• Ideally JS variables should be attachable to the Player object (via localPlayer?) to declare and set any properties affiliated with the current player. (eg player.resource.food; modder could alter it to add/remove any resources he wishes.)

• Civilisation: Celt, Greek, etc ... Chosen from session config. (--> Currently there is a concept of civ, but it's not choosable in the GUI.)

• Colour: Player colour, chosen from session config.

• Team: In session config, players can choose to be on the same team (allied at start).

• Various team settings, such as allowing teams to have "shared control" and being able to command each other's units.

• Alliance: Diplomatic settings (who's an ally, who's an enemy).

• array of players matching slot numbers, 1: enemy, 2: ally (Not choosable from GUI yet though.)

• Phase: The starting "age" for the player (defaults to village). If otherwise, the player begins with the phase upgrade already researched.

• Resource: Quantities of each of the resources gathered by the player.
  • Currently resources are tracked by the GUI; every player should have their own set of resources and the entity scripts and GUI should be updated to work with this.

• Start Location: Position on the map where the player will begin.
  • Camera should also start centered on this location.

• When a unit depletes a resource object, he should seek out another one to gather from if nearby.

• Stances
  • GUI: List stance options in GUI and allow user to select the stance for current unit.

• Units react to opponents in LOS, depending on stance (move to attack, hold ground and attack, don't attack except to retaliate). Changed to "ready" animation instead of idle.

• Units react to opponents in attack range, depending on stance (move to attack, hold ground and attack, don't attack except to retaliate).

• Equally, depending on stance's freedom of movement, capable idle units will repair or heal nearby damaged units and buildings when not otherwise engaged.

• When an entity places the foundation for a structure, he'll move over to it and aid in constructing it.

Most of the code for technologies is now done.

The modifications that technologies make still need coding. See ResourceGatherer.js for an example.

Tech pairs where only one out of two techs can be researched need implementing. An idea is to have a tech pair json file which defines the pair and have the building have that in it's template file for what is available for research. The tech pairs can then supersede each other.

• When releasing screenshots, the scenario designers would like to be able to slow down the speed of the game so they can set up a shot in "slow-mo" and have a better chance of taking an impressive-looking screenshot.

• We got around this by altering the movement speed of entities, but this was a bit of a hack since it didn't affect other actions, and units move at a crawl for any other development until changed back. So, would it be possible to alter the simulation rate while the game is running?

• Probably best to handle it as an exposed JS function with a parameter to set the game speed modifier. Then could link this to two hotkeys (and maybe also onscreen buttons with speed display) to increase/reduce the simulation speed a step at a time.

• One hotkey increases the rate of simulation by one step (slows it down, eg for easier taking of screenshots). Another hotkey decreases the rate of simulation by one step (speeds it up, eg for feature testing).

• Feature should not be available to clients, only the host (so that all players match to one speed).

• Improve to handle tree traversal and dynamic terrain.

• Integrate rmgen with engine, so random map based on session config can be generated on the fly.

• Accept parameters so that random maps can be generated based on host's settings (eg number of players, their civilisations, desired number of territories, if any, size of starting force (eg prebuilt base or "Nomad", number of starting units), etc).

• RMS Reference

• ​Map List

• Fundamental commands:

• Set terrain textures

• Define elevation

• Area definition (for territories)

• Objects in territories (resources, settlements)

• Water height (This is just height 0 in RMS, but maybe we can make it customizable)

• Support for sending chat messages to players (all, specific, alliance groups). #1583

• Taunts (including GUI)

• Making/breaking alliances

• Sending tribute

• Tributing units/territories

• hh:mm:ss clock variable from engine showing time elapsed since the session started (perhaps best handled as an event, so it's updated in real-time?). I'd suggest a separate variable for each component (eg elapsedHours, elapsedMinutes, elapsedSeconds).

• This can then be used to control various time-based occurrences in the game ... Such as moving to the next season or stage in the day/night cycle by comparing it to a duration set in session creation.

• Or maybe implement it using a ProgressTimer? set to update each game second? (Would still have to calculate the difference on the fly.)

• Initial settings and all input is recorded.

• These same inputs can be fed back into the game at a later date so that the player can view it as a replay.

• User cannot affect what happens, but can view it (alter playback speed, look at play from any players' POV, etc).

• Needs a JS Interface. Should probably start and stop recording synchronized with CGame somehow. Needs a file format (pipe network packets to a file). Needs an entry point for replay (could do something wicked with CMessagePipe and the polling framework already in place for sockets, perhaps). ​http://www.wildfiregames.com/forum/index.php?showtopic=1665

• Also would need a Game Replay user interface to handle special features (selecting control of a different player, adjusting the rate of playback, rewind, etc) as well as select a replay to load (variation of Load Game screen).

Related tickets: #3168 Lets observer change perspective #3253 Save playernames to commands.txt #3255 Update directory structure to prevent overwrites #3258 Main menu entry to start replays #3261 Session panel (pause / fast forward / rewind) #3309 Fast forwarding replays doesn't work as expected #3387 Access summary screen from replays

• Units can have a height offset relative to the terrain (eg flying units, or ones the scenario designer has decided to make float in the air, for some reason)

• XML.Entity.Traits.Anchor

• XML.Entity.Actions.Formation
• ​Formation XML
• ​Formation Mockups
• ​Formation Notes

• Allow running
• (Requires better pathfinding): attacking, and switching between formation shapes on goto.
• Redirection angle and "take closest position"
• Formations can grant units unique idle, ready, and attack animations.
• Formations will require a flag in the cell that triggers special animations for a unit in a particular location of that formation. See special formation such as these:
  • ​http://www.wildfiregames.com/~art/history/romans/pages/rnorton_testudo_01.htm
  • ​http://www.wildfiregames.com/~art/history/hellenes/pages/01_macedonian_phalanx.htm

• Remove the "decay animation". Instead, the last frame of the death animation is sustained while the corpse is lying about and while it is decaying.

• If an entity leaves a corpse, it should still decay (ie be removed from the map with a bit of eye candy) at some point to remove the entity for performance reasons (whether after a timer has elapsed, or based on some signal from the graphics engine).

• Decay: If an entity doesn't leave a corpse (eg buildings decay immediately after being destroyed) or after its corpse state has elapsed, the default will be for it to sink vertically into the terrain, ideally with also a simple fade effect, until out of sight (basically the reverse of construction). Then it'll be removed from the game.

• An upper limit on how much population limit can be increased, set at Session Creation (add to GUI), default 200.

• Ensure resource and population requirements are taken into account in determing prerequisites for entity creation.

• Units currently delay for a few seconds between reaching a target and switching to performing this action. This delay period should be reduced.

• Ranged attack projectiles should not guaranteed to hit target
• Splash damage should be implemented when the projectile hits something (target, ground, or otherwise)

• GUI: Progress bar over unit currently being created. Numbers next to units in queue.
• Sound: Units make a "unit ready" sound when creation is complete.

• Training building has a "Rally Point" property. (When the building is selected, a banner actor shows the position of its rally point.)

• Rally Point position can be set by right-clicking a terrain location with the building selected (or pressing a Rally Point UI button to get the rally cursor, then doing the same).

• Spawned units move to rally point after appearing adjacent to the building.

• Training building has a "Spawn Point" property. (The X/Z origin on the edge of the building where the entity first emerges (eg at the actor position of a door) after being created; he then walks from there to the rally point.) [Assuming it isn't satisfactory to just have units randomly appear at any side of the building. Varies from RTS to RTS.]

• Building rises out of ground as construction proceeds (#1174)
• Scaffolding not necessary but maybe as a bonus if time allows? (#1581)

• Resources can be traded for other resources at the Market from the Barter tab.
• (This is mostly an interface task, since it largely involves pressing buttons to juggle the player's resource values, without need for special events.)
• XML.Entity.Actions.Barter

• Sometimes selecting a target for an entity (eg telling a unit to gather from a tree) causes the game to freeze as a vast amount of memory is allocated to the game, and the only way out is to close the application from the task manager. Not easily replicable.

• (Uncertain whether this is coming from the pathfinder failing to plot a course to the tree, or the order being given to the entity.)

• XML.Entity.Traits.Garrison

• Visible Garrisoning (see link above) liable to be most challenging aspect of garrisoning. Some of the garrisoned entities are propped to sockets on the parent actor, and while not directly controllable, can interact with the environment (eg archers at windows or on decks of ships fire at nearby units, including units propped on other ships).

• XML.Entity.Traits.Garrison

• Maintain a "selection" of entities stored in an entity. (These do not appear in the game unless occupying prop points ... See Entity Events - Garrison - Propping.)

• Garrisoned entities can be removed (either immediately adjacent to parent, or ungarrisoned to a target using something similar to a Rally Point) and listed (each for GUI ungarrisoning/viewing).

• Actor player-coloured pennant prop appears over entity when units are garrisoned in it. (This will need X/Z location properties so artists can put it somewhere which looks vaguely appropriate from unit to unit.)

• The same principle of entities-owned-by-other-entities could also be used to imitate an entity's inventory (possessions), with objects being carried by an entity and appearing (and being droppable from) the entity's garrison UI.

• This is largely for TLA's sake (which has more RPG elements; eg they might want a unit to be able to pick up a sword, have it appear in the entity's hand socket, and increase his attack damage while he is carrying it -- this would also include having an "equipment slot" so he can only equip one melee weapon).

• However, we should keep this feature set in mind for "relics" (whether we do them in Part I or Part II), since they follow similar principles of picking up (see the Pick Up event) and carrying an item (bestowing a tech effect) and dropping (ungarrisoning) it in the world.

• (Priests and Heroes can pick up a relic and give it to a Temple's inventory, which then unlocks the relic's tech effect.)

• Traders generate resources by moving along trade routes (between player and allied Markets/Docks)
• XML.Entity.Actions.Trade

Priests and temples heal injured organic units in a radius around them.

• Citizen Soldiers repair damaged mechanical units (buildings, ships)
• XML.Entity.Actions.Repair

• XML.Entity.Actions.Graze
• (Part of animal unit AI)

• XML.Entity.Actions.TownBell

• Pressing this UI button of the Civ Centre puts the units in its province (or surrounding area, if playing a game mode where the map isn't split into provinces) into "red alert": units automatically garrison in nearby buildings, with priority on gatherers that cannot defend themselves (female citizens), and military units going for buildings that allow them to attack from the battlements (like towers).

• XML.Entity.Traits.Auras

• Trample Aura (component of Charging). Used in particular by cavalry to mow down infantry.

• Provinces Document: ​http://www.wildfiregames.com/forum/index.php?showtopic=696

• Relies on area definition (RMS), allocation of resources and settlements to Provinces.

• Ownership of areas (provinces)

• Visible borders

• Scripted behaviour for Provinces:
  • Unlocks (eg resource gathering, building)
  • Construction over NPC (gaia) settlements (see Entity Event - Construction - Sites)
  • Statistical Benefits/Disadvantages
  • Ceasefire (see Timed Events above), attrition damage (see Terrain below)
  • Entity caps (limited to a certain number of entities of certain types, particularly buildings, per province)

• Higher Ground: Units at higher elevations gain a bonus to damage in combat.

• Steepness:
  • Slopes can be too steep to traverse, though this can vary from unit to unit (foot units can climb steeper slopes than carts or siege weapons). --> Decided not to do this for now since it is expensive and would be confusing to the player; instead, just make certain terrains impassable.
  • The steepness of a slope also affects the movement rate of units (going up is slower, going down is faster).

• Additional parameter to extract folder names.

• on-demand loading = assimilate changed files without restarting game

• GUI Hotload.

• Scope/cache management

• VFS should handle newline conversion (recognize all \r\n \n \r combinations and treat them as newlines) when reading files.

• General-purpose trigger system

See ​this topic for forum discussion.

Here is the recent topic: ​http://www.wildfiregames.com/forum/index.php?showtopic=18211

• Able to set camera position at game start for each player (usually centered on player's Civ Centre if not specified ... done by RMS or baked into map?). (See Random Map Scripting above.)

• Key to lock camera on the selected unit ("follow" mode).

• Bugfix: camera shakes when unit being followed goes downhill away from camera.

• Cinematics: functions to control camera and define tracks at positions in the environment.

• Option to lock camera rotation or "snap" to saved rotation.

• Animated meshes using vertex animation (swaying trees, flags, flowing capes -- could be done with boning/skinning if absolutely necessary.

• If we could support vertex animation though, we would also be more accessable to the moding community that isn't as capable with bones and skin as developers).

• Foundation texture (eg leafy texture under trees, paved/dirt texture under buildings).

• Particle effects (smoke, flames, rubble) can be attached to sockets on a unit just like props.

• Obviously requires a working particle system.

• Start times for initial animations should be randomised. All units on the map start with their idle animations in perfect unison, for example.

• Document the needs of the revised GUI engine. Essentially do a postmortem on the current system and see where it can be improved (an initial start on that in the sections below). With any luck this'll provide a game plan to Gustav to make amends in the time before he goes to Japan, and then for his successor to take over.

• (Merge the GUI Engine Trac task into that brief, and post for feedback.)

• 01/05/06 "3 weeks to compile a wishlist for Gee, then he has 3 weeks of spare time."

• "the school is coming to an end in 3 weeks so I have to pretty much study every day until then. Although when it's over I have 3 weeks completely free before I leave for Japan."

• i18n - allow conversion from game's text into a translatable table, and then to the right format for the game to read. Possibly using GetText (​http://www.gnu.org/software/gettext/gettext.html) if it supports XML, JS, and C++?. Multiple tools for GetText translation already exist, which would simplify the translation process (not a custom format and editor). See #1389

• Includes correct handling of accented characters (many characters are replaced with (?) symbols). [The tools to do this are not quite completed/working yet.]

• l10n - perform that conversion and do the translations (into English first, and maybe another language to test the system)

• Provide a default hotkey set for each locale? See #692

Latency Adjustment (moved to #3752)

Currently the networking code is hard-coded to a 200ms turn length, with commands scheduled for execution 2 turns ahead. This causes jerky motion if the RTT for any player is >200ms.

We ought to do some combination of dynamically adjusting the turn length (which allows fine-tuning and reduces simulation CPU load, but large values will result in very variable responsiveness depending on whether the player clicks near the start or end of a turn), and dynamically adjusting the number of turns ahead (which allows more consistent responsiveness, but is a coarse adjustment).

Player disconnections

I believe if a non-host player is disconnected, it'll display a message already, saying the player left?

However, if a client does not hear from a host for ~5 seconds, it should display a "waiting for host" message to the player (with an "exit game" button on it), and then either go back to normal or go to the "connection lost" message on network timeout. (Will be done in #3264)

Auto-Reconnect

If a player is kicked, allow them to try to reconnect via a button (in the case where their (the players) internet was cut off, but the host is still running the game, and the player rejoins the same I.P. and takes control of their units).

Observer Mode

Players who can join a game, but don't have a civ to play (full or doesn't want to play), can just watch (like a real-time Replay).

They get full view of the map (no SoD or FoW). Sometimes called 'God Mode'.

• Shaders

• Simple implementation

• Shaders and materials for shiny armour and helmets, reflective, glowing and transparent objects

• Paint a small texture on the terrain such as footprints, tree roots, blood splats, burn marks, etc… This would conform to the terrain perfectly and appear to be part of the terrain. (Splats/doodads/goobers/decals.)

• Basic particle engine is in the game, provided by Will Dull, with a constant emitter.

• Needs to be picked up to allow creation of XML-based particle effects, using editor to tweak properties, and have them broadcast from locations in the game (either placed as particle objects, or attached to actors in prop sockets).

• Often particles can appear conditionally (eg smoke and flame trails on a building when it is damaged) or when a unit is splashing through water.

• Currently an entity can only be assigned one order at a time.
• The player should be able to queue multiple orders for an entity by holding down Shift when giving the entity an order. It will then complete each order in turn.

VFS requires a bit of work so it can create real directories when files are written to not-yet-existing dirs. This is needed for the XMB move, because they no longer reside in the game's data dirs that are sure to exist anyway.

.. specifically to $install/cache/XMB/mods/official .

Rationale: unclutters the XML dirs, allows easily deleting XMB, allows file writes to go into some writable dir (important when not running as admin).

Discussion: see ​http://www.wildfiregames.com/forum/index.php?showtopic=9939&pid=170413&st=0&

Requires #78.

To support Least-permission User Accounts (i.e. everything except Administrator) on Windows, no files in the game directory may be written to.

1) All output files (e.g. screenshots, savegames) should go into "My Documents\My Games\0ad" (suggestion). 2) Non-user-specific stuff that also doesn't need to be backed up (e.g. archive and XMB) should go in "Documents and Settings\All Users\Application Data\0ad". 3) this will be implemented in a portable fashion by providing sysdep routines to return "home" and "local appdata" directories.

• Control over:
  • Lighting colours (sun, terrain ambient, unit ambient)
  • Sun direction
• Real-time preview of changes
• Load preset lighting conditions
• Save new preset lighting conditions

• Allow control over height of water plane
• Select water effect (texture, shader parameters)
• (Some time in the future, allow editing of the more advanced multi-level water system we'd hopefully have)

• Allow selection of skybox texture, and (if possible) shades.

• Run the game from inside the editor, using the in-game GUI and controls (but preferably inside the Atlas window, to make it as seamless as possible). Once finished, forget all changes and return to the editor

• Edit number of players and their names, types (AI/human/other), civs, resources, starting location, etc.

UI for editing (and previewing) cinematics tracks.

Allow copy-and-paste of various things, like the selected objects.

Probably implement by converting the data to XML and saving as text into the clipboard, so that people can easily edit it in a text editor or save it to disk and then copy it back into Atlas.

Create an object in Atlas. Zoom in and move the camera so the object is near a corner of the screen. With the object-move tool, click the unit and try moving by a single pixel. It jumps towards the edge of the screen by dozens of pixels.

It seems the cause is that CCamera::GetScreenCoordinates and CCamera::GetWorldCoordinates are not exact inverses - for points near the edge of the screen, the difference is quite significant. (GetWorldCoordinates requires integer pixel values, but I don't think that'd cause errors of more than one screen-space pixel (which would be acceptable) - it's probably some difference in the algorithms used to convert between spaces.)

We are currently using really hacky code in sound/ogghack.cpp to decode OGG for sending it to OpenAL.

Background: all of our sounds are in .ogg format. At one point it was promised by Creative that their Windows implementation would support an extension that allows passing in OGG data directly to OpenAL. This would be very convenient for us and also quite efficient. Unfortunately Creative dropped the ball - they never got it working and now officially say "do it yourself".

Since ogghack.cpp was implemented during the time when they promised the extension would be available soon, it is a very basic implementation, wasteful of memory, and doesn't check for errors. This should be cleaned up and made permanent.

No serious changes should be required; we can still use the OGG Vorbis library as currently.

GameSetup.cpp checks for the presence of the GL_ARB_texture_env_crossbar extension, and complains if it's not there, in a way which implies that it's required and the game may break in unexpected ways.

That extension appears to be used only in FastPlayerColorRender, which is optional - it falls back to SlowPlayerColorRender if the required features are missing. Currently it's only testing for >= 3 texture units. It should probably be testing for GL_ARB_texture_env_crossbar too, and quietly switching to the slow method if it's unavailable.

• Problem: Currently all the GUI XML files are combined into one big GUI object, and you have to hide/unhide individual screens at appropriate times to make them appear, which is kind of nasty (and would make any editor much harder to do). This also creates problems with hotkeys (if the Esc key has a specific effect on the main menu, the Esc key will have the same effect everywhere else), as well as determining z order. For example, the dropped-down part of a combobox should usually be visible on top of everything, except when a message box is open on top of that (which seems to be an unavoidable problem if there's a single global z, instead of a localised per-GUI-screen z).

• Solution: Separate GUI XMLs (other than those reserved for special purposes, such as sprite.xml, styles.xml and setup.xml) are considered their own namespaces (local pages). To switch between them, call a function to open/close an XML screen.
  • Each page is an independent GUI window ("modal dialog") with its own local z order. When objects with the same z order overlap, whichever object is written first should have the lower z, making specification of z values largely unnecessary.
  • There could be any number of these pages; pop-up windows (like game setup or a Yes/No dialog) could also be separate pages.
  • The XMLs are added to the GUI scope via init.xml (so that each XML load is not hardcoded into the engine).
  • Pages are drawn in z order using the sequence in which they are declared in init.xml.
  • .cfg Hotkeys are grouped under page names (hotkey.mainmenu.quit, hotkey.session.return) so that they are only effective while that page is open, or can be set as part of the reserved "global" page (hotkey.global.watermark) and the hotkey has the effect on any page.
  • This will also facilitate the -quickstart allowing the game to jump straight to a default session for testing purposes (while currently loading the GUI means having to start at the main menu).

• Problem: Although VFS recognises when game assets are modified and reloads them while the game is running, the GUI does not currently support this feature.
• Solution: Adapt the GUI to be capable of auto-reloading changed assets. Such functionality should greatly improve rate of GUI development, since coordinates can be tweaked while the game is running without the need for an equivalent WYSIWYG editor.

• Problem: The GUI engine shares the same JS space as the simulation, which makes it far too easy to break synchronisation.
• Solution: Separate the GUI and simulation so that they can be loaded independently. Also the setup of the simulation should not be dependent upon the GUI (so that -quickstart works). (Ask Philip for help if any questions.)

• Problem: A common GUI need are tables or lists of buttons with similar properties, such as a construction queue, list of garrisoned unit portraits, or table of statistics. Creating,
• Solution: One solution is to be able to omit the XML and create a control instance entirely from script. A much safer solution, however, would be a "grid" control which acts as a container for an array of child objects of the same type with the same default properties from a given starting position. Only one grid would need to be declared and it would generate a specified number of child controls. It would automatically take care of lining up the buttons and fit to a specific spacing/rows/columns if specified. The child objects could then be individually accessed and their properties modified.

Background: using an existing Unit/self-test framework is deemed helpful (see 2006-05-27 meeting) Chiefly based on ​evaluation, we choose ​CxxTest as unit testing framework.

Basic usage:

• self-tests are separate Cpp files that contain only the test code. (splitting is necessary so that cxxtestgen doesn't have to parse the entire codebase)
• they are stored in a tests subdir of the source directory (simplifies finding the relevant test file for purposes of editing it)
• a workspace generated by premake enumerates all test files, uses custom build step to generate the actual test code, and then links it all into a separate program that runs all tests. Note: separating test code from the main PS executable has several advantages as ​documented
• only running certain tests will be possible (shortens iteration time when only editing one module).

TODO:

• add CxxTest to SVN
• generate workspace for the test program (via premake)
• move all existing tests over to that
• add squelchNextError feature, use it in string_s test
• make sure all tests run successfully
• filtering: modify(/replace/subclass) TestRunner and change its runWorld() so that it filters by the SuiteDescription.testName().

Motivation: error log works but isn't usually looked at (only when a problem has otherwise made itself apparent). Solution: provide an on-screen display of errors to make sure they are seen.

Problems:

• we currently have lots of unsolved errors in the log; when display on screen, those would get annoying fast. Solution: solve all of them! :P More seriously, having them in our face should provide a nice incentive to investigate and solve them. However, if need be, we can also use the logging system's severity mechanism to avoid displaying some log messages on screen.
• how to deal with a flood of errors? Make sure each line gets (say) 30 seconds of face time, then scroll it off the screen. Only display a certain limit of lines to avoid scribbling on the entire screen.

Detailed Implementation Points:

• probably best to integrate into CLogger; that which is logged there also goes onto screen.

• text must be visible to user; whether scribbling directly on the screen (i.e. adding code to Render() ) or adding a GUI control of sorts is left open. The latter is preferable of course, but is a bit harder (needs a control to be made visible/invisible).

• need to maintain a buffer of errors awaiting display. This is because errors can come up before the GUI or graphics systems are initialized; we'd want to display them as soon as possible afterwards.

• proposal: std::deque of error-description lines. push new ones onto the end. as soon as GUI is ready, display the first MAXLINES of them on screen. maintain a count of when each line went on-screen. check every frame if the top lines have gotten more than FACETIME seconds; if so, pop them off the deque and remove them from the screen.

If deque is empty, no longer clutter up the screen; disappear from view.

• this error display should be available for developers and modders. add a config option that enables/disables it.

• prefer unicode-only strings (to allow easier localization)

Thanks to Philip for discussion+planning.

After a unit finishes its dying animation, it "jumps" over to another spot when changing to the corpse actor.

Add CppDoc comments to:

• types
• timer
• posix_types
• posix
• path_util
• ogl
• mmgr
• lib_errors
• debug_stl
• config
• byte_order
• app_hooks
• allocators
• adts
• debug
• lib

• sysdep
• snd
• ia32
• gfx
• dir_watch
• cpu

• win_internal
• win
• wgl
• wdbg
• wdll_ver

• h_mgr
• handle
• cursor
• ogl_shader
• unifont
• ogl_tex
• tex
• tex_codec
• tex_internal
• snd_mgr

• zip
• vfs_tree
• vfs_optimizer
• vfs_mount
• vfs
• trace
• path
• file_io
• file_internal
• file_cache
• file
• compression
• archive_builder
• archive

.

Some possibilities for minor improvement noticed while cruising by:

• premake: document extra_params
• ogl: remove EXT_ symbols, use glext instead
• debug_stl: prefix symbols with debug_stl
• adts: move cache into separate file (also remove old cache impl)
• wgl: prevent multiple include
• move SDL files in sysdep/win to libraries/SDL
• delay_load uses NLSO ctor to register functions. require load_notify to be called at init time. (rationale: this no longer works since splitting up into several static libs)
• wsock_init: call via module init, have it do load_notify call
• ScopeTimerAccrue: change #ifdef spaghetti into templated policy class
• app_hooks, define VOID_FUNC to FUNC(.., (void) )
• look at stalk walk code; any reason not to work on Win64?
• replace ERR_TEX_CODEC_CANNOT_HANDLE with INFO_* - not an error. also ERR_SYM_SUPPRESS_OUTPUT ERR_SYM_SINGLE_SYMBOL_LIMIT
• wdbg_sym: only import RtlCaptureContext once (not every walk_stack)
• add sys_last_error (GetLastError on win32); converts to text in display_error, also show sys_last_error and errno
• app_hooks: move i18n impl out (belongs in pyrogenesis)
• move non-lib asm code into separate asm file; split out memcpy and color code from ia32_asm.asm
• rename all C headers to C++ name (e.g. cmath) Deemed not worth the effort.
• fix mem leaks No longer happens.
• move GUIbase.h errors (DECLARE_ERROR) into philip's error mechanism, remove DECLARE_ERROR, remove ps/errors.h Separate ticket.

Desmond reports:

I have a similar problem too. Ever since I updated the cache from a few months ago it does that for me.

Function call failed at dll_ver.cpp:45 (get_ver): return value was -1 (Unknown error (-1, 0xFFFFFFFF))

Call stack:

RtlFreeUnicodeString (:0)

win_get_snd_info (wsnd.cpp:178)
    dlls = (uninitialized/invalid set<string > >)
    suppress__ = 0x55 ('U')

WriteSystemInfo (util.cpp:48)
    LINE_44 = 
        t0 = 2.82072e-317
        description = 0x00000000

    N_path = "PCI\VEN_1002&DEV_4A49&SUBSYS_0D42174B&REV_00"
    hostname = "dvag"
    un = 
        sysname = [9] { 244, 218, 246, 119 ('w'), 212, 246, 45 ('-'), 0, 0 }
        nodename = [16] { 1, 0, 0, 254, 58 (':'), 83 ('S'), 0, 0, 0, 0, 0, 0, 5, 0, 0, 0 }
        release = [9] { 0, 0, 0, 0, 240, 253, 126 ('~'), 18, 59 (';') }
        version = "S"
        machine = [9] { 0, 0, 0, 0, 0, 0, 97 ('a'), 116 ('t'), 105 ('i') }

    ips = 0x40297020 -> (unavailable - internal error)

Then with a supress the game stays minimized, and the first time it does a 'rebuilding archive' then then goes to menu. A few builds ago it would show two error messages, but it dosen't do that anymore.

My sound card is a Creative Soundblaster Audigy 2 ZS. I'm running Windows XP Professional x64 in English. This 64bitness requires 64bit drivers to work, so the sound driver I'm using is a 64bit one. I'll check to see if there are any sounddriver updates after this. I haven't moved anything around. My 0ad directory is located here D:\Documents and Settings\Administrator\My Documents\0 AD. I have a C drive, but it is used for Win xp home boot and storing files.

It needs to work.

Currently XMB converts all element and attribute names to lower-case. It should keep them as they were in the file. We can then modify the entity or GUI code to change the JavaScript versions to camelCase.

• We now get warnings whenever there is a C++ file that is empty or doesn't contain any meaningful info (e.g. stuff that would end up in the compiled static library). Accordingly: decide if CGUIScrollBarStyle.cpp and CGUIList.cpp can be removed.

• Some parts of the GUI still use the old and obsolete PS_ERROR mechanism. Removing that would be nice, to avoid confusion. Also, localization would be helped a bit by not having error text hard-coded in English.

If a return-value-based approach is still desired (instead of tossing exceptions), we could simply assign numbers to the error definitions, rename them a bit, and be done. Otherwise, we can move that over to Philip's error mechanism, i.e. throwing exceptions. I can help out and/or do all the work - just let me know which is preferable. (old TODO entry: move GUIbase.h errors (DECLARE_ERROR) into e.g. Philip's error mechanism, remove DECLARE_ERROR, remove ps/errors.h.)

• tc_CalculateVariationKey: 1.72362 s (8010x): graphics/objectmanager.cpp

TODO (for some efficiency): use the pre-calculated choices for this object, which has already worked out what to do for props, instead of passing the selections into BuildVariation and having it recalculate the props' choices. Release mode: 38.6697 ms

• Gamesetup/config.cpp: 300ms

g_ConfigDB.Reload(CFG_SYSTEM); 216ms LoadGlobals(); 64ms Release mode: 60ms

• plain_transform: 500ms

flipping font texture files vertically. fix: convert them to e.g. DDS (possibly compressed, if that looks ok?!) same thing for art/textures/particles/sprite cursors also require runtime conversion. ironically enough, being DDS means they have to be converted back to TGA's BGR ordering (required by Windows for cursors) Release mode: 5ms

• CSimulation init: 5000ms Now about 600ms.

• font load: 350ms (total for 12 fonts) mostly spent building display lists. Better to store coords and render quads in immediate mode? Or use vertex arrays?

• GUI init: 900ms (having subtracted font load time)

[more will be added as they are found]

• VC2005 release mode selftest crashes.
• VC2003 string_s selftest complains due to missing debug_skip_next_error
• isnan() reports spurious failures. TODO: check IA32 FXAM bit definitions.

Territory boundaries should be rendered in the game world, flowing along the terrain. Ideally, each territory's boundary should be coloured with its owner's player colour so that at the edge between two territories owned by different players, you see two lines of different colours side by side.

Buildings should only be buildable on allied territories. In addition, buildings should have a per-territory cap (e.g. 2 barrackses per territory).

Gathering should only be allowed in allied territories or around mills built in enemy territories.

Controlling a territory should remove the Shroud of Darkness (but not Fog of War) within it for its owner.

​wxWidgets 2.7.0 released. It doesn't seem to have any major changes (except for an added 'advanced user interface' library with no documentation so I'm not quite sure what it does), so it ought to be reasonably easy to upgrade to, and it should provide continued bug fixes and things.

Shadows seem to be rendered on the screen one frame later than they should be: click on the minimap to jump somewhere (preferably while having a low framerate), and the screen is momentarily bright before all the shadows pop back in.

Basically the default for actors would be that they would cast a shadow and they would recieve a shadow. Only need flags if they don't in either case. Something like:

<not_recieve_shadows /> <not_cast_shadow />

We need a way to render player controled/obstructed (by trees, walls, terrain) units (player color/filled sillohettes), and some nifty effects to render enemy units hiding in 'trees' (some sort of transparency).

• divide GPUs into several classes (i.e. crap, latest+greatest, shader capable, etc.)
• measure perf on one sample system from each class, and then compare with performance attained during a test run.
• if there is a test run, warn of a problem similar to what Nicolai experienced: lack of VIA 4in1 driver => AGP disabled => 5FPS when it should be much higher.

Could also use this to autodetect graphics settings based on what the user's system is capable of.

During startup of the game, I got the following error (unfortunately, I wasn't able to reproduce it).

This is in Windows XP, running the game in a 800x600 window, the screen resolution is 1280x1024. The window appeared in to top-left corner of the screen.

Here's the error message from the error dialog:

Assertion failed: "pt.x >= 0 && pt.y >= 0"
Location: wsdl.cpp:802 (screen_to_client)
errno = 42 (?)
OS error = Der Vorgang wurde erfolgreich beendet.


Call stack:

screen_to_client (wsdl.cpp:802)
    screen_x = (unavailable - stored in register 0x0)
    screen_y = (unavailable - stored in register 0x0)
    x = (unavailable - stored in register 0x0)
    y = (unavailable - stored in register 0x0)
    pt = { x = 5791740, y = 179 }
    suppress__ = 0x00
    suppress__ = 0x00

OnMouseButton (wsdl.cpp:903)
    __formal = 0x0048C874 (..)
    uMsg = (unavailable - stored in register 0x0)
    screen_x = 182
    screen_y = 16
    __formal = 0x0048C874
    outstanding_press_events = 1
    x = 0x7FFDF000
    y = 0x00000001

wndproc (wsdl.cpp:1035)
    hWnd = 0x00050600 (..)
    uMsg = 0x00000201
    wParam = 0x00000001
    lParam = 1048758
    ps = 
        hdc = 0x0012FCC0 (..)
        fErase = 1
        rcPaint = 
            left = 0
            top = 0
            right = 16
            bottom = 0

        fRestore = 0
        fIncUpdate = 1
        rgbReserved = [32] 
            0x00
            0x00
            0x00
            0x00
            0x00
            0x00
            0x00
            0x00
            0x44 ('D')
            0xFC
            0x12
            0x00
            0x48 ('H')
            0xFC
            0x12
            0x00 ...


GetDC (:0)

GetDC (:0)

IsWindowUnicode (:0)

CallWindowProcW (:0)

wglSwapBuffers (:0)

GetDC (:0)

GetDC (:0)

GetWindowLongW (:0)

DispatchMessageW (:0)

SDL_PumpEvents (wsdl.cpp:1051)
    msg = 
        hwnd = 0x00000001
        message = 0x00050600
        wParam = 0x00000201
        lParam = 1
        time = 0x001000B6
        pt = { x = 12142140, y = 185 }


SDL_PollEvent (wsdl.cpp:1061)
    ev = 0x0012FEFC -> 
        type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }
        key = 
            type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }
            keysym = { type = [8] { 0x10, 0xFF, 0x12, 0x00, 0x01, 0x00, 0x00, 0x00 }, sym = 1, unicode = 0xF1F9 }

        motion = { type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }, x = 0x0000, y = 0xFF10 }
        button = 
            type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }
            button = [8] { 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00, 0x01 }
            state = [8] { 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00, 0x01, 0x00 }
            x = 0xFF10
            y = 0x0012

        active = { type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }, gain = [8] { 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00, 0x01 }, state = [8] { 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00, 0x01, 0x00 } }
        user = { type = [8] { 0x00, 0x00, 0x00, 0x00, 0x10, 0xFF, 0x12, 0x00 }, code = 1244944, data1 = 0x00000001 }


PumpEvents (main.cpp:91)
    ev = 
        ev = 
            type = 0xFC
            key = 
                type = 0xFC
                keysym = { type = 0x00, sym = 1244944, unicode = 0x0001 }

            motion = { type = 0xFC, x = 0x0012, y = 0x0000 }
            button = 
                type = 0xFC
                button = 0xFE
                state = 0x12
                x = 0x0000
                y = 0x0000

            active = { type = 0xFC, gain = 0xFE, state = 0x12 }
            user = { type = 0xFC, code = 0, data1 = 0x0012FF10 }



Frame (main.cpp:240)
    need_update = true 
    TimeSinceLastFrame = 3.62822e-038
    need_render = false
    suppress__ = 0x00
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)
    __profile = (CProfileSample)

main (main.cpp:382)
    argc = 1
    argv = 0x00384EA0 -> 0x00384EA8 -> "c:\Wildfiregames\0ad\binaries\system\pyrogenesis.exe"

__tmainCRTStartup (crtexe.c:586)
    nested = 1245020

SEH_wrapped_entry (win.cpp:338)
    ret = 1245076

RegisterWaitForInputIdle (:0)

peltast: when running, spear is on wrong prop point (holding it in the wrong hand?!)

Duplicate: fire up cantabrian generated and order the starting unit(s) to run.

bandbox selection is buggy. sometimes selection works, sometimes none are selected.

woman has a build menu, but can't build anything. should hide the build menu

on click on minimap: it apparently captures the mouse, thus one can move the mouse around and the view follows it. clicking again undoes this and unhooks the mouse. Desirable behavior would be: one click on the minimap simply moves the view there.

select the (cantabrian generated) 3 starting units, formation "wedge" => crash

when Lclick on finished farm, temple, mill, crashes. Lclick on dock,barracks,market works fine.

on maps that have no territories defined, one cannot build anywhere. Fix: retrofit territories into all maps, or disable the territory check when building on maps that have no Ts defined?

rally point not implemented --> It actually is implemented; you can select a building, click the pink flag button, and rightclick on the map to set a rally point.

when finished gathering from a tree, the tree does not disappear. It is still rendered, but at least is no longer available for selection. Same goes for rocks.

This bug occured in SVN r4384

To reproduce:

1. Launch CantabrianHighland-Mythos
   
2. In the Barracks to the left: start researching something
   
3. Hit Escape to exit the game

From ​meeting - it should be possible to specify that certain actors should float on top of water instead of following the terrain.

• Add flag to actor XML
• Add checkbox to Actor Editor
• Make game use that when sticking nonentities onto the ground

As ​here:

• Create .dae -> .pmd, .pma converter
  • for static objects (e.g. buildings)
  • for animated bipeds
  • for arbitrary skeletons (skinned, or multiple meshes linked to bones)
• Make game check for presence of updated .dae when reading .pmd/.pma files and convert it if necessary
• Things that should probably be split into further tickets in the future:
  • Create .pmd, .pma -> .dae converter so people can modify our current files
  • Use the newly available mesh information and tools to create more optimal .pmd files

Premake isn't setting up PCH properly for GCC. (lib/config.h was defining HAVE_PCH, but the project isn't being compiled with PCH so that doesn't work.)

From the ​docs, it looks like we just need to make it compile precompiled.h into precompiled.gch before doing the rest of the source files (whatever.cpp into whatever.o). That'd probably require modifying Premake's source so it handles the pchsource/pchheader settings differently to MSVC - it should ignore the pchsource and just add the pchheader as the first compiled file.

After running update-workspaces.sh (in particularly the line that calls premake with --outpath), it creates a number of mode 000 directories inside the premake directory:

$ ls -al ../premake/     
total 857
drwxr-xr-x 11 philip users    424 Sep 30 18:55 .
drwxr-xr-x 11 philip users    296 Sep 30 13:28 ..
drwxr-xr-x  7 philip users    328 Sep 30 14:52 .svn
d---------  2 philip users     48 Sep 30 18:55 0ad
d---------  2 philip users     48 Sep 30 18:55 build
-rw-r--r--  1 philip users   6165 Sep 30 13:27 extern_libs.lua
-rwxr-xr-x  1 philip users    495 Sep 30 13:27 functions.lua
d---------  2 philip users     48 Sep 30 18:55 gcc
d---------  2 philip users     48 Sep 30 18:55 home
d---------  2 philip users     48 Sep 30 18:55 philip
-rwxr-xr-x  1 philip users 563342 Sep 30 13:39 premake
-rwxr-xr-x  1 philip users 278528 Sep 30 13:26 premake.exe
-rwxr-xr-x  1 philip users  19926 Sep 30 13:26 premake.lua
drwxr-xr-x  5 philip users    376 Sep 30 13:39 src
d---------  2 philip users     48 Sep 30 18:55 trunk
d---------  2 philip users     48 Sep 30 18:55 workspaces

(I'm not sure if this is just a Premake bug - if it is, we should see if it's been fixed in recent versions, and if not then fix it and submit a patch.)

test_archive_builder.h has setUp/tearDown to initialise/shutdown VFS and related things. But it doesn't shut down cleanly - if you add another test case in that test suite, it dies with various errors when running setUp for the second time. That also affects any other test suites which are wanting to initialise the VFS so it's available to the tested code.

The tests fail, probably because of differences between vsnprintf2 (the implementation used on Windows, based on MSVC's _vsnprintf) and vsnprintf (as implemented in glibc) in handling strings that are larger than the buffer.

../../../source/ps/tests/../../../source/ps/tests/test_CLogger.h:46: Error: Expected (lines[1] == msg1), found ("******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************" != "*******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************")
../../../source/ps/tests/../../../source/ps/tests/test_CLogger.h:47: Error: Expected (lines[2] == clipped), found ("******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************" != "****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************...")
../../../source/ps/tests/../../../source/ps/tests/test_CLogger.h:48: Error: Expected (lines[3] == clipped), found ("******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************" != "****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************...")

In TestPathUtil::test_append:
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:12: Error: Expected (std ::string(dst) == std::string(correct_result)), found ("abc\\/def" != "abc\\def")
.
In TestPathUtil::test_name_only:
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:28: Error: Expected (std ::string(result) == std::string(correct_result)), found ("abc\\def\\ghi" != "ghi")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:28: Error: Expected (std ::string(result) == std::string(correct_result)), found ("def\\ghi" != "ghi")
In TestPathUtil::test_last_component:
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:34: Error: Expected (std ::string(result) == std::string(correct_result)), found ("abc\\def\\ghi" != "ghi")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:34: Error: Expected (std ::string(result) == std::string(correct_result)), found ("def\\ghi" != "ghi")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:34: Error: Expected (std ::string(result) == std::string(correct_result)), found ("abc\\def\\ghi\\" != "ghi\\")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:34: Error: Expected (std ::string(result) == std::string(correct_result)), found ("def\\ghi/" != "ghi/")
In TestPathUtil::test_strip_fn:
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:42: Error: Expected (std ::string(path) == std::string(correct_result)), found ("" != "abc\\def\\")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:42: Error: Expected (std ::string(path) == std::string(correct_result)), found ("abc/" != "abc/def\\")
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:42: Error: Expected (std ::string(path) == std::string(correct_result)), found ("" != "abc\\def\\ghi\\")
.
In TestPathUtil::test_path_package:
../../../source/lib/tests/../../../source/lib/tests/test_path_util.h:57: Error: Expected (std ::string(pp.path) == std::string(correct_result)), found ("a\\b\\/c" != "a\\b\\c")

lockfree.cpp(179): Assertion failed: "tls_key != 0"
Assertion failed: "tls_key != 0"
Location: lockfree.cpp:179 (tls_alloc)
errno = 0 (?)
OS error = ?

Call stack:

(0x084e4999) allocators.cpp:621 bucket_create(Bucket*, unsigned int)
(0x084e4be0) allocators.cpp:166 validate_da
(0x084e5239) allocators.cpp:581 pool_free_all(Pool*)
(0x082d7f14) stl_iterator_base_funcs.h:118 std::iterator_traits<__gnu_cxx::__normal_iterator< unsigned short*, std::basic_string<unsigned short, std::char_traits<unsigned short>, std::all ocator<unsigned short> > > >::difference_type std::distance<__gnu_cxx::__normal_iterator<unsi gned short*, std::basic_string<unsigned short, std::char_traits<unsigned short>, std::allocat or<unsigned short> > > >(__gnu_cxx::__normal_iterator<unsigned short*, std::basic_string<unsi gned short, std::char_traits<unsigned short>, std::allocator<unsigned short> > >, __gnu_cxx:: __normal_iterator<unsigned short*, std::basic_string<unsigned short, std::char_traits<unsigne d short>, std::allocator<unsigned short> > >)
(0x082d81c4) stl_construct.h:121 void std::__destroy_aux<__gnu_cxx::__normal_iterator<JSFunct ionSpec*, std::vector<JSFunctionSpec, std::allocator<JSFunctionSpec> > > >(__gnu_cxx::__norma l_iterator<JSFunctionSpec*, std::vector<JSFunctionSpec, std::allocator<JSFunctionSpec> > >, _ _gnu_cxx::__normal_iterator<JSFunctionSpec*, std::vector<JSFunctionSpec, std::allocator<JSFun ctionSpec> > >, __false_type)
(0x082b3f78) stl_algobase.h:387 HEntity* std::copy<__gnu_cxx::__normal_iterator<HEntity const *, std::vector<HEntity, std::allocator<HEntity> > >, HEntity*>(__gnu_cxx::__normal_iterator<H Entity const*, std::vector<HEntity, std::allocator<HEntity> > >, __gnu_cxx::__normal_iterator <HEntity const*, std::vector<HEntity, std::allocator<HEntity> > >, HEntity*)
(0x082b826a) stl_tree.h:817 std::_Rb_tree<CStrW, std::pair<CStrW const, int>, std::_Select1st <std::pair<CStrW const, int> >, CStrW_hash_compare, std::allocator<std::pair<CStrW const, int > > >::_M_insert(std::_Rb_tree_node_base*, std::_Rb_tree_node_base*, std::pair<CStrW const, i nt> const&)
(0x082b8342) stl_tree.h:917 std::_Rb_tree<CStrW, std::pair<CStrW const, int>, std::_Select1st <std::pair<CStrW const, int> >, CStrW_hash_compare, std::allocator<std::pair<CStrW const, int > > >::insert_unique(std::pair<CStrW const, int> const&)
(0x082b8433) stl_tree.h:927 std::_Rb_tree<CStrW, std::pair<CStrW const, int>, std::_Select1st <std::pair<CStrW const, int> >, CStrW_hash_compare, std::allocator<std::pair<CStrW const, int > > >::insert_unique(std::pair<CStrW const, int> const&)
(0x082b84e9) stl_tree.h:932 std::_Rb_tree<CStrW, std::pair<CStrW const, int>, std::_Select1st <std::pair<CStrW const, int> >, CStrW_hash_compare, std::allocator<std::pair<CStrW const, int > > >::insert_unique(std::pair<CStrW const, int> const&)
(0x082b8511) stl_tree.h:942 std::_Rb_tree<CStrW, std::pair<CStrW const, int>, std::_Select1st <std::pair<CStrW const, int> >, CStrW_hash_compare, std::allocator<std::pair<CStrW const, int > > >::insert_unique(std::_Rb_tree_iterator<std::pair<CStrW const, int> >, std::pair<CStrW co nst, int> const&)
(0x082b62be) stl_pair.h:69 pair
(0xb760d878) /lib/libc.so.6:0 __libc_start_main
(0x082b36a1) stl_vector.h:776 HEntity* std::vector<HEntity, std::allocator<HEntity> >::_M_all ocate_and_copy<__gnu_cxx::__normal_iterator<HEntity const*, std::vector<HEntity, std::allocat or<HEntity> > > >(unsigned int, __gnu_cxx::__normal_iterator<HEntity const*, std::vector<HEnt ity, std::allocator<HEntity> > >, __gnu_cxx::__normal_iterator<HEntity const*, std::vector<HE ntity, std::allocator<HEntity> > >)

​http://www.wildfiregames.com/forum/index.php?showtopic=10656

Every time the game brings up its Program Error window which looks somewhat like ...

Function call failed: return value was -100101 (Case 1)
Location: allocators.cpp:174 (validate_da)
errno = 0 (?)
OS error = The system cannot find the file specified.


Call stack:

validate_da (allocators.cpp:164)
    da = 0x00CEE6A0 -> 
        base = 0x00000000
        max_size_pa = 0x00000000
        cur_size = 0x00000000
        prot = 0
        pos = 0x00000000

    pos = 0xCCCCCCCC
    prot = 0xCCCCCCCC
    base = 0xCCCCCCCC
    cur_size = 0xCCCCCCCC
    max_size_pa = 0xCCCCCCCC

... I can't help but read the one piece of understandable meaningful English on it, which is nearly always "OS error = The system cannot find the file specified.", which is nearly always completely irrelevant (because it was probably set seconds ago by some unrelated operation). Then I unfailingly get confused for a while before remembering that that message is completey irrelevant, and have to start again at trying to work out what the error is complaining about.

Since it's worse than useless in most cases (because it distracts me from what the real problem is), it would be nice to remove it. If there are cases where it's actually useful, it could at least report the error number instead of the string so it doesn't stand out (and change the "OS error" text because that's distractingly English too), and we can always look up the meaning of that number in the rare cases when it's relevant to a bug report.

So, it could look more like:

Function call failed: return value was -100101 (Case 1)

Location: allocators.cpp:174 (validate_da)

(errno=0, oserr=2)

Call stack:

validate_da (allocators.cpp:164)
    da = 0x00CEE6A0 -> 
        base = 0x00000000
        max_size_pa = 0x00000000
        cur_size = 0x00000000
        prot = 0
        pos = 0x00000000

    pos = 0xCCCCCCCC
    prot = 0xCCCCCCCC
    base = 0xCCCCCCCC
    cur_size = 0xCCCCCCCC
    max_size_pa = 0xCCCCCCCC

and then it's still not very helpful but at least it's not misleading, and I can see I should just look in the code or the debugger to work out what's wrong.

VFS_MOUNT_ARCHIVABLE is never used, except for being passed to (then ignored by) vfs_mount.

But mount_is_archivable tests for VFS_MOUNT_ARCHIVES - is that where ARCHIVABLE should be used instead? (The comment in source:ps/trunk/source/lib/res/file/vfs_optimizer.cpp@4703#L161​ talks about VFS_MOUNT_ARCHIVE which doesn't even exist.)

It would be nice if the ​TDD Wiki was on Trac instead of MediaWiki, to simplify administration (user accounts, backups, etc). It should be possible to export the pages from MediaWiki, do some conversion, and import them into Trac without too much manual effort.

The assertion at source:ps/trunk/source/lib/res/file/trace.cpp@4706#L202​ fails with start_idx = 25243, t->total_ents = 25225.

trace.txt is 25326 lines, with the last run's delimeter on line 25243, with a total of 101 delimeters. 25326 - 101 = t->total_ents, and 25243 = start_idx. So it looks like the problem is that start_idx counts the divider lines, while t->total_ents doesn't, and so it fails if the number of divider lines is greater than the size of the last run in the file.

The terrain renderer draws non-planar GL_QUADS. The OpenGL spec just says

For a polygon with more than three edges, we require only that a convex combination of the values of the datum at the polygon's vertices can be used to obtain the value assigned to each fragment produced by the rasterization algorithm.

which in practice seems to result in the quad being triangulated along an arbitrary diagonal. Since there's no way to know exactly what's going to happen, and particularly since what happens isn't very good, we need to split the quads into triangles ourselves.

It is presumably sensible to split each quad along the diagonal that minimises the angle between the two triangles.

ogl.cpp does:

	p##nameARB = NULL;
	if(oglHaveVersion(version))
		*(void**)&p##nameARB = SDL_GL_GetProcAddress(#nameCore);
	if(!p##nameARB) /* use the ARB name if the driver lied about what version it supports */
		*(void**)&p##nameARB = SDL_GL_GetProcAddress(#nameARB);

But ​glXGetProcAddress never returns NULL. In the case where "the driver lied about what version it supports", it will still return a non-NULL function pointer, which will crash when it is called, even if there would have been a valid function pointer when we used the ARB name.

(NVIDIA's OpenGL implementation is different, and does (when I test it) return NULL for functions that it doesn't have. But the xorg-x11 implementation only returns NULL for non-existent glX* functions or for names not starting with "gl", and it returns non-NULL for any arbitrary gl[^X]* name.)

Hopefully it doesn't actually matter for us, since ogl.cpp seems to be the only place where we're testing the NULLness of the function pointers, and since ogl.cpp should still work correctly as long as the driver reports its GL version correctly (which it should). So I don't think there's anything we should actually do, except be aware of the issue.

premake.lua says:

package.defines = {
    "__STDC_VERSION__=199901L",
    "CONFIG_USE_MMGR",
}

sysdep.h says:

for some reason, g++-3.3 claims to support C99 (according to __STDC_VERSION__) but doesn't have the restrict keyword.

So it's our fault that we get a little confused by GCC pretending to support C99 when it really doesn't.

Claiming to support C99 is nonsensical in any case, because we're writing C++ and not C. It would probably be better to define a series of HAVE_RELEVANT_C99LIKE_FEATURE, for all the relevant C99-like features which GCC supports in its normal nonstandard C++ mode.

It should be possible to specify a sound for each animation, using code such as:

<Entity>
        ...

	<SoundGroups>
		<Walk>infantry_walk.xml</Walk>
		<Run>infantry_run.xml</Run>
		<Death>screams.xml</Death>
		<Gather_wood>axe_chopping.xml</Gather_wood>
		<Gather_stone>pickaxe.xml</Gather_stone>
	</SoundGroups>
</Entity>

These should be in a hashmap in CEntityTemplate.

When running the cinematics.pmp simulation in Atlas, it crashes after a while in the aura-handling code.

Some script calls CEntity::Kill, which calls Remove on everything in m_aurasInfluencingMe. That fails because the CAura* in there is not valid - it's a readable pointer but the contents are rubbish (seemingly being used by some other object, since it's not the default debug memory-fill pattern).

When units with a death animation die, a corpse showing the last frame of this animation should persist. However, they currently just disappear. Could be related to changes in [4810].

Let the entity code simply tell the actor to switch to a certain animation state, and not worry about any timing details itself. The actor should handle looping, randomisation, and potentially things like blending. The animation speed should be determined just by the actor XML.

Change the entity XML files to store only a Rate (units per second) for actions, rather than Speed+Rate. Update the simulation world at a fixed frequency, independent of the animation speed.

Experiment with the same for attack animations to see if it still looks alright, but at least keep projectile launches synchronised.

By default, the shadow DepthTextureBits is 16. With a GF4, that is very slow compared to other depths - e.g. on a blank grass map, with AA disabled, it is 72fps with DepthTextureBits=32 and 23fps with DepthTextureBits=16. With 4xAA enabled, it is 36fps at 32, but only 2fps at 16.

Is a similar effect visible on other graphics cards? (Use renderer.depthTextureBits in the console to change it). If so, presumably the default should be changed; otherwise, the default should depend on the hardware.

The automatic archiver runs pretty much every time I start the game (probably because I don't start the game very often and so several data files change in the meantime). It takes ages, and tends to leave multiple giant zip files hanging around, and wastes much more time than simply loading slower unarchived files, and it leaves the game window (and sometimes the whole computer) unresponsive while it's working.

I'd be much happier if it was manually activated instead - people could run it once if they care about the loading speed, and the time lost to loading loose files after the archive becomes out of date would be more than made up for by the time gained by never automatically rebuilding the archive. We can run the archiver once manually when distributing the game, and no normal user (including modders) will change enough files for it to ever be worth automatically rebuilding the archive, particularly since it happens precisely when they're trying to play the game and it stops them from doing so. (It wouldn't be so bad if it happened at installation-time for large patches.)

(Or maybe I'm misinterpreting the problem and it's just not using the mini archives like it should do (since they never look mini - they're always about 60MB), or something.)

...

Oh, it's finished rebuilding now, I'll get back to work ;-)

After deleting entities, they tend to still exist in an invisible ghostly form and attack people when you play the simulation from within the editor.

They should be deleted properly from the entity manager, instead of the current pseudo-deletion. To allow undo, serialise the entity into some form (same as when you play/reset the simulation) on deletion and then reconstruct it on undo.

The skeleton definitions should be located in mods/*/art/skeletons/*.xml, and all the files matching that pattern should be loaded. That would allow modders to add their own skeleton definitions, just by creating a new file (rather than overwriting the existing one). It would also reduce the number of animations that get cache-invalidated when a single skeleton definition is tweaked.

The currently-installed version of TracWebAdmin doesn't work with Trac 0.10, so it should be updated. (The newest version only seems to be available from SVN, not as an Egg file, which makes it slightly less trivial.)

Currently only five buttons (left, right, middle, wheel-up, wheel-down) are supported. It might be nice if users could map functions onto other buttons (e.g. my mouse has seven additional buttons).

The SoundGroup system for choosing random sounds is not yet in use. The current placeholder code enumerates all audio files in a certain subpath, which works but is slow and inflexible (see ​http://www.wildfiregames.com/forum/index.php?showtopic=10924&hl= for an overview and list of links that detail the sound system design).

Here's a list of what I think is necessary to get the SoundGroup show on the road:

• during init, create all the sound groups we'll be needing (e.g. attack grunts, death, ...) and associate them with entity classes
• .. by storing a string or index that identifies the sound group in entity.traits.audio.path instead of path (as now)
• create a script interface for the SoundGroup that allows calling it from our Javascript event handlers (this is where the performAttack logic runs). This kind of thing is pretty much boilerplate, we can refer to stuff like JSI_Sound.cpp to get an idea of what to do.
• replace newRandomSound with the corresponding new function that basically just calls CSoundGroup::PlayNext
• add a call to CSoundGroupMgr::Update in the main game loop

• production queue: at least feedback that something is being built.
  Matei: maybe a list box or something showing just a list of items in the queue, with add/delete buttons.

• Disable formations and all other non-implemented UI stuff
  (remove entirely from scripts by commenting them out)

• disable game setup stuff that's not implemented

Crash occurs in-game as well as in Atlas simulation when initiating combat. Problem seems to me to reside with Super Infantry, as other types of units do not cause such a crash. I assume this could be entity related, but scanning the template_unit_super and template_unit_super_infantry entities does not immediately present any issues with coding. grrr

​http://www.wildfiregames.com/forum/index.php?s=&showtopic=11872&view=findpost&p=192546

Currently, skeleton mismatches are only detected and reported if the number of bones differs. The PMD and PSA files ought to store their skeleton name, so mismatches can be detected properly and reported nicely.

The game probably doesn't work with the NO_GUI compiler flag, and it doesn't seem a useful feature at all, and it adds complexity to the code, so the flag should just be removed entirely.

The Collada project gets linked with libFColladaSD.a (the debug version) in Release configurations as well as Debug. It should use libFColladaSR.a instead, since that's faster.

This probably just needs some small (but not entirely trivial) changes in extern_libs.lua

By clicking on this button, all idle resource gathering units (women ?) should be selected.

In an MP game (Oasis), scrolling off the map shows that the skybox is uninitialized or filled with garbage (looks like a terrain texture, but Wijit was seeing something else).

When loading Oasis in MP, having started the game with -quickstart and -onlyPublicFiles, a VFS file load error is produced.

It turns out that "art/textures/skies/mountainous/front.dds" is the only file that is not found, after that the map loads successfully.

In MP games (specifically Oasis with 2 players), blue (the host) can move red's (the second player's) units .

Encountered by Erik whilst attempting MP on Latium map with Anton: in CEntity::UpdateXZOrientation(), apparently m_base is invalid, thus causing an access violation. Since inheritance is data-driven, we should add checks (especially in this function, likely in others as well) whether m_base is valid at all. If not, try to use reasonable defaults.

In an MP game of Oasis, the blue (host's) cavalry can be clicked upon to show unit details in the upper-left corner of the screen. Here, the range information is displayed with very many digits, thus overwriting the next field. It would be good to display all floating-point information in the GUI with one digit of precision, which can be expressed in printf via "%.1f".

Chat is a fairly important part of the game being playable. As soon as a client connects, it would be good to be able to chat (the previous mechanism having been entering any line of text not starting with : or ? into the console).

We used to have this functionality, but it was a part of the previous network code and apparently still needs work.

After loading a MP game (Oasis), the following warnings/errors show up in the console:

gui/test/functions_utility_list.js:142: TypeError: getGuiObjectByName(objectName) has no properties

setCurrItemValue(): pgSessionSetupPO (<- could also be a 0) not found

Actions to reproduce: Click on multiplayer. Join. Any ip (I tried with 127.0.0.1) where there is no server. (A message will popup saying that could connect.) Go back Host game. [Error.]

Assertion failed: "!ms_singleton" Location: Singleton.h:28 (Singleton<class CNetHost>::Singleton)

Call stack:

Singleton<CNetHost>::Singleton<CNetHost> (singleton.h:28)

this = (unavailable)

CNetHost::CNetHost (netsession.cpp:23)

this = (unavailable)

CNetServer::CNetServer (netserver.cpp:28)

this = (unavailable) pGame = 0x0EA9C030 ->

(boost::noncopyable_::noncopyable) m_World = 0x0EB1AED8 ->

(boost::noncopyable_::noncopyable) m_pGame = 0x0EA9C030 (see above) m_Terrain = 0x0EB43830 ->

m_MapSize = 0 (0x00000000) m_MapSizePatches = 0 (0x00000000) m_Patches = 0x00000000 m_Heightmap = 0x00000000 m_BaseColour =

R = [8] { 255 (0xFF), 255 (0xFF), 255 (0xFF), 255 (0xFF), 0 (0x00), 0 (0x00), 0 (0x00), 0 (0x00) } G = [8] { 255 (0xFF), 255 (0xFF), 255 (0xFF), 0 (0x00), 0 (0x00), 0 (0x00), 0 (0x00), 37 (0x25) } B = [8] { 255 (0xFF), 255 (0xFF), 0 (0x00), 0 (0x00), 0 (0x00), 0 (0x00), 37 (0x25), 193 (0xC1) } A = [8] { 255 (0xFF), 0 (0x00), 0 (0x00), 0 (0x00), 0 (0x00), 37 (0x25), 193 (0xC1), 91 (0x5B) }

m_UnitManager = 0x0EB43810 ->

m_Units = (error -100500 while analyzing vector<CUnit * >) m_NextID = 0 (0x00000000) m_ObjectManager = 0x0EB702A0 ->

(boost::noncopyable_::noncopyable) m_MeshManager = 0x0EB70260 ->

(boost::noncopyable_::noncopyable) m_MeshMap =

(error -100406 while analyzing _Container_base) comp = (CStr8_hash_compare)

_List = (error -100500 while analyzing list<pair<CStr8 const ,boost::weak_ptr<CModelDef> > >) _Vec = (error -100406 while analyzing vector<list<pair<CStr8 const ,boost::weak_ptr<CModelDef> > >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_ColladaManager = 0x0EB7025C ->

m = 0x0EB43850 ->

dll =

m_Name = 0x0150E31C -> "Collada" m_Handle = 0x00000000

set_logger = 0xFE4EF270 -> set_skeleton_definitions = 0xFD14F377 -> convert_dae_to_pmd = 0xFC3CF462 -> convert_dae_to_psa = 0xFBA1F4A5 ->

m_SkeletonAnimManager = 0x0EB70290 ->

(boost::noncopyable_::noncopyable) m_Animations = (error -100500 while analyzing map<CStr8,CSkeletonAnimDef * >) m_ColladaManager = 0x0EB7025C (see above)

m_Objects = (error -100500 while analyzing map<CObjectManager::ObjectKey,CObjectEntry * >) (too much output; skipping to next top-level symbol)

m_EntityManager = 0x0EB4E608 ->

m_entities = [16384]

{ m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } { m_entity = 0x00000000, m_refcount = 0 (0x0000) } ...

m_refd = (error -100406 while analyzing bitset<16384>) m_reaper = (error -100500 while analyzing vector<CEntity * >) m_collisionPatches = 0x00000000 m_nextalloc = 0 (0x00000000) m_death = false m_collisionPatchesPerSide = 13365591 (0x00CBF157) m_entityClassData = (error -100500 while analyzing map<size_t,map<CStrW,int > >) m_screenshotMode = false

m_ProjectileManager = 0x0EAC6428 ->

m_LastTurnLength = 0 (0x00000000) m_Projectiles = (error -100500 while analyzing list<CProjectile * >)

m_LOSManager = 0x0EB437F0 ->

m_VisibilityMatrix = 0x00000000 m_TilesPerSide = 10 (0x0000000A) m_TilesPerSide_1 = 0 (0x00000000) m_LOSSetting = LOS_SETTING_NORMAL m_FogOfWar = true

m_TerritoryManager = 0x0EB1B6A8 ->

m_Territories = (error -100500 while analyzing vector<CTerritory * >) m_TerritoryMatrix = 0x00000000 m_TilesPerSide = 1953066081 (0x74696C61) m_DelayedRecalculate = false

m_Simulation = 0x0EB6EE40 ->

m_pGame = 0x0EA9C030 (see above) m_pWorld = 0x0EB1AED8 (see above) m_pTurnManager = 0x0EB706A4 ->

m_Clients = (error -100406 while analyzing vector<CTurnManager::SClient >) m_Batches = [3]

m_Messages = (error -100406 while analyzing vector<CTurnManager::SMessageSyncEntry >) m_TurnLength = 300 (0x0000012C)

m_Messages = (error -100406 while analyzing vector<CTurnManager::SMessageSyncEntry >) m_TurnLength = 300 (0x0000012C)

m_Messages = (error -100406 while analyzing vector<CTurnManager::SMessageSyncEntry >) m_TurnLength = 300 (0x0000012C)

m_pRecord = 0x0DF402C9 ->

m_IsRecording = false m_pTurnManager = 0x00000000

m_Time = 0 (0x0000000000000000) m_DeltaTime = 0 (0x0000000000000000) m_Random =

x = [1248]

5489 (0x00001571) 1301868182 (0x4D98EE96) 2938499221 (0xAF25F095) 2950281878 (0xAFD9BA96) 1875628136 (0x6FCBD068) 751856242 (0x2CD06A72) 944701696 (0x384F0100) 2243192071 (0x85B46507) 694061057 (0x295E8801) 219885934 (0x0D1B316E) 2066767472 (0x7B305E70) 3182869408 (0xBDB6BBA0) ...

i = 624 (0x00000270)

m_GameView = 0x0EA0C8C0 ->

(Scene) (boost::noncopyable_::noncopyable) m = 0x0EB701F0 ->

(IJSObject) m_JS = 0x00000000 m_NonsharedProperties =

(error -100406 while analyzing _Container_base) comp = (CStrW_hash_compare)

_List = (error -100500 while analyzing list<pair<CStrW const ,IJSProperty *> >) _Vec = (error -100406 while analyzing vector<list<pair<CStrW const ,IJSProperty *> >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_ScriptProperties =

(error -100406 while analyzing _Container_base) comp = (CStrW_hash_compare)

_List = (error -100500 while analyzing list<pair<CStrW const ,IJSProperty *> >) _Vec = (error -100406 while analyzing vector<list<pair<CStrW const ,IJSProperty *> >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_EngineOwned = true

(boost::noncopyable_::noncopyable) Game = 0x0EA9C030 (see above) ColladaManager =

m = 0x0EB43850 (see above)

MeshManager =

(boost::noncopyable_::noncopyable) m_MeshMap =

(error -100406 while analyzing _Container_base) comp = (CStr8_hash_compare)

_List = (error -100500 while analyzing list<pair<CStr8 const ,boost::weak_ptr<CModelDef> > >) _Vec = (error -100406 while analyzing vector<list<pair<CStr8 const ,boost::weak_ptr<CModelDef> > >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_ColladaManager = 0x0EB7025C (see above)

SkeletonAnimManager =

(boost::noncopyable_::noncopyable) m_Animations = (error -100500 while analyzing map<CStr8,CSkeletonAnimDef * >) m_ColladaManager = 0x0EB7025C (see above)

ObjectManager =

(boost::noncopyable_::noncopyable) m_MeshManager = 0x0EB70260 (see above) m_SkeletonAnimManager = 0x0EB70290 (see above) m_Objects = (error -100500 while analyzing map<CObjectManager::ObjectKey,CObjectEntry * >) m_ObjectBases = (error -100500 while analyzing map<CStr8,CObjectBase * >)

ViewCamera =

m_Orientation =

_11 = 1.000000 (0x3F800000) _21 = 0.000000 (0x00000000) _31 = 0.000000 (0x00000000) _41 = 0.000000 (0x00000000) _12 = 0.000000 (0x00000000) _22 = 0.866025 (0x3F5DB3D7) _32 = 0.500000 (0x3F000000) _42 = 0.000000 (0x00000000) _13 = 0.000000 (0x00000000) _23 = -0.500000 (0xBF000000) _33 = 0.866025 (0x3F5DB3D7) _43 = 0.000000 (0x00000000) _14 = 100.000000 (0x42C80000) _24 = 150.000000 (0x43160000) _34 = -100.000000 (0xC2C80000) _44 = 1.000000 (0x3F800000) _data = [16]

1.000000 (0x3F800000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.866025 (0x3F5DB3D7) 0.500000 (0x3F000000) 0.000000 (0x00000000) 0.000000 (0x00000000) -0.500000 (0xBF000000) 0.866025 (0x3F5DB3D7) 0.000000 (0x00000000) ...

_data2d = [4]

[4] { 1.000000 (0x3F800000), 0.000000 (0x00000000), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 0.866025 (0x3F5DB3D7), 0.500000 (0x3F000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), -0.500000 (0xBF000000), 0.866025 (0x3F5DB3D7), 0.000000 (0x00000000) } [4] { 100.000000 (0x42C80000), 150.000000 (0x43160000), -100.000000 (0xC2C80000), 1.000000 (0x3F800000) }

m_ProjMat =

_11 = 4.219332 (0x408704C4) _21 = 0.000000 (0x00000000) _31 = 0.000000 (0x00000000) _41 = 0.000000 (0x00000000) _12 = 0.000000 (0x00000000) _22 = 5.671282 (0x40B57B24) _32 = 0.000000 (0x00000000) _42 = 0.000000 (0x00000000) _13 = 0.000000 (0x00000000) _23 = 0.000000 (0x00000000) _33 = 1.001955 (0x3F804010) _43 = 1.000000 (0x3F800000) _14 = 0.000000 (0x00000000) _24 = 0.000000 (0x00000000) _34 = -8.007820 (0xC1002008) _44 = 0.000000 (0x00000000) _data = [16]

4.219332 (0x408704C4) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 5.671282 (0x40B57B24) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 1.001955 (0x3F804010) 1.000000 (0x3F800000) ...

_data2d = [4]

[4] { 4.219332 (0x408704C4), 0.000000 (0x00000000), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 5.671282 (0x40B57B24), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 0.000000 (0x00000000), 1.001955 (0x3F804010), 1.000000 (0x3F800000) } [4] { 0.000000 (0x00000000), 0.000000 (0x00000000), -8.007820 (0xC1002008), 0.000000 (0x00000000) }

m_NearPlane = 4.000000 (0x40800000) m_FarPlane = 4096.000000 (0x45800000) m_FOV = 0.349066 (0x3EB2B8C2) m_ViewPort =

m_X = 0 (0x00000000) m_Y = 0 (0x00000000) m_Width = 1024 (0x00000400) m_Height = 768 (0x00000300)

m_ViewFrustum =

m_aPlanes = [10]

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

...

m_NumPlanes = 0 (0x00000000)

CullCamera =

m_Orientation =

_11 = 1.000000 (0x3F800000) _21 = 0.000000 (0x00000000) _31 = 0.000000 (0x00000000) _41 = 0.000000 (0x00000000) _12 = 0.000000 (0x00000000) _22 = 0.866025 (0x3F5DB3D7) _32 = 0.500000 (0x3F000000) _42 = 0.000000 (0x00000000) _13 = 0.000000 (0x00000000) _23 = -0.500000 (0xBF000000) _33 = 0.866025 (0x3F5DB3D7) _43 = 0.000000 (0x00000000) _14 = 100.000000 (0x42C80000) _24 = 150.000000 (0x43160000) _34 = -100.000000 (0xC2C80000) _44 = 1.000000 (0x3F800000) _data = [16]

1.000000 (0x3F800000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.866025 (0x3F5DB3D7) 0.500000 (0x3F000000) 0.000000 (0x00000000) 0.000000 (0x00000000) -0.500000 (0xBF000000) 0.866025 (0x3F5DB3D7) 0.000000 (0x00000000) ...

_data2d = [4]

[4] { 1.000000 (0x3F800000), 0.000000 (0x00000000), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 0.866025 (0x3F5DB3D7), 0.500000 (0x3F000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), -0.500000 (0xBF000000), 0.866025 (0x3F5DB3D7), 0.000000 (0x00000000) } [4] { 100.000000 (0x42C80000), 150.000000 (0x43160000), -100.000000 (0xC2C80000), 1.000000 (0x3F800000) }

m_ProjMat =

_11 = 4.219332 (0x408704C4) _21 = 0.000000 (0x00000000) _31 = 0.000000 (0x00000000) _41 = 0.000000 (0x00000000) _12 = 0.000000 (0x00000000) _22 = 5.671282 (0x40B57B24) _32 = 0.000000 (0x00000000) _42 = 0.000000 (0x00000000) _13 = 0.000000 (0x00000000) _23 = 0.000000 (0x00000000) _33 = 1.001955 (0x3F804010) _43 = 1.000000 (0x3F800000) _14 = 0.000000 (0x00000000) _24 = 0.000000 (0x00000000) _34 = -8.007820 (0xC1002008) _44 = 0.000000 (0x00000000) _data = [16]

4.219332 (0x408704C4) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 5.671282 (0x40B57B24) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 0.000000 (0x00000000) 1.001955 (0x3F804010) 1.000000 (0x3F800000) ...

_data2d = [4]

[4] { 4.219332 (0x408704C4), 0.000000 (0x00000000), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 5.671282 (0x40B57B24), 0.000000 (0x00000000), 0.000000 (0x00000000) } [4] { 0.000000 (0x00000000), 0.000000 (0x00000000), 1.001955 (0x3F804010), 1.000000 (0x3F800000) } [4] { 0.000000 (0x00000000), 0.000000 (0x00000000), -8.007820 (0xC1002008), 0.000000 (0x00000000) }

m_NearPlane = 4.000000 (0x40800000) m_FarPlane = 4096.000000 (0x45800000) m_FOV = 0.349066 (0x3EB2B8C2) m_ViewPort =

m_X = 0 (0x00000000) m_Y = 0 (0x00000000) m_Width = 1024 (0x00000400) m_Height = 768 (0x00000300)

m_ViewFrustum =

m_aPlanes = [10]

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

m_Norm =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

m_Dist = 0.000000 (0x00000000)

...

m_NumPlanes = 0 (0x00000000)

LockCullCamera = false Culling = true CachedLightEnv =

m_Elevation = 0.785398 (0x3F490FDB) m_Rotation = 5.497787 (0x40AFEDDF) m_TerrainShadowTransparency = 0.000000 (0x00000000) m_SunDir =

X = -0.270598 (0xBE8A8BD6) Y = -0.923880 (0xBF6C835F) Z = 0.270598 (0x3E8A8BD3)

m_SunColor =

X = 1.000000 (0x3F800000) Y = 1.000000 (0x3F800000) Z = 1.000000 (0x3F800000)

m_TerrainAmbientColor =

X = 0.643137 (0x3F24A4A5) Y = 0.643137 (0x3F24A4A5) Z = 0.643137 (0x3F24A4A5)

m_UnitsAmbientColor =

X = 0.643137 (0x3F24A4A5) Y = 0.643137 (0x3F24A4A5) Z = 0.643137 (0x3F24A4A5)

TrackManager =

m_Active = true m_DrawCurrentSpline = false m_DrawLines = (bool)0x8B m_ValidCurrent = false m_CurrentPath = (error -100406 while analyzing _Tree<<CStrW,CCinemaPath>::iterator) m_Paths = (error -100500 while analyzing map<CStrW,CCinemaPath >) m_PathQueue = (error -100500 while analyzing list<CCinemaPath >)

ViewScrollSpeed = 60.000000 (0x42700000) ViewRotateSensitivity = 0.002000 (0x3B03126F) ViewRotateSensitivityKeyboard = 1.000000 (0x3F800000) ViewRotateAboutTargetSensitivity = 0.010000 (0x3C23D70A) ViewRotateAboutTargetSensitivityKeyboard = 2.000000 (0x40000000) ViewDragSensitivity = 0.500000 (0x3F000000) ViewZoomSensitivityWheel = 16.000000 (0x41800000) ViewZoomSensitivity = 256.000000 (0x43800000) ViewZoomSmoothness = 0.020000 (0x3CA3D70A) ViewSnapSmoothness = 0.020000 (0x3CA3D70A) CameraDelta =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

CameraPivot =

X = 0.000000 (0x00000000) Y = 0.000000 (0x00000000) Z = 0.000000 (0x00000000)

UnitView = 0x00000000 UnitViewProp = 0xF448015D ->

m_Bounds =

m_Data = [2]

X = (unavailable - internal error)

Y = (unavailable - internal error)

Z = (unavailable - internal error)

X = (unavailable - internal error)

Y = (unavailable - internal error)

Z = (unavailable - internal error)

m_Transform =

_11 = (unavailable - internal error)

_21 = (unavailable - internal error)

_31 = (unavailable - internal error)

_41 = (unavailable - internal error)

_12 = (unavailable - internal error)

_22 = (unavailable - internal error)

_32 = (unavailable - internal error)

_42 = (unavailable - internal error)

_13 = (unavailable - internal error)

_23 = (unavailable - internal error)

_33 = (unavailable - internal error)

_43 = (unavailable - internal error)

_14 = (unavailable - internal error)

_24 = (unavailable - internal error)

_34 = (unavailable - internal error)

_44 = (unavailable - internal error)

_data = [16]

(unavailable - internal error)

_data2d = [4]

[4] { (unavailable - internal error)

m_InvTransform =

_11 = (unavailable - internal error)

_21 = (unavailable - internal error)

_31 = (unavailable - internal error)

_41 = (unavailable - internal error)

_12 = (unavailable - internal error)

_22 = (unavailable - internal error)

_32 = (unavailable - internal error)

_42 = (unavailable - internal error)

_13 = (unavailable - internal error)

_23 = (unavailable - internal error)

_33 = (unavailable - internal error)

_43 = (unavailable - internal error)

_14 = (unavailable - internal error)

_24 = (unavailable - internal error)

_34 = (unavailable - internal error)

_44 = (unavailable - internal error)

_data = [16]

(unavailable - internal error)

_data2d = [4]

[4] { (unavailable - internal error)

m_RenderData = (unavailable - internal error)

m_BoundsValid = (unavailable - internal error)

(boost::noncopyable_::noncopyable) m_Parent = (unavailable - internal error)

m_Flags = (unavailable - internal error)

m_Texture =

m_Name = { (error -100500 while analyzing string >) } m_Handle = (unavailable - internal error)

m_Material =

m_Hash = (unavailable - internal error)

m_Diffuse =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_Ambient =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_Specular =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_Emissive =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_SpecularPower = (unavailable - internal error)

m_Texture = { (error -100500 while analyzing string >) } m_VertexProgram = { (error -100500 while analyzing string >) } m_FragmentProgram = { (error -100500 while analyzing string >) } m_Alpha = (unavailable - internal error)

m_PlayerID = (unavailable - internal error)

m_TextureColor =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_pModelDef =

px = (unavailable - internal error)

pn =

pi_ = (unavailable - internal error)

m_ObjectBounds =

m_Data = [2]

X = (unavailable - internal error)

Y = (unavailable - internal error)

Z = (unavailable - internal error)

X = (unavailable - internal error)

Y = (unavailable - internal error)

Z = (unavailable - internal error)

m_Anim = (unavailable - internal error)

m_NextAnim = (unavailable - internal error)

m_AnimSpeed = (unavailable - internal error)

m_AnimTime = (unavailable - internal error)

m_BoneMatrices = (unavailable - internal error)

m_InverseBindBoneMatrices = (unavailable - internal error)

m_Props = (error -100406 while analyzing vector<CModel::Prop >) m_PositionValid = (unavailable - internal error)

m_ShadingColor =

r = (unavailable - internal error)

g = (unavailable - internal error)

b = (unavailable - internal error)

a = (unavailable - internal error)

m_SkeletonAnimManager = (unavailable - internal error)

UnitAttach = 0x00000000 CameraTargets = (error -100500 while analyzing vector<CVector3D >) ZoomDelta = 0.000000 (0x00000000)

m_pLocalPlayer = 0x00000000 m_Players = (error -100500 while analyzing vector<CPlayer * >) m_NumPlayers = 0 (0x00000000) m_GameStarted = false m_SimRate = 1.000000 (0x3F800000) GameStatus = 1551066223 m_Paused = false

pGameAttributes = 0x0EACBE10 ->

(IJSObject) m_JS = 0x02815388 -> (JSObject) m_NonsharedProperties =

(error -100406 while analyzing _Container_base) comp = (CStrW_hash_compare)

_List = (error -100500 while analyzing list<pair<CStrW const ,IJSProperty *> >) _Vec = (error -100406 while analyzing vector<list<pair<CStrW const ,IJSProperty *> >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_ScriptProperties =

(error -100406 while analyzing _Container_base) comp = (CStrW_hash_compare)

_List = (error -100500 while analyzing list<pair<CStrW const ,IJSProperty *> >) _Vec = (error -100406 while analyzing vector<list<pair<CStrW const ,IJSProperty *> >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

m_EngineOwned = true

m_SynchedProperties =

(error -100406 while analyzing _Container_base) comp = (CStrW_hash_compare)

_List = (error -100500 while analyzing list<pair<CStrW const ,ISynchedJSProperty *> >) _Vec = (error -100406 while analyzing vector<list<pair<CStrW const ,ISynchedJSProperty *> >::_Iterator<0> >) _Mask = 15 (0x0000000F) _Maxidx = 16 (0x00000010) _Max_bucket_size = 1.000000 (0x3F800000)

(Singleton<CGameAttributes>) m_MapFile = { (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >) } m_ResourceLevel = { (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >) } m_StartingPhase = { (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >) } m_GameMode = { (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >) } m_LOSSetting = 0 (0x00000000) m_FogOfWar = true m_ScreenshotMode = false m_NumSlots = 8 (0x00000008) m_Players = (error -100500 while analyzing vector<CPlayer * >) m_PlayerSlots = (error -100500 while analyzing vector<CPlayerSlot * >) m_PlayerSlotArrayJS = 0x02815380 -> (JSObject) m_UpdateCB = 0x00000000 m_UpdateCBData = 0x000055D4 m_PlayerUpdateCB = 0x00000000 (too much output; skipping to next top-level symbol)

CreateServer (scriptglue.cpp:695)

cx = 0x028BB6D8 -> (JSContext) formal = 0x028BCFC8 -> (JSObject) argc = 246895736 (0x0EB75478) argv = 0x027F4FA4 -> 0 (0x00000000) rval = 0x002FF17C -> 2147483649 (0x80000001)

js_Invoke (:0)

errno = 0 (?) OS error = Se ha intentado tener acceso a una dirección no válida.

In MP game. When playing 1 vs 1. The host sees his name fine, but his opponents name as "Player", while opponents name is Matei (as in the default configuration.). The opponent sees his and host's name fine.

In an 1v1 game of Oasis, the stone resource counter was observed to be -1. After starting at 100 and building two houses (50 stone each), it should have been zero (and probably was). After a while of further play - without any harvesting of stone - the resource counter was shown as "-1". Failed to duplicate this in SP, so it's probably related to network transmission of the value.

Need to set up a new autobuilder, so that the code can be compiled and the binaries committed to SVN from a stable consistent build environment.

Some discussion in ​this thread.

Details on the planned setup: BuildServerSetup.

How to duplicate bug: Start a game. Select a citizen-soldier. Cycle through the construction list in the GUI. Select House. Begin placing house foundations by clicking rapidly twice in one spot. When construction is complete you can see two houses overlapping in the same spot (as long as both house variations have been randomized).

Building limits "per territory" and other building limit attributes should be dictated in the entity code rather than in a script.

Example: The Hellenes have 3 Special Buildings. Each one could have a different building limit criteria. Same for any other faction and any other building type (for instance, what if we want the Romans to be able to build 3 Barracks per territory instead of the 2 dictated in the script? The Persians are supposed to be able to only build one "Kahk" Special Building per map, not per territory as is dictated in the script).

The DCDT code's license​ says

  Permission to use this software for educational, research and non-
  profit purposes, without fee, and without a written agreement is
  hereby granted, provided that the above notice and the following
  three paragraphs appear in all copies.

which means it is incompatible with the GPL, so it can't be distributed as part of the GPL release. So it has to be removed from that release, or removed from our repository entirely, or we have to get permission to use under GPL-compatible terms.

In Atlas buildings have their proper orientation (the "front" of the building should usually face towards the lower right hand side of the screen - there are some exceptions that face towards the lower left corner of the screen, but the point is, they should face at a 45 degree angle, like a diamond, from the default viewing angle), but when constructed in-game, structures face 135 degrees the wrong direction so that the "fronts" of the buildings now face the exact opposite direction from the camera.

This occurs with every Faction's buildings, no matter if they are PMD meshes or DAE Collada meshes.

Similar to the bug which made it possible to move around an enemy players units ( ​http://trac.wildfiregames.com/ticket/212 ) I've found that it's possible to control an enemy's buildings. After having had the building within LOS of one of my units I can control the building, and create units. On Latium I just get a message telling me that I cannot build since there aren't enough housing, since the enemy doesn't have any houses but just the CC, but on other maps with enough housing for the enemy player it's possible to build units.

svn_revision.cpp relies on svn_revision.txt, but that latter file is never actually generated, which breaks the build.

One possible solution: add an svn_revision.txt to SVN which just says "unknown". Don't ever update it for normal developers, because they can work out what they're doing themselves. Then just modify the autobuilder script to run svn_revision.bat and overwrite that file, so the automated builds will identity the revision correctly.

Another possible solution: make svn_revision.bat get called automatically by the VS project. But if it's called for every build, it needs to be careful to avoid triggering a load of unnecessary linking every time. Also it needs to be added on Linux too.

(I'd be happy to help with this, particularly any autobuilder or Linux stuff, depending on what the desired solution is.)

Travis and Brian reported errors in wnuma.cpp like:

'valid' : is not a member of '_PSAPI_WORKING_SET_EX_BLOCK'
'LargePageSize': identifier not found
'node' : is not a member of '_PSAPI_WORKING_SET_EX_BLOCK'

That code is inside an #if WINVER >= 0x600. It compiles fine for me in VS2008 on Vista; I guess it's a problem with an older version of the Platform SDK or something.

(I suggested to them that they just changed the code to #if 0 to get it to compile.)

Set up the access permissions so anonymous users can access the public files.

Work out which parts of Trac should be available to normal users, and make them available.

For the purposes of relicensing as open source, it'd be good to know exactly what code was written by who, so those people can be contacted to make sure they're okay with it.

Collada prop point "root" cannot hand more than one prop at a time. It is essential that collada prop points be able to hand more than one prop at a time, just like the old pmd files.

It is imperative to the game as designed that the game engine supports "static" "non-boned" Collada files, such as those created for ships and buildings, that include multiple prop-points. This is mainly for unit garrisoning purposes (units garrisoned on walls, battlements, towers, and on the decks of ships).

I am slating this for Beta, because for Alpha we could conceivably get by with just simple garrisoning capability without the propping. But for any public beta this feature should be implemented, as it is a major stated feature of the project.

It'd be nice to use proper validator tools to report errors in XML tools. So:

• Write RelaxNG Compact schemas for all the XML formats. (See existing DTDs for GUI and terrain files.)
• Add a script to convert them to RelaxNG XML.
• Change the CXeromyces API to accept a schema argument.
• Use that schema as part of the file hash, and use it to validate the document.
• Possibly refuse to load invalid documents? Or just report the error (and cache the error message in the XMB)? Currently we report an error.
• Write an external tool that validates all the XML files at once.

Editing system.cfg is nasty, because you get merge conflicts and accidental commits etc. So people should always use user-specific configuration, which looks like it's already implemented in data/profiles/default/settings/user.cfg. So: make sure that actually works, delete the non-default profiles because they're confusing and unneccessary, and update all the relevant documentation (and add a comment in system.cfg) to say to edit user.cfg.

CLogger is unnecessarily complex. Get rid of the LOG_CATEGORY and user-configurable interestingness stuff, and simplify the macros to LOG_ERROR("whatever") etc.

I first saw this when constructing a building, but had trouble reproducing it so I thought I'd wait until I saw it again, but then I saw that the same thing occurred when enemy units were attacking my units. The enemy unit were on the opposite side of a building from my units, but still hitpoints were subtracted from one of my units. It wasn't a ranged unit, and there were no other units nearby, so for some reason the animation/model wasn't "at the place where the simulation occurred" or how I should say.

The same thing with the units constructing a building. As you can see on the screenshot I attached the units that are constructing the building are at a distance from it, though using the proper animation and everything else works.

In both cases it's a building that's between the unit (or rather the animation/model) and the action/simulation that's actually taking place. Not sure if that's just a coincidence, but I thought it was worth mentioning.

Also as it's not interrupting the gameplay or anything it shouldn't be on this milestone, but I don't know on which it fits best, so someone who knows that better should move it, please.

When a config file says something like

key = value ; comment

the config parser thinks the key is associated with two values, "value" and " ; comment". (See mainlog.html, where it reports all the config values that were loaded.)

This needs to be fixed, so legitimate (comma-separated) values are kept but comments are discarded.

you can view a list of all tickets but cant actually read the content of tickets if you're not registered / logged in.

It would help to check for more information if you got a problem to be able to check what problems and possible solution and workarounds other people reported.

There were some missing include errors while I was building 0 a.d. on FreeBSD. Maybe that's because I was building with precompiled headers (doesn't seem to work nice with ccache), but I guess it's better to add them.

build/premake/src/Src/os.h: check for FreeBSD correctly source/tools/atlas/AtlasScript/ScriptInterface.cpp: use valgrind only on Linux and MacOS X (not on FreeBSD) source/ps/DllLoader.cpp: add missing const build/workspaces/update-workspaces.sh: use plain sh instead of bash (no bash-specific features are used and bash is not available on FreeBSD by default). Also substitute positional parameters correctly

Collada tries to load plugins from applicationFolderName without checking whether it is empty first, which leads to crash on FreeBSD, where it is empty. While actually there should be FreeBSD version of FUFileManager::GetApplicationFolderName implemented, here where we don't need to load any plugins (do we?) this fix should go.

in libraries/cxxtest/include/cxxtest/ValueTraits.h:281

#ifndef __APPLE__ // avoid redefinition errors on mac
    CXXTEST_COPY_TRAITS( size_t, const unsigned int ); // avoid /Wp64 warnings in MSVC
#endif

There is a check to make sure redefinition errors do not happen on mac. The check also needs to apply on linux.

In the help page for building on linux (​http://trac.wildfiregames.com/wiki/BuildInstructions#Installingthread-safeSpiderMonkey) The recommended version of spider monkey causes build errors with 0AD. I installed the latest version of spider monkey (changeset: 30318) from ​https://developer.mozilla.org/En/SpiderMonkey/Getting_SpiderMonkey_source_code#Getting_the_latest_SpiderMonkey_source_code

And copied header files 0AD wanted (jsosdep.h and jsconfig.h) that were not present in the latest Spider Monkey revision from an older version to /usr/include/js

0AD then compiled successfully. There was an error (which seemed to have no effect) when launching a single player game but I don't know if it is related to Spider Monkey or what effects the error had on the game.

exe name before realpath: /Volumes/Data/Development/Compiling/0AD/trunk/binaries/system/./pyrogenesis_dbg
exe name after realpath: /Volumes/Data/Development/Compiling/0AD/trunk/binaries/system/./pyrogenesis_dbg
returning exe name: /Volumes/Data/Development/Compiling/0AD/trunk/binaries/system/pyrogenesis_dbg
TIMER| InitVfs: 1.603 ms
TIMER| InitScripting: 7.995 ms
TIMER| CONFIG_Init: 8.452 ms
TIMER| write_sys_info: 1.098 ms
TIMER| ps_console: 17.498 ms
TIMER| ps_lang_hotkeys: 6.716 ms
TIMER| ps_gui_init: 187 us
TIMER| ps_gui_setup_xml: 4.976 ms
TIMER| ps_gui_styles_xml: 2.096 ms
TIMER| ps_gui_sprite1_xml: 61.773 ms
(current document)(0): malformed UTF-8 character sequence at offset 25199
ERROR: JavaScript Error ((current document), line 0): malformed UTF-8 character sequence at offset 25199
terminate called after throwing an instance of 'PSERROR_Scripting_LoadFile_EvalErrors'
  what():  Scripting_LoadFile_EvalErrors
Abort trap

I eventually tracked this down to gui/test/functions_page_session_status_commands.js, which was in some ISO-* encoding, instead of UTF-8. This problem exists for quite a few other files as well. Spidermonkey fails to parse them, and dies.

hi

i was building the game on a kubuntu 9.04 using the instructions on ​http://trac.wildfiregames.com/wiki/BuildInstructions however if i start the ./pyrogenesis_dbg all i get is

TIMER| InitVfs: 1.64816 ms TIMER| InitScripting: 55.6995 ms TIMER| CONFIG_Init: 30.1024 ms TIMER| write_sys_info: 1.48977 ms TIMER| ps_console: 12.7604 ms TIMER| ps_lang_hotkeys: 11.0896 ms TIMER| ps_gui_init: 45.814 us TIMER| ps_gui_setup_xml: 30.3468 ms TIMER| ps_gui_styles_xml: 4.25142 ms TIMER| ps_gui_sprite1_xml: 220.275 ms gui/test/functions_page_session.js(280): ReferenceError: addGlobalHandler is not defined ERROR: JavaScript Error (gui/test/functions_page_session.js, line 280): ReferenceError: addGlobalHandler is not defined terminate called after throwing an instance of 'PSERROR_Scripting_LoadFile_EvalErrors'

what(): Scripting_LoadFile_EvalErrors

Aborted

a windows pops up for a second and closes again.

did i do anything wrong? i even built the spidermonkey lib threadsafe for my system

cd311

libxml2 checks for thread support by not only checking if it was compiled in, but also by checking whether "_REENTRANT" or "MT" is set. Usually, AFAIK, these are set on *nix when pthread is included, but since on OS X GNU threads are included in libSystem, this does not happen.

I propose the attached fix, but if anyone has a better idea, feel free.

"lib.tests/test_secure_crt.h" fails to compile on Mac OS X, as Mac OS X does not have a included "strnlen".

A crappy fix is attached, because I fail at C.

from lib/external_libraries/openal.h, line 34:

// ALC strings (e.g. device and extension names) are typed differently
// between platforms *sigh*
#if OS_MACOSX
typedef ALCubyte* alcString;
#else
typedef ALCchar* alcString;
#endif

This is incorrect, and creates compilation errors. It is not platform dependent - using ALCubyte was changed to using ALCchar in the 1.1 spec of OpenAL. It should be identified through AL_VERSION_1_1 (Unless you can think of a better way, it's not 100% accurate)

A fix is attached.

Basically this:

NetClient.cpp
../../../source/network/NetClient.cpp: In static member function 'static bool CNetClient::OnError(void*, CFsmEvent*)':
../../../source/network/NetClient.cpp:253: warning: comparison between signed and unsigned integer expressions
../../../source/network/NetClient.cpp: In static member function 'static bool CNetClient::OnAuthenticate(void*, CFsmEvent*)':
../../../source/network/NetClient.cpp:367: warning: comparison between signed and unsigned integer expressions
NetServer.cpp
../../../source/network/NetServer.cpp: In static member function 'static bool CNetServer::OnError(void*, CFsmEvent*)':
../../../source/network/NetServer.cpp:395: warning: comparison between signed and unsigned integer expressions
Util.cpp
../../../source/ps/Util.cpp: In function 'void WriteSystemInfo()':
../../../source/ps/Util.cpp:96: warning: format '%d' expects type 'int', but argument 5 has type 'size_t'
../../../source/ps/Util.cpp:96: warning: format '%d' expects type 'int', but argument 6 has type 'size_t'
../../../source/ps/Util.cpp:96: warning: format '%d' expects type 'int', but argument 7 has type 'size_t'
../../../source/ps/Util.cpp:109: warning: format '%u' expects type 'unsigned int', but argument 3 has type 'size_t'
../../../source/ps/Util.cpp:109: warning: format '%u' expects type 'unsigned int', but argument 4 has type 'size_t'
CLogger.cpp
../../../source/ps/CLogger.cpp:171: warning: unused parameter 'category'
../../../source/ps/CLogger.cpp:188: warning: unused parameter 'category'
Parser.cpp
../../../source/ps/Parser.cpp: In member function 'bool CParserLine::ParseString(const CParser&, const std::string&)':
../../../source/ps/Parser.cpp:557: warning: comparison between signed and unsigned integer expressions
../../../source/ps/Parser.cpp:605: warning: comparison between signed and unsigned integer expressions
../../../source/ps/Parser.cpp:657: warning: comparison between signed and unsigned integer expressions
../../../source/ps/Parser.cpp:793: warning: comparison between signed and unsigned integer expressions
EntityManager.cpp
../../../source/simulation/EntityManager.cpp: In member function 'HEntity CEntityManager::Create(CEntityTemplate*, CVector3D, float, const std::set<CStr8, std::less<CStr8>, std::allocator<CStr8> >&, const CStrW*)':
../../../source/simulation/EntityManager.cpp:225: warning: comparison between signed and unsigned integer expressions
ScriptGlue.cpp
../../../source/scripting/ScriptGlue.cpp: In function 'void InitJsTimers()':
../../../source/scripting/ScriptGlue.cpp:648: warning: format '%d' expects type 'int', but argument 3 has type 'size_t'
Terrain.cpp
../../../source/graphics/Terrain.cpp: In member function 'void CTerrain::Resize(ssize_t)':
../../../source/graphics/Terrain.cpp:368: warning: comparison between signed and unsigned integer expressions
Renderer.cpp
../../../source/renderer/Renderer.cpp: In member function 'virtual CStr8 CRendererStatsTable::GetCellText(size_t, size_t)':
../../../source/renderer/Renderer.cpp:152: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:152: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:158: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:158: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:164: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:164: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:170: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:170: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:176: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
../../../source/renderer/Renderer.cpp:176: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int'
WaterManager.cpp
../../../source/renderer/WaterManager.cpp: In member function 'int WaterManager::LoadWaterTextures()':
../../../source/renderer/WaterManager.cpp:114: warning: format '%02d' expects type 'int', but argument 5 has type 'size_t'
../../../source/renderer/WaterManager.cpp:114: warning: format '%02d' expects type 'int', but argument 5 has type 'size_t'
../../../source/renderer/WaterManager.cpp:131: warning: format '%02d' expects type 'int', but argument 5 has type 'size_t'
../../../source/renderer/WaterManager.cpp:131: warning: format '%02d' expects type 'int', but argument 5 has type 'size_t'
MessagePasserImpl.cpp
../../../source/tools/atlas/GameInterface/MessagePasserImpl.cpp: In constructor 'MessagePasserImpl::MessagePasserImpl()':
../../../source/tools/atlas/GameInterface/MessagePasserImpl.cpp:37: warning: format '%d' expects type 'int', but argument 3 has type 'size_t'
ldbg.cpp
../../../source/lib/sysdep/os/linux/ldbg.cpp: In function 'void* debug_GetCaller(void*, const char*)':
../../../source/lib/sysdep/os/linux/ldbg.cpp:87: warning: unused variable 'bt_size'
../../../source/lib/sysdep/os/linux/ldbg.cpp: In function 'LibError debug_resolve_symbol_dladdr(void*, char*, char*, int*)':
../../../source/lib/sysdep/os/linux/ldbg.cpp:303: warning: format '%08x' expects type 'unsigned int', but argument 4 has type 'size_t'
../../../source/lib/sysdep/os/linux/ldbg.cpp:303: warning: format '%08x' expects type 'unsigned int', but argument 4 has type 'size_t'
item.cpp
../../../source/tools/atlas/wxJS/gui/control/item.cpp: In static member function 'static bool wxjs::gui::ControlItem::GetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/item.cpp:84: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/gui/control/item.cpp: In static member function 'static bool wxjs::gui::ControlItem::SetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/item.cpp:108: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/gui/control/item.cpp: In static member function 'static JSBool wxjs::gui::ControlItem::remove(JSContext*, JSObject*, uintN, jsval*, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/item.cpp:145: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/gui/control/item.cpp: In static member function 'static JSBool wxjs::gui::ControlItem::select(JSContext*, JSObject*, uintN, jsval*, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/item.cpp:172: warning: comparison between signed and unsigned integer expressions
radioboxit.cpp
../../../source/tools/atlas/wxJS/gui/control/radioboxit.cpp: In static member function 'static bool wxjs::gui::RadioBoxItem::GetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/radioboxit.cpp:68: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/gui/control/radioboxit.cpp:79: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/gui/control/radioboxit.cpp: In static member function 'static bool wxjs::gui::RadioBoxItem::SetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/gui/control/radioboxit.cpp:109: warning: comparison between signed and unsigned integer expressions
rect.cpp
../../../source/tools/atlas/wxJS/gui/misc/rect.cpp: In static member function 'static JSBool wxjs::gui::Rect::inside(JSContext*, JSObject*, uintN, jsval*, jsval*)':
../../../source/tools/atlas/wxJS/gui/misc/rect.cpp:464: warning: 'bool wxRect::Inside(int, int) const' is deprecated (declared at /usr/include/wx-2.8/wx/gdicmn.h:490)
../../../source/tools/atlas/wxJS/gui/misc/rect.cpp:477: warning: 'bool wxRect::Inside(const wxPoint&) const' is deprecated (declared at /usr/include/wx-2.8/wx/gdicmn.h:491)
sizer.cpp
../../../source/tools/atlas/wxJS/gui/misc/sizer.cpp: In static member function 'static JSBool wxjs::gui::Sizer::remove(JSContext*, JSObject*, uintN, jsval*, jsval*)':
../../../source/tools/atlas/wxJS/gui/misc/sizer.cpp:534: warning: 'virtual bool wxSizer::Remove(wxWindow*)' is deprecated (declared at /usr/include/wx-2.8/wx/sizer.h:513)
textline.cpp
../../../source/tools/atlas/wxJS/io/textline.cpp: In static member function 'static bool wxjs::io::TextLine::GetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/io/textline.cpp:102: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/io/textline.cpp: In static member function 'static bool wxjs::io::TextLine::SetProperty(wxjs::Index*, JSContext*, JSObject*, int, jsval*)':
../../../source/tools/atlas/wxJS/io/textline.cpp:134: warning: comparison between signed and unsigned integer expressions
../../../source/tools/atlas/wxJS/io/textline.cpp: In static member function 'static bool wxjs::io::TextLine::Enumerate(wxjs::Index*, JSContext*, JSObject*, JSIterateOp, jsval*, jsid*)':
../../../source/tools/atlas/wxJS/io/textline.cpp:186: warning: comparison between signed and unsigned integer expressions
DLLInterface.cpp
In file included from /usr/include/libxml2/libxml/parser.h:15,
from ../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp:42:
/usr/include/libxml2/libxml/xmlversion.h:429:1: warning: "ATTRIBUTE_PRINTF" redefined
In file included from /usr/include/wx-2.8/wx/wxprec.h:13,
from ../../../source/tools/atlas/AtlasUI/Misc/precompiled.h:43,
from ../../../source/tools/atlas/AtlasUI/Misc/precompiled.cpp:18:
/usr/include/wx-2.8/wx/defs.h:501:1: warning: this is the location of the previous definition
Trigger.cpp
../../../source/tools/atlas/AtlasUI/ScenarioEditor/Sections/Trigger/Trigger.cpp:1049: warning: unused parameter 'label'
DLL.cpp
../../../source/collada/DLL.cpp: In member function 'virtual void BufferedOutputCallback::operator()(const char*, unsigned int)':
../../../source/collada/DLL.cpp:78: warning: comparison between signed and unsigned integer expressions
../../../source/collada/DLL.cpp:84: warning: comparison between signed and unsigned integer expressions
test_byte_order.cpp
In file included from ../../../source/lib/self_test.h:190,
from ../../../source/pch/test/precompiled.h:24,
from ../../../source/pch/test/precompiled.cpp:25:
../../../libraries/cxxtest/include/cxxtest/TestSuite.h: In function 'bool CxxTest::equals(X, Y) [with X = long int, Y = long long unsigned int]':
../../../libraries/cxxtest/include/cxxtest/TestSuite.h:56: instantiated from 'void CxxTest::doAssertEquals(const char*, unsigned int, const char*, X, const char*, Y, const char*) [with X = long int, Y = long long unsigned int]'
../../../source/lib/tests/../../../source/lib/tests/test_byte_order.h:83: instantiated from here
../../../libraries/cxxtest/include/cxxtest/TestSuite.h:47: warning: comparison between signed and unsigned integer expressions
test_allocators.cpp
In file included from ../../../source/lib/self_test.h:190,
from ../../../source/pch/test/precompiled.h:24,
from ../../../source/pch/test/precompiled.cpp:25:
../../../libraries/cxxtest/include/cxxtest/TestSuite.h: In function 'bool CxxTest::equals(X, Y) [with X = unsigned int, Y = int]':
../../../libraries/cxxtest/include/cxxtest/TestSuite.h:56: instantiated from 'void CxxTest::doAssertEquals(const char*, unsigned int, const char*, X, const char*, Y, const char*) [with X = unsigned int, Y = int]'
../../../source/pch/test/../../../source/lib/allocators/tests/test_allocators.h:42: instantiated from here
../../../libraries/cxxtest/include/cxxtest/TestSuite.h:47: warning: comparison between signed and unsigned integer expressions

Game crashed on startup. I have a pretty crappy graphics card - Mobile Intel 945 Express Chipset Family - so that might be the cause.

-Fixing so that when Villager/builder woman has built a farm they will automatically start farming on it, like when a villager has finished chopping a tree they move to the next possible one.

-It only show a blue screen where the map is supposed to be.

​http://zyk-production.com/contributions/atlas.jpg

Would like the renderer to have distance fog implemented and able to be enabled on a map through the Atlas map editor. Some variables would include:

• Thickness
  

• Tint
  

• Distance (the distance that the background becomes 100% obscured by the fog)
  

• anything else?
  

This would be a good way to visually mask the fact that we don't have other whiz-bang graphical features like normal maps and suchness.

-When dragging the selection box over screen, only select military units if there are any, if not, then make it only choose villagers, and only let 1 building get selected at once instead of(lets say) 10, currently if you drag you get villagers,buildings and military units in one selection

-The way it is now, you have to wait a certain time before you can click the create "any unit here" or else it doesnt stack the order, making it painfully slow when a player have to build 1+ unit, it should register per click and not have what seems to be a wait function between each possible click.

-0.5 sec to 1 sec before able to queue another unit into production queue everytime is very slow.

On Linux: Delete binaries/data/cache. Run the game. Get an error:

/usr/include/boost/shared_ptr.hpp:375: T* boost::shared_ptr<T>::operator->() const [with T = RealDirectory]: Assertion `px != 0' failed.

vfs_lookup.cpp​ attempts to create the directory with mkdir. Two problems:

First: It appears to call mkdir("/.../binaries/data/") with subdirectoryName == "cache". Then it calls mkdir("/.../binaries/data/cache/mods") with subdirectoryName == "mods". It never actually creates cache, so it fails.

Second: If mkdir fails, then no error is reported, and the shared_ptr assertion occurs arbitrarily later. If it returns -1 and errno != EEXIST then the error should probably be reported immediately.

Hello, I try to compile wildfire on ubuntu jaunty following your build instructions.

Now there is an error while compiling the AtlasUI. The output is:

==== Building AtlasUI ====
precompiled.cpp
ActorEditor.cpp
AnimListEditor.cpp
ActorEditorListCtrl.cpp
PropListEditor.cpp
ActorViewer.cpp
ColourTesterFileCtrl.cpp
ColourTester.cpp
ColourTesterColourCtrl.cpp
ColourTesterImageCtrl.cpp
ActionButton.cpp
ToolButton.cpp
Canvas.cpp
ColourDialog.cpp
DraggableListCtrl.cpp
DraggableListCtrlCommands.cpp
ListCtrlValidator.cpp
QuickComboBox.cpp
QuickTextCtrl.cpp
EditableListCtrl.cpp
FieldEditCtrl.cpp
QuickFileCtrl.cpp
EditableListCtrlCommands.cpp
FileHistory.cpp
HighResTimer.cpp
SnapSplitterWindow.cpp
virtualdirtreectrl.cpp
AtlasWindow.cpp
AtlasDialog.cpp
ErrorReporter.cpp
Observable.cpp
AtlasClipboard.cpp
AtlasEventLoop.cpp
Datafile.cpp
AtlasWindowCommand.cpp
AtlasWindowCommandProc.cpp
FFmpeg.cpp
VideoRecorder.cpp
DLLInterface.cpp
SectionLayout.cpp
ScenarioEditor.cpp
Sidebar.cpp
Cinematic.cpp
LightControl.cpp
Environment.cpp
Terrain.cpp
Trigger.cpp
../../../source/tools/atlas/AtlasUI/ScenarioEditor/Sections/Trigger/Trigger.cpp:1050: warning: unused parameter 'label'
ScriptedTool.cpp
AlterElevation.cpp
PlaceObject.cpp
FlattenElevation.cpp
PaintTerrain.cpp
TransformObject.cpp
ObjectSettings.cpp
MiscState.cpp
Brushes.cpp
Tools.cpp
Linking AtlasUI
obj/AtlasUI_Debug/ColourTesterColourCtrl.o: In function `CustomColourButton::OnButton(wxCommandEvent&)':
ColourTesterColourCtrl.cpp:(.text._ZN18CustomColourButton8OnButtonER14wxCommandEvent[CustomColourButton::OnButton(wxCommandEvent&)]+0x92): undefined reference to `wxColour::IsOk() const'
obj/AtlasUI_Debug/FieldEditCtrl.o: In function `wxColour::Ok() const':
/usr/include/wx-2.8/wx/gtk/colour.h:30: undefined reference to `wxColour::IsOk() const'
/usr/bin/ld: ../../../binaries/system/libAtlasUI_dbg.so: hidden symbol `wxColour::IsOk() const' isn't defined
/usr/bin/ld: final link failed: Nonrepresentable section on output
collect2: ld returned 1 exit status
make[2]: *** [../../../binaries/system/libAtlasUI_dbg.so] Error 1
make[1]: *** [AtlasUI] Error 2
make[1]: Leaving directory `/build/korn-wildfire_0svn7029-1~getdeb1-amd64-L8XWbo/wildfire-0svn7029-1~getdeb1/build/workspaces/gcc'
make: *** [build-stamp] Error 2
dpkg-buildpackage: failure: debian/rules build gave error exit status 2

In debug.cpp, debug_WriteCrashlog tries to open crashlog.txt for writing. If it fails (e.g. it doesn't have write permission to that directory), it resets isBusy and then does WARN_RETURN, which calls debug_DisplayError, which calls debug_WriteCrashlog, and isBusy is false so it loops forever.

See ​http://ruby.pastebin.com/m53d88e4a for an example of this.

lib/file/path.h:

    const LibError PATH_ROOT_DIR_ALREADY_SET = -110200;
    const LibError PATH_NOT_IN_ROOT_DIR      = -110201;

lib/file/file.h:

    const LibError FILE_ACCESS = -110200;
    const LibError IO          = -110201;

Confusion ensues.

I use a laptop with vista OS to open the pyrogenesis.exe, then it crash.

Much to our regret we must report the program has encountered an error.

Details: unhandled exception (Access violation writing 0x00000000)

Location: unknown:0 (?)

Call stack:

(error while dumping stack: No stack frames found) errno = 0 (?) OS error = Accesso a posizione di memoria non valido.

when placing a dock if location is invalid you get this error

gui/test/functions_sim_entity.js(254): Index out of bounds. ERROR: JavaScript Error (gui/test/functions_sim_entity.js, line 254): Index out of bounds. gui/test/functions_sim_entity.js(254): Invalid parameter(s) or count ERROR: JavaScript Error (gui/test/functions_sim_entity.js, line 254): Invalid parameter(s) or count

but are still allowed to build dock even if its underwater also it changes the land around build location and the land doesnt fix its self if you delete the building

On Linux, if the game is run from outside the 'system' directory, it can't find ./libCollada_dbg.so and dies. It should probably look in the game executable's path to find the library (perhaps using sys_get_executable_name).

This is the result I got from following the instructions on the wiki om my Ubuntu 8.10 laptop

oldrocker99@oldrocker99-laptop:~/games/stargus-0.2/trunk/build/workspaces$ cd gcc oldrocker99@oldrocker99-laptop:~/games/stargus-0.2/trunk/build/workspaces/gcc$ make -j2

Building network

Building engine

precompiled.cpp NetClient.cpp In file included from ../../../source/pch/engine/precompiled.h:30,

from ../../../source/pch/engine/precompiled.cpp:18:

../../../source/scripting/SpiderMonkey.h:45:22: error: js/jsapi.h: No such file or directory ../../../source/scripting/SpiderMonkey.h:49:26: error: js/jsdbgapi.h: No such file or directory In file included from ../../../source/scripting/ScriptingHost.h:42,

from ../../../source/scripting/JSConversions.h:23, from ../../../source/scripting/JSSerialization.h:23, from ../../../source/network/NetMessages.h:31, from ../../../source/network/NetMessage.h:38, from ../../../source/network/Network.h:68, from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/SpiderMonkey.h:45:22: error: js/jsapi.h: No such file or directory ../../../source/scripting/SpiderMonkey.h:49:26: error: js/jsdbgapi.h: No such file or directory In file included from ../../../source/scripting/SpiderMonkey.h:54,

from ../../../source/scripting/ScriptingHost.h:42, from ../../../source/scripting/JSConversions.h:23, from ../../../source/scripting/JSSerialization.h:23, from ../../../source/network/NetMessages.h:31, from ../../../source/network/NetMessage.h:38, from ../../../source/network/Network.h:68, from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/JSUtil.h:20: error: ‘jsval’ does not name a type In file included from ../../../source/scripting/JSConversions.h:23,

from ../../../source/scripting/JSSerialization.h:23, from ../../../source/network/NetMessages.h:31, from ../../../source/network/NetMessage.h:38, from ../../../source/network/Network.h:68, from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/ScriptingHost.h:72: error: ISO C++ forbids declaration of ‘JSObject’ with no type ../../../source/scripting/ScriptingHost.h:72: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:73: error: ISO C++ forbids declaration of ‘JSClass’ with no type ../../../source/scripting/ScriptingHost.h:73: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:79: error: ISO C++ forbids declaration of ‘JSRuntime’ with no type ../../../source/scripting/ScriptingHost.h:79: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:80: error: ISO C++ forbids declaration of ‘JSContext’ with no type ../../../source/scripting/ScriptingHost.h:80: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:81: error: ISO C++ forbids declaration of ‘JSObject’ with no type ../../../source/scripting/ScriptingHost.h:81: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:83: error: ‘JSErrorReport’ does not name a type ../../../source/scripting/ScriptingHost.h:89: error: expected ‘;’ before ‘(’ token ../../../source/scripting/ScriptingHost.h:91: error: expected ‘;’ before ‘(’ token ../../../source/scripting/ScriptingHost.h:103: error: ISO C++ forbids declaration of ‘JSContext’ with no type ../../../source/scripting/ScriptingHost.h:103: error: ‘JSContext’ declared as an ‘inline’ field ../../../source/scripting/ScriptingHost.h:103: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:104: error: expected `;' before ‘inline’ ../../../source/scripting/ScriptingHost.h:104: error: ISO C++ forbids declaration of ‘JSContext’ with no type ../../../source/scripting/ScriptingHost.h:104: error: ‘JSContext’ declared as an ‘inline’ field ../../../source/scripting/ScriptingHost.h:104: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:105: error: expected `;' before ‘inline’ ../../../source/scripting/ScriptingHost.h:105: error: ISO C++ forbids declaration of ‘JSObject’ with no type ../../../source/scripting/ScriptingHost.h:105: error: ‘JSObject’ declared as an ‘inline’ field ../../../source/scripting/ScriptingHost.h:105: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:107: error: expected `;' before ‘void’ ../../../source/scripting/ScriptingHost.h:107: error: ‘JSObject’ has not been declared ../../../source/scripting/ScriptingHost.h:108: error: ‘JSObject’ has not been declared ../../../source/scripting/ScriptingHost.h:111: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:113: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:115: error: ‘JSNative’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSClass’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSNative’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘uintN’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSPropertySpec’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSFunctionSpec’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSPropertySpec’ has not been declared ../../../source/scripting/ScriptingHost.h:120: error: ‘JSFunctionSpec’ has not been declared ../../../source/scripting/ScriptingHost.h:122: error: ISO C++ forbids declaration of ‘JSObject’ with no type ../../../source/scripting/ScriptingHost.h:122: error: expected ‘;’ before ‘*’ token ../../../source/scripting/ScriptingHost.h:124: error: ‘JSObject’ has not been declared ../../../source/scripting/ScriptingHost.h:124: error: ‘jsval’ has not been declared ../../../source/scripting/ScriptingHost.h:125: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:127: error: ‘JSObject’ has not been declared ../../../source/scripting/ScriptingHost.h:128: error: ‘JSObject’ has not been declared ../../../source/scripting/ScriptingHost.h:130: error: ‘jsval’ has not been declared ../../../source/scripting/ScriptingHost.h:131: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:133: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:133: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:134: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:134: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:135: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:135: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:136: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:136: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:137: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:137: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:138: error: expected ‘,’ or ‘...’ before ‘value’ ../../../source/scripting/ScriptingHost.h:138: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptingHost.h:140: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:141: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptingHost.h:143: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptingHost.h:143: error: ‘JSErrorReport’ has not been declared In file included from ../../../source/scripting/JSSerialization.h:23,

from ../../../source/network/NetMessages.h:31, from ../../../source/network/NetMessage.h:38, from ../../../source/network/Network.h:68, from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/JSConversions.h:40: error: template declaration of ‘T* ToNative’ ../../../source/scripting/JSConversions.h:40: error: ‘JSContext’ was not declared in this scope ../../../source/scripting/JSConversions.h:40: error: ‘cx’ was not declared in this scope ../../../source/scripting/JSConversions.h:40: error: ‘JSObject’ was not declared in this scope ../../../source/scripting/JSConversions.h:40: error: ‘obj’ was not declared in this scope ../../../source/scripting/JSConversions.h:45: error: expected constructor, destructor, or type conversion before ‘*’ token ../../../source/scripting/JSConversions.h:52: error: template declaration of ‘T* ToNative’ ../../../source/scripting/JSConversions.h:52: error: ‘jsval’ was not declared in this scope ../../../source/scripting/JSConversions.h:59: error: template declaration of ‘bool ToPrimitive’ ../../../source/scripting/JSConversions.h:59: error: ‘JSContext’ was not declared in this scope ../../../source/scripting/JSConversions.h:59: error: expected primary-expression before ‘,’ token ../../../source/scripting/JSConversions.h:59: error: ‘jsval’ was not declared in this scope ../../../source/scripting/JSConversions.h:59: error: expected primary-expression before ‘&’ token ../../../source/scripting/JSConversions.h:59: error: ‘Storage’ was not declared in this scope ../../../source/scripting/JSConversions.h:68: error: template declaration of ‘bool ToPrimitive’ ../../../source/scripting/JSConversions.h:68: error: ‘JSContext’ was not declared in this scope ../../../source/scripting/JSConversions.h:68: error: expected primary-expression before ‘,’ token ../../../source/scripting/JSConversions.h:68: error: ‘jsval’ was not declared in this scope ../../../source/scripting/JSConversions.h:68: error: expected primary-expression before ‘*’ token ../../../source/scripting/JSConversions.h:68: error: ‘Storage’ was not declared in this scope ../../../source/scripting/JSConversions.h:77: error: ‘ToPrimitive’ declared as an ‘inline’ variable ../../../source/scripting/JSConversions.h:77: error: template declaration of ‘T ToPrimitive’ ../../../source/scripting/JSConversions.h:77: error: ‘JSContext’ was not declared in this scope ../../../source/scripting/JSConversions.h:77: error: ‘cx’ was not declared in this scope ../../../source/scripting/JSConversions.h:77: error: ‘jsval’ was not declared in this scope ../../../source/scripting/JSConversions.h:86: error: ‘ToPrimitive’ declared as an ‘inline’ variable ../../../source/scripting/JSConversions.h:86: error: template declaration of ‘T ToPrimitive’ ../../../source/scripting/JSConversions.h:86: error: ‘jsval’ was not declared in this scope ../../../source/scripting/JSConversions.h:91: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:96: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:101: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:110: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:111: error: expected constructor, destructor, or type conversion before ‘*’ token ../../../source/scripting/JSConversions.h:112: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:115: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:116: error: expected constructor, destructor, or type conversion before ‘*’ token ../../../source/scripting/JSConversions.h:119: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:120: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:123: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:124: error: expected constructor, destructor, or type conversion before ‘*’ token ../../../source/scripting/JSConversions.h:127: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:128: error: expected constructor, destructor, or type conversion before ‘*’ token ../../../source/scripting/JSConversions.h:131: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:132: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:135: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:136: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:137: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:140: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:141: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:142: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:145: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:146: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:147: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:150: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:151: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:152: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:173: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:174: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:175: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:178: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:179: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:180: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:183: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:184: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:185: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:195: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:196: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:197: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:200: error: expected initializer before ‘<’ token ../../../source/scripting/JSConversions.h:201: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:202: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:211: error: ‘jsval’ does not name a type ../../../source/scripting/JSConversions.h:212: error: expected `)' before ‘v’ ../../../source/scripting/JSConversions.h:214: error: expected constructor, destructor, or type conversion before ‘ToJSVal’ ../../../source/scripting/JSConversions.h:218: error: ‘jsval’ does not name a type In file included from ../../../source/network/NetMessages.h:31,

from ../../../source/network/NetMessage.h:38, from ../../../source/network/Network.h:68, from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/JSSerialization.h:37: error: ‘jsval’ does not name a type ../../../source/scripting/JSSerialization.h:42: error: expected `)' before ‘data’ ../../../source/scripting/JSSerialization.h:56: error: expected type-specifier before ‘jsval’ ../../../source/scripting/JSSerialization.h: In member function ‘jsval_ser::operator CStr8() const’: ../../../source/scripting/JSSerialization.h:62: error: ‘ToPrimitive’ was not declared in this scope ../../../source/scripting/JSSerialization.h:62: error: expected primary-expression before ‘>’ token ../../../source/scripting/JSSerialization.h:62: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h: In member function ‘virtual size_t jsval_ser::GetSerializedLength() const’: ../../../source/scripting/JSSerialization.h:76: error: ‘ToPrimitive’ was not declared in this scope ../../../source/scripting/JSSerialization.h:76: error: expected primary-expression before ‘>’ token ../../../source/scripting/JSSerialization.h:76: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h: In member function ‘virtual uint8_t* jsval_ser::Serialize(uint8_t*) const’: ../../../source/scripting/JSSerialization.h:92: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h:92: error: ‘JSVAL_TO_INT’ was not declared in this scope ../../../source/scripting/JSSerialization.h:103: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h:103: error: ‘JSVAL_TO_DOUBLE’ was not declared in this scope ../../../source/scripting/JSSerialization.h:108: error: ‘ToPrimitive’ was not declared in this scope ../../../source/scripting/JSSerialization.h:108: error: expected primary-expression before ‘>’ token ../../../source/scripting/JSSerialization.h:108: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h: In member function ‘virtual const uint8_t* jsval_ser::Deserialize(const uint8_t*, const uint8_t*)’: ../../../source/scripting/JSSerialization.h:122: error: ‘m_data’ was not declared in this scope ../../../source/scripting/JSSerialization.h:122: error: ‘JSVAL_FALSE’ was not declared in this scope ../../../source/scripting/JSSerialization.h:125: error: ‘JSVAL_TRUE’ was not declared in this scope ../../../source/scripting/JSSerialization.h:131: error: ‘INT_TO_JSVAL’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘uint32’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: expected `;' before ‘_v1’ ../../../source/scripting/JSSerialization.h:141: error: expected `;' before ‘_v2’ ../../../source/scripting/JSSerialization.h:141: error: ‘_v1’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v1’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v1’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v1’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v2’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v2’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v2’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v2’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v1’ was not declared in this scope ../../../source/scripting/JSSerialization.h:141: error: ‘_v2’ was not declared in this scope ../../../source/scripting/JSSerialization.h:142: error: ‘class ScriptingHost’ has no member named ‘GetContext’ ../../../source/scripting/JSSerialization.h:142: error: ‘JS_NewDoubleValue’ was not declared in this scope ../../../source/scripting/JSSerialization.h:149: error: ‘ToJSVal’ was not declared in this scope ../../../source/scripting/JSSerialization.h:149: error: expected primary-expression before ‘>’ token In file included from ../../../source/network/Network.h:68,

from ../../../source/network/NetSession.h:30, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/network/NetMessage.h: At global scope: ../../../source/network/NetMessage.h:165: error: ‘JSContext’ has not been declared ../../../source/network/NetMessage.h:166: error: ‘uintN’ has not been declared ../../../source/network/NetMessage.h:167: error: ‘jsval’ has not been declared In file included from ../../../source/scripting/SynchedJSObject.h:54,

from ../../../source/ps/Player.h:22, from ../../../source/ps/GameAttributes.h:21, from ../../../source/network/NetSession.h:32, from ../../../source/network/NetClient.h:30, from ../../../source/network/NetClient.cpp:28:

../../../source/scripting/ScriptableObject.h:37: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptableObject.h:38: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:38: error: ‘jsval’ has not been declared ../../../source/scripting/ScriptableObject.h:47: error: expected identifier before ‘*’ token ../../../source/scripting/ScriptableObject.h:47: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:47: error: ISO C++ forbids declaration of ‘jsval’ with no type ../../../source/scripting/ScriptableObject.h:47: error: ‘jsval’ declared as function returning a function ../../../source/scripting/ScriptableObject.h:48: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:54: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:74: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptableObject.h:78: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:78: error: ‘jsval’ has not been declared ../../../source/scripting/ScriptableObject.h:96: error: ‘jsval’ does not name a type ../../../source/scripting/ScriptableObject.h:100: error: ‘JSContext’ has not been declared ../../../source/scripting/ScriptableObject.h:100: error: ‘jsval’ has not been declared ../../../source/scripting/ScriptableObject.h:112: error: ‘GetFn’ in class ‘IJSObject’ does not name a type ../../../source/scripting/ScriptableObject.h:118: error: expected `)' before ‘Getter’ ../../../source/network/NetClient.cpp:699: error: expected `}' at end of input ../../../source/network/NetClient.cpp:699: error: expected unqualified-id at end of input make[1]: * [obj/network_Debug/NetClient.o] Error 1 make: * [network] Error 2 make: * Waiting for unfinished jobs.... In file included from ../../../source/scripting/SpiderMonkey.h:54,

from ../../../source/pch/engine/precompiled.h:30, from ../../../source/pch/engine/precompiled.cpp:18:

../../../source/scripting/JSUtil.h:20: error: ‘jsval’ does not name a type make[1]: * [obj/engine_Debug/precompiled.h.gch] Error 1 make: * [engine] Error 2

I am not a coder, but I've compiled a number of programs successfully. If there's a simple thing I'm doing wrong, let me know.

If a model fails to load (e.g. the Collada library wasn't compiled), the code currently prints an error message and returns null and then the game crashes a bit later. It should be handled better, e.g. not crashing and not rendering the unit, or (preferably) replacing the model with some kind of 'error' mesh so the problem is clearly visible.

From the annotations added in r7072:

../../../source/ps/Interact.cpp: In function ‘LibError LoadUnitUIThunk(const VfsPath&, const FileInfo&, uintptr_t)’:                                               
../../../source/ps/Interact.cpp:985: warning: too many arguments for format                                                                                        
../../../source/ps/Filesystem.cpp: In member function ‘PSRETURN CVFSFile::Load(const VfsPath&)’:                                                                   
../../../source/ps/Filesystem.cpp:60: warning: format ‘%d’ expects type ‘int’, but argument 6 has type ‘LibError’                                                  
../../../source/ps/ConfigDB.cpp: In member function ‘bool CConfigDB::Reload(EConfigNamespace)’:                                                                    
../../../source/ps/ConfigDB.cpp:287: warning: format ‘%lld’ expects type ‘long long int’, but argument 6 has type ‘LibError’                                       
../../../source/ps/XML/XMLWriter.cpp: In member function ‘bool XMLWriter_File::StoreVFS(const char*)’:
../../../source/ps/XML/XMLWriter.cpp:92: warning: format ‘%d’ expects type ‘int’, but argument 5 has type ‘LibError’
../../../source/simulation/EntityTemplate.cpp: In member function ‘bool CEntityTemplate::LoadXml(const CStr8&)’:
../../../source/simulation/EntityTemplate.cpp:320: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/renderer/Renderer.cpp: In static member function ‘static CRenderer::RenderPath CRenderer::GetRenderPathByName(const CStr8&)’:
../../../source/renderer/Renderer.cpp:713: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/simulation/Technology.cpp: In member function ‘bool CTechnology::LoadElEffect(XMBElement, CXeromyces&, const CStr8&)’:
../../../source/simulation/Technology.cpp:300: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/gui/CGUI.cpp: In member function ‘void CGUI::LoadXmlFile(const std::string&)’:
../../../source/gui/CGUI.cpp:1111: warning: format ‘%s’ expects type ‘char*’, but argument 6 has type ‘PSRETURN’
../../../source/gui/CGUI.cpp: In member function ‘void CGUI::Xeromyces_ReadObject(XMBElement, CXeromyces*, IGUIObject*)’:
../../../source/gui/CGUI.cpp:1398: warning: too few arguments for format
../../../source/lib/ogl.cpp: In function ‘void dump_gl_error(GLenum)’:
../../../source/lib/ogl.cpp:276: warning: format ‘%04f’ expects type ‘double’, but argument 2 has type ‘GLenum’
../../../source/scripting/ScriptGlue.cpp: In function ‘JSBool GetBuildTimestamp(JSContext*, JSObject*, uintN, jsval*, jsval*)’:
../../../source/scripting/ScriptGlue.cpp:1070: warning: unknown conversion type character ‘w’ in format
../../../source/scripting/ScriptGlue.cpp:1070: warning: too many arguments for format
../../../source/scripting/ScriptGlue.cpp:1079: warning: unknown conversion type character ‘w’ in format
../../../source/scripting/ScriptGlue.cpp:1079: warning: too many arguments for format
../../../source/scripting/ScriptGlue.cpp: In function ‘JSBool GetTrigger(JSContext*, JSObject*, uintN, jsval*, jsval*)’:
../../../source/scripting/ScriptGlue.cpp:1381: warning: unknown conversion type character ‘w’ in format
../../../source/scripting/ScriptGlue.cpp:1381: warning: too many arguments for format
../../../source/lib/res/graphics/ogl_shader.cpp: In function ‘LibError Ogl_Shader_reload(Ogl_Shader*, const VfsPath&, Handle)’:
../../../source/lib/res/graphics/ogl_shader.cpp:157: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:173: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp: In function ‘LibError do_load_shader(Ogl_Program*, const VfsPath&, Handle, const CXeromyces&, const XMBElement&)’:
../../../source/lib/res/graphics/ogl_shader.cpp:279: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:287: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:295: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp: In function ‘LibError Ogl_Program_reload(Ogl_Program*, const VfsPath&, Handle)’:
../../../source/lib/res/graphics/ogl_shader.cpp:348: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:369: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:378: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:394: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/res/graphics/ogl_shader.cpp:400: warning: use of ‘h’ length modifier with ‘s’ type character
../../../source/lib/sysdep/os/unix/x/x.cpp: In function ‘wchar_t* sys_clipboard_get()’:
../../../source/lib/sysdep/os/unix/x/x.cpp:173: warning: format ‘%d’ expects type ‘int’, but argument 4 has type ‘long unsigned int’
../../../source/lib/sysdep/os/unix/x/x.cpp:173: warning: format ‘%d’ expects type ‘int’, but argument 6 has type ‘long unsigned int’
../../../source/lib/sysdep/os/unix/x/x.cpp:184: warning: format ‘%d’ expects type ‘int’, but argument 3 has type ‘Atom’
../../../source/lib/sysdep/os/unix/x/x.cpp:184: warning: format ‘%d’ expects type ‘int’, but argument 4 has type ‘Atom’

secure_crt.cpp emulates sprintf_s, vsprintf_s, swprintf_s, vswprintf_s using the snprintf functions. These have the wrong behaviour when the buffer is too short (based on MSDN's documentation), and ought to be made consistent.

Problem: 'picking' (selecting via cursor) units doesn't work as expected because the bounding box is calculated from the primary geometry of the entity and doesn't include props. Example: celt trader; clicking on either horse or rider should select the unit but currently doesn't.

(Jason's brief explanation of actor and props:) The way most of our actors in the game works is... there is geometry and there are various 'nodes' that are throughout the model - they can be parented to a bone, or in the cases of most static meshes - parented to the 0,0,0 coordinate. These nodes have a name and in the actor xml file it says.. associate this prop point (node) to display this actor (then it points to another actor) Example: <prop actor="props/units/heads/rome_monte_a.xml" attachpoint="helmet"/> The prop actor also has a model; for example.. within that prop actor it defines: <mesh>props/helmet/rome_helmet_d.pmd</mesh>

To solve this, I'd first recommend using Atlas to create an otherwise empty map with a celt trader to serve as a test. The picking/selection code itself is ok, what we need to do is change the bounding box to include props and then the selection will be as desired.

"props" are in turn actors containing a CModel, which again can have props. In this case (trader) I think the rider is a prop.

CModel has a CalcBounds which constructs bounds from all model vertices (possibly also looping over the animation). What we need to do is also recurse over m_Props and expand the parent's bounds to that of the child prop. these "bounds" (m_ObjectBounds) are CBound objects, which are expanded via operator+= (allowing both bounds or a vector as argument). suggestion for what to do: 1) make CalcObjectBounds and CalcAnimatedObjectBound take a parameter of the bounding box to calculate (or expand.. in that case, no longer do SetEmpty there) 2) CalcBounds (the master function that dispatches to one of the above impl functions) must be recursive, expanding the bounds for each prop's bound (m_Props[i].m_Model->ObjectBounds)

​http://img30.imageshack.us/img30/2858/infintefield.jpg

The amount of the field went to 588.9999 which made it infinite. It was just this one and the others around it was working fine.

• cant seem to change priority from major.

This has been reported on two Linux systems, both running OpenALsoft. To narrow down the cause, I added some printfs [source ID, is it a valid source name, #processed buffers]:

ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
ade7938 1 0
adcede0 1 1

so the first time it's actually wanting to unqueue a buffer, OpenAL reports an error. After continuing from the error, it doesn't come up again during the program run.

alSourceUnqueueBuffers fails with "invalid value" iff a buffer cannot be unqueued because it's still playing. However what we do is ask how many buffers have finished before removing exactly that many. I have no idea what could be causing this, especially because the Windows OpenAL implementation does not raise this error. Help is very welcome.

The concept of "Batch" training:

Essentially training many units (of a same entity type) at once. This would be in multiples of 5. So basically a player can train anywhere from 5-30 units at once and once the batch is done training, ideally when the formations have been implemented, all the units in that batch (say: 20 hoplites) should march to the rally point as a pre-grouped battalion in formation.

We were thinking there could be an economy of scale applied as well, akin to gaining a discount in either unit training cost or training time (or both) the more that you queue into one batch.

So, the task can be broken down into multiple components:

A.) Implementing the batching itself. The player procedure to queue a batch should be the shift-click the unit icon, each shift-click adding 5 units to the batch up to 30 units. If the shift key is let go at any time that batch is done queuing. Any further shift-clicking would create a new batch.

B.) The economy of scale discount. The formula for the discount could follow a simple bell curve, so that 15-unit batches would be ideal, but the player can continue to queue more units into the batch, up to 30, and still get a discount, but not as lucrative. Whether the discount is in train time or cost (or both) should be made available to do based upon testing.

C.) Grouped into a Formation. This is contingent upon formations being implemented first. The newly created batch, once it is done training, should proceed to the rally point as a grouped formation of units.

Every faction except the Iberians has combat ships. there are three types of ships:

• Bireme This is the smallest battle ship in the game. It's slow, and has not much ramming power, altough they can damage a trireme critical. The manouvrebility is quite high.
• Trireme The most common ship in use. It's the fastest type of all ships, wich makes an ultimate ramming power. A trireme can hunt down another trireme when at highest attack speed.
• Quinquireme The biggest ship in the game. This massive ship is as fast as the bireme. But due their mass they crack other smaller ships like planks. they turn very slow and have a low acceleration. Only the Romans and the Carthaginians have this giant.

The Punic(Carthaginian), Hellenes, Roman and Phoenican(Persian) ships in the game will be able to ram each other. The celtic "trireme" can only use range, but they have a high armor level.

Hello! I was playing the "Badlands" map. crashed after sending units to an unclaimed settlement

pyrogenesis_dbg: /usr/include/boost/shared_ptr.hpp:315: T* boost::shared_ptr<T>::operator->() const [with T = CTerrainProperties]: Assertion `px != 0' failed.

Program received signal SIGABRT, Aborted.


(gdb) backtrace
#0  0xb7f7a410 in __kernel_vsyscall ()
#1  0xb7788085 in raise () from /lib/tls/i686/cmov/libc.so.6
#2  0xb7789a01 in abort () from /lib/tls/i686/cmov/libc.so.6
#3  0xb778110e in __assert_fail () from /lib/tls/i686/cmov/libc.so.6
#4  0x083747c9 in boost::shared_ptr<CTerrainProperties>::operator-> (this=0xbf9b305c) at /usr/include/boost/shared_ptr.hpp:315
#5  0x08374647 in CTerrain::IsPassable (this=0x9584120, loc=@0xbf9b3130, entity=@0xbf9b30f2) at ../../../source/graphics/Terrain.cpp:117
#6  0x0834c316 in AStarGoalLowLevel::IsPassable (this=0x8cf4558, loc=@0xbf9b3130, entity=@0xbf9b3140) at ../../../source/simulation/AStarEngine.cpp:508
#7  0x0834c1f1 in AStarGoalLowLevel::GetNeighbors (this=0x8cf4558, loc=@0xb549760, entity=@0xbf9b31ba) at ../../../source/simulation/AStarEngine.cpp:543
#8  0x0834cf4a in CAStarEngine::FindPath (this=0x8cf4438, src=@0xbf9b3278, dest=@0xbf9b32e4, entity=@0xbf9b328c, radius=0)
    at ../../../source/simulation/AStarEngine.cpp:234
#9  0x082f5f57 in CPathfindEngine::RequestLowLevelPath (this=0x8cf4438, entity=@0xbf9b3302, destination=@0xbf9b32e4, contact=false, radius=0, 
    orderSource=CEntityOrder::SOURCE_PLAYER) at ../../../source/simulation/PathfindEngine.cpp:533
#10 0x083478ab in CEntity::ProcessGotoWaypoint (this=0xa840808, current=0x8cf7a70, contact=false)
    at ../../../source/simulation/EntityStateProcessing.cpp:732
#11 0x082cc2d7 in CEntity::UpdateOrders (this=0xa840808, timestep=300) at ../../../source/simulation/Entity.cpp:582
#12 0x082cc940 in CEntity::Update (this=0xa840808, timestep=300) at ../../../source/simulation/Entity.cpp:457
#13 0x082b65cc in CEntityManager::UpdateAll (this=0x9588b80, timestep=300) at ../../../source/simulation/EntityManager.cpp:450
#14 0x083162f2 in CSimulation::Simulate (this=0x9561030) at ../../../source/simulation/Simulation.cpp:179
#15 0x083166e3 in CSimulation::Update (this=0x9561030, frameTime=0.033608004450798035) at ../../../source/simulation/Simulation.cpp:106
#16 0x0829f6a5 in CGame::Update (this=0x9588898, deltaTime=0.033608004450798035, doInterpolate=true) at ../../../source/ps/Game.cpp:230
#17 0x0820f421 in Frame () at ../../../source/main.cpp:271
#18 0x0820f882 in RunGameOrAtlas (argc=1, argv=0xbf9b3774) at ../../../source/main.cpp:397
#19 0x0820f906 in main (argc=1, argv=0xbf9b3774) at ../../../source/main.cpp:410

Please ignore.

Hello again.. I was playing the "Latinum" map. Crashed after trying to build a building (dock) on the very border of the map.

#0  0x08374786 in CRenderableObject::SetDirty (this=0x0, dirtyflags=2)
    at ../../../source/graphics/RenderableObject.h:82
#1  0x083732e7 in CTerrain::FlattenArea (this=0x9578708, x0=62.6870041, 
    x1=80.6870041, z0=564.263, z1=582.263)
    at ../../../source/graphics/Terrain.cpp:467
#2  0x082dc932 in CEntity::FlattenTerrain (this=0xc102d98)
    at ../../../source/simulation/EntityScriptInterface.cpp:932
#3  0x082e84ea in CNativeComplexFunction<CEntity, false, jsval_t, &(CEntity::FlattenTerrain(JSContext*, unsigned int, int*))>::JSFunction (cx=0x8723558, 
    obj=0xc0f8c50, argc=0, argv=0xc1d5860, rval=0xbfa566b0)
    at ../../../source/scripting/ScriptableComplex.inl:468
#4  0x085277cd in js_Invoke (cx=0x8723558, argc=0, flags=0) at jsinterp.c:1187
#5  0x0853838e in js_Interpret (cx=0x8723558, pc=0xa385fd0 ":", 
    result=0xbfa573d8) at jsinterp.c:3586
#6  0x08527843 in js_Invoke (cx=0x8723558, argc=1, flags=2) at jsinterp.c:1207
#7  0x08527b9e in js_InternalInvoke (cx=0x8723558, obj=0xb24e780, 
    fval=156876080, flags=0, argc=1, argv=0xbfa5750c, rval=0xbfa57510)
    at jsinterp.c:1284
#8  0x084f0c98 in JS_CallFunction (cx=0x8723558, obj=0xb24e780, fun=0x8a54860, 
    argc=1, argv=0xbfa5750c, rval=0xbfa57510) at jsapi.c:4146
#9  0x082c0707 in CScriptObject::DispatchEvent (this=0xa344f2c, 
    Context=0xb24e780, evt=0xbfa57a5c)
    at ../../../source/simulation/ScriptObject.cpp:126
---Type <return> to continue, or q <return> to quit---
#10 0x0832803c in IEventTarget::_DispatchEvent (this=0xa344cf0, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:49
#11 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa5792e8, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#12 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6ee1d8, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#13 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6ed0f8, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#14 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6ec3e0, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#15 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6eb648, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#16 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6ea918, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#17 0x083281b9 in IEventTarget::_DispatchEvent (this=0xa6e9c78, 
    evt=0xbfa57a5c, target=0xaebbff8)
---Type <return> to continue, or q <return> to quit---
    at ../../../source/scripting/DOMEvent.cpp:66
#18 0x083281b9 in IEventTarget::_DispatchEvent (this=0xaebbff8, 
    evt=0xbfa57a5c, target=0xaebbff8)
    at ../../../source/scripting/DOMEvent.cpp:66
#19 0x08328339 in IEventTarget::DispatchEvent (this=0xaebbff8, evt=0xbfa57a5c)
    at ../../../source/scripting/DOMEvent.cpp:86
#20 0x0834868f in CEntity::ProcessContactActionNoPathing (this=0xaebbf90, 
    current=0xc1749c8, timestep_millis=300)
    at ../../../source/simulation/EntityStateProcessing.cpp:516
#21 0x082cc2bc in CEntity::UpdateOrders (this=0xaebbf90, timestep=300)
    at ../../../source/simulation/Entity.cpp:577
#22 0x082cc968 in CEntity::Update (this=0xaebbf90, timestep=300)
    at ../../../source/simulation/Entity.cpp:457
#23 0x082b65cc in CEntityManager::UpdateAll (this=0x95ab9e0, timestep=300)
    at ../../../source/simulation/EntityManager.cpp:450
#24 0x0831631a in CSimulation::Simulate (this=0x9581140)
    at ../../../source/simulation/Simulation.cpp:179
#25 0x0831670b in CSimulation::Update (this=0x9581140, 
    frameTime=0.023770732805132866)
    at ../../../source/simulation/Simulation.cpp:106
#26 0x0829f6a5 in CGame::Update (this=0x95aab10, 
    deltaTime=0.023770732805132866, doInterpolate=true)
    at ../../../source/ps/Game.cpp:230
---Type <return> to continue, or q <return> to quit---
#27 0x0820f421 in Frame () at ../../../source/main.cpp:271
#28 0x0820f882 in RunGameOrAtlas (argc=1, argv=0xbfa58014)
    at ../../../source/main.cpp:397
#29 0x0820f906 in main (argc=1, argv=0xbfa58014)
    at ../../../source/main.cpp:410

Currently, non-support of S3TC causes ugly unhelpful output like in #306. (This particularly matters on Linux, where (I think) non-support is the default for non-proprietary drivers.)

The message should be displayed in a much more user-friendly way (perhaps a dialog box? or perhaps it should carry on loading the game and display a warning on the menu screen?).

Also, the message should help people resolve the problem - on Linux, it should tell them to run driconf and enable the "enable S3TC texture compression" option. (That option makes the driver lie that it fully supports S3TC, when actually it only supports decompression (which is all we need).)

Also, maybe we could detect the situation where we're running on Linux and the driver really does support decompression even though it doesn't claim to fully support S3TC, since the game should actually run fine with S3TC in that case and we don't need to tell the user to do anything.

There are some GUI components that get called for some missing settings, this could be a result of couple of reasons:

• Actual settings that are missing from the component
• Faulty calling to those settings

Another thing could be that the call to the settings is OK.
Meaning, the code is trying to probe the component for the wanted setting, and if it doesn't find it, then it just doesn't use that feature for that component.

To find the faulty settings run the program, open the main log (mainlog.html), and follow the warnings.

Universal Fire Crackling Sound SDD# 177

Specifications

MONO
5 Seconds
Non-Looping

5 sounds req'd

Seige Unit Select SDD# 295

Specifications

MONO
2-Seconds
Non-Looping

5 Sounds Req'd

Onager Destruction SDD# 296

Specifications

MONO
2-Seconds
Non-Looping

5 Sounds Req'd

Destruction - Gate SDD# 399

Specifications

MONO
2-Seconds
Non-Looping

5 Sounds Req'd

ps/CStr.cpp has its own UTF-8 encoding/decoding logic in ToUTF8/FromUTF8. The implementation should be changed to use the functions in lib/wchar.h, to remove the duplication of functionality. (The CStr API ought to remain unchanged.)

gui/GUItypes.h does some stupid trick with Perl (which I used to think was clever) to automatically regenerate its own code. It should be simplified by removing all the Perl and just leaving the TYPE(...) macro lines, and leaving the GUITYPE_IGNORE_CGUISpriteInstance bit (which is used in GUIutil.cpp; but get rid of the other IGNOREs).

On Linux, crashlog.txt by default seems to get put in binaries/system/ which is often not a writable directory. It should probably go in ~/.config/0ad/logs/ (and print a message telling the user where it is).

The Collada target is a run-time dependency for pyrogenesis and test, so it ought to get built automatically when people run make pyrogenesis and make test. Currently it doesn't.

It's not a build-time dependency for the executables, and shouldn't be used when linking, so Premake's package.links does not seem to be a suitable place to put it. Is there another way? or maybe Premake needs to be patched to support non-linking dependencies?

Profiler indicates the game is spending ~14% of its process's cycles in CLOSManager::GetStatus and CLOSManager::GetUnitStatus, when sitting idle in the default map, which is far too much.

To avoid unnecessarily restricting script code, it should be possible for script components to use new Number(42), new String("foo"), etc, and have them correctly (de)serialized with no information loss. ​MDC has a list of the standard classes, and the simple common ones should be supported.

The code for this is currently in CBinarySerializer::HandleScriptVal

It should be possible for scripts to create and use classes (function Thing(x) { this.x = x; }; Thing.prototype.foo = ...; y = new Thing(1); etc), and have them correctly (de)serialized.

This is probably infeasible in the general case, but it would be useful to support the simple case: classes are defined as top-level functions (not defined or changed at run-time inside other functions), no prototype chains, etc, so the code can simply serialize the object like a plain Object plus a class name that gets looked up in the appropriate script's scope when deserializing.

Currently C++ components have to say:

ICmpPosition* cmpPosition = static_cast<ICmpPosition*>
  (context.GetComponentManager().QueryInterface(GetEntityId(), IID_Position));

which is horrid.

JS just says

var cmpPosition = Engine.QueryInterface(this.entity, IID_Position);

which is nicer. C++ has got to have more type declarations, and there's tradeoffs with other desirable features of the code, but it should be possible to do better than the current syntax.

Atlas needs to be updated to support the new entity system. In particular, the object creation listbox needs to be populated with entities from the new system, and it needs to create them properly (along with the static preview before the object is placed, possibly by extracting just its Visual component somehow?)

The map saving/loading code needs to deal with the new entity system. It should just save the template, position, orientation, player ID, and random variation selections, for each entity.

Entity templates are a kind of tree structure, defined by a series of XML files (each inheriting from a parent file and overriding some parts). To save component code from having to worry about detecting and reporting invalid input (missing fields, incorrect data types, etc) there should be a centralised validation system. Each component reports its requirements, and the validator checks each template against them.

It would probably be sensible to use libxml2's RelaxNG support for this, since it provides a reasonable schema language for checking the structure and data types. The entity template can be serialised into XML (after applying all the inheritance rules etc) and then validated. If validation fails then the entity won't be loaded. It'd be particularly nice to use the RelaxNG Compact syntax, but libxml2 doesn't support that (it just has a half-finished RNC->RNG converter in its source control repository that is of unknown quality), so I don't know how feasible that is.

For efficiency (if it turns out that it matters), the validation status of entity templates could perhaps be cached at the same time as they're converted to XMB.

We use wcscasecmp but it does not exist on OS X, causing errors. We should either define that function ourselves on OS X (possibly a bad idea since it might conflict with a system version in the future, and we would not bother with the locale support that the standard expects), or define a similar function with a new name (probably in lib/wchar.cpp, just doing ASCII-case-insensitive comparisons) and use that on all platforms.

There needs to be an easy way to do unit testing of script components.

The tests themselves should be written as scripts. Each component should be tested in isolation, with all engine functionality and other components probably implemented in the script as mock objects, so that the test cases don't require complex scene setups.

It should be integrated with the current CxxTest system, but maybe it would also be useful to be able to run the tests as part of the normal executable, so they can be used by non-C++ programmers and integrated with the debugger (#410).

Similar to #415, except for native (C++) components. The tests will need to be written in C++ themselves. Mock objects might be much more painful to do; not sure what the best solution is.

Add a review request flag to Trac

CppDoc is old and unmaintained and has bugs and sometimes crashes and doesn't support JS. Doxygen seems to work better, and is more common, and accepts a similar comment style to CppDoc. So we should probably move to Doxygen.

r7193 changed how we link with pthreads. First thing to check (by testing the build): does the -pthread option work on OS X? Second thing: premake.lua defines _REENTRANT explicitly on OS X (because it wasn't defined by the included headers); is that unnecessary now that we use -pthread?

CParamNode::GetChild returns pointers which could be NULL (if the child doesn't exist). In theory that should never happen (template validation ought to catch the errors first), but it's fragile and crash-prone.

Instead, it should return a dummy CParamNode object (by reference) with a bool Ok() method to allow explicit checking of whether the child really existed. GetChild should return another dummy object, ToInt/ToString/etc should log an error message and return 0/""/etc.

Position is critical to a lot of things, and needs to be implemented. (This includes orientation, and some of these features, and needs to be designed with garrisoning in mind.)

(Haven't tried to reproduce.)

Did something to get a crash in the game, while loading a map, triggered by the MRU file list in editor. The GUI froze completely.

(thread 1)
#0  0x00007f227c59adc9 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0     
#1  0x00007f22752d6c7d in g_main_context_wait () from /usr/lib/libglib-2.0.so.0           
#2  0x00007f22752d8ee1 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0               
#3  0x00007f2276f96a07 in gtk_main () from /usr/lib/libgtk-x11-2.0.so.0                   
#4  0x00007f22780af478 in wxEventLoop::Run () from /usr/lib/libwx_gtk2u_core-2.8.so.0     
#5  0x00007f22781008e4 in wxDialog::ShowModal () from /usr/lib/libwx_gtk2u_core-2.8.so.0  
#6  0x00007f2278a9ca9b in wxDebugReportPreviewStd::Show () from /usr/lib/libwx_gtk2u_qa-2.8.so.0
#7  0x00007f22797adfb5 in AtlasDLLApp::OnFatalException (this=0x2da95f0) at ../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp:254
#8  0x00007f2277a1fe4c in wxFatalSignalHandler () from /usr/lib/libwx_baseu-2.8.so.0                                                     
#9  <signal handler called>                                                               

(thread 3)
#0  0x00007f227c59cc81 in sem_wait () from /lib/libpthread.so.0                                                          
#1  0x00000000008e0d9d in MessagePasserImpl::Query (this=0xf1fc20, qry=0x7f2272c75530) at ../../../source/tools/atlas/GameInterface/MessagePasserImpl.cpp:166
#2  0x00007f22797af050 in AtlasMessage::QueryMessage::Post (this=0x7f2272c75530) at ../../../source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp:796
#3  0x00007f22797b17a9 in ScenarioEditor::OpenFile (this=0x2ec5530, name=@0x7f2272c755d0) at ../../../source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp:581
#4  0x00007f22797b1943 in ScenarioEditor::OnMRUFile (this=0x2ec5530, event=@0x7f2272c756d0) at ../../../source/tools/atlas/AtlasUI/ScenarioEditor/ScenarioEditor.cpp:611
#5  0x00007f2277a1b7b9 in wxEvtHandler::ProcessEventIfMatches () from /usr/lib/libwx_baseu-2.8.so.0                                                                     
#6  0x00007f2277a1c993 in wxEventHashTable::HandleEvent () from /usr/lib/libwx_baseu-2.8.so.0                                                                           
#7  0x00007f2277a1ca87 in wxEvtHandler::ProcessEvent () from /usr/lib/libwx_baseu-2.8.so.0       

The Makefiles generated by Premake try to detect when the .lua file changed, and re-run premake. That doesn't work because it's not the right path (and it might try to run an incorrect version of Premake instead). Maybe it could be changed to use the right relative path, but it would probably be safer to just print a message telling the user to run update-workspaces.sh again.

(We already use a custom fork of Premake so it'd be adequate to just hard-code these changes into Premake's source.)

The new serialization and CParamNode code needs tests and documentation.

Need to have some integration between the GUI and simulation. Rather than try to get the 'real' GUI working, just concentrate for now on getting a rough approximation of the right data being shown and updated, and messages passing from the GUI back to the simulation.

I got an error while compilation with AtlasUi in "obj/AtlasUI_Debug/DLLInterface.o"

Debug/precompiled.h ../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp ../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp:48:2: error: #error libxml2 must have threading support enabled make[1]: * [obj/AtlasUI_Debug/DLLInterface.o] Erreur 1 make: * [AtlasUI] Erreur 2

I'm on Ubuntu Karmic 9.10 tell me if you need more information this is my first report and my english sucks !

CRenderer::RenderSubmissions() calls RenderTransparentModels() twice, once before drawing water and once after, so that underwater transparent models will blend properly.

That seems like a terrible idea. We have lots of transparent objects (trees) so it will kill performance; and the overdraw makes transparent objects look different, so when we fix the performance bug all the art will be wrongly calibrated.

Possible temporary solution: don't support transparent objects underwater. Possible better solution: do the overdraw only on models which intersect the water plane. Ideal solution: render transparent objects correctly always (but I don't know how that'd be implemented).

The new simulation system should let you select entities properly. Inputs should be similar to the old system - click to select, double-click to select all of type on screen, triple-click to select all of type in world, drag to select in area, shift to add to current selection, plus predefined groups, or something like that. Also it should handle own vs enemy units sensibly. See ps/Interact.cpp and gui/session_new/input.js.

It needs to be possible for the GUI to push commands onto a command queue, which (eventually) will be network-synchronised, and then entities need to respond to the commands.

Also it needs some kind of security checking, so you can't move other players' units around.

Also it'd be nice if it could do things like play sounds immediately when the command is pushed, rather than after the network turn latency.

See SimulationSyntax. Need to extend it to cover writing script wrappers for interfaces, defining components in C++, and defining components in JS. Also need to cover all the simulation-specific things you might want to do inside a component (CmpPtr etc).

Some of the code uses jsval as a template type parameter, particularly in ToJSVal/FromJSVal. Since jsval is a typedef of unsigned long (or something like that), it conflicts with the normal integer type and things will break. (The current code doesn't even compile on 32-bit Linux).

So there should be a special type (maybe a ScriptVal or something) which just wraps a jsval, and can be used as a distinct type for templates.

JS won't automatically run the GC until it's run out of memory, which might be an inconvenient time to freeze the simulation. We should probably call it manually at regular intervals, when it's least likely to cause disruption (e.g. during a frame when it's not also busy doing all the simulation computation), to make it more predictable.

When I start a new game, it doesn't matter the map I get only a huge circle: here's a ​screenshot and a ​video with the bug. And the output of pyrogenesis_dbg:

TIMER| InitVfs: 6.28272 ms
TIMER| InitScripting: 2.62648 ms
TIMER| CONFIG_Init: 10.9219 ms
TIMER| write_sys_info: 1.12136 ms
TIMER| ps_console: 5.00708 ms
TIMER| ps_lang_hotkeys: 4.32716 ms
TIMER| common/setup.xml: 13.6533 ms
TIMER| common/styles.xml: 1.8468 ms
TIMER| common/sprite1.xml: 41.6079 ms
TIMER| common/init.xml: 14.132 ms
TIMER| pregame/mainmenu.xml: 75.108 ms
TIMER| common/global.xml: 1.93192 ms
TIMER| InitRenderer: 32.0211 ms
TIMER| SimulationInit: 2.881 ms
TIMER| Init_miscgamesection: 26.0635 ms
SND| alc_init: success, using ALSA Software
TIMER| common/setup.xml: 1.759 ms
TIMER| common/styles.xml: 2.11024 ms
TIMER| common/sprite1.xml: 81.4024 ms
TIMER| common/init.xml: 18.9571 ms
TIMER| loading/loading.xml: 3.22424 ms
TIMER| common/global.xml: 1.27692 ms
TIMER| common/setup.xml: 2.15872 ms
TIMER| common/styles.xml: 2.22512 ms
TIMER| common/sprite1.xml: 77.4852 ms
TIMER| common/init.xml: 23.1116 ms
TIMER| session/session.xml: 68.2606 ms
TIMER| session/manual.xml: 3.00416 ms
TIMER| common/global.xml: 10.2518 ms
ERROR: GUI page 'page_session.xml': Failed to call init() function
GAME STARTED, ALL INIT COMPLETE
gui/common/functions_page_session.js(174): Error: Size only accepts strings or GUISize objects
ERROR: JavaScript Error (gui/common/functions_page_session.js, line 174): Error: Size only accepts strings or GUISize objects
TIMER| shutdown Scheduler: 7.92 us
TIMER| shutdown mouse stuff: 49.04 us
TIMER| shutdown Pathfinder: 180.919 us
TIMER| shutdown game scripting stuff: 14.8534 ms
TIMER| shutdown actor stuff: 11.28 us
TIMER| shutdown TexMan: 5.6 us
TIMER| shutdown Renderer: 582.64 us
TIMER| shutdown ScriptingHost: 7.24432 ms
TIMER| shutdown ConfigDB: 2.08 us
TIMER| shutdown CSocketBase: 229.68 us
TIMER| shutdown CNetLogManager: 45.999 us
TIMER| shutdown I18N: 7.72 us
TIMER| resource modules: 38.0077 ms
TIMER TOTALS (25 clients)
-----------------------------------------------------
  js_timer 19: 0 c (0x)
  js_timer 18: 0 c (0x)
  js_timer 17: 0 c (0x)
  js_timer 16: 0 c (0x)
  js_timer 15: 0 c (0x)
  js_timer 14: 0 c (0x)
  js_timer 13: 0 c (0x)
  js_timer 12: 0 c (0x)
  js_timer 11: 0 c (0x)
  js_timer 10: 0 c (0x)
  js_timer 9: 0 c (0x)
  js_timer 8: 0 c (0x)
  js_timer 7: 99.6559 Mc (598x)
  js_timer 6: 226.475 Mc (598x)
  js_timer 5: 190.2 Mc (106x)
  js_timer 4: 54.9052 Mc (106x)
  js_timer 3: 74.5003 Mc (106x)
  js_timer 2: 128.114 Mc (106x)
  js_timer 1: 1.84467e+10 Gc (106x)
  js_timer 0: 0 c (2x)
  tc_2: 0 c (0x)
  tc_1: 0 c (0x)
  tc_transform: 2108.03 kc (14x)
  tc_plain_transform: 3394.96 kc (14x)
  tc_png_decode: 0 c (0x)
-----------------------------------------------------
TIMER| shutdown misc: 638.841 us

I am building 0ad on Slackware Linux 13 x86_64 from subversion with Revision: 7232

Need to remember to add the GPL header block to all the new source files.

Need to merge the code currently in the Hg branch into SVN. (Now seems as good a time as any.)

When the player chooses to build a building, they need some way to pick a location for it.

The implementation should be somewhat similar to the Atlas entity preview feature. Probable approach is: GuiInterface.js should provide some functions ShowEntityPreview(templateName, x, z, angle, player) and HideEntityPreview(). Those functions would construct a local entity with the preview|templateName magic template, and move it around, and delete it when it's hidden. Show would return a flag indicating if that's a valid build location (based on terrain, obstructions, etc). (Also it'd have to change the colour to indicate obstruction, and deal with FOW etc). The GUI input-handling scripts would just call those functions to show the previews, and then send the parameters to a 'construct' command when the player clicks.

When starting a new game Pyrogenesis crashes. It seems to be loading the map and all as I get ~ a second before the game crashes when I can see the game, and hear the music, it's at most a second though as the game crashes almost immediately and the music start to repeat the first < second or so. See below for call stack + output from VC++

Call stack:

 	wrap_oal.dll!0b641d5a() 	
 	[Frames below may be incorrect and/or missing, no symbols loaded for wrap_oal.dll]	
 	winmm.dll!76b454e3() 	
 	winmm.dll!76b5adfe() 	
 	winmm.dll!76b5aee9() 	
 	winmm.dll!76b5af02() 	
 	kernel32.dll!7c80b729() 	
>	pyrogenesis_dbg.exe!CreateDirectories(const boost::filesystem::basic_path<std::basic_string<wchar_t,std::char_traits<wchar_t>,std::allocator<wchar_t> >,boost::filesystem::wpath_traits> & path={...}, unsigned int mode=0)  Line 116 + 0x13 bytes	C++

Output from VC++:

TIMER| common/global.xml: 17.62 ms
GAME STARTED, ALL INIT COMPLETE
First-chance exception at 0x0b641d5a in pyrogenesis_dbg.exe: 0xC0000094: Integer division by zero.
Unhandled exception at 0x0b641d5a in pyrogenesis_dbg.exe: 0xC0000094: Integer division by zero.
The program '[116860] pyrogenesis_dbg.exe: Native' has exited with code 0 (0x0).

Please tell me if there's anything else you want to know. I've attached my system_info.txt.

As in ​this post, the game doesn't look in the per-user path for local.cfg, and it should do.

(Adding to Trac because I'd forgotten about it entirely and someone mentioned it, and I don't think it's been fixed yet.)

Ideally, GetChild would return a CParamNode& instead of a CParamNode* (now that it will never return NULL), to simplify the usage syntax (GetChild("x").GetChild("y") instead of GetChild("x")->GetChild("y")).

Assertion failed: "hModule"
Location: wdlfcn.cpp:55 (dlopen)

Call stack:

dlopen (wdlfcn.cpp:50)
    so_name = 0xCCCCCCCC
    flags = 1244820 (0x0012FE94)
    hModule = 0xCCCCCCCC
    pathname = 
        m_path = (error -100500 while analyzing string<﻾td::char_traits<char> >)


DllLoader::LoadDLL (dllloader.cpp:70)
    this = 0x00BAD56B -> 
        m_Name = 0x0824548B -> (unavailable - internal error)

        m_Handle = 0x8B0C428D


ATLAS_Run (atlas.cpp:41)
    args = 0x0012FEC8 -> 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    flags = 3017173665 (0xB3D66AA1)

ATLAS_RunIfOnCmdLine (atlas.cpp:70)
    args = 0x00000002

RunGameOrAtlas (main.cpp:379)
    argc = 2 (0x00000002)
    argv = 0x01F17690 -> 0x01F1769C -> "h:\0ad\binaries\system\pyrogenesis_dbg.exe"
    ran_atlas = true 
    args = 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    res = 2.63516e-299 (0x01F1A5C0CCCCCCCC)

main (main.cpp:416)
    argc = 1245112 (0x0012FFB8)
    argv = 0x00506C74 -> 0xC7E04589 -> (unavailable - internal error)


__tmainCRTStartup (crtexe.c:410)

mainCRTStartup (crtexe.c:393)

CallStartupWithinTryBlock (wseh.cpp:352)
    ret = 1245060 (0x0012FF84)

00000000
    ret = 2147328000 (0x7FFDA000)


errno = 0 (?)
OS error = 0 (no error code was set)

After I click continue I get the following error message:

The Atlas UI was not successfully loaded and therefore cannot be started as requested.
Location: Atlas.cpp:46 (ATLAS_Run)

Call stack:

ATLAS_Run (atlas.cpp:41)
    args = 0x0012FEC8 -> 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    flags = 3017173665 (0xB3D66AA1)

ATLAS_RunIfOnCmdLine (atlas.cpp:70)
    args = 0x00000002

RunGameOrAtlas (main.cpp:379)
    argc = 2 (0x00000002)
    argv = 0x01F17690 -> 0x01F1769C -> "h:\0ad\binaries\system\pyrogenesis_dbg.exe"
    ran_atlas = true 
    args = 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    res = 2.63516e-299 (0x01F1A5C0CCCCCCCC)

main (main.cpp:416)
    argc = 1245112 (0x0012FFB8)
    argv = 0x00506C74 -> 0xC7E04589 -> (unavailable - internal error)


__tmainCRTStartup (crtexe.c:410)

mainCRTStartup (crtexe.c:393)

CallStartupWithinTryBlock (wseh.cpp:352)
    ret = 1245060 (0x0012FF84)

00000000
    ret = 2147328000 (0x7FFDA000)


errno = 0 (?)
OS error = 0 (no error code was set)

And if it's useful here's the output, perhaps it can provide some clues{{{ 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\pyrogenesis_dbg.exe', Symbols loaded. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\ntdll.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\kernel32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\opengl32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\msvcrt.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\advapi32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\rpcrt4.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\secur32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\gdi32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\user32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\glu32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\ddraw.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\dciman32.dll' 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\libxml2.dll', Binary was not built with debug information. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\wsock32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\ws2_32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\ws2help.dll' 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\iconv.dll', Binary was not built with debug information. 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\zlib1.dll', Binary was not built with debug information. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\WinSxS\x86_Microsoft.VC80.CRT_1fc8b3b9a1e18e3b_8.0.50727.4053_x-ww_e6967989\msvcr80.dll' 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\OpenAL32.dll', Binary was not built with debug information. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\shell32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\shlwapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\winmm.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\ole32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\oleaut32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\WinSxS\x86_Microsoft.VC90.DebugCRT_1fc8b3b9a1e18e3b_9.0.30729.1_x-ww_f863c71f\msvcp90d.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\WinSxS\x86_Microsoft.VC90.DebugCRT_1fc8b3b9a1e18e3b_9.0.30729.1_x-ww_f863c71f\msvcr90d.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\WinSxS\x86_Microsoft.Windows.Common-Controls_6595b64144ccf1df_6.0.2600.5512_x-ww_35d4ce83\comctl32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\dsound.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\version.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\wintrust.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\crypt32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\msasn1.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\imagehlp.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\wdmaud.drv' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\wdmaud.drv' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\wdmaud.drv' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\msacm32.drv' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\msacm32.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\midimap.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\setupapi.dll' 'pyrogenesis_dbg.exe': Unloaded 'C:\WINDOWS\system32\setupapi.dll' HPET: rev=1 vendor=8086 bits=64 period=429B17F freq=1.43182e+007 wdlfcn.cpp(55): Assertion failed: "hModule" 'pyrogenesis_dbg.exe': Loaded 'H:\0ad\binaries\system\dbghelp.dll' First-chance exception at 0x0052471b in pyrogenesis_dbg.exe: 0xC0000005: Access violation reading location 0x0824548b. First-chance exception at 0x0052471b in pyrogenesis_dbg.exe: 0xC0000005: Access violation reading location 0xc7e04589. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\uxtheme.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\Program Files\RocketDock\RocketDock.dll', Binary was not built with debug information. 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\psapi.dll' 'pyrogenesis_dbg.exe': Loaded 'C:\WINDOWS\system32\MSCTF.dll' Atlas.cpp(46): The Atlas UI was not successfully loaded and therefore cannot be started as requested. First-chance exception at 0x0052471b in pyrogenesis_dbg.exe: 0xC0000005: Access violation reading location 0xc7e04589. The thread 'whrt_UpdateThread' (0x2be7c) has exited with code 0 (0x0). The thread 'Win32 Thread' (0x2c1d0) has exited with code 0 (0x0). The program '[180652] pyrogenesis_dbg.exe: Native' has exited with code 0 (0x0).

}}} mainlog.html is not updated, I've attached system_info.txt

When pressing the Copy button in the Program Error window when running the game (with -editor) in VC++ Express 2008, I get an error message from VC++:

Debug Assertion Failed! Program: h:\0ad\binaries\system\pyrogenesis_dbg.exe File: f:\dd\vctools\crt_bld\self_x86\crt\src\tcscpy_s.inl Line: 30 Expression: (L"Buffer is too small" && 0) For information on how your program can cause an assertion failure, see the Visual C++ documentation on asserts. (Press Retry to debug the application)

So I do, another dialog box:

pyrogenesis_dbg.exe has triggered a breakpoint [Break][Continue]

I press break, and below is the call stack:

>	pyrogenesis_dbg.exe!SetClipboardText(const wchar_t * text=0x02e90068, void * * hMem=0x0012f400)  Line 35 + 0x14 bytes	C++ 

	pyrogenesis_dbg.exe!sys_clipboard_set(const wchar_t * text=0x02e90068)  Line 55 + 0xd bytes	C++ 

	pyrogenesis_dbg.exe!error_dialog_proc(HWND__ * hDlg=0x00090978, unsigned int msg=273, unsigned int wParam=1002, long lParam=395626)  Line 183 + 0x9 bytes	C++

 	user32.dll!7e418734() 	

 	user32.dll!7e423ce4() 	

 	user32.dll!7e423b30() 	 

	user32.dll!7e43e577() 	 

	user32.dll!7e418734() 	

 	user32.dll!7e418734() 	

 	user32.dll!7e43e577() 	 

	user32.dll!7e43e577() 	

 	user32.dll!7e43e18a() 	

 	MSCTF.dll!747313d4() 	

 	MSCTF.dll!747313d9() 	

 	MSCTF.dll!7473057a() 	

 	user32.dll!7e418734() 	

 	user32.dll!7e418816() 	

 	user32.dll!7e4189cd() 	

 	user32.dll!7e4193e9() 	 

	user32.dll!7e4193a8() 	

 	user32.dll!7e431b7c() 	

 	user32.dll!7e4249c4() 	

 	user32.dll!7e424a06() 	

 	user32.dll!7e43b190() 	

 	pyrogenesis_dbg.exe!sys_display_error(const wchar_t * text=0x02040000, unsigned int flags=6)  Line 265 + 0x1d bytes	C++

 	pyrogenesis_dbg.exe!CallDisplayError(const wchar_t * text=0x02040000, unsigned int flags=6)  Line 405 + 0xd bytes	C++

 	pyrogenesis_dbg.exe!debug_DisplayError(const wchar_t * description=0x0012f9bc, unsigned int flags=6, void * context=0x00000000, const wchar_t * lastFuncToSkip=0x01f1a640, const wchar_t * pathname=0x00c1b0a0, int line=55, const char * func=0x00c1b098, unsigned char * suppress=0x00d48993)  Line 480 + 0xd bytes	C++

 	pyrogenesis_dbg.exe!debug_OnAssertionFailure(const wchar_t * expr=0x00c1b108, unsigned char * suppress=0x00d48993, const wchar_t * file=0x00c1b0a0, int line=55, const char * func=0x00c1b098)  Line 565 + 0x35 bytes	C++

 	pyrogenesis_dbg.exe!dlopen(const char * so_name=0x0012fe48, int flags=10)  Line 55 + 0x21 bytes	C++

 	pyrogenesis_dbg.exe!DllLoader::LoadDLL()  Line 82 + 0x10 bytes	C++

 	pyrogenesis_dbg.exe!ATLAS_Run(const CmdLineArgs & args={...}, int flags=1)  Line 43 + 0xa bytes	C++

 	pyrogenesis_dbg.exe!ATLAS_RunIfOnCmdLine(const CmdLineArgs & args={...})  Line 73 + 0xb bytes	C++

 	pyrogenesis_dbg.exe!RunGameOrAtlas(int argc=2, const char * * argv=0x01f17690)  Line 393 + 0x9 bytes	C++

 	pyrogenesis_dbg.exe!main(int argc=2, char * * argv=0x01f17690)  Line 419 + 0xd bytes	C++

 	pyrogenesis_dbg.exe!__tmainCRTStartup()  Line 586 + 0x19 bytes	C

 	pyrogenesis_dbg.exe!mainCRTStartup()  Line 403	C

 	pyrogenesis_dbg.exe!CallStartupWithinTryBlock()  Line 356 + 0x5 bytes	C++

 	pyrogenesis_dbg.exe!wseh_EntryPoint()  Line 384	C++

 	kernel32.dll!7c817077() 	

 	pyrogenesis_dbg.exe!std::_Tree<std::_Tmap_traits<void const *,CModelDefRPrivate *,std::less<void const *>,std::allocator<std::pair<void const * const,CModelDefRPrivate *> >,0> >::insert(std::_Tree<std::_Tmap_traits<void const *,CModelDefRPrivate *,std::less<void const *>,std::allocator<std::pair<void const * const,CModelDefRPrivate *> >,0> >::const_iterator _Where=(...,..., const std::pair<void const * const,CModelDefRPrivate *> & _Val=(...,...)  Line 691 + 0x1f bytes	C++

 	e04d8d52()

The output seems to be the same as the other error (and I assume at least some of the above information is at least as much related to that error as this error :P If nothing else it was that error that caused the error box to appear in the first place.) ​http://trac.wildfiregames.com/attachment/ticket/453 <-- the other error, am getting too tired to make nicer link =) Please tell me if I've missed some information and I'll try to include it.

DllLoader is used to load the Atlas and Collada libraries at run-time. In a Debug build, it tries to load AtlasUI_dbg.dll (./libAtlasUI_dbg.so on Linux). In a Release build, it loads AtlasUI.dll (./libAtlasUI.so).

(Ignore the comment in DllLoader saying "it is critical that compiler options are the same between app and SO" - that is untrue, since the libraries are designed to be safely mixed with different compiler options. We just want to use the Debug library with Debug builds in order to help debugging and error-detection.)

Because Atlas is a bit awkward and slow to compile, it's disabled by default on Windows, so AtlasUI_dbg.dll won't be created. But AtlasUI.dll will exist, because we put a pre-compiled binary in SVN (for non-programmers to use). Rather than fail to load Atlas in Debug builds because there's no matching library, it should fall back on AtlasUI.dll and load that one instead.

So in Debug builds on Windows, DllLoader needs to look for the Debug library then the Release library; in all other cases it should just look for the single appropriate library.

In all cases, it shouldn't trigger internal error messages in dlopen: it should check that the file exists before trying to open it, and give a nice friendly error message if it can't find anything. (Or maybe it should try to open it first, and change the Windows emulation of dlopen in wdlfcn.cpp to silently return a failure code that the caller can check.)

When running the public build, the game still tries to mount the internal directory and expects it to fail silently, but it looks like it actually mkdirs the internal directory if it didn't already exist. That breaks when the game is installed in a read-only location - it shouldn't try to create any directories there.

When I run the game on Linux (with no non-standard directories, and with a read-only root), g_VFS->Mount(L"screenshots/", paths.Data()/L"screenshots/") ends up calling vfs_Lookup which calls wmkdir("screenshots") which fails because that's relative to the cwd (the system directory) and read-only.

The ~/.local/share/0ad/screenshots/ directory is created before this happens, and it tries creating the bogus screenshots directory even if that proper one exists.

After the new simulation system new reasons why this doesn't work were added, I don't know enough about them to explain it, but I thought I should at least update the description to reflect that the reason it is broken have changed.

At least that's what I assume. I don't know enough of the inner workings of the engine to tell if it's something different, but I think the first time I saw it was pretty soon after the change.

In any case, as far as I can tell most things work (at least the maps I've tried don't display any error during the creation phase, i.e. the time while the "command" window is displayed), however the textures don't show up on cantabrian_highlands.j, and I get errors after the map creation has finished on other maps, so it should preferably be looked into by someone with more knowledge of the system than me.

Many of the models are in the PMD format and are not editable by most 3d editing programs. A converter would be needed to ensure high modability in this game, as well to make edits to models.

An alternate solution is for someone to just convert all PMD to DAE.

Make --with-spidermonkey-tip the default.

Make update-workspaces.sh run the new SM build script.

Compile the new version for Windows.

Rather than forcing everyone to download from SVN, there should be a few downloadable-over-HTTP snapshots provided:

• Linux/OSX SVN working copy, so people can run svn up.
• Windows SVN working copy, so people can run TortoiseSVN update.
  • (These differ because of line endings. Also, we should ideally have a separate Linux SVN URL that excludes the Windows binaries, but that's not critical.)
  • (These need to be done for a sufficiently old version of the SVN client, so it'll work for users with those old versions.)
• Linux/OSX data files (just the binaries/data/ directory), to be used by most distro packages.
• Linux/OSX build files (everything that's not binaries/data/), to be used by people who are creating packages and by source-based distros.
• Windows data files plus executables (like Linux data plus binaries/system/, and with Windows line endings), to be used by non-developers on Windows.

These don't all need to be updated constantly, but hopefully it'll be fairly frequent and so it should be automated as much as possible. Maybe use Wine to generate the Windows working copies.

In the non-SVN distributions, svn_revision.txt should be set to something appropriate.

Windows files should be .7z (because .zip is far too inefficient). Linux files should be .tar.gz or maybe .7z/.tar.lzma (if it's efficient enough to be worthwhile).

Put the files on releases.wildfiregames.com so we can move the hosting around easily.

addendum to [7341]

you forgot some files..

In the simulation code, Math.random() needs to be replaced with an implementation that is network-synchronised and gets serialized.

Projectiles need to be created with the correct actor (probably just coming from the attacker's actor's relevant prop point, not from the entity template, since that's easier with the current data files).

Need to reimplement CTerrain::GetExactGroundLevel using fixed-point instead of floating-point, so it can be safely used by CCmpTerrain::GetGroundLevel.

I am always getting the following error:

svn: OPTIONS of '​http://svn.wildfiregames.com/public/ps/trunk': timed out waiting for server (​http://svn.wildfiregames.com)

subversion client details:

svn, version 1.6.6 (r40053)

compiled Mar 20 2010, 18:04:14

Copyright (C) 2000-2009 CollabNet. Subversion is open source software, see ​http://subversion.tigris.org/ This product includes software developed by CollabNet (​http://www.Collab.Net/).

The following repository access (RA) modules are available:

• ra_neon : Module for accessing a repository via WebDAV protocol using Neon.
  • handles 'http' scheme
  • handles 'https' scheme
• ra_svn : Module for accessing a repository using the svn network protocol.
  • with Cyrus SASL authentication
  • handles 'svn' scheme
• ra_local : Module for accessing a repository on local disk.
  • handles 'file' scheme

When I run on Linux with force_s3tc_enable (without libtxc_dxtn which implements S3TC compression, so I've only got decompression), there's a number of rendering errors that I assume are caused by us attempting to use run-time compression. Need to look in more detail to work out the full list of problems and the exact causes, but I see:

• Some textures are bogus when zoomed out (probably the .dds files are missing mipmaps, and they can't be created dynamically - we ought to ensure all texture files have mipmaps).
• The minimap is entirely black.
• Terrain texture previews in the terrain-picker in Atlas are black.

Currently the graphics settings are determined by the defaults in system.cfg, modified by hand-written local.cfg, modified by what the hardware supports (based on GL extensions).

It ought to be based partly on performance, e.g. don't enable fancywater and pretty shadows by default on rubbish slow Intel devices (perhaps based on some benchmarking on first run, or a hardcoded list of common known devices?), and the user should be able to configure things in the UI and preferably see the changes dynamically and then save their settings.

The new ​JS_New function looks like it does what ScriptInterface::CallConstructor attempts to emulate. We should update to a SpiderMonkey version that includes that function, and then use it (and make sure it doesn't change the behaviour in a way that breaks anything).

Water doesn't animate with the new simulation system. It should. (Probably just forgetting to send update messages to it.)

Assertion failed: "SDL_SetGamma failed"
Location: GameSetup.cpp:142 (SetVideoMode)

18:01 <@Philip-> janwas: Should the warning about gamma be disabled, because it's a bit annoying and doesn't seem to be useful?
18:03 < janwas> Philip-: hm, that check has failed (in the intervening years, no one was able to find the underlying problem), so it can probably be commented out :) (preferably with a note that it sometimes fails, usually on multimon systems)

If Atlas isn't run from binaries/system/ then it fails to find files. Fix it.

When located in a path containing spaces, the Windows builds fail with some error message. If I remember correctly, the problem is we pass an absolute path to the cxxtestgen.exe tool but don't correctly put quotes around it. It should perhaps be fixed by modifying our patched copy of Premake, so that it emits correct .vcproj files with the necessary quotes.

Someone should verify that this is really the cause of the errors, and fix it.

The actor viewer (the one that's integrated into Atlas) doesn't work with the new simulation system. It needs to be substantially changed so it renders entity previews, not actors, and does all the necessary bookkeeping to interact with the sim components.

Somebody just got the error:

Details: unhandled exception (Unknown (0xC06D007E))
Location: wdll_delay_load.cpp:358 (__delayLoadHelper2)

when running on 64-bit Windows 7. This looks like it's probably the same error as somebody else (whose name I forget) reported a while ago, where I think we discovered that the delay-loader was failing to load version.dll. But I don't think we ever found or solved the cause of that problem, and it appears it's affecting more people now.

Need to work out how to integrate audio with the new simulation code.

Automatically prepend simulation scripts with "use strict"; so they run under ES5's tighter rules, to discourage bad code and potentially allow more efficient execution.

​http://www.wildfiregames.com/forum/index.php?showtopic=12996&hl=

I noticed that the 0ad-unix-build.tar contains some .bat/.dll/.exe/.lib/.vcproj files and win-specific directories that could be removed reducing tar size from 37 to 25 MB, adding to the source/tools/dist/build.sh script some --exclude options:

tar cf $PREFIX-unix-build-temp.tar --exclude='*.bat' --exclude='*.dll' --exclude='*.exe' --exclude='*.lib' --exclude='*.vcproj' --exlude='win' --exclude='win32' --exclude='include-win32' ${PREFIX}/{source,build,libraries/{cxxtest,fcollada,spidermonkey-tip,valgrind},binaries/system/readme.txt,*.txt}

I don't know if this is exactly the correct list to exclude; should double-check exactly which files are getting included and remove the unused ones from SVN or from the tar.

See ​https://bugs.launchpad.net/getdeb.net/+bug/556356 comment 4. It'd probably be nice if the build tarball was sufficient for compiling and running all the tests, and if the data tarball only included game data and not test data, assuming we can actually split them cleanly in that way.

We should have some documentation along the lines of XML.Entity for the new simulation system, aimed at people who are writing XML by hand (not writing scripts or C++) or who want an overview of the features supported by entities.

This is probably a precursor to (and obsoleted by) #422 - it's good to have readable offline documentation even if it's duplicated in a cleverer editing tool, and also it's easier than implementing the cleverer editing tool. The documentation should be generated from annotations in the RelaxNG schema so it can be kept up to date and reused. Maybe we also want some extra higher-level documentation for components, derived from the Doxygen or something? The output should probably just be plain HTML files, which can be uploaded somewhere and linked from the wiki.

I am attaching some files from debian package that may be useful to other distributions and that could eventually included in the 0ad source:

• 0ad.desktop : desktop file for menu (it assumes the executable is in /usr/lib/games/0ad/system/pyrogenesis )
• 0ad.png : png icon file (converted from upstream ps.ico and optimized with optipng and advdef)
• 0ad : script that call /usr/lib/games/0ad/system/pyrogenesis to easily run the game from terminal
• 0ad.6 : man page with options taken from upstream readme.txt

Update (2012-04-10):

• an updated png (maybe too big) is available, see ​http://www.wildfiregames.com/forum/index.php?showtopic=15902
• updated man page (0ad.6) based on updated and now more complete readme.txt is available at ​http://anonscm.debian.org/viewvc/pkg-games/packages/trunk/0ad/debian/

CCmpObstructionManager​ deals with line/circle/square intersections, for use with pathfinding and building placement etc. Currently squares aren't implemented at all (it pretends they're circles) - they should be added.

"Squares" are non-axis-aligned rectangles, so I think it's just some slightly ugly but straightforward geometry. Everything needs to be done with fixed-point maths (to get reliably repeatable results), so care needs to be taken with overflow, which makes it a bit trickier.

There ought to be some test cases for the code (see e.g. test_Position.h for the general framework for testing simulation components).

What currently happens:

$ LANG=bogus ./pyrogenesis_dbg
terminate called after throwing an instance of 'std::runtime_error'
  what():  locale::facet::_S_create_c_locale name not valid
Aborted

This can happen on misconfigured systems where the locales aren't set up correctly. It would be better to give a much more informative warning and fall back to a default - e.g. perl says:

$ LANG=bogus perl
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE = "",
        LC_ALL = (unset),
        LANG = "bogus"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").

which is more helpful.

We need to do this fairly early (before any attempts to access the filesystem), so maybe in EarlyInit in GameSetup.cpp.

Need to get the CppDoc and Doxygen output and the entity XML documentation uploaded somewhere and linked from the wiki, with an easy way to keep them updated.

CppDoc and Doxygen can probably just be run on the web server itself, given an SVN checkout. The entity XML stuff requires the .rng file which comes from compiling and running the game, which a pain so that'll probably have to be done offline and uploaded.

See ConstructSquareOnGround in simulation2/helpers/Render.cpp (used to render the outlines of selected units). Currently it constructs a 4-point square, and moves each point vertically to match the terrain. That looks a bit rubbish if a unit with a large square outline is on bumpy land.

The straight edges ought to be split into smaller segments, each of which is fitted to the terrain, so that the whole shape fits better. Rather than a fixed number of segments, I guess they should be split so each segment is slightly smaller than a tile (4 units), or whatever looks reasonable without wasting lots of vertexes.

Trained units spawn around the footprint, but some buildings have a larger obstruction shape so the units are unable to spawn. Need to fix either the code or the data files.

Currently you can't place a new building if there's any unit in the way, and it immediately blocks all movement through its area.

From ​here:

what ought to happen ![...] - units are ignored when placing the foundation, then it tells friendly units to get out of the way and blocks them from walking over the foundation any more, and a builder can get the building to above 0% build progress once there's no units in the way (including enemy units), and after that point the foundation blocks enemy units walking over it too. (That avoids your own units blocking the building, but also avoids the exploitation of foundations to block enemies).

When running ./pyrogenesis_dbg -editor on OS X, it starts up the editor window and then the window freezes (though all the code threads continue running).

In the current design, pyrogenesis_dbg starts up first, via SDL's main function, which calls [NSApplication run] and eventually ends up in our main (source/main.cpp).

We then detect the -editor flag, load libAtlasUI_dbg.dll, start up a new thread, and run Atlas_StartWindow (source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp) in the new thread, which calls wxEntry to start the wxWidgets event loop.

As far as I can tell, the problem is that we have two Cocoa event loops, one in SDL and one in wxWidgets, and the GUI thread doesn't get the input events it needs. (See ​here etc for some stuff about Cocoa threading.)

If that's true, we need to do, uh, something to fix it. (I have no idea what, but it might involve a redesign of the game/editor startup process). If that's not true (I'm not an OS X developer so I may easily be mistaken), we need to work out what the problem really is and then do something to fix it.

Need to convert all the XML, and probably implement dummy components to contain the new data, and have some way to verify that there's no accidental data loss.

Running from E:\0ad-test-é\, argv[0] is something like win1252, not utf8:

Function call failed: return value was -100703 (Invalid UTF-8 sequence)
Location: utf8.cpp:81 (RaiseError)

Call stack:

RaiseError (utf8.cpp:74)
    err = 3272664 (0x0031EFD8)
    perr = 0x02FACFB9 -> 3452816640 (0xCDCDCD00)

UTF8Codec::Decode (utf8.cpp:120)
    srcPos = 0x0031F1D8 -> 0xCCCCCCCC
    srcEnd = 0x0031F1F8 -> [8] { 0 (0x00), 0 (0x00), 0 (0x00), 0 (0x00), 204 (0xCC), 204 (0xCC), 204 (0xCC), 204 (0xCC) }
    err = 0x00000000
    size = 1 (0x00000001)
    offsets = [5] { 0 (0x00000000), 0 (0x00000000), 12416 (0x00003080), 925824 (0x000E2080), 63447168 (0x03C82080) }
    u = 23346901 (0x01643ED5)

wstring_from_utf8 (utf8.cpp:217)
    src = 0x6F8CB5CA -> (error -100500 while analyzing string<﻾td::char_traits<char> >)
    err = 0x0031FCD0 -> 3276112 (0x0031FD50)
    srcEnd = 0x00000031
    srcPos = 0xCCCCCCCC
    dst = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

Paths::Paths (paths.cpp:31)
    this = 0xCCCCCCCC
    args = 0x0031FD14 -> 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    subdirectoryName = 0xCCCCCCCC

InitVfs (gamesetup.cpp:506)
    args = 0x0031FD14 (see above)
    hooks = 
        override_gl_upload_caps = 0xCCCCCCCC
        get_log_dir = 0xCCCCCCCC
        bundle_logs = 0xCCCCCCCC
        translate = 0xCCCCCCCC
        translate_free = 0xCCCCCCCC
        log = 0xCCCCCCCC
        display_error = 0xCCCCCCCC

    readonlyConfig = 
        m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

    mods = (error -100500 while analyzing vector<CStr8 >)
    LINE_507_ = 
        m_t0 = -3689348814741910324 (0xCCCCCCCCCCCCCCCC)
        m_description = 0xCCCCCCCC

    logs = 
        m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

    modLoosePath = 
        m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

    modArchivePath = 
        m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

    paths = 
        m_root = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

        m_rdata = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

        m_data = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

        m_config = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

        m_cache = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)

        m_logs = 
            m_path = (error -100500 while analyzing basic_string<wchar_t,char_traits<wchar_t> >)


    cacheSize = 3435973836 (0xCCCCCCCC)

Init (gamesetup.cpp:855)
    args = 0x00000001
    flags = 49970672 (0x02FA7DF0)
    setup_vmode = 204 (0xCC)
    hooks = 
        override_gl_upload_caps = 0xCCCCCCCC
        get_log_dir = 0xCCCCCCCC
        bundle_logs = 0xCCCCCCCC
        translate = 0xCCCCCCCC
        translate_free = 0xCCCCCCCC
        log = 0xCCCCCCCC
        display_error = 0xCCCCCCCC

    setup_gui = 204 (0xCC)
    quality = 3435973836 (0xCCCCCCCC)
    missing = 0xCCCCCCCC
    windowed = 204 (0xCC)

RunGameOrAtlas (main.cpp:384)
    argc = 1 (0x00000001)
    argv = 0x02FA7DF0 -> 0x02FA7DF8 -> "E:\0ad-test-é\binaries\system\pyrogenesis_dbg.exe"
    ran_atlas = false
    args = 
        m_Args = (error -100500 while analyzing vector<pair<CStr8,CStr8> >)
        m_Arg0 = { (error -100500 while analyzing string<﻾td::char_traits<char> >) }

    res = 1.6976e-314 (0x00000000CCCCCCCC)

main (main.cpp:421)
    argc = 3276292 (0x0031FE04)
    argv = 0x00FAD184 -> 0xC7E04589 -> (unavailable - internal error)


__tmainCRTStartup (crtexe.c:410)

mainCRTStartup (crtexe.c:393)

CallStartupWithinTryBlock (wseh.cpp:357)
    ret = 3276240 (0x0031FDD0)

00000000
    ret = 23186032 (0x0161CA70)

00000000
    ret = 0 (0x00000000)

RtlInitializeExceptionChain (:0)


errno = 0 (?)
OS error = 50 (The request is not supported.
)

Currently, shadows and geometry have a visible pop in/out at the edges of the screen due to culling too early.

I'm not sure what the previous comment is referring to with shadow maps becoming too low-detail with a larger culling frustum, but one way to fix shadows popping in at the edge of the screen is to widen the culling frustum so that objects aren't culled as early. The downside is that objects will get rendered even though they appear off screen, which is somewhat inefficient. A better solution would be to decouple shadow map rendering from model rendering; as it is now, a shadow map is only rendered if its associated model is to be rendered (I think).

Needs more testing, plus various fixes like:

• Map selections should be reflected to other players
• Chat should work, to some extent
• Players should have persistent configurable names
• Display "waiting for other players" dialog after loading
• Notify players about disconnections
• Notify players about lag

and probably more - it doesn't need to be perfect at all, it just needs to be vaguely understandable and usable and needs to tell people what's going on.

18:32 < janwas> booom in CNetSession::SetPlayerSlot
18:32 < janwas> because m_psession == 0
18:33 < janwas> oops, that happened because i fat-fingered "host", aborted that, then did join

The state hashing is very useful for detecting OOS, but very slow (especially in debug builds) which makes it hard to use when testing. It should be faster.

When running fullscreen on a multi-monitor system, the mouse can easily leave the game window and move onto the other screen, which breaks edge-of-screen scrolling, and if you accidentally click outside then the game minimises which is quite irritating.

When fullscreen, the mouse ought to be restricted to the window (presumably via ClipCursor).

Documentation for enet_initialize says:

Must be called prior to using any functions in ENet.

We don't call it at all. We probably should. (It does nothing on Unix, but calls WSAStartup and timeBeginPeriod on Windows; need to make sure that doesn't interfere with our own wsock or timer code.)

jd823592 on IRC reports problems, having compiled the game in Debug mode on MSVS2008 9.0.30729.1 SP on WinXP SP2 (and having run update-workspaces and 'clean'):

'pyrogenesis_dbg.exe': Loaded 'C:\0ad\binaries\system\libpng14d.dll'
LDR: LdrpWalkImportDescriptor() failed to probe c:\0ad\binaries\system\libpng14d.dll for its manifest, ntstatus 0xc0150002
'pyrogenesis_dbg.exe': Unloaded 'C:\0ad\binaries\system\libpng14d.dll'
First-chance exception at 0x7c812a6b in pyrogenesis_dbg.exe: 0xC06D007E: Module not found.
Unhandled exception at 0x7c812a6b in pyrogenesis_dbg.exe: 0xC06D007E: Module not found.

failing in tex_png.cpp:209 (a call to a libpng function).

Currently ctrl+backspace/delete is ignored completely, which is better than how some programs react to it (i.e. placing an "unknown character" box instead of removing the entire word - I'm looking at you Photoshop =) ). It would be nice to have it working like this: Ctrl+Backspace = remove the word to the left of the cursor (i.e. the entire word left of the cursor, nothing to the right of the cursors), Ctrl+Delete the same but on the right side of the cursor.

As you can see by the priority it's not something too important, but it would be nice to have it working, and it can be a good relatively simple task for someone wanting to get more familiar with the engine.

I've got an updated repository and workspace. I've successfully build both the release and debug executables in a fully patched VC2008 (9.0.21022.8 RTM) under Windows 7. I am able to run the release executable, but unable to run the debug executable -- either from within VC++ or outside.

Solution: Since there is no VC90 version DebugCRT.manifest, the program picks up the VC80 version which has the wrong assemblyIdentity/version (recently and correctly changed in ticket #510). The fix is easy, just add a Microsoft.VC90.DebugCRT.manifest to the binaries\system directory. I've attached the file which is based on the existing non-debug version, Microsoft.VC90.CRT.manifest.

See source/lib/sysdep/dir_watch.h and source/lib/sysdep/os/osx/dir_watch.cpp. The interface needs to be implemented on OS X, so we can automatically reload changed files (for quick testing during development).

See linux/dir_watch_fam.cpp and win/wdir_watch.cpp for possible inspiration.

When doubleclicking a unit, you should select all units of the same type which are visible at once. This behavior was described in input.js comments, but not yet implemented.

With this ticket I supply a patch, implementing this feature.

See ​this suggestion for a blur/smooth tool.

I expect it can be implemented like:

• Copy source/tools/atlas/AtlasUI/ScenarioEditor/Tools/AlterElevation.cpp to SmoothElevation.cpp, and make the necessary name changes.
• See binaries/data/tools/atlas/scripts/section/terrain.js and add the new tool to the list in there.
• See source/tools/atlas/GameInterface/Messages.h and add the new one (like where it has AlterElevation)
• See source/tools/atlas/GameInterface/Handlers/ElevationHandlers.cpp and add the new one.

As for actually implementing the smooth functionality, it'll probably need some thinking and experimenting to discover what's a good smoothing algorithm that behaves usefully and intuitively.

Ideally the same tool would roughen too (when right-clicking), as basically the exact opposite of smoothing.

There's a few things which it'd be good to do:

• Test ​these packages on all (or at least most) of the various platforms. (Can we easily get VMs for testing?)
• Use ​PPAs to build Ubuntu packages, so we can keep them regularly updated ourselves.
• Update Gentoo ebuilds to install into proper locations (not /opt).
• Add installation instructions for all these packages.

Self-extracting archives are not ideal - we should probably have a (very basic) NSIS-based installer. Select an install location, choose whether to install source code, create a start menu entry and uninstaller (which should probably clean up the cache etc files too).

I use glXSwapIntervalSGI to activate vsync in Linux. This is found in GL/glxext.h. Which needs GL/glx.h. I put those inlcudes to the according places where i found the wgl.h and wglext.h Windows ones.

More information about this extension: ​http://www.opengl.org/registry/specs/SGI/swap_control.txt

The problem is that glx.h inludes X11/X.h which defines Cursor. So i replaced the Cursor stuff in cursor.cpp with GameCursor... I have no other idea how to solve this. Maybe using namespaces?

The other mystery is that ogl_HaveExtension("GLX_SGI_swap_control") returns false. Do I have to register it somewhere? I did not get the magic of ogl_HaveExtension in ogl.cpp. But it runs without checking for me. This should be temporary.

The ifdefs are confusing with OS_WIN, OS_MACOSX and OS_MAC. Is there OS_UNIX or OS_LINUX? So I could ensure a better separation. Maybe this extension could be usefull in OS X too, but I can't test it.

The ability to create Rally Points for buildings by right clicking on the ground after selecting a building. All newly created units from that building will walk towards the rally point soon after creation.

I tried 0ad pre alpha 3. Despite my fairly modern computer (3 years old, 2gb ram, 256-512mb graphics card), it was running slow (moving mouse had about a second's delay). So I've checked out the latest source and tried to run it, with no change. It gave the follow message:

Your graphics drivers do not support S3TC compressed textures. ...[snip]...
Installing the libtxc_dxtn library will fix these problems.

So I followed the link in the message, compiled and installed the library, and things got FAR WORSE!! Moving the mouse had a 5 second delay, and for only 1/4 of the distance I moved the mouse (non-technical term would be jumpy). I couldn't even reach the exit button, so had to force quit the game and uninstall the dxtn library.

I've already taken advice to disable water and shading etc etc. But with no change, because... this slowness is happening on the main menu, not even within the game where most of the effects are.

It's not just this computer. Another computer, with a different OS, different specs, has the same speed issues. I love the concept of an AOE like game, and looking forward to when multiplayer is actually possible, but at it's current state, I'd be lucky to find anyone with a computer fast enough to show me what the game is like.

What am I doing wrong? What information can I provide you with to help get this speed up? I'm very keen to be of any help to get this sorted, so please demand anything from me :-P I am a web developer, and I have slightly dabbled in C before, so if you want any small source changes on my end to gather debug data, let me know what they are.

Keep up the great work, and look forward to solving this speed issue.

ps/VideoMode.cpp does SDL_GL_SetAttribute(SDL_GL_SWAP_CONTROL, ...) to set vsync.

On Windows we emulate the SDL API, and don't support that attribute. ps/GameSetup/GameSetup.cpp uses WGL_EXT_swap_control instead. That OS-specific code ought to be moved into lib/sysdep/os/win/wsdl.cpp to implement the standard SDL attribute.

When a building's rally point is a long way away, it's easy to lose the little flag marker. It'd probably be nice if we could display some kind of line between the building and rally point once you select the building, so you can follow it visually to find the end.

I don't know what the line should look like. It shouldn't be too distracting (people will often select buildings for other reasons and not care about the rally point), but it shouldn't be too subtle to notice.

See ​call stack (from Brian), particularly:

        pyrogenesis_dbg.exe!_ia32_asm_CAS()  + 0x10 bytes       
>       pyrogenesis_dbg.exe!ModuleInit(volatile int * initState=0x00c55a7c, long (void)* init=0x004b4ff0)  Line 44 + 0x10 bytes C++
        pyrogenesis_dbg.exe!ApicIds()  Line 172 + 0xf bytes     C++
        pyrogenesis_dbg.exe!VerifyRunningOnCorrectProcessors(unsigned long affinity=1)  Line 252 + 0x5 bytes    C++
        pyrogenesis_dbg.exe!os_cpu_SetThreadAffinityMask(unsigned int processorMask=1)  Line 275 + 0x9 bytes    C++
        pyrogenesis_dbg.exe!os_cpu_CallByEachCPU(void (unsigned int, unsigned int)* cb=0x00414380, unsigned int cbData=12933656)  Line 293 + 0xb bytes  C++
        pyrogenesis_dbg.exe!InitApicIds()  Line 160 + 0xf bytes C++
        pyrogenesis_dbg.exe!ModuleInit(volatile int * initState=0x00c55a7c, long (void)* init=0x004b4ff0)  Line 46 + 0x5 bytes  C++
        pyrogenesis_dbg.exe!ApicIds()  Line 172 + 0xf bytes     C++

The first ModuleInit call locks the state while calling init() which then tries to lock the state itself.

Need some basic descriptions of the maps in public/maps/scenarios. Also need to give them better names.

Currently, players in the game are shown as "Player 1", "Player 2", etc. The names ought to match the names given in the (multiplayer) game setup screen.

Possibly the names should be entirely a GUI thing - the simulation code will only care about player ID numbers and will never know the names.

The maximum zoom should perhaps be the current default. Shift+wheel should rotate the camera. Q, E should rotate. Simplify the other rotation controls. Fix the jerky zoom. Add a dev command to unlock the camera movement.

Enforce Population Limit when training units. Units can be queued up for training past the population limit, but training will start only when there are enough population slots available.

Steps to reproduce:

• Optionally set up a .cfg file to start the game in windowed mode with a certain xres/yres (else the default is 1024x768).
• Set display resolution to smaller than that size.
• Start the game.

The initial window is shrunk to fit on the screen, but the game renders everything at the configured size instead (so the menu screen is partially off the side of the screen). It should detect this kind of situation and update things to fit.

(ps/VideoMode.cpp is probably a relevant place to look for this.)

Because we don't want to allow military or merchant ships to fish, we should have dedicated fishing ships. The first we will need will be for the Hellenes. The history department might have relevant information and images. If not, then research will need to be done first. It will require a new model and texture. It might be possible to reuse an existing rigging set from another ship; otherwise that will need to be made also.

Some entities (farms and fish) should not obstruct other entities. They must still register collisions, so that they can be gathered from. This means that their size cannot simply be set to zero. Some possibly relevant files are:

CCmpObstruction.cpp, CCmpObstructionManager.cpp

There may be other files that need to be modified.

Kimball reports messages like:

../../../source/lib/sysdep/cpu.cpp:38: error: invalid conversion from ‘volatile int64_t*’ to ‘volatile intptr_t*’
../../../source/lib/sysdep/cpu.cpp:38: error:   initializing argument 1 of ‘bool cpu_CAS(volatile intptr_t*, intptr_t, intptr_t)’
../../../source/lib/timer.h: In member function ‘void TimerUnit::AddDifferenceAtomic(TimerUnit, TimerUnit)’:
../../../source/lib/timer.h:180: warning: dereferencing type-punned pointer will break strict-aliasing rules

which looks like it's perhaps using the ARCH_AMD64 branch on a 32-bit target. (Maybe we should add cassert(sizeof(void*) == 8) etc to verify the arch selection earlier?)

Reportedly:

$ g++ -v
Target: i686-apple-darwin10
Configured with: /var/tmp/gcc/gcc-5664~38/src/configure --disable-checking --enable-werror --prefix=/usr --mandir=/share/man --enable-languages=c,objc,c++,obj-c++ --program-transform-name=/^[cg][^.-]*$/s/$/-4.2/ --with-slibdir=/usr/lib --build=i686-apple-darwin10 --program-prefix=i686-apple-darwin10- --host=x86_64-apple-darwin10 --target=i686-apple-darwin10 --with-gxx-include-dir=/include/c++/4.2.1
Thread model: posix
gcc version 4.2.1 (Apple Inc. build 5664)

$ gcc -dumpmachine
i686-apple-darwin10

$ echo $HOSTTYPE
x86_64 

Changing premake.lua to correctly use x86 doesn't seem to change much (the ARCH_AMD64 comes from the source code, not the build system).

I'm not sure what we need to do to help debug this.

Give builders the ability to repair damaged buildings.

See source/tools/dist/0ad.nsi (NSIS installer script). It adds the map editor to the Start menu, and attempts to use AtlasUI.dll's icon, but that file has no icon. We need to somehow get it to use the right icon (currently found in binaries/data/tools/atlas/icons/ScenarioEditor.ico), either by associating that icon with the DLL or by using that icon by itself.

My patch ignore height of terrain.

Per discussion ​here, it would be good to try disabling the skybox and rendering flat black when looking outside the terrain.

Water reflections should still use the skybox as normal.

See CRenderer's calls to RenderSky. I think the background colour is set by Render in GameSetup.cpp, so that should probably be changed to always set it to black.

Someone reports startup errors:

ERROR: SetVideoMode failed: 2960x1050:24 1 ("No video mode large enough for 2960x1050")
ERROR: SetVideoMode failed: 2960x1050:24 1 ("No video mode large enough for 2960x1050")
terminate called after throwing an instance of 'PSERROR_System_VmodeFailed'
  what():  System_VmodeFailed

Probably CVideoMode should determine when fullscreen mode fails, then try windowed mode instead of giving up entirely.

When the game is running in fullscreen, the mouse should be grabbed so it cannot move outside the game. I have multiple monitors, so when I try to pan the camera by putting the cursor along the edges, it moves right into the next monitor, which is a major annoyance.

Attached is a patch to fix the issue.

The ideal functionality for mouse grabbing would probably be to have it always grabbed in fullscreen, and optionally grabbed during gameplay in windowed mode (but not on menus, dev console, text entry, etc). I can also implement this if desired (it'd be a good way for me to start getting my feet wet).

When the game is paused, water still appears to flow. I've noticed this in games before and it always bugs me. Fix attached.

PROBLEM: after compiling and running test_dbg stack call trace lead us to the problem in lib/sysdep/os/linux/lcpu.cpp file while calling to sched_getaffinity function which fails with errno 22 (EINVAL) if you choose Launch debug and test errno after calling to this function.

DESCRIPTION: problem comes not from the game, but from incompatibility of the Linux kernel with NUMA support and glibc. The whole description with solution you may find by the link in the section below.

SOLUTION: ​http://sourceware.org/ml/libc-help/2009-10/msg00017.html

When a training queue contains a item at the front that hasn't started yet, and the player doesn't have a high enough population limit to start training it, it won't start training. Currently the player is never notified about that - the queue looks normal, with no indication the player needs more houses (or more corpses). We need some kind of notification (I don't know what - maybe a message in the tooltip in the training queue? and/or a message sent to the player's screen? and/or make the population counter flash red? etc).

Units can be ordered to move to a certain place by right clicking on the mini-map.

When I start Atlas in Ubuntu 10.04 it doesn't load maps, when I try to load someone it shows "Failed to run rmgen (error code: 255)".

At now users with new boost library are not able to build 0ad because of error like:

../../../source/lib/file/vfs/vfs_path.h:68: error: ‘is_basic_path’ is not a template
../../../source/lib/file/vfs/vfs_path.h:69: error: explicit specialization of non-template ‘boost::filesystem::is_basic_path’

​http://www.wildfiregames.com/forum/index.php?showtopic=13327

I think it might be a bug in boost but the default api will be changed soon: ​http://live.boost.org/doc/libs/1_44_0/libs/filesystem/v3/doc/index.htm "Version 3 is a major revision with many new and improved features. It breaks some Version 2 code.

Version 2 is the default version for Boost release 1.44 only. Version 3 will become the default starting with release 1.45."

The problem is the renaming of namespaces (unfortunatelly I haven't found any official info about it) and the compiler can't find "is_basic_path".

I think the attached patch will not break building with older boost (I've tested it only with 1.44).

I've got an error while building 0ad:

../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp: In member function ‘virtual bool AtlasDLLApp::OnInit()’:
../../../source/tools/atlas/AtlasUI/Misc/DLLInterface.cpp:191:28: error: ‘wxHandleFatalExceptions’ was not declared in this scope

It is because the wxHandleFatalExceptions' declaration in wx-2.8/wx/utils.h depend on config in wx/setup.h where I have "#define wxUSE_ON_FATAL_EXCEPTION 0". I can see in utils.h:

#if wxUSE_ON_FATAL_EXCEPTION

// should wxApp::OnFatalException() be called?
WXDLLIMPEXP_BASE bool wxHandleFatalExceptions(bool doit = true);

#endif // wxUSE_ON_FATAL_EXCEPTION

I've disabled debugger's checking locally like in attached patch but I think it is not the best solution.

When camera is being moved either by mouse or by keyboard it can be placed outside the terrain, so that the user will see only a black ambient. The right camera handling should never move its position outside the terrain.

See ​discussion.

• Create a metadata format for expressing DDS compression options.
• Create sensible metadata for each texture currently in the game.
• Create a basic tool for visually testing the compression options.
• Make the engine read the metadata when loading a non-DDS texture, and automatically compress and cache the file on first load.
• Make the release packager perform the compression in advance.
• Make the release packager do XMB and Collada conversions too, and make this conversion pipeline less ad-hoc, and pack everything into a zip.

The camera outline clips correctly on the left, overlaps the borders on the bottom and right, and clips prematurely on the top. glScissor is being passed incorrect values. Attached is a patch which fixes the issue.

This patch pertains to the water rendering when 'fancy' mode is not enabled, and is purely aesthetic.

Currently, in non-fancy water mode a single fresnel value approximation is being calculated based on camera orientation and applied to the whole map. This is incorrect behaviour, as at low angles near water should be largely transparent (low fresnel) while distant water should appear opaque (high fresnel). Attached is a replacement that calculates fresnel per-vertex and addresses other water lighting issues in low-detail mode.

Thank-you for your time,

Robin

There is no user friendly way to return to the main menu from the map selection screen, other than to select a map to be loaded and then cancel from there. There needs to be a "Cancel" or "Return to Main Menu" or equivalent button.

The main menu suffers from the same problem as identified in ticket # 558. There needs to be an easily identifiable "Exit Game" button or similar parchment mouse-over graphic button, not just a small close button positioned out of the way. Running fullscreen on a 1080p display, this button is not even within the user's field of view.

The tooltips for the main menu mouse-over parchment graphics pop up at inappropriate places. For example, pointing at or around the "zero" to the left of the A.D. will popup either the single player, multi-player, or options tooltip.

The appropriate gui/pregame/sprites.xml and mainmenu.xml values need to be adjusted.

For the options graphic, make the following changes:

• sprites.xml
  • pgOptionsOver and pgOptionsDisabled
    • real_texture_placement="58 87 220 251"
• mainmenu.xml
  • pgOptionsBt
    • size="80.5664063% 53.6458333% 95.99609375% 74.609375%"

We have passive, idle, and skittish.

Add violent, aggressive and defensive behaviors to animalAI.js

Possibly generalize the code so it can be applied to opponent AI (#707)

IVFS::GetRealPath claims to "retrieve the real (POSIX) pathname underlying a VFS file", but it seems to work incorrectly with mods. In particular, it always returns a path like .../data/mods/public/... (or always internal if that mod is loaded), even when the VFS file comes from a different mod.

CXeromyces::GetXMBPath tries to get the mod name from this, which doesn't work since it always returns public (or always internal).

Getting the mod name seems like useful functionality - by separating caches per mod, we can avoid one mod maliciously polluting the cache with a file with the same size/timestamp as another mod's but with different content (which would probably cause OOS errors after the first mod is disabled).

I don't know what GetRealPath should return if the file is in an archive. (It probably doesn't matter here, since archives ought to already contain cached files and we won't need to dynamically generate them.)

I implemented basic logic for checking victory conditions (only "conquest" for now). There is:

• new code in player class for buildings count tracking;
• EndGameManager javascript component, which check defeat/victory conditions by timer and changes player states;
• new code in session.js, which check state of current player and ends the game if player have been defeated/won:
• mockup gui page for game summary.

the commit changed something in population counts, one of the tests didnt like that.

Hotkey system needs to be totally reviewed. The main thing to correct is to reduce confusing things like defining hotkeys in Hotkey.cpp and Hotkey.h. They won't work until a new string is added to default.cfg. Some other notes will be added during the process of enhancement.

The gatehouse acts as a single square rather than two squares and a rectangle.

Edit: This has been pushed back to a later release, but see the comments below for issues that have been fixed.

Need to convert and cache and archive them all. Only exceptions are fonts and cursors (and blends?) since those don't use the new system.

If I load up a map like Arcadia and click and drag on the minimap, the view follows the cursor (as it should). After dragging over a mountain, the view becomes increasingly zoomed in, which is annoying.

If you do a full zoom, you're not able to get to the top of the map, you have to zoom out a little. And if you zoom in again, the camera will automatically move back.

I used 'top' here, instead of 'north' because if you rotate the camera, it happens on every border on top.

On many Linux systems, we need the user to force the enabling of S3TC in their drivers. (We no longer require the drivers to do compression, so we don't need libtxc_dxtn). Currently we tell them to run driconf and do it themselves.

Perhaps the game could automatically edit the ~/.driconf file and add an application-specific entry to set force_s3tc_enable. Given the ​documentation it sounds like the intention is that the file is a standard format that can be edited by multiple tools, so we probably shouldn't suffer from incompatible format changes. But maybe this would still be too fragile and/or evil?

The following happens ocasionally after 20 minutes or so of playing, with lots of soldiers (30-40) and few other units.

Now, if I select as many soldiers as I can (selection limit) and then try move them to another location, the game crashes. Remember, this does not happen every time (I'd say every tenth attempt, but that's just guessing) and is quite hard to reconstruct exactly. It just happens.

Judging from the crashlog, I'd say it's a bug in the sound manager, but I don't have any experience with the code, so don't listen to me :)

As far as I an tell, it doesn't matter on which map I'm playing, and I haven't yet found any other patterns.

When watching an object through a water reflection, but this object is outside the camera, unwanted behaviour is observed. I uploaded a video to show the bug.

​http://www.vimeo.com/15241286

Related: #504

To avoid problems like ​this, we should just remove the code that uses libbfd, since it's not particularly useful - it's just used for basic stack traces, which rarely provide enough useful information for debugging any problems, and we can easily get Linux users to run in gdb if we want more details.

In the map editor, sometimes you can get the camera lost (badly rotated or scrolled off the side of the world) so it's hard to recover. There should be some way (hotkey? button? menu command?) to reset the camera to an on-world position and default orientation.

Also there should probably be a way to reset to the game's default camera orientation without changing position, so designers can see the same perspective players will usually see. Maybe this should be combined with the recovery thing - always reset to default orientation, and if the camera's still off the side of the map then move it back on.

I was trying to install this via your ppa, but you have created no version for Maverick yet. As it comes out on 10 October, many users will upgrade and find the ppa is broken. It would be great if you could fix this on time.

I updated the FontBuilder tool to support Windows and *nix (in the same scripts). Now it checks system platform before loading Freetype and Cairo libraries. Only difference is library names vary but interface seems to be identical.

You probably noticed this bug when trying to set the sound.mastergain config variable to 0.0, in order to mute sound. The sound would be inexplicably loud. Incorrect conditional in Config.cpp caused gain to remain at default value of 1.0

Using the mouse to rotate produces unexpected results and makes wall/gate building much harder. It would be nice to have the 'q' rotate 45 degrees anti-clockwise, and the 'w' key rotate 45 degrees clockwise, when a building is selected for construction.

I'm going to attempt to write a patch in my free time, but if it hasn't been done yet, please go ahead and do this yourself. From what I understand, it should be very easy.

• Adds player data to map files

• Adds support for civ data including "subfactions"

• Adds support for other map types (random, saved?, etc)

• Player options displayed in setup

• Player entities initialized by script (instead of engine)

• Generalizes XML loading for scripts

• Civ info displayed in dialog

See discussion ​here.

If an enemy unit is in fog-of-war/shroud-of-darkness areas, you shouldn't be able to hear them.

This could probably be handled by making CCmpSoundManager::PlaySoundGroup detect if the source entity is visible to the current player (see e.g. CCmpVisualActor::Interpolate using ICmpRangeManager::GetLosVisibility) and then ignoring the sound if not.

When a group of units is ordered to do anything other than walk, the formation controller disbands and each unit performs the order individually. If the formation controller had queued orders, they get lost.

Instead, units ought to stay associated with the formation controller while they're individually carrying out the action (attacking, gathering, etc). The formation controller should detect when its order has been completed (target died, resource exhausted, etc) and then regroup the units and move onto its next order.

When a group of units are ordered to attack/gather/repair a target, they should walk in formation until they're 'close' to the target (a fixed distance, plus their attack range if it's an attack order?), before splitting up to individually reach the target.

Shouldn't be allowed to build buildings in fog/shroud areas. Construction previews should correctly highlight red when obstructed.

Units, which pursue the enemy, should only go as far as they last saw the unit. If when they get there, they don't see it, they return to where they were originally. If they see the unit, or other enemy unit, they pursue until they kill it, or lose it. Again, in both cases, they should return to where they started off, not remain where they lost/killed the enemy unit.

When a selected enemy unit disappears into the FOW or black zone, it should be unselected automatically. You can't follow a unit you can't see.

Discovered stone mines glow when they should be dimmed out because I have no units within the area

The rendering of fog-of-war / shroud-of-darkness is currently based on vertex colours, and it looks quite ugly. It'd be nice to have much smoother transitions between the different regions. See e.g. ​AoE3's rendering (AoM is very similar).

When a building (or tree etc) is visible in a fog-of-war region, no new information should be revealed to the player - they shouldn't see when its hitpoints change, or when it's destroyed, etc. Perhaps they shouldn't be able to select the building at all (even just to see its name).

Probably this would have to be implemented by replacing buildings with 'mirage' entities when they fall into FoW, which are clones of the original buildings with all the dynamic/interactive bits removed. (These would have to be serialised in saved games, and therefore network-synchronised, but only rendered for the appropriate player.)

We also have to deal with the case of players ordering units to attack or gather from a mirage entity - when it comes back into vision, the units should automatically switch to attacking/gathering the real entity (or realise that it's been destroyed).

In the default box formation used for moving groups of units, the units should be put into appropriate places depending on their type.

Suggested layout:

   ^   ^   ^   ^   ^
  [S] [S] [S] [S] [H]
  [C] [I] [I] [I] [C]
  [C] [I] [I] [I] [C]
  [C] [A] [A] [A] [C]

(for super-units, heroes, cavalry, infantry, archers). The layout obviously needs to adapt depending on what arbitrary selection of units are in the group.

Also, units should be assigned to a 'nearby' slot of the appropriate type (rather than simply using numerically ordered slots), to minimise the amount of reshuffling when units enter or leave the formation.

If a selection consists entirely of female citizens, move orders shouldn't cause them to move in strict formation (since they're not trained soldiers). They should probably stay in their initial layout, or move into some random loose group or something.

When they're in a group with soldiers, probably they should then be in formation (the soldiers are telling them where to stand), or maybe they should hang around in a loose group behind the soldiers.

The game needs a way to select units to add to a group of already selected units.

All of the following happen when holding the SHIFT key:

• Clicking a single unselected unit should add them to the group of already selected units

• Clicking a single selected unit should remove them from the group of already selected units

• Dragging over a group units where any/all of them in the dragzone are unselected should select the unselected ones, while keeping the selected ones selected

• Dragging over a group of units where all of them in the dragzone are selected should remove them all from the group.

The latter two options should show the dragging zone as normal, and use the unit selection rings to indicate what will happen. e.g. for the last one, unselecting a group of units should remove the unit selection ring as you dragzone over them.

... on my hudson compiling 0ad snaps. it runs on f13.

Apply the patch for ​this once it's available.

I checked the sourcecode of 0ad with cppcheck (a static code analysis tool). It has found a couple of possible error and style problems. I attachet the log to this ticket.

Would someone want mails from my Hudson instance when this happens? Or would you like support to set up a public build server on your infrastructure? Currently I use Hudson the most, but I could help set up whatever.

After latest nvtt changes 0ad no longer compile correctly:

test_test.cpp
test_CStr.cpp
/usr/bin/ld: warning: libnvcore.so, needed by ../../../binaries/system/libnvtt.so, not found (try using -rpath or -rpath-link)
/usr/bin/ld: warning: libnvmath.so, needed by ../../../binaries/system/libnvtt.so, not found (try using -rpath or -rpath-link)
/usr/bin/ld: warning: libnvimage.so, needed by ../../../binaries/system/libnvtt.so, not found (try using -rpath or -rpath-link)
../../../binaries/system/libnvtt.so: undefined reference to `nv::AlphaBlockDXT5::index(unsigned int) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::isSingleColor() const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::pixels()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::Image()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::createImage(unsigned int, unsigned int) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::BoxFilter::BoxFilter()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setHeight(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::String::setString(char const*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::FloatImage(nv::Image const*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setLinearSize(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setTexture2D()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::~Image()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::fastDownSample() const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Quantize::FloydSteinberg_BinaryAlpha(nv::Image*, int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::init(nv::Image const*, unsigned int, unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::AlphaBlockDXT5::setIndex(unsigned int, unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::height() const'
../../../binaries/system/libnvtt.so: undefined reference to `vtable for nv::KaiserFilter'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setPixelFormat(unsigned int, unsigned int, unsigned int, unsigned int, unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `vtable for nv::TriangleFilter'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::allocate(unsigned int, unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setTextureCube()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::downSample(nv::Filter const&, nv::FloatImage::WrapMode) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::hasDX10Header() const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::swapBytes()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::Image(nv::Image const&)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::normalizeNormalMap(nv::FloatImage*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setFourCC(unsigned char, unsigned char, unsigned char, unsigned char)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::StringBuilder::StringBuilder()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::mem::free(void const*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::ColorBlock()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::resize(nv::Filter const&, unsigned int, unsigned int, nv::FloatImage::WrapMode) const'
../../../binaries/system/libnvtt.so: undefined reference to `nvAbort(char const*, char const*, int, char const*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::splatX()'
../../../binaries/system/libnvtt.so: undefined reference to `vtable for nv::BoxFilter'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Quantize::FloydSteinberg(nv::Image*, unsigned int, unsigned int, unsigned int, unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::StringBuilder::StringBuilder(char const*)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setWidth(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setMipmapCount(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::TriangleFilter::TriangleFilter()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::AlphaBlockDXT5::evaluatePalette(unsigned char*) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Quantize::BinaryAlpha(nv::Image*, int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::swizzleDXT5n()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::createNormalMap(nv::Image const*, nv::FloatImage::WrapMode, nv::Vector4 const&, nv::Vector4 const&)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Filter::~Filter()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::KaiserFilter::KaiserFilter(float)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::width() const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::Image::scanline(unsigned int) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::ColorBlock::splatY()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::DDSHeader()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::BlockDXT1::evaluatePalette(nv::Color32*) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::KaiserFilter::setParameters(float, float)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setPitch(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::StringBuilder::reset()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::createImageGammaCorrect(float) const'
../../../binaries/system/libnvtt.so: undefined reference to `nv::mem::malloc(unsigned int)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::StringBuilder::~StringBuilder()'
../../../binaries/system/libnvtt.so: undefined reference to `nv::StringBuilder::copy(nv::StringBuilder const&)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::FloatImage::toLinear(unsigned int, unsigned int, float)'
../../../binaries/system/libnvtt.so: undefined reference to `nv::DDSHeader::setNormalFlag(bool)'
collect2: ld returned 1 exit status
make[1]: *** [../../../binaries/system/pyrogenesis_dbg] Errore 1
make: *** [pyrogenesis] Errore 2
make: *** Attesa dei processi non terminati....
test_CLogger.cpp

Patch to implement Garrisoning in buildings Features Implemented: Auto Unloading of units when the building's health reduces to a particular amount Filtering unit classes on garrisoning Fixed "GetPosition() called when IsInWorld() is false" error Unload All Auto Healing garrisoned units

Features Not Implemented: Special cases for Ships Garrison flag appearing on top of the building Playing Garrisoning sound

Other things to be done: Modify XML to set the class/max number of units which can be garrisoned in a building, heal rate etc

Re r8319:

I'm on 64-bit Linux with 4GB RAM. In ChooseCacheSize, I get values like total == 3924, available == 303. (I've apparently got around 1.2GB of OS caches/buffers, which are not counted as available - the available figure is always pretty low). That results in ChooseCacheSize choosing its "below min-spec" path (which is silly) and returning 2304 MB (which is also silly, and if this were a 32-bit OS (with PAE) it could be trying to allocate more than the entire virtual address space which sounds problematic).

The calculation seems strange anyway, and lacking in rationale. We don't have 500MB of (public) game data (it's more like 250MB), the 'game' isn't 400MB (it depends on the amount of data and it often looks more like 200MB for me (including file cache)), we don't load all the data at once (the point of biomes was partly to cut the amount of data required per map by only using a well-defined subset). Also we have lots of other caches that reduce the need for everything to be kept in the file cache (I'm assuming it's only useful if a file gets loaded twice) and that contend with the file cache for RAM (it seems better to spend the RAM on higher-level caches to save on loading cost), so it's not clear that a large file cache has any benefit and/or doesn't cause any harm. Also it's contending with the OS disk cache, which will cache the compressed .zip file and therefore make better use of RAM than our own file caching.

Also it looks like OperatingSystemFootprint is going to give a highly uninformative assertion failure message as soon as somebody runs the game on Windows 8 (and even when we update the code, people will still try running old releases), which sounds like a bad idea.

So it seems that this change is adding a lot of complexity and unpredictable non-deterministic system-dependent performance-affecting behaviour, without it being clear that it's any better than the old fixed-size cache. So I'd be much happier if either it was made much simpler again, or if it was made clear to me why I'm missing the point and it's actually a good idea :-)

When closing the editor I always get a:

Mesa: User error: GL_INVALID_OPERATION in glDeleteShader

I compiled mesa and 0ad with debug symbols and put a break on output_if_debug. I get this backtrace:

Breakpoint 1, output_if_debug (prefixString=0x2919dcc "Mesa: User error", outputString=0xbfffcb6c "GL_INVALID_OPERATION in glDeleteShader", 
    newline=1 '\001') at main/imports.c:809
809	{
(gdb) bt
#0  output_if_debug (prefixString=0x2919dcc "Mesa: User error", outputString=0xbfffcb6c "GL_INVALID_OPERATION in glDeleteShader", 
    newline=1 '\001') at main/imports.c:809
#1  0x02759a16 in _mesa_error (ctx=0xb6fc4c78, error=1282, fmtString=0x2928cec "%s") at main/imports.c:998
#2  0x0276f4fe in _mesa_lookup_shader_err (ctx=0xb6fc4c78, name=1, caller=0x291c4b9 "glDeleteShader") at main/shaderobj.c:170
#3  0x0276e2df in delete_shader (ctx=0xb6fc4c78, shader=3221212012) at main/shaderapi.c:449
#4  0x084470d3 in Ogl_Program_dtor (p=0x8829540) at ../../../source/lib/res/graphics/ogl_shader.cpp:430
#5  0x08444582 in h_free_idx (idx=2, hd=0x8829528) at ../../../source/lib/res/h_mgr.cpp:613
#6  0x08444d6e in Shutdown () at ../../../source/lib/res/h_mgr.cpp:838
#7  0x084646f1 in ModuleShutdown (initState=0x861fe8c, shutdown=0x8444cd2 <Shutdown()>) at ../../../source/lib/module_init.cpp:71
#8  0x08444e4b in h_mgr_shutdown () at ../../../source/lib/res/h_mgr.cpp:860
#9  0x082d24c1 in Shutdown () at ../../../source/ps/GameSetup/GameSetup.cpp:681
#10 0x083b259b in AtlasMessage::fShutdown (msg=0xb6ff5d40)
    at ../../../source/tools/atlas/GameInterface/Handlers/GraphicsSetupHandlers.cpp:103
#11 0x083b2552 in AtlasMessage::fShutdown_wrapper (msg=0xb6ff5d40)
    at ../../../source/tools/atlas/GameInterface/Handlers/GraphicsSetupHandlers.cpp:87
#12 0x0839f799 in BeginAtlas (args=..., dll=...) at ../../../source/tools/atlas/GameInterface/GameLoop.cpp:211
#13 0x082d4a59 in ATLAS_Run (args=..., flags=1) at ../../../source/ps/GameSetup/Atlas.cpp:54
#14 0x082d4afb in ATLAS_RunIfOnCmdLine (args=..., force=false) at ../../../source/ps/GameSetup/Atlas.cpp:77
#15 0x081e4a29 in RunGameOrAtlas (argc=2, argv=0xbffff424) at ../../../source/main.cpp:431
#16 0x081e5107 in main (argc=2, argv=0xbffff424) at ../../../source/main.cpp:504
(gdb) c
Continuing.
Mesa: User error: GL_INVALID_OPERATION in glDeleteShader
TIMER| resource modules: 9.81219 s
TIMER TOTALS (6 clients)
-----------------------------------------------------
  xml_validation: 24.7842 Mc (3x)
  tc_linkProgram: 62.8229 Mc (7x)
  tc_compileShader: 215.246 Mc (11x)
  tc_transform: 15.5265 Mc (140x)
  tc_plain_transform: 6956.99 kc (126x)
  tc_png_decode: 7304.01 kc (2x)
-----------------------------------------------------
TIMER| shutdown misc: 552.861 us
JS engine warning: leaking GC root '' at 0xb7220110
JS engine warning: 1 GC root remains after destroying the JSRuntime at 0xb7485c48.
                   This root may point to freed memory. Objects reachable
                   through it have not been finalized.
[Thread 0xb7dd9b70 (LWP 27344) exited]
lost connection to FAM[Thread 0xa4127b70 (LWP 27347) exited]

Program exited normally.

If you keep clicking units to add to a production queue, you can do it forever. The GUI limits to 16, but behind the scenes, it allows an infinite amount of units to be queued. This causes some gameplay issues when playing multiplayer, where a player can queue hundreds and not have to go back to it.

Obviously! I can't kill units or buildings, but I can control them and build things, depleting the other persons resources. Not to mention revealing the whole map right away.

If they are not able to hunt the meat icon shouldn't show up when hovering over animals. Eventually they should be able to hunt chickens and other domestic animals.

There isn't currently a way for scripts to change the color of an image. However, the GUI engine does allow "special" sprites to be defined at run-time. Currently only stretched images are supported, so I've modified this to add support for solid color sprites. This is useful for among other things, player colors, which may change from one game to the next, thus eliminating the need to hardcode N number of sprites ahead of time.

Example usage:

// Get an image object from the GUI
var imageObject = getGUIObjectByName("colourThing");
imageObject.sprite = "colour: 255 0 0 200"; // RGBA

The colour is specified in integer RGBA format like other GUI objects (0-255).

This should be useful for any GUI object with a sprite property.

One caveat as Philip notes, is that the sprite will be stored based on its name so every time a new sprite is created like this, it will be cached separately if even one character is different (even if it would appear identical).

Resolution: Update mesa driver (see comment 4 - #comment:4)

• Gates that have animation to open and close, along with sounds to accompany it

• Gates that have obstruction zones over the towers only, not the gate part

• Gates that can be locked, so nothing gets in or out

• Allow Walls, Gates, and Towers to overlap, so that complex wall systems, that link together, are able to be built.