* Melchior FRANZ -- Thursday 16 October 2008:
> Could you put some debug messages in
> $FG_ROOT/gui/dialogs/file-select.xml,
Since the last Nasal update it's now really broken, just like
the ufo's model selection dialog. You can hack around that by
adding these lines somewhere
* Erik Hofman -- Monday 20 October 2008:
> I find it rather annoying to have to move away from ALUT just because
> Apple can't play nice. Bad for them, let them straighten it out if they
> don't obey the the specs. I vote against this.
The problem seems to be that the original ("Creative") OpenA
* Melchior FRANZ -- Monday 20 October 2008:
> - (non-gui) fonts (i.e. texture fonts on instruments etc.)
Whoops, no. That's already converted to OSG. Only "non-gui"
fonts like the frame-rate counter and the ATC/multiplayer
messages are concerned. But those aren't actually
* James Turner -- Monday 20 October 2008:
> I also realised there are probably some other pieces of the code that
> still rely on sg or ssg types or functions,
- gui
- (non-gui) fonts (i.e. texture fonts on instruments etc.)
- joystick
- networking
and whatever they depend on.
m.
* James Turner -- Monday 20 October 2008:
> Anyone care to speculate on how much work it would be to
> migrate from pu to osgWidget?
AFAIK that's not an option ATM. osgWidget is very new and was
committed only a few weeks ago. It still only offers basic
widgets, like plain buttons, text, menu, (ta
It's probably not well known that livery files (or other "overlay"
files) have no defined structure. They are just XML files that are
copied ("laid") over the property tree on request. What you put in
there, and where, is entirely your business. It doesn't have to do
with liveries and textures at a
* Melchior FRANZ -- Sunday 19 October 2008:
> Patch attached ...
Oh, and uncomment the / part, so that
the livery is correctly transmitted via MP. And the
property only needs to be there, but it's not actually
read, so it doesn't have to get initialized. But that's just
co
* Melchior FRANZ -- Sunday 19 October 2008:
> What was probably broken were your "material" animations
Patch attached ... s/object/object-name/
m.
Index: Models/pc7.xml
===
RCS file: /var/cvs/FlightGear-0.9/data
* Erik Hofman -- Sunday 19 October 2008:
> I looked at a number of aircraft that use it and somehow it
> doesn't apply to they way I have set op the PC-7.
Your Nasal parts were correct and functional. (They are just
some trivial helpers, anyway.) What was probably broken were
your "material" anim
aircraft.livery and aircraft.livery_update are now only
wrappers for more generic classes, which one can also
use directly instead. They allow to manage more than just
livery XML files. One could allow users to choose a
livery, seat covers, flight attendant uniforms, ... :-)
each of them separatel
* Alasdair Campbell -- Saturday 18 October 2008:
> fgfs --aircraft=CitationX --airport=EGPF
No segfault here with fg/osg/head. Only lots of
"CullVisitor::apply(Geode&) detected NaN, [...]".
m.
-
This SF.Net email is sponsore
* Torsten Dreyer -- Saturday 18 October 2008:
> "texcon" to convert the *.?af textures to bmp and gimp for
> the .rgb.
> Search the net for texcon01.zip - runs smoothly under wine.
Or use plib's own ./tools/src/af2rgb/af2rgb without wine. ;-)
m.
* Tatsuhiro Nishioka -- Friday 17 October 2008:
> Needless to say,"someone" who wouldn't bother working around
> it is going to report. ;-)
>
> Go for it!
The bo105 doesn't use texture repetition, so this wasn't a priority
for me. I just feel better reporting bugs that I have observed and
investi
* Torsten Dreyer -- Friday 17 October 2008:
> I consider this a bug in the osg ac3d loader.
Yes, it is a bug and needs to get fixed in OSG. I wouldn't bother
working around it. The spec and the reference implementation
by the AC3D makers are clear. So, who's going to report the
bug to the osg list
* Melchior FRANZ -- Saturday 13 September 2008:
> [l-key file select dialog]
> * Rob Shearman, Jr. -- 9/10/2008 12:48 PM:
> > if I select a subdirectory [...] the screen fails to refresh
>
> I'll look into that next week.
And "next week" is the week after this
* robin424 -- Wednesday 15 October 2008:
> You know the problem, we discussed it before, since YASim is not
> JSBSim i can't use the "text" condition only numeric condition.
Sounds like jsbsim is a major millstone around your neck.
What about fixing jsbsim? :-P
m.
-
Oh, and what I really wanted to say in my last mail:
* Curtis Olson -- Wednesday 15 October 2008:
> Look for docs-mini/README.multiscreen in the CVS source.
That's a bad place. It belongs to $FG_ROOT/Docs/. (Only!)
m. :-)
* Curtis Olson -- Wednesday 15 October 2008:
> Look for docs-mini/README.multiscreen in the CVS source.
In the past we had "identical" files in $FG_SOURCE/docs-mini/
and $FG_ROOT/Docs/. The problem with that approach, however,
is that both being in sync was rather the exception than the
norm. And
* Detlef Faber -- Thursday 09 October 2008:
> I want to give the option to select a different logo with the same
> livery.
That's now implemented (though not well tested yet):
var logo_dialog = gui.OverlaySelector.new("Select Logo",
"Aircraft/foo/Models/Logos",
"sim/model/lo
* Melchior FRANZ -- Tuesday 14 October 2008:
>
> some/model.ac
err ... some/model.xml
^^^
Of course, it only works for XML-wrapped submodels. Otherwise
there wouldn't be a lot of properties to overri
* Vivian Meazza -- Tuesday 14 October 2008:
> The catapult steam needs customizing for each catapult length.
> So you need at least different versions for each carrier, and
> possibly more than 1.
You can now define an overlay branch that's applied to a loaded
submodel before the animation proper
* James Turner -- Monday 13 October 2008:
> I believe Syd's point was that he *had* changed the code, but was
> nevertheless encountering the error he described.
May be. And I have replaced all .rgba? by foo.png and
it just worked.
m.
---
* Curtis Olson -- Monday 13 October 2008:
> Just real quickly, ".rgba" signifies to our .rgb loader
> than the texture contains an alpha channel.
That would be a rather bad idea, but it doesn't look at
the extension at all, and neither do the loaders of
plib/osg/gimp/blender/kde. All SGI loaders/w
* Melchior FRANZ -- Sunday 12 October 2008:
> Only a few people's emails were concerned, and I suggest that
> those unsubscribe and subscribe again. This seems to have
> fixed it for me (so far :-).
OK, I take it back. Not even *that* worked. Sigh.
sf.net is seriou
* Melchior FRANZ -- Sunday 12 October 2008:
> The usage of cmdarg() in listener callback functions was deprecated
> since a while. Now it's no longer supported and will cause an error
> message.
Correction: it won't cause an error, but (still) return the last
cmdarg() property.
* Detlef Faber -- Sunday 12 October 2008:
> I've noticed that Livery over MP seems to have stopped working with
> recent CVS.
Can you update (data & fg) and try again? I hope it's fixed now.
m.
-
This SF.Net email is sponsor
The usage of cmdarg() in listener callback functions was deprecated
since a while. Now it's no longer supported and will cause an error
message.
Where some of you still used something like:
setlistener("/whatever", func {
var foo = cmdarg().getValue();
);
you must now use a regular fun
* Detlef Faber -- Sunday 12 October 2008:
> Strange thing. R22 works here too, but the bo105 doesn't.
My local bo105 differs from the CVS version. Now that I switched
back to an unmodified version, I see the same breakage. I found
the reason and will work on a fix.
m.
---
* Melchior FRANZ -- Sunday 12 October 2008:
> But not the only problem. bo105 and R22 work, but the s76
> doesn't, even if one makes the path relative. I'm still
> investigating ...
Ah, yes. Another bug in the s76. :-)
Check out Gijs' wiki page:
http://wiki.
* Melchior FRANZ -- Sunday 12 October 2008:
> A bug. In that model. ;-)
But not the only problem. bo105 and R22 work, but the s76
doesn't, even if one makes the path relative. I'm still
investigating ...
m.
* Detlef Faber -- Sunday 12 October 2008:
> I've noticed that Livery over MP seems to have stopped working with
> recent CVS.
Works here for the bo105 and the R22. And it *would* work for the
s76 without the good old leading-slash bug.
--- Sikorsky-76C/Models/s76c.xml30 May 2008 23:46:28
As explained in one of the recent threads, the mail duplicates
that are coming in since a few days are caused by a problem at
sf.net. This is reported and still unfixed:
https://sourceforge.net/tracker/?func=detail&atid=21&aid=2147569&group_id=1
Only a few people's emails were concerned, a
* gerard robin -- Saturday 11 October 2008:
> Does the number of http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-de
* Vivian Meazza -- Friday 10 October 2008:
> I've no idea how it got into the list of properties.
Maybe it's for the (steerable) carrier "aircraft"? Using
8 bytes more over MP than required is probably not such a
big issue. Demanding string comparisons for every check
is a bit annoying, even thoug
* Vivian Meazza -- Friday 10 October 2008:
> You need to pass the launchbar state over mp?
The "you" is inappropriate. I didn't say that I need it,
and I didn't add it to the transmitted properties. Yet
it is there! Why don't you read the source?!
$ grep launch src/MultiPlayer/multiplaymgr.cxx
Better make that ...
setlistener("/gear/launchbar/state", func(n) {
setprop("/fdm/jsbsim/launchbar/launch-bar-state", n.getValue() ==
"Engaged");
}, 0, 0);
... because I seem to remember that the property is written to
quite frequently (also a bug, if you ask me), and the second
0 make
* gerard robin -- Friday 10 October 2008:
> So from the boolean value coming from JSBSim i have to listen it,
> and to convert with a specific Nasal script , which is tricky.
You are slightly overrating the trickiness factor ... :-}
setlistener("/gear/launchbar/state", func(n) {
setp
Added:
+ {111, "surface-positions/wing-pos-norm", SGPropertyNode::FLOAT},
+ {112, "surface-positions/wing-fold-pos-norm", SGPropertyNode::FLOAT},
wing(s)-pos-norm is apparently not used anywhere yet,
but wing-fold-pos-norm is:
$ lsprop|grep wing-fold-pos-norm
$FG_ROOT/Aircraft
* gerard robin -- Thursday 09 October 2008:
> On jeudi 09 octobre 2008, Melchior FRANZ wrote:
> > If these are the commonly used properties for this
> > purpose, then we should just add them to the list in
> > src/Multiplyer/mulitplaymgr.cxx.
> Wing position, not sure t
* gerard robin -- Thursday 09 October 2008:
> surface-positions/wing-fold-pos-norm and
> surface-positions/wings-pos-norm
If these are the commonly used properties for this
purpose, then we should just add them to the list in
src/Multiplyer/mulitplaymgr.cxx. Generic MP properties
should only be
* Detlef Faber -- Thursday 09 October 2008:
> I want to give the option to select a different logo with the same
> livery. (e.g One Navy Aircraft with the skull logo and another navy
> Aircraft with dragon logo).
Use aircraft.formation for now. It's very similar to aircraft.livery.
I'll add a gene
Just to make that clear: I don't say: "don't touch 'my' pages".
If you find typos, bad grammar (lots of that to be found),
wrong content or other mistakes, just fix them. And if you
want to add a section or two, just do it.
But don't come along and "fix" my style, because (1) mine
is better, and (
I've just added a chapter about Nasal's concept of
OOP (object oriented programming). Some developers
still shy away from classes, and I hope that this
now changes. :-)
http://wiki.flightgear.org/index.php/Nasal_scripting_language
I'll review and improve that over the next days,
and maybe add
A few days ago I noticed that someone had modified my
AO tutorial page on the wiki. I had all images centered,
and suddenly they were all left aligned. This looked
horrible! I had to roll back.
Just a few minutes ago I found that my Nasal page was
modified as well: all blank lines before section a
* James Turner -- Tuesday 07 October 2008:
> Nine copies of this message so far, and counting.
http://sourceforge.net/tracker/?func=detail&atid=21&aid=2147569&group_id=1
m.
-
This SF.Net email is sponsored by the Moblin
* gerard robin -- Monday 06 October 2008:
> i hope that there is not many typo error like this
Nope, that was it. Occasionally I run lsprop (./scripts/tools/)
like so:
$ lsprop >/dev/null
and this warns about such garbage. It's really only a
cosmetic issue and has no negative effects. It's a
Hi,
* Erwan MAS -- Sunday 05 October 2008:
> I resubmit my patch .
I had seen it the first time, but after glancing over them I
decided that the patches have problems and that for some of
that I considered other developers "responsible" (e.g. Curt
for the net_fdm, one of the MP people for the MP
* James Turner -- Sunday 05 October 2008:
> PNG *can* be lossy, and for all I know the default settings in some
> tools may even save it lossily, but most PNGs are lossless. It's
> basically RLE + zlib + some other tricks.
I've never heard of that. I think it can only be lossless. There
is a l
* Heiko Schulz -- Sunday 05 October 2008:
* * Melchior FRANZ -- Sunday 05 October 2008:
> > - "timed" animation broken
>
> Hu? never noticed!
The magenta ring that cockpit tutorials use is meant to
blink. It does in fg/plib.
> > - "shader" animation
* Syd -- Sunday 05 October 2008:
> I would like to convert all the scenery textures to png's , if I get
> the go ahead here.
OK from me.
To Alex: PNG is lossless.
m.
-
This SF.Net email is sponsored by the Moblin Your Move
* Christian Schmitt -- Sunday 05 October 2008:
> I clearly vote for the Concorde here.
>From the Concord developer:
"Note that with OSG, many 2D hotspots are missing inside the Concorde
(that works with Plib). Perhaps because of clipping of 2D panels by 3D
flatness."
Though that doesn't really
* Durk Talsma -- Sunday 05 October 2008:
[wiki/"Changes since 1.0.0"]
We should really have had this from the beginning. It's hard
to remember all the details afterwards. I couldn't even remember
my contributions without looking at my git log. A lot of work
went into making things work again, and
* Heiko Schulz -- Sunday 05 October 2008:
> I think we don't should change the base package.
We should definitely add one of the airships.
> I will try to get most of my projects now into CVS, [...]
Ask on IRC, or one of the developers per mail (you find their
names in the cvs-logs list), or p
* Jon S. Berndt -- Thursday 02 October 2008:
> Dave C. is still out there, but he's really busy.
Dave is a bad example, as he probably really left fgfs
development because of an "incident". All others that have
recently been cited as someone who left because of an
allegedly rough climate here AFAI
* gerard robin -- Thursday 02 October 2008:
> I was on going to open directories on CVS for it, and i just notice
> that there is yet, an other Stampe-SV4 in it, which is not mine.
[...]
> AFAIK duplicate models are not whish.
Well, I said at some time that duplicating aircraft should be avoided.
* Oliver Schroeder -- Thursday 02 October 2008:
> No, it's from the "CVS Resources" page on flightgear.org.
>
> "cvs -d :pserver:[EMAIL PROTECTED]:/var/cvs/FlightGear-0.9
> co data" creates the "data" subdir.
Yes, but it doesn't say that you should set FG_ROOT to the directory
that contains a sub
* gerard robin -- Wednesday 01 October 2008:
> An other remark/question , how could we activate the "turn to wind" and
> elevator up/down with that keyboard feature.
> Today i have it, included into my naval Aircraft which is the wrong way , i
> would be better to have it within FG out of any Ai
* Oliver Schroeder -- Tuesday 30 September 2008:
> Here is a config for the Saitek Pro Flight Yoke
Committed, thanks. I only added for the gear. There
are aircraft where the landing gear only moves as long as a
button is pressed (b29, for example).
m.
--
* gerard robin -- Saturday 04 October 2008:
> what is "the ugly hack" .
That was something only remotely related -- code that ignores clear
instructions and has unpredictable consequences, while it doesn't
have a single advantage. Sounds bad, doesn't it? :-}
> I promise to prepare a example to
In a discussion about a desirable keyboard reorganization,
John D. suggested vi-like keyboard commands that consist
of more than one stroke:
http://www.mail-archive.com/flightgear-devel%40lists.sourceforge.net/msg13356.html
I've now committed a first version of a Nasal-based
solution. Tat grea
* gerard robin -- Sunday 28 September 2008:
> The OSG animation particles models could be very accurate
> within XML, but unfortunately there is missing a lot of
> features ( more than a lot :) ) which are there within
> OSG native model.
OK, so you identified where the brainpower should have
* Durk Talsma -- Saturday 04 October 2008:
> Given that we have an OSG branch that has undergone significant
> development this year, and a PLIB branch, with little or no
> development, I would strongly urge that the main release would
> be OSG based.
I agree. The PLIB branch was only kept alive f
* Martin Spott -- Friday 03 October 2008:
> Maintaining this branch for such a long time, including the pile of
> work that had been spent to get the 1.0 release out, has been the
> biggest distraction and waste of developer resources since I'm
> monitoring this project
Yeah, emphasis more on mon
* Melchior FRANZ -- Thursday 02 October 2008:
> * Tatsuhiro Nishioka -- Thursday 02 October 2008:
> > I didn't change the key names, so you type ":rt028x"
> > and 028X is set.
Oh, and thanks for fixing that.
m.
PS: sf.net mail seems to be broken. Only every othe
* Tatsuhiro Nishioka -- Thursday 02 October 2008:
> I didn't change the key names, so you type ":rt028x" and 028X is set.
> Which do you guys prefer, :rt028X/Y or :rt028x/y? I prefer the latter
> though.
Lower case letters were intended and approved by the
"TACAN man" on irc. I just didn't realiz
* gerard robin -- Thursday 02 October 2008:
> Sorry I send that message again, because i guess it has been lost
No, it didn't get lost. And I have replied to it, but like my first
message about multikey it never made it to the list. Will probably
come later. Here's my response again:
* gerard
* Erik Hofman -- Thursday 02 October 2008:
> Would it be feasible to reserve the 'a' for aircraft specific and 'x'
> for experimental features?
'x' for experimental seems fine. 'a' for aircraft is a problem, as I
think we should keep using that for the autopilot. There are a lot
of autopilot sett
* Oliver Schroeder -- Thursday 02 October 2008:
> I noticed that fgjs is broken, it looks into
> "/usr/local/share/FlightGear/Input/Joysticks/template.xml" instead
> of "/usr/local/share/FlightGear/data/Input/Joysticks/template.xml".
fgjs is correct. Don't use the "data" version. I wonder why pe
* Melchior FRANZ -- Wednesday 01 October 2008:
> I have planned to offer a list of options per level on request
> (TAB key).
Done. toggles a list of options.
m.
-
This SF.Net email is sponsored by the Moblin You
* Nicolas -- Wednesday 01 October 2008:
> In the window, FG could display the available letters. So the
> user can see the different choices.
Yes, that's what I meant when I wrote
| I have planned to offer a list of options per level on request
| (TAB key).
You can also use :? to get a
Just added:
D >>> Dialogs
Da autopilot
Dc cloud
Df fuel&payload
Dg GPS settings
Di instrument settings
Dm multiplayer pilot list
Dpa position in air
Dpg position on ground
Dr radio settings
Ds sound configuration
Dt time settings
Dv view options (not i
Committed, thanks. I only renamed it to fg-completion.zsh,
because the bash file was already named fg-completion.bash.
m.
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux
* gerard robin -- Wednesday 01 October 2008:
> we will now need a multi help map behind the keyboard :) .
:-P
I have planned to offer a list of options per level on request
(TAB key). But I also expect that people quickly memorize the
few combinations that they are interested in and just ignore
As you may have noticed, I have a few times committed stuff
to files called "multikey" in the last days. (Couldn't come
up with something better sounding :-). I also posted to the
list about it two and a half days ago, but this message didn't
make it to the list yet. I expect that it will come in a
* Melchior FRANZ -- Sunday 28 September 2008:
> The change wasn't/isn't even necessary (see above).
Another reason for the patch was that we could use OSG's
model embedded particles in the same scenery. Now that
we have XML configured OSG particles, this reason is
obsolete, to
* Tim Moore -- Sunday 28 September 2008:
> You can call it whatever you like :) The consensus is not
> universally negative.
Fact is: there was no consensus at all. IIRC two people on IRC
agreed with you and later Curt on the list. None of them knew
that the patch was based on wrong assumptions.
* Heiko Schulz -- Saturday 27 September 2008:
> Can some explain why it loads the wrong file?
That's an intentional bug, a.k.a. (mis)feature. You can also call
it poor design. It was introduced after a discussion in this thread
(where my objection was overruled):
http://www.mail-archive.com/fl
* Melchior FRANZ -- Tuesday 23 September 2008:
> The implementation is new, the idea isn't: John D. has suggested
> such a feature. (I had started implementing it, but never finished.)
FWIW: I'm about to finish that, so no need to adapt your version. :-)
BTW: there's n
* James Turner -- Wednesday 24 September 2008:
> Well, in the future, the key args for getActiveRunwayForUsage are what
> I said - the usage type (landing or takeoff, maybe some other flags in
> the future for weird stuff, like helis) and the inbound / outbound
> radial, [...]
OK, then let's
* James Turner -- Wednesday 24 September 2008:
> This could either be an addition member on the airport hash
yes
> [...] a new function exposed to Nasal makes sense [...]
IMHO no
>active_runway_for_airport(, ... other optional args )
The airportinfo() function is there to return
* Sébastien MARQUE -- Sunday 14 September 2008:
> I've created a system, which I think is new in FG as I've never
> seen it elsewhere.
The implementation is new, the idea isn't: John D. has suggested
such a feature. (I had started implementing it, but never finished.)
Having something like this gl
* Bohnert Paul -- Saturday 23 August 2008:
> Run FlightGear Multiplayer I noticed the following error;
> Nasal runtime error: floating point error in math.sin()
> at /usr/local/share/FlightGear/data/Nasal/geo.nas, line 169
> called from: /usr/local/share/FlightGear/data/Nasal/multiplayer.nas,
* gerard robin -- Tuesday 19 August 2008:
> When loading a JSB FDM Aircraft i get twice the same heading information.
[...]
> I just wonder why :)
No idea. Doesn't happen here. Maybe some local modification?
m.
-
This SF.Net
[l-key file select dialog]
* Rob Shearman, Jr. -- 9/10/2008 12:48 PM:
> if I select a subdirectory [...] the screen fails to refresh
I'll look into that next week.
m.
-
This SF.Net email is sponsored by the Moblin Your Move
* Melchior FRANZ -- 9/5/2008 12:47 PM:
> * Manfred Janßen -- 9/5/2008 11:39 AM:
>> WARNING: PUI: Too many live puInterfaces open at once!
>
> A few people have reported this error message, but unfortunately none of
> them were capable of debugging it. Those who are never get
* Manfred Janßen -- 9/5/2008 11:39 AM:
> WARNING: PUI: Too many live puInterfaces open at once!
A few people have reported this error message, but unfortunately none of
them were capable of debugging it. Those who are never get it. :-)
> var values = screen.window.new(10,150,4,1);
> values
* gerard robin -- 9/4/2008 1:25 PM:
> On jeu 4 septembre 2008, Melchior FRANZ wrote:
> =>You can sell the bo105 for one billion dollars.=<
> Oh yes, sure, not any doubt, bo105 by Melchior Franz will be
> like Gioconda by Leonardo di ser Piero da Vinci.
Bah! We both know tha
* gerard robin -- 9/4/2008 12:05 PM:
> This must be said again and again, i guess that, here, nobody would
> accept that somebody else could make money with our own work.
Umm, no. Actually, everyone here who has understood the GPL
*does* accept that others make money with their work. Because
tha
* Manfred Janßen -- 9/4/2008 9:58 AM:
> Now it runs, I got an error that screen wasn't defined, solved it by
> renaming param_print.nas to xparam_pring.nas ;-)
Eeew! Had you put your local extension into the directory for local
extensions (%APPDATA%\flightgear.org\Nasal\), then you wouldn't
have h
* Manfred Janßen -- 9/4/2008 9:37 AM:
> values = screen.window.new(10,150,4,1);
I assume the "var" got lost here in the copying process. Not that
it's the cause of your problem ...
> values.write("Heading: " ~ sprintf("%6.2f", Heading);
closing parenthesis missing
> values.write("Altit
* Melchior FRANZ -- 9/3/2008 6:26 PM:
> - There's a fake-cvs server that Curt could install alongside of
> GIT. It mimics CVS and allows to access the GIT repo with a CVS
> client as if it were a true CVS. I guess that sooner or later
> there will be the same for SVN.
Oh,
Hey,
looks like I'm late to the party, and others have already answered
the questions, but here's my sermon:
* Frederic Bouvier -- 9/3/2008 2:05 PM:
> there is another thing that is unclear to me. How GIT currently
> interface with CVS ( and tomorrow SVN ) ?
> How do you merge content from CVS i
* Melchior FRANZ -- 9/3/2008 10:49 AM:
> So you can at the moment not commit files/content with a
> size >2GB after compression(!).
Err, >2GB before compression, as it will use the same file
routines for turning content into blobs. But the difference
doesn't matter, as suc
* Thomas -- 8/29/2008 10:46 PM:
> I'm more concerned about the 2 GB repo size limit listed in the "Known
> issues" in the release notes.
That's only worded badly. The 2GB limit isn't for the whole repository,
but for single "blobs" *in* a repository. So you can at the moment not
commit files/conte
Just for the record: KDE, one of the biggest F/OSS projects out
there, switched to SVN a few years ago. Now there are plans to
switch to one of the distributed SCM systems. SVN's capabilities
seem to be no longer fit for the job. The project will probably
switch to GIT. Here's an article about it.
* Frederic Bouvier -- 8/27/2008 12:26 PM:
> It is just that nobody explained me the benefits of using GIT over a
> well known system such as CVS and SVN. I am aware of the serious lacks
> of CVS, that's why I am advocating switching to SVN.
Half of the fgfs developers are already using GIT for sg/
* Melchior FRANZ -- 8/26/2008 3:03 PM:
> But it would make me a bit nervous if an aircraft developer commits
> several pointless updates of 5MB sound files. GIT can't compress that.
> We'd collect the whole pile on our disks. How much would disk space
> requirements grow e
* gerard robin -- 8/26/2008 2:32 PM:
> The specific air refueling and or fuel.nas makes problems if the Aircraft has
> some specific tank management, for instance Crusader has a specific "transfer
> tank" which feed the others, Blackbird has an other process in order to
> manage the balance
* Curtis Olson -- 8/21/2008 5:46 PM:
> I also really like how svn handles group and user authentication ...
> it does it outside of the unix account system which makes the system
> much easier to manage.
Unix permissions are well understood and not hard to handle. They are
the most straightforwar
* gerard robin -- 8/26/2008 11:23 AM:
> Won't it be possible to get with a generic Nasal script?
>
> with the closest Carrier:
> The Catapults position with heading.
> The wire positions Left/right position.
Could be problematic. The involved subsystems are handled at different
places in the ma
301 - 400 of 1910 matches
Mail list logo
