterate, Inkscape deserves a more "grown-up"
> version numbering scheme for users. I very much think Mental's comment
> about we should have a public version number and then stick with our
> current numbering scheme for development purposes. It's akin to
> code-names internally and solid numbers publicly. So, are date formats
> appealing? YY.MM or YYYY, some arbitrary number we just agree on
> starting with, bump the decimal place from with the current numbering,
> or what? If we get too many opinions again, I'll propose handing it
> over to the board...
> Thoughts or comments?
I have a dbus branch with several bugfixes and some additional seemingly
The code is battle-proven by my Inkmacs project:
I also have a fairly small patch to enable embedding of Inkscape in
other applications using the Xembed protocoll. I don't think it's clean
enough to go into trunk yet except maybe as a configurable
experimental option, but I could polish it if theres interest.
It works by providing a --socket-id command line option. When Inkscape
starts it then creates a gtkplug toplevel instead of the ordinary
toplevel, and embeds itself there. This is similar to how many other
applications do it.
What's missing is a dbus function to create more xembedded
windows. Currently new windows after the first embedded one become
ordinary toplevels again.
> Got Input? Slashdot Needs You.
> Take our quick survey online. Come on, we don't ask for help often.
> Plus, you'll get a chance to win $100 to spend on ThinkGeek.
> Inkscape-devel mailing list