Dear Bogdan,

thank you for you answers, commented further below.

On Mo, 24.10.2011, 21:40, Bogdan Marinov wrote:
> On Thu, Oct 20, 2011 at 5:28 PM, Georg Zotti <[email protected]>
> wrote:
>> Dear core team,
>> [snip]
>
>> Where in the branch shall I put plugin data files, i.e. some demo
>> sceneries?
>> [snip]
> This is a good question. As Alex Wolf said, you can put the files in a
> resource file. There is a problem, though - this will embed the files
> in Stellarium's executable file. While this is OK for small files like
> button textures, adding everything to the executable will continue to
> bloat it in the long run. We need a standardized file locations and
> install scripts for the plug-in resources.

OK, so the resource file is a clear no-go here. I speak of 3D models,
which are large, exchangeable and loadable via a GUI list similar to
"Landscapes". I think an install script is not necessary, just unpack and
copy into your Stellarium data folder. The question is, shall I add a
folder "modules" into the main folder, in which I now add "scenery3d"
(because it's a plugin and plugin data go into subfolders of "modules"),
or shall I add a "scenery3d" folder directly, because it's a concept
similar to "landscapes".

>> For walkaround with keyboard controls, I had to modify
>> StelMovementMgr::getCallOrder(), thereby breaking the pure "plugin"
>> concept. However, I feel that plugins should be allowed keyboard
>> interaction intercepting the main application.
>
> There is something like that done in the Oculars plug-in. I haven't
> looked at your code yet, so I don't know if you have used the same
> mechanism.

Ah, I have to look inside there!

>> A few months ago, there was discussion on the list on the buttons
>> behaviour of plugins. Has there been a decision/solution yet?
>
> No. Timothy Reaves wrote some code in this direction, but in the
> discussion, different people turned out to have different ideas on
> what exactly it should look like. I was going to write a resume of the
> positions and post it here or on the Wiki, but got distracted with
> "real life" issues, including the lack of a permanent Internet
> connection.
Sounds difficult...

>> It would be
>> helpful to have a right-click behaviour like "open config panel" -
>> currently I must use 2 buttons.
>
> Overriding the right click will be tricky, because it is currently
> used to de-select objects.
Then maybe with double-click? Or can a button "grab" the right-click event
before forwarding to the main app?

> You can have a look at my branch for plug-in GUI improvement -
> https://code.launchpad.net/~daggerstab/stellarium/plugins-gui-improvement
> Using the same method you can add a command GUI on the screen similar
> to the existing toolbars.

OK, must look inside.

>> [Manual is outdated]
> "Outdated" is an understatement. It was decided to migrate the manual
> to the Wiki, supposedly to make it easier to use/edit/translate. Alex
> Wolf did a lot of work on importing it in the Wiki, but a lot of the
> information is still either not up-to-date, incomplete or misleading.
Yes, following Alex' response I added some details on recent changes in
Landscapes.

> One of the reasons of the migration is that there is a MediaWiki
> extension that can export a collection of wiki pages as a PDF file.
> Unfortunately, it doesn't work with our site because of SourceForge's
> hosting policies. The Wiki was supposed to be moved together with the
> site on a seprate server When We Enable the DSS Background, but I
> don't know what's the current state of this, as Fabien has dropped off
> the radar.
Uh, that's bad to read! Who else in the team is OpenGL expert now?

>> [accuracy, documentation]
> That will be somewhat hard. Stellarium is optimized for fast
> rendering, not accuracy. There are some fudge factors included in the
> code, as well as some outright "nobody will care about that". Hight
> accuracy generally means more calculations and a lower
> frame-per-second rate. When Fabien was working on the Nokia N900 port,
> he even made some parameters to use a lower precision format (float
> instead of double).

I had seen this change, but did not know exactly why. OK, it's not a
problem for landscapes, but definitely for long-term computations. It will
be an important point to consider for me.

> Also, I think that the person responsible for some of the orbital
> modeling code no longer hangs around. :)
Yes, what a pity!

>> [unmerged/"dead"(?) branches]
> Only the branch owners _can_ do that. :)  There are also some
> apparently abandoned branches and merge proposals that should have
> been done ages ago. I'll try to comment/update my branches anyway.

OK, thanks!

Kind regards,
Georg

-- 
DI Dr Georg Zotti
Archaeoastronomy / ASTROSIM
VIAS-Vienna Institute for Archaeological Science
University of Vienna



------------------------------------------------------------------------------
The demand for IT networking professionals continues to grow, and the
demand for specialized networking skills is growing even more rapidly.
Take a complimentary Learning@Cisco Self-Assessment and learn 
about Cisco certifications, training, and career opportunities. 
http://p.sf.net/sfu/cisco-dev2dev
_______________________________________________
Stellarium-pubdevel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/stellarium-pubdevel

Reply via email to