historic_bruno

Do you mean to say it worked, or didn't work, or it's in progress? For the wxWidgets/iconv error, it may be helpful to see what wxWidgets' configure scripts are doing. This is logged in config.log in the build directory, "libraries/osx/wxwidgets/wxWidgets-2.9.4/build-release/config.log". Any version of OS X to this point (including 10.8) should have libiconv available in /usr/include/iconv.h and /usr/lib/libiconv.dylib, but maybe MacPorts or something is confusing it. I have a possible fix that involves using --with-libiconv-prefix=/usr to override whatever configure is finding.

I think we've been making a mistake not to let people translate the game over the past years. Not all the text verbatim, but some words and phrases we know we'll need. What are the odds that we'll find all these people again when we actually need and want translations? Better to put a system in place now, and let people work on it gradually over time. Depending on how much effort that would require and how easily it could be maintained, of course.

Update: I have a fresh Mountain Lion (10.8.2) test system now and the build process works fine, so I guess it's MacPorts/Homebrew that is causing these issues. I will try to see what can be done about that (other than uninstalling MacPorts).

Oh right, if I recall correctly trees are a special case of selection boxes (they tend to have very large branch/leave polys that are annoyingly easy to select) so it doesn't change with animation. In fact there's a ticket that describes this exact problem and the reason it exists: http://trac.wildfiregames.com/ticket/1032

By bounding box, you mean the one used for pathfinding obstructions? That can't be model/actor dependent since it affects the simulation. It might be possible to work around that by having multiple bounding boxes that are somehow synced with animations, but it could still look bad if the user had locally modified actors. The selection box should conform to the animated model and that's ok because it's not part of the simulation.

I don't see a problem with it, but I think currently effort would be better spent updating our own wiki http://trac.wildfiregames.com/wiki

Ok, that looks like the same error as wraitii, can you paste the output of ./libraries/osx/wxwidgets/bin/wx-config --libs ?

Hi, if you haven't seen it yet, I'd start with our introduction for programmers on Trac: http://trac.wildfiregames.com/wiki/GettingStartedProgrammers That should at least point you in the direction of some interesting links. The other thing I would do is spend time in #0ad-dev on IRC as it suggests. The activity level varies but if you have questions and stick around for a while, you will likely be answered. Try getting the game's source code from SVN and compiling it, then play around for a while to get a feel for how it works. At that point you might find something interesting to work on, or at least understand the context of any ticket you want to work on.

Strange, you're on Mountain Lion? I wonder if Homebrew/MacPorts could be interferring somehow, assuming you have those installed. Can you try passing -v to clang and verify that it links libiconv for wxWidgets, and also that the SDK paths are setup properly? Maybe post the full verbose build output. I'm on Lion using the latest Xcode release, I see libiconv in both the 10.7 and 10.8 SDKs, so it should work.

I agree with this, it wouldn't be too hard to change the implementation, it could use the OnHealthChanged message handler to adjust the preview building's height. So I don't think worrying about that should be the biggest concern, but rather what makes the game most playable and intuitive

The animation system should support interpolation between the last frame of one animation and the first of another, along with a specified transition time. That is for cases where a custom transition animation isn't needed, for anything more complex (an animation state machine), I would like to see the logic moved out of UnitAI if posible.

Good, people ignoring them is why they never get fixed

Indeed, I forgot to mention that Luckily Cmake has an OS X package available: http://www.cmake.org...s/software.html

Very nice, I have to say, the Mauryans will have some of the best looking units in the game...

This Premake fork seems to have some support for schemes: https://bitbucket.org/liamdevine/premake-dev-xcode4/src/65e4fb2ace8aebce03368bbe8986e48a8043198c/src/actions/xcode?at=default

osx-build branch has been merged into SVN! http://trac.wildfire...changeset/13148 (CMake is still a required dependency to build NVTT, get the OS X package here) If you're on OS X and want to test it, my advice is: delete all *.dylib and *.a in binaries/system run ./clean-workspaces.sh in build/workspaces run ./build-osx-libs.sh in libraries/osx -- it will download and build a lot of libraries from source run ./update-workspaces.sh in build/workspaces do a clean build ("make clean && make" in gcc workspace, if using gcc, or Product > Clean in Xcode) (I only include the cleaning steps to be safe since the build system has changed, those aren't typically needed) That will build a loose binary. The process if you want to build a bundle takes longer but is simpler: delete all *.dylib and *.a in binaries/system run ./build-osx-bundle.sh in build/workspaces -- this will take a while... but it does everything else

--disable-collada isn't a valid option. If you want to skip the FCollada build, you should probably just remove/comment the line in update-workspaces.sh where --collada is added to the Premake arguments and the line that builds FCollada. I don't know why it's failing, but it's not required if you use archive build to create a public.zip, which is part of the experimental Android Port build instructions.

Adding to the wishlist: Catch JS exceptions using the debugger (if possible) Allow something like assert() to optionally break into the debugger when a condition is met. I believe it was quantumstate who proposed a JS assert(). In release builds it obviously wouldn't break into the debugger, but log a message instead.

Any chance you could create a Github repo or something for the script(s) you're using? Or post them here. Even if a work in progress, it would be interesting to see them and people would probably make suggestions or find bugs.

Here's a solution I've found as a workaround, it involves manually creating the proper schemes: Open the Xcode 4 workspace. In the top left corner of the Xcode window, you'll see a Scheme selector. Click that and choose "Manage Schemes..." All the schemes that are listed are autocreated by Xcode and most of them aren't useful, they can be deleted or hidden (by unchecking "Show"). Create a new scheme with the (+) button, name it something like "pyrogenesis (release)" and set the target to "pyrogenesis". Double click the new scheme to open its settings. Select Build in the left column. Add (+) two new targets: AtlasUI and Collada. In the left column, choose Run and set the build configuration to "Release". Do the same for Test and Analyze. Now copy that scheme using the Duplicate Scheme button, naming it something like "pyrogenesis (debug)". For each of Run, Test, Profile, Analyze and Archive in the left column, set the build configuration to "Debug". There should now be two schemes pyrogenesis (release) and pyrogenesis (debug) that have the correct build configuration. You can also create another scheme for the test target, one for Atlas, etc. The schemes only specify build/test/profile/etc. configurations, so you can delete them or create as many as you want, multiple users can have their own different schemes or they can be shared. You could create a scheme that runs Atlas, or passes autostart options to test a given map, etc. It should be possible to generate release and debug schemes with Premake, they are XML files that get stored in pyrogenesis.xcworkspace/xcshareddata/xcschemes/*.xcscheme

While we're nitpicking The call stack and values list boxes should be cleared if we're not in the debugger, as that data is no longer valid It should detect when the connection to the server is lost and reset the UI Hotkeys would definitely be nice Would it be possible to arrange the files in a hierarchical/tree list? I would also like to have some settings: host IP and SetSettingSimultaneousThreadBreak This may be wishful thinking and very hard to implement, but how about hovering tooltips to show a variable's value? I like the find (Ctrl+F) feature in the source view, too bad it also opens Firefox's find box :/ Requests for the debugger itself: Local variables, if possible, would be quite helpful Include variable types/classes How about a "BreakAll" command, that breaks into all script interfaces? And "Break id" to break into a single script interface? ClearBreakpoints to clear all existing breakpoints

See http://trac.wildfiregames.com/attachment/ticket/1682/license-updates.patch

Yeah, that's because Premake is not properly using schemes There should be two schemes: release and debug (matching the Premake concept of configurations), but instead each project gets turned into a scheme by Xcode and debug is the default build configuration for some reason. I don't know if we can have separate projects combined under two schemes, in a single workspace, anyone with Xcode 4 experience know about that?

Impressive! MIT license would also work, a lot of the scripts in source/tools are licensed under MIT. The Ace scripts appear to be BSD, so I don't see any problems assuming mmayfield45 chooses a free software license for their own code

Heh, that's pretty cool, did you enable user feedback reporting for both of them? It might be nice to have data from these rarely used devices