Forum: wxRuby wxruby is dead?

385940666f9ef223e65a5032a340295b?d=identicon&s=25 Carlo Bertini (waydotnet)
on 2010-11-29 09:09
I wanted to know whether the development of the project is still active
or not.
many bugs in the bug tracker are open from 25/03/2009 ....

thank you very much

Carlo Bertini
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2010-11-29 11:07
(Received via mailing list)
Hi Carlo

On 29/11/10 08:09, Carlo Bertini wrote:
> I wanted to know whether the development of the project is still active
> or not.
> many bugs in the bug tracker are open from 25/03/2009 ....


I wouldn't call a project with a stable, complete release across half a
dozen platforms 'dead' - but it's true that little or no new code has
been committed in the past 12 months or so since the release of 2.0.1. I
led the development of 2.0 but have been doing other non-programming
things, but am around on the mailing list along with other devs.
Hibernation maybe?

cheers
alex
04d1dcca8d15e7640920df71d50ee23c?d=identicon&s=25 Don Wilde (Guest)
on 2010-11-29 16:19
(Received via mailing list)
On Mon, Nov 29, 2010 at 1:41 AM, Alex Fenton <alex@pressure.to> wrote:

>
> I wouldn't call a project with a stable, complete release across half a
> dozen platforms 'dead' - but it's true that little or no new code has been
> committed in the past 12 months or so since the release of 2.0.1. I led the
> development of 2.0 but have been doing other non-programming things, but am
> around on the mailing list along with other devs. Hibernation maybe?
>
> Alex, if that is the case it would be very good if we (and I include myself
as a sacrificial offering) could get the docs updated to match the new
code
base. It is very frustrating to discover that half the 'overviews' are
gone
completely, and wxRuby is vastly different from wxC++ in the way things
like
Device Contexts and such are instantiated. I have been able to get my
code
to work well (after reading and re-RTFM-ing), but I'll bet it would be
even
better if I knew the little bits of tribal knowledge that you have
stored in
your collective brains.
00109e19a784b64f81b483a5dbec690a?d=identicon&s=25 Chauk-Mean Proum (chauk-mean)
on 2010-12-01 23:22
(Received via mailing list)
Hi all,

2010/11/29 Alex Fenton <alex@pressure.to>:
> dozen platforms 'dead' - but it's true that little or no new code has been
> committed in the past 12 months or so since the release of 2.0.1. I led the
> development of 2.0 but have been doing other non-programming things, but am
> around on the mailing list along with other devs. Hibernation maybe?
>
> cheers
> alex

Indeed, there have not been many bugfixes. We all have non-programming
activities :-)
Nevertheless, during the last months, I've committed fixes to the
stable branch to ensure that wxRuby builds properly with an up-to-date
environment :
- ruby-1.9.2 / rubygems-1.3.7
- SWIG-2.0
As Alex wrote, wxRuby is not dead but maybe in hibernation !

cheers,
Chauk-Mean.
00109e19a784b64f81b483a5dbec690a?d=identicon&s=25 Chauk-Mean Proum (chauk-mean)
on 2010-12-01 23:47
(Received via mailing list)
Hi Don,

2010/11/29 Don Wilde <dwilde1@gmail.com>:
> completely, and wxRuby is vastly different from wxC++ in the way things like
> Device Contexts and such are instantiated. I have been able to get my code
> to work well (after reading and re-RTFM-ing), but I'll bet it would be even
> better if I knew the little bits of tribal knowledge that you have stored in
> your collective brains.

I've seen that you faced some documentation issues.
Could you create some bug reports for tracking them ?
Also, documentation patches are always welcome
(http://wxruby.rubyforge.org/wiki/wiki.pl?DocumentingWxRuby).

cheers,
Chauk-Mean.
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2010-12-01 23:47
(Received via mailing list)
Hi Don

On 29/11/10 15:13, Don Wilde wrote:
> Alex, if that is the case it would be very good if we (and I include
> myself as a sacrificial offering) could get the docs updated to match
> the new code base. It is very frustrating to discover that half the
> 'overviews' are gone completely, and wxRuby is vastly different from
> wxC++ in the way things like Device Contexts and such are
> instantiated. I have been able to get my code to work well (after
> reading and re-RTFM-ing), but I'll bet it would be even better if I
> knew the little bits of tribal knowledge that you have stored in your
> collective brains.

Yes, I agree a big part of a library's usefulness is whether it's well
documented. I set up the current doc system for this reason, but it's
certainly not what it could be. The tricky bit is documenting a very
large library from scratch when about 90% of it (getter/setter
methods...) is shared with wx C++. What's needed is something that can
bring together 1) the existing, good documentation from wx C++ 2) the
info SWIG provides about the C++ to Ruby mapping and 3) manual
annotations, in a way that's manageable for a small development team.

I guess we've tended to prioritise adding samples to document the
implementation of wxRuby, since these serve as test cases, demo code and
partial docs at once. But additions to the current docs are also welcome
(in Textile format). I think there's also a need to document SWIG
knowledge - which is probably the really hard bit of 'tribal knowledge'
I have stored in my brain.

a
04d1dcca8d15e7640920df71d50ee23c?d=identicon&s=25 Don Wilde (Guest)
on 2010-12-02 05:01
(Received via mailing list)
Hi, Alex et al -

On Wed, Dec 1, 2010 at 2:33 PM, Alex Fenton <alex@pressure.to> wrote:

>> get my code to work well (after reading and re-RTFM-ing), but I'll bet it
> the C++ to Ruby mapping and 3) manual annotations, in a way that's
>
Anything that deals with multiple languages and interfaces is going to
be
tough. While I've played with C++, I've never gone deep with it, so docs
that come from 'here's how it works in C++' as a starting point are
sometimes baffling especially WRT DeviceContexts. I have gotten insights
from the C++ wxWidgets book, and, in fact, figured out my re-paint on
exposure solution after digging there and thinking hard.

This is not in any way to say that it is your "fault" that the docs are
the
way they are, except in a good way. :D

I love the examples from a 'neato' perspective, but some of them drive
me
nuts because I have to erase all the loops and random color insertions.
Besides the fun toys, examples that demonstrate one thing at a time are
needed.

Now that I've swum through the seaweed, I'll see if I can extract a
simple
example and submit that as a start.

AFA doc patches, usually where I get bitten is where there is no
ruby-side
doc at all (such as the evt_xyz names and what class they inherit from).
I'm
not good enough yet to be able to surf the C++ and SWIG interfaces to
figure
it out, but I'll keep learning.

I love using this Wx library with Ruby; I'm blowing the boss's mind and
he's
stopped muttering about Intel's commitment to C++ and Qt, at least for
the
moment.
Ef9964ca3deee9f1c2b1c961306e3af8?d=identicon&s=25 Fred L. (ffred)
on 2010-12-20 08:34
Hello,
I'm a beginner with Ruby and even more with wxRuby and it was a question
I was planning to ask too, because the "news" on wxRuby website are
rather old and the roadmap is dated from 2009.

wxWidgets 2.9.2 is about to come shortly (as to the wxWidgets roadmap)
and the "main release goals" of v3.0 are all marked "done" or "almost
done".

there are nice things in the wxRuby Roadmap, but maybe it could be
actualised and a message could be add about that "paused" progression of
wxRuby.. at least to let the newcomers know exactly what to expect.

of course I completely understand that you can't keep all your time on
wxRuby development, I also got very little time to learn Ruby/wxRuby
right now. but maybe a message could also be add to ask some help or new
developpers...?

Fred
Ef9964ca3deee9f1c2b1c961306e3af8?d=identicon&s=25 Fred L. (ffred)
on 2010-12-21 11:03
just to let you see what we could get with wxWidgets 3.0 :
(sorry if it's an "old" info, but I just discovered it yesterday)
http://www.wxdesigner-software.de/WoWoW30.html
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2010-12-22 13:46
(Received via mailing list)
On 20/12/2010 07:34, Fred L. wrote:
> I'm a beginner with Ruby and even more with wxRuby and it was a question
> I was planning to ask too, because the "news" on wxRuby website are
> rather old and the roadmap is dated from 2009.
>
> wxWidgets 2.9.2 is about to come shortly (as to the wxWidgets roadmap)
> and the "main release goals" of v3.0 are all marked "done" or "almost
> done".

That's really handy to know - part of the reason I got stalled was b/c I
couldn't get a development environment working on OS X after upgrading
to 10.6. But it sounds like 3.0 is much closer and should be usable now.

Like you say, there are a lot of things that are exciting in there; some
of the architectural changes will help the redesign of wxruby 3.0
(though I'm not promising I have time to get started on this
immediately).
> there are nice things in the wxRuby Roadmap, but maybe it could be
> actualised and a message could be add about that "paused" progression of
> wxRuby.. at least to let the newcomers know exactly what to expect.

That's a good idea - as per this thread, perhaps it's worth saying that
it's not abandoned, just snoozing.

happy xmas to all
alex
Ae84e0bb7cf816006c0267961c814899?d=identicon&s=25 Ken Spyce (k13oharts)
on 2010-12-26 19:42
I was wondering if we're going to have the PropertyGrid class anytime
soon...
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2010-12-29 02:18
(Received via mailing list)
Hi Ken

On 26/12/2010 18:42, Ken Spyce wrote:
> I was wondering if we're going to have the PropertyGrid class anytime
> soon...
>

I did have a go at porting this class a while ago but I ran into
difficulties - can't quite remember what the problem is but I couldn't
work round it. It might be worth another go with the newest version of
SWIG if there's interest in the class.

alex
00109e19a784b64f81b483a5dbec690a?d=identicon&s=25 Chauk-Mean Proum (chauk-mean)
on 2010-12-29 02:27
(Received via mailing list)
Hi Alex,

2010/12/29 Alex Fenton <alex@pressure.to>:
> It might be worth another go with the newest version of SWIG if
> there's interest in the class.

Have you tried wxRuby with SWIG-2.0.0/2.0.1 on OS X ?

Cheers,
Chauk-Mean.
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2010-12-29 10:58
(Received via mailing list)
Hi Chauk-Mean

On 29/12/2010 01:26, Chauk-Mean Proum wrote:
> Have you tried wxRuby with SWIG-2.0.0/2.0.1 on OS X ?

No, not yet. After upgrading to OS X 10.6 I found it hard to get a ruby
+ wxWidgets build that would work together for wxruby. Having seen your
new instructions for Ubuntu I'm going to have a sweep through the bug
list in the new year - catch up with you on -dev soon.

alex
D24691bdb2ed83327cd763d00c57f858?d=identicon&s=25 T. N. T. (tnt)
on 2011-08-24 18:17
The answer seems to be yes. The 'current' wxruby for linux/GTK requires
a libwx_gtk2u_media_2.8.so that isn't inluded in Ubuntu since years...
FXRuby also doesn't work.
06f6780c99d4a8dd71f2b474082ea9ce?d=identicon&s=25 Alex Fenton (Guest)
on 2011-08-25 14:19
(Received via mailing list)
On 24/08/11 17:17, T. N. T. wrote:
> The answer seems to be yes. The 'current' wxruby for linux/GTK requires
> a libwx_gtk2u_media_2.8.so that isn't inluded in Ubuntu since years...

No. There hasn't been a huge amount of new dev activity, but that
package should work on 10.04 LTS Ubuntu, I think.

Given the variation in distro structure, the expectation is that many
people will need to compile for Linux anyway.

a
2d352e436babbf3d80d4a6171220cd5b?d=identicon&s=25 Chloro Form (tony44)
on 2011-09-13 00:25
I'm sure that some people invested a lot of work into wxRuby and I
appreciate that a lot. After having invested a lot of time into wxRuby
(as a user) over the past 1.5 years, I can say that I believe I have an
insight about what works and what doesn't.

So is wxRuby dead?
I don't think that question is relevant. The code certainly runs,
there's a number of ports and when it works, it works fine (at least on
win and linux, there are a number of flakes and differences with OSX)

Should it be used for new projects?
As long as they're not supposed to run on OSX, maybe.

Is the documentation complete and up to date?
no. If you're not up to checking the wxwidget docs and then looking at
the port source to wxruby, I would strongly discourage from starting a
new project with wxruby. there are enormous gaps in the documentation.

Is it bugfree to an acceptable degree?
I had no problems in single-threaded designs. Everything that involved a
second thread (be it 1.8 or 1.9) was a disaster. And yes, I know the ins
and outs of MT programming, I've been making my living as a C/C++ dev
for embedded RT systems for 10+ years now.

Is there an active community?
no. this is the only forum I found on the web. no offense, but it's
anything but active.

Is it being maintained or are reported bugs being fixed?
no. Over two years, I have not seen a single update or bug fix.

wxRuby is a nice toy. Don't expect it to be anything else.
85991f138ede6236f35eb98da22b7b01?d=identicon&s=25 Marvin Gülker (quintus)
on 2011-09-13 10:26
(Received via mailing list)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Am 13.09.2011 00:25, schrieb Tony Meier:
> Is the documentation complete and up to date? no. If you're not up
> to checking the wxwidget docs and then looking at the port source
> to wxruby, I would strongly discourage from starting a new project
> with wxruby. there are enormous gaps in the documentation.

I'm sure the wxRuby devs appreciate each and every documentation
patch. The docs are written in Textile and the textile files are
included in the wxRuby source tarball, so there shouldn't be a problem.

> Is it bugfree to an acceptable degree? I had no problems in
> single-threaded designs. Everything that involved a second thread
> (be it 1.8 or 1.9) was a disaster. And yes, I know the ins and outs
> of MT programming, I've been making my a C/C++ dev for embedded MT
> systems for 10+ years now.

I absolutely agree with the multi-threading problem in wxRuby.
Threads, be it 1.8's green threads or 1.9's native threads, are a pain
in wxRuby. I learned how to manage multi-process programs due to this,
but that has the mayor drawback of not working properly on Windows
(although *this* is not a wxRuby problem, but a general Ruby one).

> Is there an active community? no. this is the only forum I found on
> the web. no offense, but it's anything but active.

It could be more, indeed.

> Is it being maintained or are reported bugs being fixed? no. Over
> two years, I have not seen a single update or bug fix.

There haven't been releases, that is true. But there *have* been some
commits in the SVN trunk (check my post on Ruby-Talk relating to
that), and they have been this year.

> wxRuby is a nice toy. Don't expect it to be anything else.

wxRuby is one of the last useful GUI toolkits for Ruby. There are
others, sure, but from what I hear from others is this:

* FxRuby is dead.
* Tk is hardly usable, and looks as if it came right from the stone age.
* Ruby-Gtk is unusable as well and not documented.
* Shooes doesn't fit well for bigger applications and comes with it's
own Ruby.
* Some other GUI toolkits are platform-specific.

So, probably the only other considerable GUI toolkit is QtRuby. If
wxRuby isn't going to be a bit more active, I may look into that.

Vale,
Marvin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJObxL9AAoJELh1XLHFkqha6+oH/0cofyQYQyo92hkGcRtPvBoK
P5dteP2ou9TwdMkoKiaP0kSA/0h3sdSP4pPU6RATAGqYFB8nqH/SgmTKuRC+vyxQ
orbI8Zw3mrGC6UUXA6dsez4E2H0aQIpnU7XkuchDNb0UN112/9PLs/HHVvRvsJ/F
ZHUQffIvRfcLDM0E6a7AfUXmNtU6FYHvs1IhLHqHxYaNgomsm6D2AB6YYnRtLeyB
L2J4c2dq5MdUFJ+D55TtTBSgqeG29+pOCWy/biU7uUxelloNYRo5syjbs1GcBK/F
/L2Su/kFTTumlnQfmoyH/7CwurIR0MCnGJ36HVXQLmfrFmmjLp7KOUT3jflIxJw=
=J1QS
-----END PGP SIGNATURE-----
This topic is locked and can not be replied to.