Messages

301

Mac / Log of an attempt to compile a universal binary for 10.5.6 for 2.3-dev (trunk)

« on: March 15, 2009, 12:05:10 am »

Alright, I took some detailed notes while trying to get this put together. I'm hoping this will be of use to the devs.

The first round:

Log of compilation of Ufo-AI for Mac

12:10 Start installing latest Xcode (312 2621)

Odd errors seen in the logs:
Mar 11 12:36:19 Kleiman-ibook Installer[489]: run preupgrade script for DeveloperToolsCLI
Mar 11 12:36:20 Kleiman-ibook michael[751]: Upgrade scripts will be run.
Mar 11 12:36:20 Kleiman-ibook Installer[489]: Removing obsoleted files
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/^Husr/sbin/vmmap64
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/ usr/libexec/gdb/plugins/MacsBug/MacsBug_plugin
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/ usr/libexec/gdb/gdb-powerpc-apple-darwin
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/ usr/libexec/gdb/gdb-i386-apple-darwin
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/ usr/libexec/gdb/gdb-for-codewarrior
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/^Husr/bin/malloc_history
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/^Husr/bin/leaks
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/^Husr/bin/heap
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/^Husr/bin/atos
Mar 11 12:36:20 Kleiman-ibook runner[493]: trying to delete: /Developer/PLibrary/Developer/3.0/distcc/distcclaunchdconfig

Yes, those are ^H's in the filenames. Go figure.

Large number of missing man pages complained by makewhatis

After install:

Kleiman-ibook:~ michael$ df -l
Filesystem   1024-blocks     Used Available Capacity  Mounted on
/dev/disk0s3    42745372 36480732   6008640    86%    /
/dev/disk0s5    15597564 14497592   1099972    93%    /Volumes/Time Machine Drive

Just under 6 GB (Finder says 5.73).

Step 2: Install MacPorts 1.7.0
Step 3: Install latest X11 (2.3.2.1)

Mar 11 16:22:51 Kleiman-ibook kernel[0]: Finder[129] Unable to clear quarantine `X11-2.3.2.1.pkg': 30

Can someone explain what that means?

Step 4: "sudo port selfupdate"

At this point, I have a "standardized" development environment.

Kleiman-ibook:~ michael$ sudo port selfupdate

MacPorts base version 1.700 installed
Downloaded MacPorts base version 1.700

The MacPorts installation is not outdated so it was not updated
Kleiman-ibook:~ michael$ df -l
Filesystem   1024-blocks     Used Available Capacity  Mounted on
/dev/disk0s3    42745372 36655848   5833524    87%    /
/dev/disk0s5    15597564 14604000    993564    94%    /Volumes/Time Machine Drive
/dev/disk1s2       85860    78380      7480    92%    /Volumes/X11-2.3.2.1
Kleiman-ibook:~ michael$

(TM snapshot done)

Now, follow the instructions on the wiki:

Kleiman-ibook:~ michael$ sudo port install libsdl_ttf libsdl_mixer

NB: Installing libsdl_mixer also installs gettext

Mar 11 20:21:48 Kleiman-ibook ReportCrash[60141]: thread_get_state() returned 268435459: (ipc/send) invalid destination port
Mar 11 20:21:48 Kleiman-ibook ReportCrash[60141]: Failed to create VMUSymbolicator for mach_kernel[0]
Mar 11 20:21:48 Kleiman-ibook ReportCrash[60141]: Formulating crash report for process mach_kernel[0]
Mar 11 20:21:48 Kleiman-ibook ReportCrash[60141]: Saved crashreport to /Library/Logs/CrashReporter/mach_kernel_2009-03-11-202148_Kleiman-ibook.crash using uid: 0 gid: 0, euid: 0 egid: 0

Yes, that showed up in the console log -- while compiling smpeg for libsdl_mixer.

"libjpeg" does not exist in macports ; the name is "jpeg".
"libpng3" does not exist in macports. A look into the fink database shows that "libpng3 1.2.29-1" has a web page at http://www.libpng.org/pub/png/libpng.html; checking there shows that it is "libpng" (no 3) and lists 1.2.35 as the current version. That version is fetched from port as libpng

Kleiman-ibook:macports michael$ date ; sudo port install libsdl_ttf +universal libsdl_mixer +universal jpeg +universal libpng +universal ; date

Woot! A complete recompile of everything for universal still triggers this:
Mar 12 00:10:41 Kleiman-ibook ReportCrash[26792]: thread_get_state() returned 268435459: (ipc/send) invalid destination port
Mar 12 00:10:42 Kleiman-ibook ReportCrash[26792]: Failed to create VMUSymbolicator for mach_kernel[0]
Mar 12 00:10:42 Kleiman-ibook ReportCrash[26792]: Formulating crash report for process mach_kernel[0]
Mar 12 00:10:42 Kleiman-ibook ReportCrash[26792]: Saved crashreport to /Library/Logs/CrashReporter/mach_kernel_2009-03-12-001041_Kleiman-ibook.crash using uid: 0 gid: 0, euid: 0 egid: 0

It is repeatable! Congrats, you found a real kernel bug!

Kleiman-ibook:development michael$ mkdir -p ufoai/trunk
Kleiman-ibook:development michael$ cd ufoai/trunk
Kleiman-ibook:trunk michael$ svn co https://ufoai.svn.sourceforge.net/svnroot/ufoai/ufoai/trunk .
Error validating server certificate for 'https://ufoai.svn.sourceforge.net:443':
 - The certificate is not issued by a trusted authority. Use the
   fingerprint to validate the certificate manually!
Certificate information:
 - Hostname: *.svn.sourceforge.net
 - Valid: from Wed, 12 Nov 2008 04:25:27 GMT until Tue, 12 Jan 2010 04:25:27 GMT
 - Issuer: Equifax Secure Certificate Authority, Equifax, US
 - Fingerprint: 04:b2:70:e9:ba:cf:70:fc:e8:8a:22:86:14:13:51:97:1b:6a:de:38
(R)eject, accept (t)emporarily or accept (p)ermanently?

Ok, so no one has even mentioned what fingerprint we should see.
Come on folks, HTTPS: is only half of the security solution, and is almost useless by itself.
It prevents a stream from being corrupted during transmission, but does nothing to protect against a man-in-the-middle.
Without something to verify the certificate -- either a trusted authority, *or knowing what the fingerprint should be* -- it doesn't do what everyone thinks it does.

Gaah. Even firefox now rejects "self signed" certificates with that big bad police officer.

svn: PROPFIND request failed on '/svnroot/ufoai/ufoai/trunk'
svn: PROPFIND of '/svnroot/ufoai/ufoai/trunk': Could not read status line: connection was closed by server. (https://ufoai.svn.sourceforge.net)

Sigh. Documenting what you see as you go runs into trouble. :-)

./configure --disable-uforadiant  --enable-universal
make && make lang

There are a LOT of warnings on this make:

 * [UFO] src/ports/unix/unix_main.c
 * [UFO] src/ports/macosx/osx_main.m
src/ports/macosx/osx_main.m:107: warning: cannot find interface declaration for ‘SDLApplication’
src/ports/macosx/osx_main.m:107: warning: cannot find interface declaration for ‘SDLApplication’

 * [DED] src/ports/macosx/osx_main.m
src/ports/macosx/osx_main.m:107: warning: cannot find interface declaration for ‘SDLApplication’
src/ports/macosx/osx_main.m:107: warning: cannot find interface declaration for ‘SDLApplication’
 * [DED] ... linking -rdynamic (-lz -lcurl -lssl -lcrypto -lz -lcurses  )
 * [MAP] src/tools/ufo2map/ufo2map.c
In file included from /opt/local/include/jpeglib.h:28,
                 from src/tools/ufo2map/common/imagelib.h:31,
                 from src/tools/ufo2map/lighting.h:28,
                 from src/tools/ufo2map/ufo2map.c:33:
/opt/local/include/jconfig.h:12:1: warning: "HAVE_STDLIB_H" redefined
In file included from src/tools/ufo2map/common/../../../shared/shared.h:30,
                 from src/tools/ufo2map/common/shared.h:32,
                 from src/tools/ufo2map/lighting.h:25,
                 from src/tools/ufo2map/ufo2map.c:33:
src/tools/ufo2map/common/../../../shared/../../config.h:71:1: warning: this is the location of the previous definition
In file included from /opt/local/include/jpeglib.h:28,
                 from src/tools/ufo2map/common/imagelib.h:31,
                 from src/tools/ufo2map/lighting.h:28,
                 from src/tools/ufo2map/ufo2map.c:33:
/opt/local/include/jconfig.h:12:1: warning: "HAVE_STDLIB_H" redefined
In file included from src/tools/ufo2map/common/../../../shared/shared.h:30,
                 from src/tools/ufo2map/common/shared.h:32,
                 from src/tools/ufo2map/lighting.h:25,
                 from src/tools/ufo2map/ufo2map.c:33:
src/tools/ufo2map/common/../../../shared/../../config.h:71:1: warning: this is the location of the previous definition

These errors continue for almost all of [ MAP ] . Odd, I didn't tell it to "make maps".

Kleiman-ibook:trunk michael$  contrib/scripts/map-get
Usage: map-get [options] command

Commands:
 upgrade - make sure all maps are up to date
 generate DST - generate or update repository dir DST
                (probably not what you want)

Options:
  -h, --help     show this help message and exit
  -v, --verbose 
Kleiman-ibook:trunk michael$  contrib/scripts/map-get upgrade
getting list of available maps
WARNING ufo2map version mismatch                   
Continue? [Y|n]n

Alright, this is a flaw. The "How to compile" page says to use this script, but it doesn't work!


Time:
Macports took 3 hours to download, compile and install the needed libraries.
Svn took 2 hours and 20 minutes just to check out a copy of the Ufo AI tree

Dealing with Time Machine blowing up (backup too large for the backup drive; having to cut both the newly installed stuff AND enough of the old stuff that a future backup can backup the new stuff): Days :-). (Small backup drive.)

Final disk space:
/dev/disk0s3    42745372 41292664   1196708    98%    /

Starting:
/dev/disk0s3    42745372 36480732   6008640    86%    /

That's just about 5 gig. Wow.


===
The second round:

Fixing the maps (thank you geever): It compiles.

I have a .dmg. And I can run the program.

Attempting to run a skirmish, results in this in the console:

Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: R_LoadProgram: 'default' loaded.
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: R_LoadProgram: 'warp' loaded.
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: SDL_ttf version 2.0.9 - we need at least 2.0.7
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: ...registering 10 fonts
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Could not find cvar 'cl_3dmap'
Mar 14 13:04:34: --- last message repeated 3 times ---
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Could not find cvar 'g_actorspeed'
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Change gametype to 'Skirmish mode'
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Could not find rank 'rifleman'
Mar 14 13:04:34: --- last message repeated 7 times ---
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: ------- Loading game.dylib -------
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: LoadLibrary (./base/game.dylib)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: ==== InitGame ====
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: SV_ParseAssembly: warning - cvar 'rm_drop' value doesn't seam to be a valid tile id '' - set to default '+drop_firebird'
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: CM_LoadMap: "-africa/af_ +main +drop_firebird" "-40 -16 0 -40 -16 0"
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Map:africa/af_main  Offset:(-40, -16, 0)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: wpMins:(128, 128, 0) wpMaxs:(215, 159, 4)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Shifted wpMins:(88, 112, 0) wpMaxs:(175, 143, 4)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Tile bounds: (88, 112, 0) to (175, 143, 4)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Source bounds: (128, 128, 0) to (215, 159, 4)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Done copying data.
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Loaded routing for tile africa/af_main in   0.0s
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Map:africa/af_drop_firebird  Offset:(-40, -16, 0)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: wpMins:(128, 128, 0) wpMaxs:(143, 151, 3)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Shifted wpMins:(88, 112, 0) wpMaxs:(103, 135, 3)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Tile bounds: (88, 112, 0) to (103, 135, 3)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Source bounds: (128, 128, 0) to (143, 151, 3)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Done copying data.
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Loaded routing for tile africa/af_drop_firebird in   1.0s
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: checksum for the map '+africa': 2906337093
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: ufo script checksum 285034083
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Created AI player (team 0)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Created AI player (team 7)
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: -------------------------------------
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Connecting to localhost...
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Map: +africa
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: load material file: 'materials/africa.mat'
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Starting the game...
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: michael has joined team 0
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: (player 0) It's team 1's round
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: michael has taken control over team 1.
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Didn't find parent menu "main"
Mar 14 13:04:01 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: music change to PsymongO4 (from van_theme)
Mar 14 13:04:02 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: 0   ufo                                 0x00198d04 Sys_Mkdir + 180
Mar 14 13:04:03 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Error: Received signal 6.
Mar 14 13:04:03 Kleiman-ibook [0x0-0x2e12e1].net.sourceforge.UFOAI[12328]: Wrote keys.cfg
Mar 14 13:04:03 Kleiman-ibook com.apple.launchd[112] ([0x0-0x2e12e1].net.sourceforge.UFOAI[12328]): Exited with exit code: 1


Alright: Attempting to start a skirmish resulted in the game crashing. It detects and traps a signal 6. Had that signal actually arrived, Mac Os would have given me a nice, easy to submit debugging trace; as it is, I can only report this. And the truth is, I don't know if it's that the africa map from map-get is not compatible with the mac compiled code, or if it's something serious.

Alright, lets go to a campaign game. Playing around with options, ufopedia, etc

Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Campaign data loaded - size 4131324 bytes
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...techs: 192
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...buildings: 22
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...ranks: 16
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...nations: 8
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...cities: 7
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Sanity check for script data
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...... no mappart for building 'building_aliencontainment2' given
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...... no mappart for building 'building_hangar2' given
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...... no mappart for building 'building_large_ufo_hangar2' given
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...... no mappart for building 'building_workshop2' given
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...buildings failed
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...tech ok
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...aircraft ok
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: ...items ok

...

Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/base_AA51_launcher
Mar 14 14:26:47: --- last message repeated 24 times ---
Mar 14 14:25:33 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/base_ammo_AA51
Mar 14 14:26:48: --- last message repeated 14 times ---

...

Mar 14 14:26:12 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/craft_ammo_shiva_1
Mar 14 14:26:51: --- last message repeated 373 times ---
Mar 14 14:26:12 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/craft_ammo_shiva_2
Mar 14 14:26:51: --- last message repeated 23 times ---
Mar 14 14:26:12 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/craft_ammo_s
Mar 14 14:26:21 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: hiva_2
Mar 14 14:26:21 Kleiman-ibook [0x0-0x2e52e5].net.sourceforge.UFOAI[12423]: Can't find pic: techs/craft_ammo_shiva_2
Mar 14 14:26:51: --- last message repeated 398 times ---

The shiva pics complaints (1 and 2) show up thousands of times in the logs, in batches of 400.

I never advance the clock past the starting moment. I'm just looking around in the in-game docs, moving the map, playing around with flat vs. 3d world view, building new places in the base, equiping the aircraft, etc. I'm expecting that the first ufo encounter will crash when the map loads.

After playing around, (about an hour), the world map gets messed up. I'm no longer seeing continents, land and water, etc. Instead, colored rectangles -- darn, I should have taken a screen shot.

I'm not sure how much of these problems are from code issues, and how much is from compilation issues.

Kleiman-ibook:trunk michael$ svn info
Path: .
URL: https://ufoai.svn.sourceforge.net/svnroot/ufoai/ufoai/trunk
Repository Root: https://ufoai.svn.sourceforge.net/svnroot/ufoai
Repository UUID: 39b98707-f80d-0410-bc7e-c6b02f5409da
Revision: 23524
Node Kind: directory
Schedule: normal
Last Changed Author: bayo-fr
Last Changed Rev: 23524
Last Changed Date: 2009-03-14 09:19:22 -0700 (Sat, 14 Mar 2009)


Ohh,here's something:

Kleiman-ibook:trunk michael$ ls ~/.ufoai/2.3-dev/base/
total 372
  0 Users/        4 keys.cfg      364 ufoconsole.log
  4 config.cfg        0 save/
  0 history        0 stats.log
Kleiman-ibook:trunk michael$

Yep, that console.log file is a duplicate of what got written to the system.log.

302

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 10:53:47 pm »

I don't know that doc. Maybe it should be modified not to mention any version numbers as they will change by time..

It is http://ufoai.ninex.info/wiki/index.php/Compile_for_Mac

303

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 06:05:25 pm »

Please note that 2.2 maps are not compatible with 2.3.
You can use map-get. For "different ufo2map version" you can say continue in most cases.
The best is compiling them ofc. but that still takes much time.

-geever

Be aware that the "Compile for Mac" page on the wiki pointed me at those 2.2 maps.
You might want to adjust that.

Also: I thought you said that the mac compiler did not generate compatible maps, and you had to use the precompiled maps.

304

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 06:01:10 pm »

Thank you, adjusting compile script now ...

305

Mac / Request: A "make clean" option in the installers

« on: March 14, 2009, 07:40:55 am »

I'd like to request a "make clean" option in the installers.

In the other thread, I described what happened when I tried to compile the 2.3 developer version on the mac. Due to an error, one step -- "make lang" -- somehow got missed, and then would not clearly get put back into the final app package. Manual attempts to "make copydata" did not clearly recopy the data. Checking timestamps did not show that any attempt to force a fresh recopy of the data were successful.

Now, normally I'd just be certain by blowing the .app and -universal directories away, and rebuild them. But there is no "clean" target. And looking at the Makefile shows this:

bundle_dirs:
        @mkdir -p UFOAI.app/base
# These are part of the svn repository as directories
#       @mkdir -p UFOAI.app/Contents/MacOS
#       @mkdir -p UFOAI.app/Contents/Frameworks
#       @mkdir -p UFOAI.app/Contents/Libraries

So, while ideally I would just blow away UFOAI.app, and rebuild it, I find that part of it is "stuck" -- it came with the original download, and is assumed to be there. So there is no easy way to force the .app to be rebuilt from scratch.

In fact, looking over the makefile setup, it seems (please correct me if I'm wrong), that the system tries to first make the .app, and then it creates ufoai-2.3-dev-macosx-$TARGET (ppc, x86, or universal) from that. But if part of the build was made as one type, and the final "assembly" is made as another type, that doesn't necessarily mean it will match up. In that case, cleaning everything out and starting over makes sense.

But right now I can't see any way to do that "clean and start over" without going all the way back to a fresh download of the whole thing. It took me longer to download the code than to compile the code.

306

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 05:16:53 am »

Foo.

Kleiman-ibook:installer michael$ cd src/ports/macosx/installer/
Kleiman-ibook:installer michael$ make TARGET_CPU=universal
cp -v ../../../../base/0pics.pk3 UFOAI.app/base/0pics.pk3
../../../../base/0pics.pk3 -> UFOAI.app/base/0pics.pk3
cp -v ../../../../base/0models.pk3 UFOAI.app/base/0models.pk3
../../../../base/0models.pk3 -> UFOAI.app/base/0models.pk3
cp -v ../../../../base/0snd.pk3 UFOAI.app/base/0snd.pk3
../../../../base/0snd.pk3 -> UFOAI.app/base/0snd.pk3
cp -v ../../../../base/0music.pk3 UFOAI.app/base/0music.pk3
../../../../base/0music.pk3 -> UFOAI.app/base/0music.pk3
cp -v ../../../../base/0maps.pk3 UFOAI.app/base/0maps.pk3
../../../../base/0maps.pk3 -> UFOAI.app/base/0maps.pk3
cp -v ../../../../base/0media.pk3 UFOAI.app/base/0media.pk3
../../../../base/0media.pk3 -> UFOAI.app/base/0media.pk3
cp -v ../../../../base/0ufos.pk3 UFOAI.app/base/0ufos.pk3
../../../../base/0ufos.pk3 -> UFOAI.app/base/0ufos.pk3
cp -v ../../../../base/0materials.pk3 UFOAI.app/base/0materials.pk3
../../../../base/0materials.pk3 -> UFOAI.app/base/0materials.pk3
cp -v ../../../../base/0base.pk3 UFOAI.app/base/0base.pk3
../../../../base/0base.pk3 -> UFOAI.app/base/0base.pk3
cp: ../../../../base/i18n/[^.]*: No such file or directory
make: *** [copydata] Error 1

I don't think I can solve that one. Did I "make lang"? Whoops.

But worse: After doing "make lang", it should have redone that language copy, right?

440 translated messages.
msgfmt -v -o radiant/i18n/ru/LC_MESSAGES/uforadiant.mo src/po/uforadiant-ru.po
74 translated messages, 13 fuzzy translations, 367 untranslated messages.
Kleiman-ibook:trunk michael$ cd src/ports/macosx/installer/
Kleiman-ibook:installer michael$ make TARGET_CPU=universal
Installing UFOAI.app/Contents/Libraries/libSDL-1.2.0.dylib...done
Installing UFOAI.app/Contents/Libraries/libpng12.0.dylib...done
Installing UFOAI.app/Contents/Libraries/libjpeg.62.dylib...done
...

Nope! So now I've got an incomplete .app package that I have to blow away and start over. Sheesh.

307

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 05:02:37 am »

I have a workaround, but I do not know if it is correct:

curl http://voxel.dl.sourceforge.net/sourceforge/ufoai/ufoai-2.2-maps.zip > ufoai-2.2-maps.zip
unzip ufoai-2.2-maps.zip
cp 0maps.pk3 base
unzip -d base -u base/0maps.pk3
make pk3

That "-u" option on unzip -- it tells unzip that any map found on the disk should be left alone. I have no idea if that is correct or not -- supposedly, the mac cannot generate working maps, yet it does.

In any event, it's letting me move forward. No clue if the result will be correct or not. Developer help please?

308

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 04:25:54 am »

Ok, this is a show stopper. I need help at this point.

curl http://voxel.dl.sourceforge.net/sourceforge/ufoai/ufoai-2.2-maps.zip > ufoai-2.2-maps.zip
unzip ufoai-2.2-maps.zip
mv 0maps.pk3 base
make pk3

Kleiman-ibook:trunk michael$ ls base/0maps.pk3
20 base/0maps.pk3

Only 20?

cd base; zip -u9  0maps.pk3 maps/africa.ump maps/alienbase.ump maps/bomber_city.ump maps/bridge.ump maps/cemetery.ump maps/city_disco.ump maps/community_centre.ump maps/construction.ump maps/country.ump maps/druglord.ump maps/eaglenest.ump maps/farm.ump maps/forest.ump maps/frozen.ump maps/gasstation.ump maps/ice.ump maps/industrial.ump maps/italy.ump maps/japan.ump maps/laboratory.ump maps/mart.ump maps/military_convoy.ump maps/oriental.ump maps/shelter.ump maps/spedition.ump maps/stadium.ump maps/transport.ump maps/tropic.ump maps/ufocrash.ump maps/village.ump
  adding: maps/africa.ump (deflated 79%)
  adding: maps/alienbase.ump (deflated 68%)
...

Ok, first comment: If I downloaded the maps, why does it need to add more?

zip -d base/0maps.pk3 maps/office06d.bsp maps/stadium06d.bsp ... (very long list of files)
deleting: maps/office06d.bsp
deleting: maps/stadium06d.bsp
deleting: maps/pdi08d.bsp
...

I can do nothing at this point. Help me please.

309

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 04:07:14 am »

Aha! After unzipping, the file goes into base/. Continuing ...

310

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 04:00:22 am »

May I suggest not storing this file in .zip form?

Kleiman-ibook:trunk michael$ ls -l *maps*
 72160 -rw-r--r--  1 michael  michael  73888236 Jan 11  2008 0maps.pk3
 71324 -rw-rw-r--  1 michael  michael  73032542 Mar 13 19:22 ufoai-2.2-maps.zip

A savings of 800K, on a 7 MB file, maybe that should just be stored as a .pk3 file?

311

Mac / Re: Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 03:54:03 am »

Trying to figure out what the "make pk3" wants, hoping that will tell me what's up with the maps

Kleiman-ibook:trunk michael$ make -n pk3
make -f build/data.mk pk3
zipinfo:  cannot find or open base/0pics.pk3, base/0pics.pk3.zip or base/0pics.pk3.ZIP.
cd base; zip -u9  0pics.pk3 pics/airfight/forest1.jpg ...

zip -d base/0pics.pk3
zipinfo:  cannot find or open base/0models.pk3, base/0models.pk3.zip or base/0models.pk3.ZIP.
cd base; zip -u9  0models.pk3 models/aircraft/craft_arm_plate/plating.md2 ...

It's not doing nothing (the -n flag) -- it's still trying to assemble stuff.

312

Mac / Trouble getting 2.3 development to compile -- What do I do with the maps?

« on: March 14, 2009, 03:29:19 am »

I'm trying to get 2.3 development to build on a mac.

The contrib/scripts/map-get doesn't work

Kleiman-ibook:trunk michael$  contrib/scripts/map-get
Usage: map-get [options] command

Commands:
 upgrade - make sure all maps are up to date
 generate DST - generate or update repository dir DST
                (probably not what you want)

Options:
  -h, --help     show this help message and exit
  -v, --verbose 
Kleiman-ibook:trunk michael$  contrib/scripts/map-get upgrade
getting list of available maps
WARNING ufo2map version mismatch                   
Continue? [Y|n]n

Alright, this is a flaw. The "How to compile" page says to use this script, but it doesn't work!
Manually fetching ...

curl http://voxel.dl.sourceforge.net/sourceforge/ufoai/ufoai-2.2-maps.zip > ufoai-2.2-maps.zip

*** BUT NOTHING TELLS ME WHERE TO PUT THIS ***

What do I do with this .zip file after I fetch it?
I am stuck. I may have a nice "simple" "script" for getting this all compiling on a virgin mac, but I have no clue what to do with the maps.

313

Discussion / Re: Wormhole travel vs. FTL engines

« on: March 11, 2009, 08:00:47 pm »

Light does vary in speed by a lot. I think the normal measured variation is 0.00000000000001 percent c.

Wow. I was not aware of that; thank you.

Your opinion of what is odd is obviously different to mine. Psychology isn't a subject I know much about (though I have my opinions) but from what I can see the reason you think of these as odd is because of the laws you formulated for yourself while growing up.

Fair enough; my personal idea of "odd" is different than probably 95% of the population.

When I call something "odd" in this respect I'm referring to what most (>66%) of the population regard as "odd" based on normal activities and observations in every day life.

For about 20 years now, we've grown up with G.R. as a given, with adults having learned enough of it in this country that they understand it does strange things in strange situations. In other countries that level of learning isn't there, and the concept of "odd" differs.

... but I find you correcting me patronizing since I wasn't referring to them as they are completely out of context for this discussion!

Fair enough. I was not trying to be patronizing, and I apologize for it.

Mass can't reach c. Mass and information can't exceed c.

Right. One of the big surprises for me in physics was that light can travel faster than C, and relatively easily. Apparently (if I remember correctly), in a wave guide you can get the light waves to twice the speed of C if the information carried by those waves drops to half C; apparently the actual constant is C squared, not C itself.

The point was that while you cannot exceed C while staying in normal space, nothing prevents you from doing odd things with space. Warping space was one of the things that I had read of; you indicated that it turned out not to work. Massive spinning objects (hello rotating super massive black holes) was another; you just indicated that the power consumption is ZPM-level (sigh, no stargates.) Wormholes are an often discussed idea, but what I've read indicates that the current belief is that they are probably either ultra-rare, unstable, self-collapsing, or just plain "untravel-able" by anything bigger than a photon (so at best they can transmit information, not matter).

So the only thing left may be an M-space bridge, but it would have to be portable and able to run both ways.

Please stop using the term "assume." Theory is a preferred word.

Now we're into nitpicking. "Assumptions", to me, are what you place at the start of any theory. Given those assumptions, the theory states what is predicted.

S.R. and G.R. each made two assumptions; then, based on those assumptions being true, some things come out to be true. The assumptions of S.R. are untestable -- you cannot get a true inertial reference frame, but it turns out that the predictions of S.R. match local observations for non-accelerating reference frames. G.R., on the other hand, DOES match predictions to observations, so the presumption is that the two assumptions made by G.R. are in fact true.

(as a side note: S.R. is not the only theory that manages to match observation and prediction. Lorenz, if I recall correctly, came up with a different set of assumptions that also match, and turns out to be 100% equivalent to S.R., but has very different assumptions about why and what's happening than S.R. does. For example, does an object only appear to change size with speed, because the observer cannot see identical instants at different locations, or is there in fact a true reference frame, and objects really do change in size, but since your rulers also change, you cannot tell.)

That is what the term "assume"/"assumption" means to me. If it has a different meaning to you, then lets agree to disagree.

314

Discussion / Re: Wormhole travel vs. FTL engines

« on: March 11, 2009, 04:03:13 am »

I wasn't going to post again since I would probably just be repeating myself, but I felt the need to correct several points.

but one of Einstein's assumptions was that the speed of light was constant from (almost any) reference frame.

Actually Einstein theorized that the speed of light through a vacuum was an average. Some photons travel slower then c and some travel faster but dividing the photon count by the net speed will always result in 186,282.4 miles per second (rounded up).

This doesn't agree with anything that I learned in special relativity, and I can't say that I know G.R. inside and out.

What I do know is that S.R. makes two assumptions:
1. The speed of light in a vacuum is constant in any inertial reference frame
2. The laws of physics are constant in any inertial reference frame

No average; no "some photons go faster and some slower". Absolute constants.

G.R. relaxes the constraints from inertial reference frame (no acceleration) to "Gaussian".

That, along with the assumption that the laws of physics are the same churns out a bunch of odd results that match experimentations.

Some ideas are hard to picture but the vast majority of experiment results make complete sense in physics. Try to imagine the problem in an abstract way, without applying assumptions that we formulate in everyday life.

You get such odd results as "F = GMm/r^2" being hard to even state. Mass as seen by who? Distance as seen by who? Different observers will see different speeds, and different distances. It isn't even clear that this law can be stated under G.R.

You get such odd results that we now take as a given -- for example, if two people are walking towards each other, they have different concepts of "now" across the universe. You might even find that many many light years away, a planet blows up, and to one person that event is in the past, while to the other person it is in the future.

You get the "Light cone", which has "Elsewhere, but cannot say past or future"; you get "Past" (or "future"), but cannot say if here or not here, etc.

You get the whole "No, our satellites do not have the same clocks as the GPS receivers that listen to them".

Heck, you get "Sorry, the top floor's clocks aren't synched with the ground floor's clocks".

You get "Sure, if you're on a train, and you throw a baseball, the baseball goes as fast as both the pitch and the train. As long as the train isn't really really fast".

We grew up with these oddities, but they are oddities.

"Cannot travel faster than light" is NOT one of the conclusions.
"Cannot accelerate up to the speed of light" is.

Space itself can, and does, "move" faster than light -- see the superluminal expansion of space itself.

Woah. Stop right there. Cannot travel faster then light IS a conclusion of special relativity. If you can't accelerate to c how would you get faster then c?

An object moving slower than light cannot accelerate faster than light.
An object moving faster than light cannot decelerate below light speed.

An object moving at the speed of light itself has an oddity. Light moving in the same direction (parallel to itself) moves at the same speed as light heading in the opposite direction. The only way this can happen is if photons takes zero time to move from emitter to absorber.

The idea that you can move a bubble of space around faster than light, while nothing in that bubble has to move "fast" (don't ask relative to what) is both theoretically possible under G.R., and under (or at least was) research.

Ha ha! Good ol' warp drive eh? In theory it is possible. Manipulate space-time to contract ahead of an object and expand behind it. The travel time / distance would resolve as a speed higher then c while local speed never approached even relativistic velocities.

But our mate special relativity still plays a part. An observer outside the bubble would see the object going at a lower speed then c and the object would see the universe traveling at a speed of less then c yet the D/T= >c so it's a paradox right? Well no it isn't, but it's a little hard to explain so I'll give an analogy. If two people, named A and B could see each other and started walking in opposite directions, A sees B getting smaller and B sees A getting smaller. An apparent paradox that makes perfect sense in everyday situations.

In practice this warp is impossible though. The "bow-wave" would have to be created with very dense matter or energy and this matter or energy cannot exceed c so the bubble behind it is limited to c as well, meaning it isn't FTL travel.

So this idea has been disproven? Thank you; I was not aware of that.

The other theory for this uses S-T disturbing objects, such as a quickly rotating singularity to distort space time to propel the object. This would require a network of these S-T disturbing objects in the same fashion as a train requires track. The track would be placed by a slower then c ship meaning the initial journey would be at normal speeds. Oh, and each track object would need it's own star for power.

So you need to build your road network, and once built, travel within your empire along the roads is fast. That doesn't sound any different than existing historical systems.

Note that it still permits FTL travel. Just not FTL expansion.

315

Discussion / Re: anyone else playing a 2.3 development build?

« on: March 09, 2009, 04:49:11 am »

Fair enough. GCC had that ".9" numbering on a not-quite-ready-but-not-really-compatible release recently.

So this is 2.3.-1 then, right? :-)