MacOsX Version issues

Programmers discuss here anything related to FreeOrion programming. Primarily for the developers to discuss.

Moderators:Committer, Committer

Message
Author
User avatar
pd
Graphics Lead Emeritus
Posts:1924
Joined:Mon Mar 08, 2004 6:17 pm
Location:52°16'N 10°31'E

#16 Post by pd » Sat Jan 15, 2005 12:06 am

wow, great news!

User avatar
Tyreth
FreeOrion Lead Emeritus
Posts:885
Joined:Thu Jun 26, 2003 6:23 am
Location:Australia

#17 Post by Tyreth » Sat Jan 15, 2005 6:53 am

Hi, good work. You can transfer your instructions from the talk page to the main page if you wish, or add a section here for Mac with some tips:
http://www.freeorion.org/index.php/Compile

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

#18 Post by frankie » Tue Apr 25, 2006 2:26 pm

LaplaceOperator wrote: i'm doing nearly everything in xcode (I could send you my Xcode-Project if you like..). except the boost-stuff. Yes i had problems too with compiling boost 1.30.2
Has there been any progress on the Mac side since 2004? FO v0.2 doesn't work with newer versions of SDL.

I am trying to compile FO myself, but the number of dependencies is staggering, most of which aren't precompiled for Mac, and Boost 1.33.1 gave me errors in make.

LaplaceOperator
Space Squid
Posts:60
Joined:Wed Sep 08, 2004 3:20 pm
Location:Switzerland

#19 Post by LaplaceOperator » Tue Apr 25, 2006 6:31 pm

No. Not really. Shortly after the release of 0.3 I tried to update the version. But I couldn't get the Graphviz working. As soon as you wanted to browse the Tech-Tree the application crashed somewhere inside GraphViz because of some structures that weren't initialized properly. And I couldn't find out exactly why they weren't. The GraphViz Code is really ugly with lots of macros that make the code confusing. The rest of the App was working though. But I didn't wanted to put something on the Wiki that was only working partially. And as i didn't have much time to further investigate (as i'm studiying physics and mechanical engineering now at the same time), I kinda gave up. This was with GraphViz 2.4 (which does not work;-) as stated in Wiki ). Now there is GraphViz 2.8 which I haven't tried out yet.

Boost 1.33.1 gave you errors in make? what errors? the only thing that doesn't work for me is the python stuff, because bjam seems to not find it. But so far this has not been a problem (as last time i checked python support was not included yet). The only thing you need to compile from there is Boost.signals and Boost.filesystem. The Rest is just Headerfiles.

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

#20 Post by frankie » Tue Apr 25, 2006 6:58 pm

LaplaceOperator wrote: Boost 1.33.1 gave you errors in make? what errors?
...
The only thing you need to compile from there is Boost.signals and Boost.filesystem. The Rest is just Headerfiles.
Hmm, make complained that some parts of Boost didn't compile. I didn't save the output, so I don't remember exactly what.
I do have signals and filesystem folders in /usr/local/include so maybe that's okay. But still another dozen-ish things to compile between here and FO. :(

cbedgar
Space Krill
Posts:5
Joined:Fri Oct 08, 2004 4:09 am
Location:San Diego

I will see what I can do

#21 Post by cbedgar » Thu May 11, 2006 4:19 am

O. K., I am back, and I am downloading the latest Mac programming tools to see if I can get FO 0.3 to compile. My mac programming skills, actually all my programming skills, are a little rusty however. I did most of my work for Mac OS system 6 and earlier, when the mac was still 9inch and black and white.

I never did get FO 0.2 to compile; I just figured out how to get the version Laplace Operator compiled to run. I will see if I can do better with 0.3.

Keep up the good work with the mainline branch! Got to move a little faster. At a point release ever two years we will get to 1.0 about 2020. :-P

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

Re: I will see what I can do

#22 Post by frankie » Thu May 11, 2006 1:25 pm

cbedgar wrote:I did most of my work for Mac OS system 6 and earlier
Greetings, fellow geezer. You can pretty much forget everything from 1980s MacOS, especially for a linuxy project like FO. Hope you're up on bash and gcc...

LaplaceOperator
Space Squid
Posts:60
Joined:Wed Sep 08, 2004 3:20 pm
Location:Switzerland

0.3-RC4 for MacOsX10.4

#23 Post by LaplaceOperator » Wed Aug 30, 2006 7:32 pm

Hello there!

Its been quite a while since i've last been to these forums. And even more time passed since my last attempt to bring freeorion to mac os x. But now the wait is (almost) over. I have managed to compile the latest source checkout and all the dependent libraries as universal mach-o files. I put up a compressed disk-image that contains the FreeOrion.app and an PackageInstaller for the libraries. As i have no ppc mac at the moment (working on a new i386 laptop), i wasnt able to check whether the ppc code runs as well as the i386 one does. The Disk image can be downloaded here: http://n.ethz.ch/student/schwadri/FreeO ... d_libs.dmg. The game is now an application bundle and comes with all the artwork and sounds inside. so just drag it to your application folder.
Please let me know if you encountered any problems running on i386 or ppc machines. Oh and btw, i would like to include a nice background-image to the Diskimage. Do you know where i can find the latest splash image uncut?

Cheers, LaplaceOperator

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

Re: 0.3-RC4 for MacOsX10.4

#24 Post by frankie » Wed Aug 30, 2006 8:04 pm

LaplaceOperator wrote:I have managed to compile the latest source checkout and all the dependent libraries as universal mach-o files. I put up a compressed disk-image that contains the FreeOrion.app and an PackageInstaller for the libraries. As i have no ppc mac at the moment (working on a new i386 laptop), i wasnt able to check whether the ppc code runs as well as the i386 one does.
Very nice indeed. Just downloaded and tried it on my G4. FO installed & launched perfectly, very nice UI. I started a game, queued some techs, took a peek at production. Sadly, while I was clicking through the list of buildables it crashed. I have emailed the log file to you.

LaplaceOperator
Space Squid
Posts:60
Joined:Wed Sep 08, 2004 3:20 pm
Location:Switzerland

#25 Post by LaplaceOperator » Thu Aug 31, 2006 8:27 am

hey! I didn't even consider that fo would be running on a G4 ;-), but i'm very pleased to hear that it works there too (at least partially). Do you remember what exactly you were doing when it crashed? What exactly were you clicking? Was it a specific buildable? From the crash log it looks like freeorion was loading a graphic file at that moment. And btw. how fast or slow is it on a G4 ?

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

#26 Post by frankie » Thu Aug 31, 2006 1:52 pm

LaplaceOperator wrote:Do you remember what exactly you were doing when it crashed? What exactly were you clicking? Was it a specific buildable? From the crash log it looks like freeorion was loading a graphic file at that moment. And btw. how fast or slow is it on a G4 ?
The crash is 100% repeatable with the following steps:
  1. start a single player game, default settings
  2. go to Production, click to show any list of buildables
  3. click an item, description pops up, click another, see it, then one more
  4. crash occurs before the third item displays
Speed seems quite playable, although I haven't gotten past this step yet. There's maybe a 1/10th second pause between my click and the game's response.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts:12689
Joined:Wed Oct 08, 2003 1:33 am
Location:Munich

#27 Post by Geoff the Medio » Thu Aug 31, 2006 7:42 pm

frankie wrote:
  1. start a single player game, default settings
  2. go to Production, click to show any list of buildables
  3. click an item, description pops up, click another, see it, then one more
  4. crash occurs before the third item displays
Is it always the third item clicked, regardless of order of items clicked, or is it always the same item, which happens to be the third on the list? If the latter, it might be missing art (as LaplaceOperator said).

This doesn't crash on my Windows build.
LaplaceOperator wrote:Do you know where i can find the latest splash image uncut?
There's a low resolution of the previous colour scheme on pd's image website, but I can't find the newest version not cut up. It shouldn't be too difficult to recombine the separated images in a bitmap editor though... pd also probably has it.

User avatar
tzlaine
Programming Lead Emeritus
Posts:1092
Joined:Thu Jun 26, 2003 1:33 pm

#28 Post by tzlaine » Thu Aug 31, 2006 8:15 pm

Thanks again, Laplace. Did you have to change any of the source? If so, could a get a patch?

User avatar
frankie
Space Krill
Posts:9
Joined:Tue Apr 25, 2006 3:18 am

#29 Post by frankie » Thu Aug 31, 2006 9:06 pm

Geoff the Medio wrote:Is it always the third item clicked, regardless of order of items clicked, or is it always the same item, which happens to be the third on the list?
Always the third item out of any, in any order. Seems like the art is all there. The button-click sound effect also works fine, right before FO dies.

As a mediocre outdated programmer I'm inclined to suspect a memory boundary error. Something isn't being reset properly when it gets rid of the previously displayed item.

Or probably something else I wouldn't even be able to guess.

User avatar
tzlaine
Programming Lead Emeritus
Posts:1092
Joined:Thu Jun 26, 2003 1:33 pm

#30 Post by tzlaine » Fri Sep 01, 2006 3:25 am

frankie wrote:
Geoff the Medio wrote:Is it always the third item clicked, regardless of order of items clicked, or is it always the same item, which happens to be the third on the list?
Always the third item out of any, in any order. Seems like the art is all there. The button-click sound effect also works fine, right before FO dies.

As a mediocre outdated programmer I'm inclined to suspect a memory boundary error. Something isn't being reset properly when it gets rid of the previously displayed item.

Or probably something else I wouldn't even be able to guess.
Can you produce a stacktrace?

Post Reply