gnustep-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Google Summer of Code


From: Riccardo Mottola
Subject: Re: Google Summer of Code
Date: Sat, 11 Feb 2012 20:58:34 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20120211 Firefox/10.0 SeaMonkey/2.7

Sebastian Reitenbach wrote:

On Tuesday, February 7, 2012 09:42 CET, Fred Kiefer<address@hidden>  wrote:

On 07.02.2012 09:21, Sebastian Reitenbach wrote:
On Monday, February 6, 2012 09:54 CET, Fred Kiefer<address@hidden>   wrote:

Looks like GSoC 2012 was announced at FOSDEM (Did anybody notice?). Here
is the timeline for it
http://www.google-melange.com/gsoc/events/google/gsoc2012

If GNUstep is interested we will have to come up with an application
until the 9th oh March. We could either do this as a stand alone project
(of course together with Etoile!) or under the GNU umbrella project as
we did the last two years.
I really would like to volunteer as a mentor again, I learned a lot from
Niels during his project. But we should only invest any effort here if
we see suitable students. Last year we were not able to attract such
students and so the whole application to GSoC was a waste of time.

Please ask around, if any students you know are interested.
I don't know anyone specifically, but I think we should give some ideas what 
could be
done, and what would be good to have.

For example, a lot of applications in GAP should be checked if still work with 
latest releases,
or could be enhanced with new features. Maybe a good job for a beginner with 
basic
objective-c skills, and just want to learn, starting from reading others code, 
and enhancing
existing applications, before sometime starting an own application.
This may be a bit too unspecific to be accepted as a GSoC project. We
should turn it into something more specific as for example, get
FlexiSheet fully implemented :-)
ah, OK, probably better then. I don't know what is OK for google, and what not.


Another thing I'd really like to have is some more cross desktop integration, 
for example,
allowing .desktop files, used in KDE and others,  to work. I'd really like to 
define Firefox or
something similar as my default browser. (until Vespucci is production ready ;)
We already once had a Google Summer of Code student to work on cross
desktop integration. Sadly not much came from that.
I remember writing .desktop support ages ago. The file specification may
have changed in between, most certainly it has, but it should be really
easy to update our file generation to match the current standard. What
is currently broken?
Well, I have a couple of .desktop files around on my GWorkspace Destop. Double 
clicking
them, doesn't do anything. I'd expect them to start the application configured 
in Exec=, or open the
URL from URL=, and use the icon defined in Icon= ...
but nothing happens when I click on such icon.

As for specifying a default browser, this should be as easy as to write
a GNUstep wrapper, that is just a .plist file and to copy it to where
make_services will find it. There must already be a lot of these
wrappers out there, where do we collect them? Maybe we should set up
some space in our source code repository to collect them?
They are in GWorkspace apps_wrappers subdirectory. But this approach generally 
has a flaw:
For how many applications do we want to create wrappers, when/where do we stop? 
;)
We obviously cannot do so for every application. Further, the paths to the 
application can be on
different places on different OS, for example /usr/bin and /usr/local/bin ...

On the other hand, many applications install a .desktop file in 
/usr/local/share/applications/
(at least which is the path for me on OpenBSD), and icons too. Packages that do 
that then run
update-desktop-database from the desktop-file-utils package on install. 
Afterwards it shows
up in the users menu, under the defined categories.

IIRC, the Makefiles support creation of .desktop files, from the info taken 
from the App bundle.

AFAIK, there doesn't exist something the other way around, allowing other 
application to create
an App Wrapper automatically. Even if that would exist, you'd still have to get 
others to make use of
it, which I think is then the harder part.

I'd also really like to have an applications menu in GWorkspace, built from the 
information from those
.desktop files in /usr/local/share/applications, that would allow me to browse 
all installed applications
and just start them from the menu ;)
I don't know about that. I don't want an application menu in GWorkspace. An app-icon however could handle that, click-opening with some fast options including a search field for example. Perhaps it could somehow query NSWorkspace for registered apps, this woul include applications and wrappers of course. How to integrate this smoothly in the UI is a mystery. I'd see it initially as an extension or an add-on to the "Fiend". I know I don't like most interfaces aorund currently, but I don't know how to make it better, what the most "openstep" way would be to deal with the load of apps.
Supporting this really standard stuff would prevent us from 
creating/maintaining a truckload of
Apps Wrappers. I actually created some of those apps wrappers for about 20 or 
so applications
but Riccardo refused to add them to the Apps wrappers, he said, this is not a 
kitchen sink, and
it should only contain really common used apps. Which I understand and is fine 
with me.
But on the other side, creating and maintaining own apps wrappers, is also a 
bit cumbersome.

Exactly. AN alternative would be to start a GAP repository where to dump dozens and dozens of ready-made wrappers. These still need to be customized by packagers due to different names (ooffice/soffice or firefox/iceape) and locations.

Riccardo



reply via email to

[Prev in Thread] Current Thread [Next in Thread]