anygui-devel — Discussions about the development of the Anygui package

[Anygui-devel] An idea: Anygui Core

[Magnus L. H. <ma...@he...>]

Just an idea for making Anygui more useful, even if we don't get
another full release: How about paring it down to an anygui-core
package? If, by chance, a full release is made at some time in the
future, it could be based on anygui-core.

I'm thinking the core could simply contain the event dispatching,
reference system, front-end/back-end sync mechanism, and back-end
import mechanism -- but make no assumptions about the specific API
(i.e., the GUI classes/widgets) and contain no such classes or
specific back-ends.

This is code that could be useful for others who wish to develop
something along the lines of Anygui (e.g., what James Thompson talked
about), and it would be much easier to debug and test (and document)
than the rest of the code.

Just an idea. If I get the time, I might do this myself. If anyone
else is interested, help (or, at least, interest ;) is always welcome.

-- 
Magnus Lie Hetland                   Fall seven times, stand up eight
http://hetland.org                                 [Japanese proverb]

Re: [Anygui-devel] Questions about anygui

[Magnus L. H. <ma...@he...>]

James Thompson <ja...@aj...>:
>
> I was reading the archives and came across the Dec post where Magnus
> said the project is pretty much dead and not to waste time on it.

Unfortunately, that does seem to be the state of things, yes.

> Use something like wx instead.  In our project
> (www.gnuenterprise.org) our forms runtime software allows the user
> to select the front end of their choice.  wx, gtk, qt, win32 native,
> curses, etc, etc.

Right. Very Anygui-like :)

> We already have a common library for our applications but several of
> our apps only work in wx (our designer, navigator, etc).  Now I'm
> tossing about the idea of refactoring our tools to work more along
> the lines of python version of netbeans or eclipse (which has a thin
> GUI wrapper that seems in the spirit of anygui).  So a base GUI
> framework using plugins would replace the separate apps, it would
> also provide me with a base application for in house custom apps.

Souns like a good idea to me.

> In our project lots of people use the native windows driver, the wx
> driver, and the gtk driver.  I'd personnally like to get our qt
> driver up and 
> 
> So a few questions..  Does anygui from cvs.....

Just a note about the CVS code base: It's not really in any kind of
stable state. Several features don't really work as they should for
certain back-ends, there are probably lots of bugs etc.

It might be better to take the code (and ideas) you need from it, and
write a custom library for your application. (Or, if you'd like to
base yourself on Anygui, any patches would certainly be appreciated.)

I just wanted to be clear on that Anygui as it is is not really usable
for any stable or commercial system, no matter what features you
need...

>   allow for the interecption of events on a per keystroke level?

No. But the event dispatching system is quite nifty, and adding this
sort of thing wouldn't be too hard. (The problem is, of course, that
you'd have to add it to all back-ends.)

>   what about events caused by things like opening a dropdown box?

No.

>   support some type of drag and drop?

No.

>   support cut n paste on the various platforms?

Well... Sort of. Anygui is cut 'n' paste-agnostic. It works, but
Anygui doesn't know about it.

>   have more complete API docs?

No, sorry. (You could try to run some automated API doc generator on
parts of it...) If you have questions, I'd be happy to help figure
things out (and perhaps document a few things in the process).

>   allow access to the underlying widget set so that if I wanted to use 
>     the base framework in house w/ custom QT widget I'd be able to
>     write my plugin that uses qt hook that widget in.

Yeah, that should be possible.

>   is it lighter weight than wx?

Hm. I guess, in some ways. It's not really big. On the other hand,
you'd have to have something like Qt or wx in the background, of
course. And Anygui is much slower for many things (such as resizing),
if that matters.

> And why did development stall?  Were there issues that arose that
> made anygui seem like a deadend?  Or was it something else?

Many reasons, I guess. I think the basic, simple reason is enthusiasm
was used up before a tangible result (such as a functioning 0.2
release) could boost it again. The reason for this, I guess, might
partially be the "second system" syndrome -- a full refactoring and
lots of fancy features. Maybe. Also, the amount of development needed
(largely because of all the back-ends and all the manual testing
needed of all of them) was, perhaps, too much compared to the (at that
time) current user base/demand (not all that high).

So... As it turned out, I think Anygui was a successful
proof-of-concept that simply aimed a bit too high.

> What I'm thinking about doing is creating the base application with
> anygui.  

Well... In that case you should be prepared for lots of bugs and
missing features (as I mentioned). It may be that you could get some
developers here help you with specific bugs, but I can't guarantee
anything.

> Then create a plugin that link in our existing widget drivers (as
> they don't allow any input without checking with the back end
> first).  If anygui gives me the access I'd need to the events I need
> then I could eventual phase out the existing widgets w/ anygui
> replacements.

Well, it seems interesting :)

> Take Care,
> James

-- 
Magnus Lie Hetland                   Fall seven times, stand up eight
http://hetland.org                                 [Japanese proverb]

[Anygui-devel] Questions about anygui

[James T. <ja...@aj...>]

I was reading the archives and came across the Dec post where Magnus said the 
project is pretty much dead and not to waste time on it.  Use something like 
wx instead.  In our project (www.gnuenterprise.org) our forms runtime 
software allows the user to select the front end of their choice.  wx, gtk, 
qt, win32 native, curses, etc, etc.

We already have a common library for our applications but several of our apps 
only work in wx (our designer, navigator, etc).  Now I'm tossing about the 
idea of refactoring our tools to work more along the lines of python version 
of netbeans or eclipse (which has a thin GUI wrapper that seems in the spirit 
of anygui).  So a base GUI framework using plugins would replace the separate 
apps, it would also provide me with a base application for in house custom 
apps.  In our project lots of people use the native windows driver, the wx 
driver, and the gtk driver.  I'd personnally like to get our qt driver up and 
running.  Anygui seems designed with just this environment in mind unlike wx.

So a few questions..  Does anygui from cvs.....

  allow for the interecption of events on a per keystroke level?
  what about events caused by things like opening a dropdown box?
  support some type of drag and drop?
  support cut n paste on the various platforms?
  have more complete API docs?
  allow access to the underlying widget set so that if I wanted to use 
    the base framework in house w/ custom QT widget I'd be able to
    write my plugin that uses qt hook that widget in.
  is it lighter weight than wx?

And why did development stall?  Were there issues that arose that made anygui 
seem like a deadend?  Or was it something else?

What I'm thinking about doing is creating the base application with anygui.  
Then create a plugin that link in our existing widget drivers (as they don't 
allow any input without checking with the back end first).  If anygui gives 
me the access I'd need to the events I need then I could eventual phase out 
the existing widgets w/ anygui replacements.

Take Care,
James

Re: [Anygui-devel] anygui devel

[Magnus L. H. <ma...@he...>]

Sharky On PTNet <sha...@gm...>:
>
> Hi all,
> 
>    Is this project stoped?

I guess I'd have to say so, yes.

>    I want to begin learn who to use some GUI for python and anygui is
> the best because i can use all gui's like gtk, curses, Qt, etc... but
> i see that this project don't have much upgrades :-(
> 
>    Is this version (0.1.1) stable and work? Can i start learning with AnyGui?

I wouldn't recommend that you invest too much time and effort in it. I
think choosing one of the more standard ones (like wx or qt) would be
a better choice; they're available on all platforms anyway these days.

(BTW: To further post to this list, please subscribe -- otherwise your
messages will have to be manually accepted.)

> Thanks for your time,
> Sharky @ PTNet
> (Helio Pereira)
> 
> PS.: Sorry my pour english :-P

-- 
Magnus Lie Hetland       Fallen flower I see / Returning to its branch
http://hetland.org       Ah! a butterfly.           [Arakida Moritake]

Re: [Anygui-devel] cvs refuses connection, snapshot is empty archive

[Magnus L. H. <ma...@he...>]

m.evans <m....@rd...>:
>
> I've tried accessing cvs, and I've tried downloading the nightly
> snapshot. anonymous cvs refused access, and when I downloaded the
> snapshot, it proved to be an empty archive. Are there any other up
> to date archives I can get anygui by?

Yes, I see there's a problem with the snapshot thing. It seems to be a
problem on the SourceForge end (perhaps an update on how things work
that I didn't notice).

The anonymous CVS works well for me, though. Perhaps you could try
again? It has happened to me that it doesn't work for a while and then
starts working again...

(If you can't get it to work I could always wrap up a tarball for
you.)

Oh -- and, in general, to post to this list one has to be a member.
(Otherwise I have to manually accept your emails.)

-- 
Magnus Lie Hetland       Fallen flower I see / Returning to its branch
http://hetland.org       Ah! a butterfly.           [Arakida Moritake]

[Anygui-devel] anygui devel

[Sharky On P. <sha...@gm...>]

Hi all,

   Is this project stoped?
   I want to begin learn who to use some GUI for python and anygui is
the best because i can use all gui's like gtk, curses, Qt, etc... but
i see that this project don't have much upgrades :-(

   Is this version (0.1.1) stable and work? Can i start learning with AnyGui?


Thanks for your time,
Sharky @ PTNet
(Helio Pereira)

PS.: Sorry my pour english :-P

[Anygui-devel] cvs refuses connection, snapshot is empty archive

[m.evans <m....@rd...>]

I've tried accessing cvs, and I've tried downloading the nightly snapshot.
anonymous cvs refused access, and when I downloaded the snapshot, it proved to
be an empty archive. Are there any other up to date archives I can get anygui by?

Re: [Anygui-devel] Trying to get the latest cersion

[Magnus L. H. <ma...@he...>]

Ellianef Jean <jel...@ya...>:
>
> Hi there,

Hi!

In order to send to this list you should subscribe. (I've accepted
this message manually.)

> I would like to give a try to anygui, but i'm facing
> problems:
> The doc I read when installing 0.1.1 is not at the
> correct level, so the progs I wrote didn't work.

I'm not sure what you mean. Did you read the docs included in 0.1.1?
Or the ones on the Web site (which is clearly marked as unreliable ;)?

> When trying to download the cvs version 0.2a.0 I can't
> login, the cvs said anonymous is not allowed.

Strange. That would be a SourceForge issue, I guess.

> If I try to download from cvs by http I get html pages
> instead of code.

Have you tried the snapshot from the download page?

> I've tried both windows and linux cvs commands.
> Do you have a compressed file of the latest version
> and could send it to me ?

There's a link on the download page of anygui.sf.net:

  http://www.idi.ntnu.no/~mlh/anygui/anygui.tar.gz

Be warned, though, that the code is incomplete and is not being
actively developed.

-- 
Magnus Lie Hetland       Fallen flower I see / Returning to its branch
http://hetland.org       Ah! a butterfly.           [Arakida Moritake]

[Anygui-devel] Trying to get the latest cersion

[Ellianef J. <jel...@ya...>]

Hi there,
I would like to give a try to anygui, but i'm facing
problems:
The doc I read when installing 0.1.1 is not at the
correct level, so the progs I wrote didn't work.
When trying to download the cvs version 0.2a.0 I can't
login, the cvs said anonymous is not allowed.
If I try to download from cvs by http I get html pages
instead of code.
I've tried both windows and linux cvs commands.
Do you have a compressed file of the latest version
and could send it to me ?
Many thanks.  =20


=09

=09
	=09
Vous manquez d=92espace pour stocker vos mails ?=20
Yahoo! Mail vous offre GRATUITEMENT 100 Mo !
Cr=E9ez votre Yahoo! Mail sur http://fr.benefits.yahoo.com/

Le nouveau Yahoo! Messenger est arriv=E9 ! D=E9couvrez toutes les nouveau=
t=E9s pour dialoguer instantan=E9ment avec vos amis. A t=E9l=E9charger gr=
atuitement sur http://fr.messenger.yahoo.com

Re: [Anygui-devel] A first look

[<ti...@se...>]

> I should warn you that this project isn't very active at present.

APIs for GTK, curses etc are rather stable, right? So when one has
caught up with the backend implementation there's not that much to
do for a while, until the next major version.


> > * There is no screenshot on the curses-backend on the website. Is it
> > working satisfactory?
>
> It works OK in the latest (quite old) release. Not quite sure how well
> it works in the CVS version -- it hasn't been tested properly for a
> while.

I'll try the old release and the CVS version and see what I find.
I'll try GTK too.


> Actually, just grab a CVS snapshot -- the talked-about interim release
> wouldn't be any different.

I think tarballs or installation programs would mean more people trying
anygui out. People are lazy and many don't know how to use cvs.


> > * How come there is not that much of interest in a project like this?
> >   I find it quite weird. Bad marketing?  :)
>
> Hard to tell. The project kind of ran out of steam (perhaps in part
> because of the "second system syndrom")

What's the second system syndrom?


> and as long as it's not actively developed I guess it won't attract
> much interest.

So let's get active  :)

My main interests are gtk, qt and curses, so perhaps I can help there.
After figuring out how it all works...


> Never thought about it. Not much is going on, so I don't think there
> would be much talk on that channel... :]

When I'm on freenode I'll be in #anygui. You should be too  ;)

www.anygui.org points to http://www.anygui.org/anygui.sf.net
That's a faulty url. Does de...@an... still work?

Tim

Re: [Anygui-devel] A first look

[Magnus L. H. <ma...@he...>]

Tim Terleg?rd <ti...@se...>:
>
> I read about anygui in the Python Cookbook yesterday. I got interested,
> so here I am. I have a few questions, of course  :)

:)

I should warn you that this project isn't very active at present.

> * There is no screenshot on the curses-backend on the website. Is it
> working satisfactory?

It works OK in the latest (quite old) release. Not quite sure how well
it works in the CVS version -- it hasn't been tested properly for a
while.

> * How up to date are the backends? Is GTK v2 working, or only v1?

Only GTK v2, in fact.

> * An interim release were announced to happen soon. Where is it?

Nowhere to be seen ;)

Actually, just grab a CVS snapshot -- the talked-about interim release
wouldn't be any different.

> * How come there is not that much of interest in a project like this?
>   I find it quite weird. Bad marketing?  :)

Hard to tell. The project kind of ran out of steam (perhaps in part
because of the "second system syndrom") and as long as it's not
actively developed I guess it won't attract much interest.

> * What are Anygui's goals? To provide a documented/specified subset of
>   GUIs for small apps or a general toolkit for any app?

Small.

> * I was surprised when I didn't find anything about IRC on the website.
>   I though all open source projects had their own channels. Is there
>   one? If not, why not start one?

Never thought about it. Not much is going on, so I don't think there
would be much talk on that channel... :]

> Tim

-- 
Magnus Lie Hetland              "Wake up!"  - Rage Against The Machine
http://hetland.org              "Shut up!"  - Linkin Park

[Anygui-devel] A first look

[<ti...@se...>]

I read about anygui in the Python Cookbook yesterday. I got interested,
so here I am. I have a few questions, of course  :)

* There is no screenshot on the curses-backend on the website. Is it
working satisfactory?

* How up to date are the backends? Is GTK v2 working, or only v1?

* An interim release were announced to happen soon. Where is it?

* How come there is not that much of interest in a project like this?
  I find it quite weird. Bad marketing?  :)

* What are Anygui's goals? To provide a documented/specified subset of
  GUIs for small apps or a general toolkit for any app?

* I was surprised when I didn't find anything about IRC on the website.
  I though all open source projects had their own channels. Is there
  one? If not, why not start one?

Tim

Re: [Anygui-devel] Status request for bug 730572 and patch 730573

[Magnus L. H. <ma...@he...>]

Michael P. Dubner <du...@mi...>:
>
> Hello,
> 
> in simple words: subj.
> 
> Can anyone answer me why bug 730572 still not fixed in CVS regarding of 
> the fact patch 730573 submitted more that year ago?

Heh -- nothing that a bit of nagging can't fix ;)

We never really used the bug tracker in the development (there's a
separate bug file for the developers) so now that development has
quieted down I guess nobody checked for submitted bugs...

I've checked in your patch now -- please see if things are better now.

[snip]
> I'm using anygui for some of my daily used admin proggies...

Just for the record: The CVS version isn't really considered "stable".
Of course, submitted patches like these make it better, so -- thanks!
And sorry this wasn't fixed before :)

-- 
Magnus Lie Hetland              "Wake up!"  - Rage Against The Machine
http://hetland.org              "Shut up!"  - Linkin Park

[Anygui-devel] Status request for bug 730572 and patch 730573

[Michael P. D. <du...@mi...>]

Hello,

in simple words: subj.

Can anyone answer me why bug 730572 still not fixed in CVS regarding of 
the fact patch 730573 submitted more that year ago?

I've just reinstalled my system, and of cause got new version of anygui 
from CVS. This annoying bug stil here.
I'm using anygui for some of my daily used admin proggies...

--
Best regards,
Michael Dubner (du...@mi...)
Brainbench MVP/HTML+JavaScript (http://www.brainbench.com)
PS: Sorry for my English

[Anygui-devel] Unit testing

[Uriel <ur...@bi...>]

Hi again, I decided to look again into this, I have reworked my
framework; and now it's even easier to write tests, in many cases it's
even "automatic", also taking advantage of this I have added a quite few
new test and now the whole thing can be called from the command line
with a parameter to set the backend... it's still far from complete,
but starts to look like might be useful.

On the down side, the whole thing is more messy, and I'm not sure if the
OO approach I have taken is a good idea at all, I would prefer
something more "procedural" or "functional"...

Please let me know what you think, I have tried to run the tests with
gtk, tk, and wx(pyqt in debian sid seems to be broken at the moment, and the
curses backend didn't start), and already found a few bugs in the gtk
backend, it seems that it doesn't support setting some parameters after
widget creation and fails silently; in tk for some reason the Exit
button doesn't work, will have to investigate...

[Warning: the code is ugly and hacky, you have been warned ;)]
http://www.binarydream.org/~uriel/anygui/anygui_utest_2004-04-15.tar.gz

Best wishes

uriel

Re: [Anygui-devel] Disgruntled would-be user

[Magnus L. H. <ma...@he...>]

Greg Ewing <gr...@co...>:
>
> Magnus:
> 
> > For example, my ability to be fascinated by new things -- as
> > opposed to my inability to finish things...
[snip]
> So the next time someone complains about you not
> finishing something, you can say, "Well, Leonardo
> never finished anything either, and he wasn't
> *half* as much a genius as I am!". Or something.

:D

Thanks for making my day :--)

-- 
Magnus Lie Hetland  "Oppression and harassment is a small price to pay
http://hetland.org   to live in the land of the free."  -- C. M. Burns

Re: [Anygui-devel] Disgruntled would-be user

[Greg E. <gr...@co...>]

Magnus:

> For example, my ability to be fascinated by new things -- as
> opposed to my inability to finish things...

Many of the highly creative people from history seem
to have suffered from this problem -- it seems to
come with the territyr. Leonardo da Vinci
apparently had it *really* badly, for example.

So the next time someone complains about you not
finishing something, you can say, "Well, Leonardo
never finished anything either, and he wasn't
*half* as much a genius as I am!". Or something.

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
gr...@co...	   +--------------------------------------+

Re: [Anygui-devel] Interim release?

[Uriel <ur...@bi...>]

hi!

Thanks for your prompt reply, and sorry for the delay...

On Sun, Apr 11, 2004 at 11:15:54PM +0200, Magnus Lie Hetland wrote:
> Right -- that's exactly the kind of thing I started to write (you can
> find it in my sandbox directory in the nondist package in CVS). But I
> didn't really get anywhere (just started to fiddle with it) -- if
> you're working on something like this, keep it up!
I think to recall taking a look at it, but for some reason I can't
recall I didn't like it, but maybe I'm thinking of something else, I
will take a look again...

 
> Yes. I thought about another possibility for a while, that would work
> even if Anygui didn't really work at all. (It would be good to be able
> to run the tests even if, say, anygui.Window has a bug in it, right?)
> That was using a local HTTP server to display successive HTML
> documents describing the test, trying to launch the test, and having
> the Pass/Fail thing being done through (local) CGI. Just an idea. Some
> early code for this, too, can be found in my sandbox directory.
I think the option with "Pass" and "Fail" buttons seems the simples, I
personally don't like the idea of running a whole web app for the rare
cases when the GUI tests wont work at all, which should be so critical
that probably would make running any test quite pointless... also, IMHO
the simpler the system is, the better, and most important of all, it
should be really easy to run and to write new test cases...

 
> > The tests can also be "deferred", when testing something at "runtime",
> > a button "Run" is presented, otherwise the test is run automatically,
> > I'm not sure now why I did this, but it made sense at the time =)
> 
> Who knows. Might be a good idea :)
Ok, I think now I remember why I did that, it's a way to test "dynamic"
update of attributes, for example to test if you can change the position
of a button once the app is running instead of passing it to the
constructor; I guess that some kind of threading with a timer or
something might work, but I hate threads :) (but I love Limbo/Alef/Plan9 style
threading: http://plan9.bell-labs.com/who/rsc/thread/ ;))

 
> > I have no clue if this is really a good idea, but you might find it
> > useful: http://www.binarydream.org/~uriel/anygui_utest/
> Looks good.
Did you take a look at the code? do you really think it's a good idea?
dunno, I wasn't very sure, but if you say it looks good... :)
 
> Would you be interested in trying to clean up the tests in the
> distribution, using this framework? 
That was my original plan, I'm really busy at the moment, in the middle
of deploying one system, with a deadline for another... the typical ;)
But I will give it a try next time I have some spare time to play with
it... just don't hold the release for it! just tar the whole thing and
put it up, the sooner the better!

Oh, and remember to announce it in the home page so the project doesn't
look so dead ;) BTW, who takes care of the web site? I'd say it could
use some updating and clean up, I might give that a try too...

> (If you want check-in privileges, just email me your SourceForge username.)
hmmmm... if I do some work I will probably import the whole thing into
arch, cvs really gets to my nerves :) But as it's so self contained it
should be easy to merge...

Best regards

uriel

Re: [Anygui-devel] Interim release?

[Magnus L. H. <ma...@he...>]

Uriel <ur...@bi...>:
>
> Hi
> 
> I have been following this project for some time, and I think it's great
> to have a pythonic GUI framework, I was hopping to do some work in
> anygui for my own projects, but other projects that didn't need it
> took priority, still I hope to contribute something...

Excellent :)

> > What the heck -- why don't we just release?
> 
> Exactly what I thought while reading the previous thread, you lose
> nothing by making a release... just call it experimental-pre-alpha or
> whatever, and if someone complains, screw them... at least it get
> something out of the door for people to try, test and play with.

Exactly. I've been thinking that people can use the CVS version, but I
guess that's quite a bit of a threshold for many users.

Unless someone pipes in soon, saying that this is a Bad Idea(tm) I
guess I'll make a release.

> You mentioned in another email that there is a need to have a better
> way to run tests;

Yes, indeed. I've started a tentative framework for automated testing
here (permitting you to press a 'Passed' or 'Failed' button for each
test in sequence -- still semi-manual)

> while I was evaluating anygui some months ago, I tried to run the
> tests in the test/ directory, but it didn't really work very well,
> they are not very complete in their coverage, and a bit cumbersome
> to run...

Yes, I know. Sadly, full automatization is very hard, given the
graphical and interactive nature of the package.

BTW: Quite a few of the tests aren't even in use yet (such as the
canvas test).

> so just to try anygui I wrote a small testing framework using
> pyunit, the "framework" itself is ~50 lines, and I wrote a few tests
> for Button in another ~50 lines. 

Wow -- great!

> The idea was, IIRC, that each test should be isolated, and that due to
> the nature of GUIs, most of the evaluations had to be done by a human(to
> determine if what was supposed to show up in the screen actually did),

Indeed.

> so every test runs inside a frame and shows 3 buttons after the test
> have been run: Fail, Pass, Exit, the tests are run one after the other,

Right -- that's exactly the kind of thing I started to write (you can
find it in my sandbox directory in the nondist package in CVS). But I
didn't really get anywhere (just started to fiddle with it) -- if
you're working on something like this, keep it up!

> and the user only has to click the button depending on his evaluating of
> the results in the screen.  (probably a label explaining what one should
> expect from each test should be added)

Yes. I thought about another possibility for a while, that would work
even if Anygui didn't really work at all. (It would be good to be able
to run the tests even if, say, anygui.Window has a bug in it, right?)
That was using a local HTTP server to display successive HTML
documents describing the test, trying to launch the test, and having
the Pass/Fail thing being done through (local) CGI. Just an idea. Some
early code for this, too, can be found in my sandbox directory.

> The tests can also be "deferred", when testing something at "runtime",
> a button "Run" is presented, otherwise the test is run automatically,
> I'm not sure now why I did this, but it made sense at the time =)

Who knows. Might be a good idea :)

> I have no clue if this is really a good idea, but you might find it
> useful: http://www.binarydream.org/~uriel/anygui_utest/

Looks good.

Would you be interested in trying to clean up the tests in the
distribution, using this framework? (If you want check-in privileges,
just email me your SourceForge username.)

-- 
Magnus Lie Hetland  "Oppression and harassment is a small price to pay
http://hetland.org   to live in the land of the free."  -- C. M. Burns

Re: [Anygui-devel] Interim release?

[Uriel <ur...@bi...>]

Hi

I have been following this project for some time, and I think it's great
to have a pythonic GUI framework, I was hopping to do some work in
anygui for my own projects, but other projects that didn't need it
took priority, still I hope to contribute something...

> What the heck -- why don't we just release?
Exactly what I thought while reading the previous thread, you lose
nothing by making a release... just call it experimental-pre-alpha or
whatever, and if someone complains, screw them... at least it get
something out of the door for people to try, test and play with.


You mentioned in another email that there is a need to have a better way
to run tests; while I was evaluating anygui some months ago, I
tried to run the tests in the test/ directory, but it didn't really work
very well, they are not very complete in their coverage, and a bit
cumbersome to run... so just to try anygui I wrote a small testing
framework using pyunit, the "framework" itself is ~50 lines, and
I wrote a few tests for Button in another ~50 lines. 

The idea was, IIRC, that each test should be isolated, and that due to
the nature of GUIs, most of the evaluations had to be done by a human(to
determine if what was supposed to show up in the screen actually did),
so every test runs inside a frame and shows 3 buttons after the test
have been run: Fail, Pass, Exit, the tests are run one after the other,
and the user only has to click the button depending on his evaluating of
the results in the screen.  (probably a label explaining what one should
expect from each test should be added)

The tests can also be "deferred", when testing something at "runtime",
a button "Run" is presented, otherwise the test is run automatically,
I'm not sure now why I did this, but it made sense at the time =)

I have no clue if this is really a good idea, but you might find it
useful: http://www.binarydream.org/~uriel/anygui_utest/

Thanks for your great work and best wishes

uriel

P.S.: Hi laura, I'm also a (lurking) 9fan ;)

Re: [Anygui-devel] GtkLabel is not defined

[Magnus L. H. <ma...@he...>]

Donn <do...@te...>:
>
> Magnus,
> Thank you for your response. I fetched the CVS version and it runs at least! 

Excellent!

> I shall have some fun tinkering with this beast and discovering how
> it works.

I'm glad to hear it.

> Apologies for not registering, the links from the web-page do not
> make that obvious.

No problem. We just added that restriction a while ago to avoid spam;
I haven't mentioned it on the Web pages. (I guess I'll put a notice
there.)

> Thanks for a great product!
> (And some great Python info on your site!)

Thanks :)

> Best regards,
> Donn Ingle

-- 
Magnus Lie Hetland  "Oppression and harassment is a small price to pay
http://hetland.org   to live in the land of the free."  -- C. M. Burns

Re: [Anygui-devel] GtkLabel is not defined

[Donn <do...@te...>]

Magnus,
Thank you for your response. I fetched the CVS version and it runs at least! 
I shall have some fun tinkering with this beast and discovering how it works.
Apologies for not registering, the links from the web-page do not make that 
obvious.

Thanks for a great product!
(And some great Python info on your site!)
Best regards,
Donn Ingle

On Saturday 10 April 2004 22:29, Magnus Lie Hetland wrote:
> Donn <do...@te...>:
>
> [snip]
>
> > I want to use Python as a language to teach kids about coding and
> > anygui would be just brilliant to cut-down the sheer amount of
> > download this and download that that gui programming seems to demand
> > on Linux.
>
> Sounds good. I think teaching is one of the areas where Anygui might
> have some potential.
>
> > Well, I would say thanks for a brilliant product, but I still gotta see
> > it!
>
> Indeed ;)
>
> [snip]
>
> >     _gtk_class = GtkLabel
> > NameError: name 'GtkLabel' is not defined
>
> Right. You're probably using Anygui 0.1.1, and that uses an old
> version of PyGTK (I think). In the new version of the GTK back-end,
> GtkLabel is not used -- gtk.Label is used instead.
>
> If you've followed some of the discussions on this list (or visited
> the Web site) you've probably noticed that there havent been any
> releases for a while. The trouble you're experiencing is fixed in the
> CVS version, but we haven't relased that yet. You're perfectly welcome
> to download and use it, though, of course. The GTK back-end might not
> be 100% up to scratch yet, but that is mainly because of a lack of
> testing, so if you use it and notice bugs, just tell us, and we can
> hopefully fix them as we go along. That should also get us closer to
> an actual release :)
>
> If you'd like to go with a more stable version, you could use Anygui
> 0.1.1 with PyGTK 1, but I don't think I'd recommend that solution
> myself...
>
> Good luck -- and just nag us about bugs and the like.
>
> Oh, and you have to subscribe to this list to post to it, normally.
> (I'll gladly accept your postings manually, but it might be just as
> practical for you to subscribe, I guess.)

-- 
at a translucent feeling ,
 a moth , sonic 
 and harmful 

[Anygui-devel] Interim release?

[Magnus L. H. <ma...@he...>]

What the heck -- why don't we just release?

So things don't work perfectly... But if we can recommend that people
try out the CVS version, why not give it to them as a release,
available through the standard download channels at SourceForge and so
forth? It might even induce a few more people to have a look and
report bugs they find...

In any case, it doesn't cost us (or me) much. In order to avoid
getting this mixed up with planned releases, I suggest we call it an
"interim release" (possibly the first of several) and number it with
the release date, e.g. Anygui interim release 20040410 (or
anygui-20040410.tar.gz). The release notes would be quite clear about
the nature of the release, of course.

By making rpms and Windows binary installers and the like, this would
at least make the code more available for fooling around :)

What do you think?

-- 
Magnus Lie Hetland  "Oppression and harassment is a small price to pay
http://hetland.org   to live in the land of the free."  -- C. M. Burns

Re: [Anygui-devel] GtkLabel is not defined

[Magnus L. H. <ma...@he...>]

Donn <do...@te...>:
>
[snip]
> I want to use Python as a language to teach kids about coding and
> anygui would be just brilliant to cut-down the sheer amount of
> download this and download that that gui programming seems to demand
> on Linux. 

Sounds good. I think teaching is one of the areas where Anygui might
have some potential.

> Well, I would say thanks for a brilliant product, but I still gotta see it!

Indeed ;)

[snip]
>     _gtk_class = GtkLabel
> NameError: name 'GtkLabel' is not defined

Right. You're probably using Anygui 0.1.1, and that uses an old
version of PyGTK (I think). In the new version of the GTK back-end,
GtkLabel is not used -- gtk.Label is used instead.

If you've followed some of the discussions on this list (or visited
the Web site) you've probably noticed that there havent been any
releases for a while. The trouble you're experiencing is fixed in the
CVS version, but we haven't relased that yet. You're perfectly welcome
to download and use it, though, of course. The GTK back-end might not
be 100% up to scratch yet, but that is mainly because of a lack of
testing, so if you use it and notice bugs, just tell us, and we can
hopefully fix them as we go along. That should also get us closer to
an actual release :)

If you'd like to go with a more stable version, you could use Anygui
0.1.1 with PyGTK 1, but I don't think I'd recommend that solution
myself...

Good luck -- and just nag us about bugs and the like.

Oh, and you have to subscribe to this list to post to it, normally.
(I'll gladly accept your postings manually, but it might be just as
practical for you to subscribe, I guess.)

-- 
Magnus Lie Hetland  "Oppression and harassment is a small price to pay
http://hetland.org   to live in the land of the free."  -- C. M. Burns

[Anygui-devel] GtkLabel is not defined

[Donn <do...@te...>]

Hello,
I have looked in the mail archives and only found 2 references to this 
problem. I would like to know if this has been solved since those posts (Jan 
2003).
I am running Fedora Core 1 and have not done any funny stuff like upgrading 
GTK or Python.
The install seems smooth enough, but a run gives the error listed at the end.

I want to use Python as a language to teach kids about coding and anygui would 
be just brilliant to cut-down the sheer amount of download this and download 
that that gui programming seems to demand on Linux. 

Well, I would say thanks for a brilliant product, but I still gotta see it!
Regards
Donn
South Africa

[root@donn test]# python test.py
Traceback (most recent call last):
  File "test.py", line 3, in ?
    from anygui.Utils import log
  File "/usr/lib/python2.2/site-packages/anygui/__init__.py", line 87, in ?
    _backend_passthrough()
  File "/usr/lib/python2.2/site-packages/anygui/__init__.py", line 59, in 
_backend_passthrough
    mod = _dotted_import('anygui.backends.%sgui' % name,)
  File "/usr/lib/python2.2/site-packages/anygui/__init__.py", line 41, in 
_dotted_import
    mod = __import__(name, globals(), locals(), [])
  File "/usr/lib/python2.2/site-packages/anygui/backends/gtkgui.py", line 64, 
in ?
    class Label(ComponentMixin, AbstractLabel):
  File "/usr/lib/python2.2/site-packages/anygui/backends/gtkgui.py", line 65, 
in Label
    _gtk_class = GtkLabel
NameError: name 'GtkLabel' is not defined

-- 
an ion plague ;
 secret adapts a moon