Update README.md - #1
Closed
MegaMech wants to merge 1 commit into
Closed
Conversation
PurpleHato
referenced
this pull request
in PurpleHato/Shipwright
Jun 4, 2022
louist103
pushed a commit
that referenced
this pull request
Jun 22, 2022
* Fixed soh filters * add more makefile changes * almost ready * more updates * update * update * Update Makefiles to handle both platforms * Allow for overriding the CXX and CC executables * Restore original structure while supporting custom CXX flags * Remove some platform specific libs * Dynamic target name * Make X11 paths package-agnostic * Remove changes to `gfx_opengl.cpp` * Use OpenGL2 on MacOS instead of OpenGL3 * make it actually render something * render at least the first texture, still need to figure out the second one * Let’s use OpenGL 3 again * maybe this works to get the right texture? link's eyes still look off a bit * did this work? * set the platform to macos * actual numbers are right, but logic is ugly XXX/TODO, i know * add zlib to ldflags for ZAPDUtils * A bit of cleanup * Revert unneeded changes * Remove GL_CHECK * Fix issues with z64 branch * use an std::map instead of a giant array * three point filter fix (#2) * Fix mac compilation * fix audio for 64 bit * revert audio heap size, keep bigger pools * Add more Apple specific checks to our modifications * Add building instructions for macOS * Remove unecessary step from building instructions * Add missing SDL2 & GLEW to Linux LDLIBS * Update BUILDING.md Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Update soh/.gitignore to include other arch binaries Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Use right platform name for debugging window Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Fix stormlib on macos (arm64) * Simplify some of the ifdef checks * Revert an older no longer necessary fix * Remove remaining unecessary deviations * Update building instructions after StormLib changes * Feature: Use OpenGL 4.1 (#1) * Further tweak the BUILDING * Tidy up * reword -j message * Add Jenkins CI Support (#2) * Fix type issues * add target <appbundle> and <filledappbundle> add makefile targets to create an .app `filledappbundle` creates the target with the .otr included this should perhaps be moved to Application Support though * pull gcc's rpath from otool output * move make target to the end so it's not default * Add Jenkins and make exe in par with other platforms * Actually save build artefacts * Fix artefact path * Remove x11 mentions and linking (not used) * Update building instructions for generating app * use appsupport directory * Add new app icon * Update target to match macOS types * Update more audio types * fix null deref in Audio_PlayFanfare * Remove old import from z64 * address final nit with apple ifdefs Co-authored-by: KiritoDev <36680385+KiritoDv@users.noreply.github.com> Co-authored-by: Jeffrey Crowell <github@crowell.biz> Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com>
victormlourenco
pushed a commit
to victormlourenco/Shipwright
that referenced
this pull request
Jun 27, 2022
leggettc18
referenced
this pull request
in leggettc18/Shipwright
Jul 7, 2022
Testing out item replacement
Kenix3
added a commit
that referenced
this pull request
Jul 26, 2022
* Moved gfx effects to the gpu and removed loadtexbyname on some textures * Readded empty if * Added nintendo switch support * Fixed compilation errors and window closing issue on linux * Fixed kaleido on 64 bit * Fixed audio on nintendo switch * Implemented performance mode switcher * Added suggestions from pr review * Fixed ftruncate errors * Fixed compilation errors * Added missing include rule * Fixed libstorm path * Removed ftruncate extern to use a correct posix version * Merge branch 'develop' from HarbourMasters into znx (#3) * Add Support for macOS (#441) * Fixed soh filters * add more makefile changes * almost ready * more updates * update * update * Update Makefiles to handle both platforms * Allow for overriding the CXX and CC executables * Restore original structure while supporting custom CXX flags * Remove some platform specific libs * Dynamic target name * Make X11 paths package-agnostic * Remove changes to `gfx_opengl.cpp` * Use OpenGL2 on MacOS instead of OpenGL3 * make it actually render something * render at least the first texture, still need to figure out the second one * Let’s use OpenGL 3 again * maybe this works to get the right texture? link's eyes still look off a bit * did this work? * set the platform to macos * actual numbers are right, but logic is ugly XXX/TODO, i know * add zlib to ldflags for ZAPDUtils * A bit of cleanup * Revert unneeded changes * Remove GL_CHECK * Fix issues with z64 branch * use an std::map instead of a giant array * three point filter fix (#2) * Fix mac compilation * fix audio for 64 bit * revert audio heap size, keep bigger pools * Add more Apple specific checks to our modifications * Add building instructions for macOS * Remove unecessary step from building instructions * Add missing SDL2 & GLEW to Linux LDLIBS * Update BUILDING.md Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Update soh/.gitignore to include other arch binaries Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Use right platform name for debugging window Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Fix stormlib on macos (arm64) * Simplify some of the ifdef checks * Revert an older no longer necessary fix * Remove remaining unecessary deviations * Update building instructions after StormLib changes * Feature: Use OpenGL 4.1 (#1) * Further tweak the BUILDING * Tidy up * reword -j message * Add Jenkins CI Support (#2) * Fix type issues * add target <appbundle> and <filledappbundle> add makefile targets to create an .app `filledappbundle` creates the target with the .otr included this should perhaps be moved to Application Support though * pull gcc's rpath from otool output * move make target to the end so it's not default * Add Jenkins and make exe in par with other platforms * Actually save build artefacts * Fix artefact path * Remove x11 mentions and linking (not used) * Update building instructions for generating app * use appsupport directory * Add new app icon * Update target to match macOS types * Update more audio types * fix null deref in Audio_PlayFanfare * Remove old import from z64 * address final nit with apple ifdefs Co-authored-by: KiritoDev <36680385+KiritoDv@users.noreply.github.com> Co-authored-by: Jeffrey Crowell <github@crowell.biz> Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> * Move macOS to parallel stage * fix linux crash * Ignore all named saves Necessary after new save format * DPad items * Move UI location arrays from static to global to accomodate customizable UI * FIX: Increased the maximum actor limit for draw distance Fix issue #529 * FIX: Remove duplicates of gNewDrops from the enhancements menu * Get rid of erroneous int casts * casts jya_cobra's shadowTexture pointer to uintptr_t instead of s32. * Fix ability to assign equipment you don't own * Fix title cards for market potion shop and bombchu shop * Invalidate message item icon texture before drawing * Add margins for dpad * Fixed some instances where the original OOT save is erroneously preserved or recreated, leading to another conversion next time you launched the game * Headphones setting in game should now work. Resolves #13 * Map Select name fixes * Change Power to Upgrades * Rename Soh* files * Update names * Add more files * Add more files * Add toggles for disabling heart drops and random drops (#507) * Adds toggles for disabling random drops and fixed heart drops * Changed the "No Fixed Heart Drops" toggle to the more useful "No Random Heart Drops" toggle * Adds an explanation that the no heart drops option is similar to Hero Mode * Moved difficulty options into their own sub-menu Moved time saver options into their own sub-menu Moved clutter reducing options into their own sub-menu * fixed the tag name of mweep speed * Moved Skip Text to Experimental * Fix testing masks of DPad * Fix audio, boot commands, and save staes. * add menu from https://github.com/PurpleHato/Shipwright/tree/ResetMenu * implement reset * Update README.md * command on mac * special char doesn't work with default imgui font * change name based on PR comment * missed a couple spots * change sunset in save editor so dampe is out * Update to latest `docking` imgui (#582) * Update to latest `docking` imgui commit: 67410d53f739b6a0df138e2252f0e5136b42062f * Fix modified import for SDL * Update imgui_impl_sdl.cpp * Add Imgui include path for better imports * Controller Navigation Fix-Up (#544) * Controller Navigation Fix-Up * [MISC] Fix typo * Use Macro for __FILE__ & __LINE__ when possible (#559) * First batch some overlay * Almost all overlay * effect & gamestate * kaleido stuffs * more overlay * more left over from code folder * remaining hardcoded line and file * Open & Close _DISP __FILE__ & __LINE__ clean up * Some if (1) {} remove * LOG_xxxx __FILE__ , __LINE__ cleaned * ASSERT macro __FILE__ __LINE__ * mtx without line/file in functions * " if (1) {} " & "if (0) {}" and tab/white place * LogUtils as macro * GameState_, GameAlloc_, SystemArena_ & ZeldaArena_ * Revert "GameState_, GameAlloc_, SystemArena_ & ZeldaArena_" This reverts commit 0d85caa. * Like last commit but as macro * Fix matrix not using macros * use function not macro * DebugArena_* functions GameAlloc_MallocDebug BgCheck_PosErrorCheck as macros removed issues with ; in macro file * MM Bunny Hood enhancements: no age restrictions & wear across entrances (#560) * Remove age restriction on MM Bunny Hood * Handle mask slot changing; init Link wearing mask * Reset mask state after selling * Use unordered maps in some hot paths (#566) * Use unordered maps in some hot paths * Address PR comments * Cheat: Infinite Epona Boost (#577) * Cheat: Infinite Epona Boost * Edit for readability * update pulseaudio for 44.1khz audio (#587) * use pulseaudio defaults * spaces/tabs * pulseaudio values tuned for n64 audio at 44.1khz * update tlength * remove one hardcoded samplerate Co-authored-by: briaguya <briaguya@alice> * Restoration Item Sliders (#595) * Allows players to set custom restoration values for potions and milk * Fixed a dumb oversight * implemented fairy sliders * use ini window resolution (#410) * use ini window resolution * use arguments for directx window size * default 640x480 not 320x240 * kick off a build * and revert * default config 640x480 not 320x240 * add todo Co-authored-by: briaguya <briaguya@alice> * 3D Drops & Projectiles update (#548) * Fix remaining rupee and keys * fix tabs -> spaces * more fix about tabs -> spaces * Better up <-> down mouvement for hearts * remove useless comment * Cosmetics menu rework (#589) * - * DPad/some margin fix(left side) * various ImGui stuff and fixes * disabled kaleido menu (non working) * fix win being dumb * same for C btn this time * Fix windows build * Beating hearts fix * Default win size and some placement fix * Fix Dpad Ammo & C notes, Kaleido, white dog * some texts fix and tab move * Add stone of agony, some fixes and build correctly * precise item place with Dpad * Gamecube -> GameCube Co-authored-by: Baoulettes <iMacWin10> * [CI] Create an Appimage for 64bit builds (#570) * [CI] Create AppImage file * [CI] Updated Jenkinsfile * [MISC] fix indentation * [MISC] Fix indentation * [MISC] Fix indentation * [CI] 7z appimage and readme * [CI] AppImage icon file (#597) * [CI] Add patchelf * [BUILD] Include Hi-Res icon * Fix portability use of std::clamp (#596) * Add N64 weird frames and OOB Bombchus cvars (#602) * Add gN64WeirdFrames and gBombchuOOB cvars * Rename gBombchuOOB to gBombchusOOB * Fix pipeline * [BUILD] Do not strip ZAPD (#598) * [BUILD] Do not strip ZAPD * [BUILD] Reuse Icon Asset * [APPIMAGE] Handle paths with spaces * [macos] Distribute optimized builds (#599) * Cosmetics texts fixes (#604) * branch init + few fixes * Reset button * Several texts fixes and default color to GameCube * space issues * some more extra space that was not needed. * space yes last time ... * Arrows being able to be rainbow some more texts fixes * more texts fixes * c-buttons some plurial fix * Fixes bug in demo effect due to removed display list set. * King dodongo room crash (#613) * first attempt * force camera setting to be at least 0 (dirty!) * - * reverted something to just include fix * Invalidate text box icon before drawing (#607) * [APPIMAGE] Set lib path for ZAPD (#611) * New drop runtime fix (#614) * inital rework * fixed default rotation for 2D sprites * fix tab/space issues * Fix Ganon's Castle title cards (#606) * Checkered room crash workaround (#615) * Increase poly and vertex count for dyna actors * Add TODO * Only apply fix in Forest Temple * Enable globally * Match case of 'Shipwright' in Linux build (#619) * Fix 0xabababab crash (#617) * Unconditionally setup the normal skybox * Only call Skybox_Setup once ever * Equipment upgrade text fix and any pause slot enhancement fix (#620) * Fix upgrade name text not rendering * Use cvar and don't render equip help for empty item slots * Fix rendering logic * Fix incorrect item name rendered for one frame * Reorder comparison * Remove extra indent * Remove accidental changes * ImGui grammer, consistency, and clarity fixes (#625) * ImGui grammer, consistency, and clarity fixes Co-Authored-By: Stormghetti <56653191+Stormghetti@users.noreply.github.com> * Adds more new lines for users on smaller displays Co-Authored-By: Stormghetti <56653191+Stormghetti@users.noreply.github.com> Co-authored-by: Stormghetti <56653191+Stormghetti@users.noreply.github.com> * New drops shadows & rotation fixes (#627) * fixes * forgot to add heart pieces in rotation logic * Dodongo's Cavern blue warp crash fix (#622) * Doodong's Cavern blue warp crash fix * >= not > * Don't waste a line of space * add support for clang compiler (#592) * hacks to align strings for clang... wow just wow * start work to getting built with clang * fix issues with struct constructors, all builds, doesn't link still * fix some narrowing issues that clang complains about * fix compliation of zapd * fix null deref in VersionInfo * builds with clang * make stringbuilding use StringHelper instead of addition * fix linking * add CLANG SHIP overlay on clang built versions * doesn't need to be volatile * mark unknown strings as extern * rename some stuff * can't align extern * hopefully fix compilation for everythign * expandtab * allow setting LD * Revert "allow setting LD" This reverts commit 711aba6. maybe to use lld it should be a LDFLAG? * -Wno-deprecated-declarations is required for newer versions of clang on macOS 13 beta sdk, the version of apple clang requires this * Add jenkins support for clang * Forward CXX flags to stormlib compilation * Move GCC only flags to check * use exports to set multiarch setup * Fix Jenkins forever * use make instead of cmake --build add some flags to build with clang-11 as well * address review coments - rework extraction to allow multi thread - misc readability cleanup * update makefile to add WARN on linux+clang Co-authored-by: David Chavez <david@dcvz.io> * Fix develop * Fixes grey screen issue + tooltip for 2 handed shield * Don't close controller after SDL has quit (#642) * Don't close controller after SDL has quit * Don't check if controller can rumble if null * Fixes Barinade's set whenever boss fight is reloaded (#639) * Reimplements unused Barinade reset function. * Removes the Jellyfish from Barinade's body on reset. * add the ability to pull graves during the day (#637) * change max internal res multiplier to 3x, use float and % instead of int (#638) * change max internal res multiplier to 3x, use float and % instead of int * Update libultraship/libultraship/ImGuiImpl.cpp Co-authored-by: Kenix3 <kenixwhisperwind@gmail.com> Co-authored-by: briaguya <briaguya@alice> Co-authored-by: Kenix3 <kenixwhisperwind@gmail.com> * Fixes crash on Biggoron trade failure screen wipe Resolves #621 * fixes center docking (#652) * testing out item replacement (#416) * skip learning song of storms * don't set flag when getting goron tunic as child * Initiates prelude check when master sword unloads. Not quite how N64 rando does it but so far it's the only way I've found to make it trigger without also triggering the time travel again. * Stops Shadow Temple lore prompts from appearing in rando. * Skips cutscene of royal tomb explosion in rando. Explosion sound doesn't play correctly and I think the debris appears in the wrong place, but the functionality is here. * Improves visual of exploding gravestone. * Adds some comments explaining the rando differences * Skip ruto text box in jabu blue warp For rando * skip intro cutscene in dodongo's cavern * load spoiler files on boot, fix spoilerfile existing check when making new saves * name entry dropped spoiler logic * make sure to actually init the cvar * no chime on load * uncomment * Skip ganondrof cutscene Skip to scream part of the death animation, skipping the text boxes etc. For rando * Update z_boss_ganondrof.c * skip owl flight cutscenes in rando * Fixes skipped text so it only applies to shadow temple. Earlier fix inadvertently applied to some other text as well, changed logic so that only specified sceneNums and textIds can have this enabled, and text skipped by sceneNum can have the skip overriden by textId if needed. Currently there are no overrides so the textId section of the logic is commented out to avoid compilation errors. * Adds a default to the switch case statements that leaves the randoSkipText variable unchanged, just in case. * TEST: Text for item * Adding ganon flavor text * ADD: AMMO Count * format ganon text/hint text * Autoskip the tower cutscene if settings call for tower collapse. * ganon hint text logic * Improved prelude after time travel fix * swapped the sizes between ganon hint text and ganon text, as they were set to the wrong things. * this is all i did * not the cleanest code ever but it's working * ADD: GS Count * ADD: Wallter (crash for now) * TWEAK: Wallet check * FIX: Use DrawItem instread of DrawUpgrade... b-baka! * Fixes some vanilla bugs introduced by rando code. * Added cutscene skip for zelda escaping Using the debug cutscene skipping function. Also added a conditional so the bridge doesn't spawn closed when cutscene is ready to trigger * ADD: X Spacing + Placeholders for song * ADD: default case for items * TWEAK: Spacing * FIX: Light Arrow * ADD: Ammo Option * use groups instead * ADD: More spacing logic * songs and names * TWEAK: Color on wallet * colors * Added flags cutscene before nabooru fight * ADD: ChromaKey text * First attempt skip cs after nabooru defeat * Better implementation for specific rando cutscene skips * use pulseaudio defaults * spaces/tabs * move color push/pop to stop crash * make the colors work again * the real bottle fix * pulseaudio values tuned for n64 audio at 44.1khz * update tlength * remove one hardcoded samplerate * Cleaned up and fixed zelda escape skip The if statement is a freaking monster, but unless we want to skip more cutscenes in the same way later, this is the most compact way of doing it that I know of. * Revert one line to match original nothing functional * another hint line that breaks autonewline logic * don't autospawn epona if we don't have the song/ocarina * Trying to use iron knuckle death effects not working yet * Streamlined OoT cutscene skip for future additions Also cleaned up if statement in general * Made if statement more readable Also added clarity for what cutscene was skipped * Fixed typo in comment * Janky nabooru defeat cs skip * altar text formatting (gonna need help shortening some of the french ones) * more altar text formatting * english altar text formatting complete * make gtg blocking guard check for card not bridge * FIX: Typo! * FIX: Uppercases * FIX: Typo * TWEAK: Alter + some names * TWEAK: More caps! * ADD: Missing string TWEAK more uppercases and namefixe s * Hide nabooru death by covering her in flames * bandaid fix for death crash issue * Twinrova defeat cs skip Skips the animation and manually calls the function to show the "beam" around the sisters * fix crash * fix caps to match * fix great fairy reward mashing/shielding issue * TWEAK : Typo clé to Clé * TWEAK: Some Altar hints TWEAK: Some capitals * TWEAK: Unmatching text + some cap again * TWEAK: More tweaks * fix build * remove extra json.hpp, add hint * Update randomizer_item_tracker.cpp * TWEAK: Double Defense with RedParticles instead of white * make sure we don't optimize out the check to ensure a spoilerfile exists * vanilla ganon boss key hint formatting * TWEAK: FR- better way of the hero text * fix * and again * Initializes dungeonsDone items in gSaveContext to 0. * Replaces sizeof calculation with a NUM_DUNGEONS constant. * Fixes Saria's Gift on the LW Bridge from getting killed when holding shield. * More airtight fix for Saria's Gift on the Bridge. * Lifts one of the conditions in the if statement a little higher to prevent unnecessary lookups of getItemId. * Invalidate text box icon before drawing * Fixes the case where Saria's gift is an Ice Trap. We still get the Ice Trap once, but never again. This does mean you can now hold R while walking in to avoid the ice trap, but everything else seems to work fine. * Initial commit Might need changing when we change the settings in the future * Fixes Door of Time opening cutscene after warping with prelude. * Initial waterfall skip Very rudimentary way of doing things but it seems to work so 🤷 * inital rework * fixed default rotation for 2D sprites * fix tab/space issues * 3d drops rando merge fix again * Allows Impa to appear in the Lullaby check post drawbridge escape. * Changes Ganon's Trials Count setting to a checkbox The checkbox is whether or not to skip all of them. Leaving the box unchecked will mean doing all of them. Eventually this will be switched back to a slider once we implement the logic for which trials start out completed. * Sets all Ganon's Trials to incomplete in new saves. Fixes https://github.com/briaguya-ai/rando-issue-tracker/issues/131 * fix castle guards when oot throw cutscene has already played in rando * Properly removes the beams when trials are cleared. * Removes Question Mark from Skip Ganon's Trials UI. * Adds a todo comment about when to change back to slider. * make deku seeds check for bullet bag * Various tweaks TWEAK: Altar Text TWEAK: Hint names TWEAK: Replace more problematic œ to oe * upgrade ocarina on both child and adult equips * FIX: Jabu Item * update equipped hookshot/longshot when obtained as other age * add hint * don't give the bgs check without the claim check * Skips Darunia Cutscene in Fire Temple * Added a TODO note about not skipping the cutscene. There is a setting we will want to have eventually that will require this cutscene to not be skipped since it is used during a glitch. * remove todo * restore fast ocarina option in imgui that was lost in merge * Fixes grey screen issue + tooltip for 2 handed shield * update to use dg instead of g for textures in item tracker * TWEAK: Default color for cosmetic RAND button was not the corect one * fix texture crash, remove unused item tracker code * don't open mask shop until we get zelda's letter * Update README.md * Prevents "correct" chime under incorrect conditions. * Fixes typo in conditional and adds "bonk" sound effect. "Bonk" sound is NA_SE_SY_OCARINA_ERROR and it plays when conditions for the Door of Time have not been met after playing Song of Time. This is only possible in rando's "Intended" Door of Time option, in which the Ocarina of Time and all 3 spritual stones are required to open the door, instead of the vanilla requirements of just having the song of time. * remove modify dpad equips toggle, replace with checks for dpad menu * remove extra check * add ability to hold c-up to assign to dpad when dpad menuing is enabled * disable d-pad navigation on item menu when holding c-up to equip * dpad+c-up stuff for equipment menu * ADD: Checbox for songs colors * TWEAK: RandoColors for normal songs * kind of quick and dirty but it works * TWEAK: Clarity of the tooltip Co-authored-by: briaguya <briaguya@alice> Co-authored-by: Christopher Leggett <chris@leggett.dev> Co-authored-by: aMannus <mannusmenting@gmail.com> Co-authored-by: PurpleHato <linkvssangoku.jr@gmail.com> Co-authored-by: Dog <5172592+Dog@users.noreply.github.com> Co-authored-by: Vague Rant <vaguerant@users.noreply.github.com> Co-authored-by: Baoulettes <perlouzerie@hotmail.fr> Co-authored-by: Ada <60364512+GreatArgorath@users.noreply.github.com> * Cosmetics hotfixes (#640) * Initial branch creation * Revert Main Game so it do not conflict later * should fix window build, made namespace for Cosmetics * forgot to edit one title * Ability to add hidden window (usefull for Rainbow) Fix building issues * Line break, unused bool remove * add descriptive todo for death crash bandaid (#655) Co-authored-by: briaguya <briaguya@alice> * Removed legacy audio mode and fixed ganon sound bug (#657) * Free Camera (#337) * wip free cam * Almost done, needs collision still * Added free cam behind cvar * added WIP collision * Fixed & implemented "Manual mode" from WW & TP * Fixed camera not rotating when Link is moving * fixed initialized camera rotation * Fixed camera getting stuck + made it smoother * reduced deadzone * fixed epona camera height + added WW z-target free camera * Adjusted player camera height & fixed fov * Fixed camera roll * fixed fov when moving the camera while in z-target * Camera resets to Auto when going through doors or changing maps * Fixed building * touch * more touch work * Added WIP mouse support to the free cam * gui stuff * fixed building * fixed building error * ok fixed building for real this time * oops * Fix compilation issues * removed mouse stuff that magically appeared in this branch * smoothed out stick values & removed remains of mouse support * re-added manual camera when pressing Z * reduced minimum Y position of camera * Addressed dcsv's nitpicks * part 2 * oops Co-authored-by: David Chavez <david@dcvz.io> * Rando: Allows Malon's Item Check to be obtained by pulling out the Ocarina. [FIXED PR] (#672) * Fixes using the Ocarina to get the check from Malon. Still some cleanup to do here. For some reason the player can shield before receiving the check. It doesn't set the flag if the player does that so they can still try again, but would prefer a different solution if possible. * Prevents Shielding from blocking the Item_Give from happening. * Code Cleanup and comments explaining the new rando flow. * Removes inventory check when pulling out Ocarina This allows OI to properly give the check, which is important for Glitched logic later down the line. Talking to Malon still requires the Ocarina in your inventory. * Prevents non-malon textboxes from triggering the check. Also adds a comment explaining the condtional for getting the check from talking to Malon since it got pretty long. * Actually fixes checking for text boxes. * Relocates a comment for improved clarity. * Fix Rando Water Temple Softlock (#665) We use 3DS logic to generate item placement, but didn't have this specific door in Water Temple unlocked from the beginning like 3DS does. This meant that if people took specific paths through the temple, they could softlock themselves by missing a key. * Rando: GtG and carpenter prompts skip (#663) * Skip gtg and carpenter prompt For rando. Tested and just works. * Fixed missing break * Hide debug overlay behind gDebugEnabled (#660) * Introduce App Directory Path (#572) * Introduce app directory path concept * macos: Remove hacky way of using applicaiton directory * Update the new SaveManager * Address stack user after return * Remove unecessary property * Use std::string for filepath * Improve clang specific detections * Use new path system for imgui files * Improve helper for getting relative paths * fix hidden wnd (#744) * Split damage multiplication into its own PR (#656) * Split damage multiplication into its own PR * Found a more elegant implementation of the powers char*[] * Fixes Maps, Compasses, and Boss Keys in Vanilla. (#751) * .xiF slebaL * Update Keese labels * Fixed soundfont issues * Skip warp song cutscenes in rando (#664) Does it by skipping to the last part of the cutscene data. Tested on all songs, both adult and child. * don't spawn blocking mido after we've already shown him the sword/shield (#675) Co-authored-by: briaguya <briaguya@alice> * Controller Configuration UI and JSON Config (#760) * Initial controller hud ui * Reverted fbdemo changes * Moved config to json and implemented controller config * fix build on linux, gitignore new config file * fix build * Fix compilation and file directory paths * Call save on cvar save * Fixed cvar loading and added deck slots to the config * Changed control deck port 0 to use a physical device by default * Added gyro and rumble & fixed loading errors * Save config on toggle menubar * fix linux build * Fixed drift calculation * Controller config now saves when pressing F1 * Removed ExitGame hook from ImGuiImpl * Moved mappings to a map * Added GetKeyName * untranslate scancodes * Fixed hud layout on keyboard device * Fixed keyboard read on hud * Fixed crash when reloading controllers * Removed ConfigFile and changed file extension * Changed Dummy to Disconnected and fixed filters * Removed function leftover * Changed ControllerHud to InputEditor Co-authored-by: briaguya <briaguya@alice> Co-authored-by: David Chavez <david@dcvz.io> * Enough! My ship sails in the morning. * Fixed menubar items position (#763) * Fixed menubar items position * Reverted tooltip tab position * Fixes macOS randomizer functionality with App Directory (#761) * Fixes macOS randomizer functionality with App Directory * Fix windows build * Update soh/soh/Enhancements/randomizer/3drando/rando_main.cpp * Update soh/soh/Enhancements/randomizer/3drando/spoiler_log.cpp * Revert band-aid fix Co-authored-by: Kenix3 <kenixwhisperwind@gmail.com> * Fix migration cvar deletion path (#765) * Various controller fixes (#771) * Fix controller * Also fix rumble strength being a bool * Remove ControllerHud.cpp * Downgrade platform toolset back to previous version * Fix gyro * Fix bug that makes binding axes difficult and clear buttons before reading * Exaggerate gyro display and adjust stick binding threshold * Initialize drift thresholds * git subrepo push soh subrepo: subdir: "soh" merged: "75ccbade8" upstream: origin: "https://github.com/HarbourMasters/soh.git" branch: "master" commit: "75ccbade8" git-subrepo: version: "0.4.1" origin: "???" commit: "???" * One more change from PR review * Fix some paths * Fix merge conflict messup * More merge conflict fixes * And another conflict fix * And another fix * Remove reference to removed build files * Add full path to switch cmake Co-authored-by: David Chavez <david@dcvz.io> Co-authored-by: KiritoDev <36680385+KiritoDv@users.noreply.github.com> Co-authored-by: Jeffrey Crowell <github@crowell.biz> Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> Co-authored-by: sholdee <102821812+sholdee@users.noreply.github.com> Co-authored-by: briaguya <briaguya@alice> Co-authored-by: rozlette <Rozelette@users.noreply.github.com> Co-authored-by: PurpleHato <linkvssangoku.jr@gmail.com> Co-authored-by: Christopher Leggett <chris@leggett.dev> Co-authored-by: Kenix3 <kenixwhisperwind@gmail.com> Co-authored-by: vaguerant <vaguerant@users.noreply.github.com> Co-authored-by: earthcrafterman <banddstudios@gmail.com> Co-authored-by: louist103 <35883445+louist103@users.noreply.github.com> Co-authored-by: briaguya <70942617+briaguya-ai@users.noreply.github.com> Co-authored-by: qurious-pixel <62252937+qurious-pixel@users.noreply.github.com> Co-authored-by: Baoulettes <perlouzerie@hotmail.fr> Co-authored-by: Sirius902 <10891979+Sirius902@users.noreply.github.com> Co-authored-by: modestposer <modestposer@gmail.com> Co-authored-by: Ada <60364512+GreatArgorath@users.noreply.github.com> Co-authored-by: Stormghetti <56653191+Stormghetti@users.noreply.github.com> Co-authored-by: Sirius902 <3645979-Sirius902@users.noreply.gitlab.com> Co-authored-by: MelonSpeedruns <melonspeedruns@outlook.com> Co-authored-by: aMannus <mannusmenting@gmail.com> Co-authored-by: Dog <5172592+Dog@users.noreply.github.com> Co-authored-by: Nicholas Estelami <NEstelami@users.noreply.github.com> Co-authored-by: ChristopherJTrent <ChristopherJTrent@outlook.com> Co-authored-by: agamache <aaroncgamache@gmail.com> Co-authored-by: M4xw <m4x@m4xw.net> * Fixed switch compilation * Fixed responsive on switch imgui * Remove config save hook * Added physical slots into controller name * Add `ifdef` for switch specific code * Add `ifdef` for switch specific code * Added applet mode detection * Replaced homebrew icon * Fixed randomness on applet mode view * Fixed clkrst initialization * Switch profiles cleanup * Cleaned code based on dcvz comments * Remove unused hooks in Mercury * Fixed compilation * Use IMGUI_IMPL_OPENGL_LOADER_CUSTOM * Fix up StormLib modifications * Handle touch events at ImGui/SDL level * Enable opening menu with - * Load Nintendo Switch font * Updates to window initiation * Handle virtual keyboard via SDL * Print OTR missing message to screen * Rename makefile for Switch * Some more additional fixes * Branch creation plus edit Changed ItemWidth & made label invisible (else issues) Added a submit button next to the text field * Removed set_texture_filter on imgui draw * Readded nintendo switch font * Fixed randomizer crash * Fix issue with strdup windows * [SWITCH] Update Jenkins file (#9) * [SWITCH] Update Jenkins file * [SWITCH] Dockerfile.switch Can be combined with other Dockerfile * [SWITCH] Add entrypoint file * Update Jenkinsfile * [SWITCH] Dockerfile mtab link link `/proc/self/mounts /etc/mtab` Co-authored-by: David Chavez <david@dcvz.io> * Cleaned code based on PR comments * Fixed switch compilation * Cleaned full texture cache instead of doing it per char * Randomizer only saves when its not a vanilla save * Disabled input when menubar is opened * Removed return from SetupFont * Cleaned code based on comments * Cleaned up random texts on switch error screen * Killed * Removed debug and changed compilation flags Co-authored-by: KiritoDv <nohomoboi01@gmail.com> Co-authored-by: Felipe Guaycuru <guaycuru@gmail.com> Co-authored-by: David Chavez <david@dcvz.io> Co-authored-by: Jeffrey Crowell <github@crowell.biz> Co-authored-by: BountyChocolate123456 <101743444+BountyChocolate123456@users.noreply.github.com> Co-authored-by: sholdee <102821812+sholdee@users.noreply.github.com> Co-authored-by: briaguya <briaguya@alice> Co-authored-by: rozlette <Rozelette@users.noreply.github.com> Co-authored-by: PurpleHato <linkvssangoku.jr@gmail.com> Co-authored-by: Christopher Leggett <chris@leggett.dev> Co-authored-by: Kenix3 <kenixwhisperwind@gmail.com> Co-authored-by: vaguerant <vaguerant@users.noreply.github.com> Co-authored-by: earthcrafterman <banddstudios@gmail.com> Co-authored-by: louist103 <35883445+louist103@users.noreply.github.com> Co-authored-by: briaguya <70942617+briaguya-ai@users.noreply.github.com> Co-authored-by: qurious-pixel <62252937+qurious-pixel@users.noreply.github.com> Co-authored-by: Baoulettes <perlouzerie@hotmail.fr> Co-authored-by: Sirius902 <10891979+Sirius902@users.noreply.github.com> Co-authored-by: modestposer <modestposer@gmail.com> Co-authored-by: Ada <60364512+GreatArgorath@users.noreply.github.com> Co-authored-by: Stormghetti <56653191+Stormghetti@users.noreply.github.com> Co-authored-by: Sirius902 <3645979-Sirius902@users.noreply.gitlab.com> Co-authored-by: MelonSpeedruns <melonspeedruns@outlook.com> Co-authored-by: aMannus <mannusmenting@gmail.com> Co-authored-by: Dog <5172592+Dog@users.noreply.github.com> Co-authored-by: Nicholas Estelami <NEstelami@users.noreply.github.com> Co-authored-by: ChristopherJTrent <ChristopherJTrent@outlook.com> Co-authored-by: agamache <aaroncgamache@gmail.com> Co-authored-by: M4xw <m4x@m4xw.net>
leggettc18
added a commit
that referenced
this pull request
May 8, 2023
mstergianis
pushed a commit
to mstergianis/Shipwright
that referenced
this pull request
Mar 26, 2024
…lter Changes gCheckTrackerHideFilteredAreas to default to true
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 17, 2026
…too fast")
The auto own-CSAB path free-ran each OoT3D CSAB at a fixed rate regardless of how
fast the N64 game logic was advancing the actor's animation, so walkers/runners
(Kakariko carpenters etc.) looked sped-up. Now SoH3D_UpdateAnimAuto drives the
CSAB at the SAME fractional progress as the live N64 anim:
csab_frame = (n64_curFrame / n64_animLength) * csab_duration
threaded from skelAnime->{curFrame,animLength} captured in SoH3D_SkelAnimeDraw.
Stub idles (animLength <= 4, the 2-frame N64 fidget stubs) and unknown durations
fall back to free-run (no meaningful progress to lock to; see memory
n64-idle-stub-no-phaselock). The SkelAnime-less raw choke point leaves animLength
0 -> free-run.
Verified live: En_Daiku_Kakariko model 2002 maps object_daiku_Anim_000C44 ->
dk2_hashiru ("run"); n64frame advances 0..18 per the 18-frame N64 run cycle, the
OoT3D run CSAB now locked to it. animdbg print extended with the N64 playhead +
[PHASE-LOCK]/[free-run] tag.
Known interaction: procedural-motion actors (cucco gCuccoAnim) keep curFrame at 0
(flap is written to jointTable directly), so they pin at CSAB frame 0 under phase-
lock. That's orthogonal — those actors need the retarget path (cucco issue HarbourMasters#1).
Bumps libultraship to the sky-bug segment diagnostic.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 19, 2026
…ecull` soh3d_model.cpp makeCgroup sets faceCull = (mat->cull == 1) so the renderer culls back faces on single-sided OoT3D materials (matches N64 G_CULL_BACK); the synthetic sun/moon billboard quad stays double-sided. soh3d.c adds REPL `facecull <0|1> [flip]` to toggle culling + the front-face winding convention live. Bumps libultraship to d272d6a9 (the GL+Vk backface-cull implementation). Fixes the BACKLOG HarbourMasters#1 "camera under terrain shows the terrain underside (flip look)" artifact — see the libultraship commit for the quantitative Vulkan verification. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019zb4KYjr5uG4hGKtGMLwGo
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 19, 2026
…atertight steps The wall-integration rework (9c28b90) — texture each staircase like the wall it abuts, via findPatchWall + per-patch inset/sawtooth groups — produced three playtest defects (issue HarbourMasters#1): cyan see-through gaps + stretched/smeared brick between steps; a whole staircase as one flat ochre slab; stair faces overlapping/covering the original wall. Per the user directive ("the wall should be preserved original; minimal vertical caps, no inset, no far base, never over the wall"), revert to the SVG-stone baseline: each kaidan ramp becomes real watertight treads+risers on its own footprint, textured with the embedded tiled stone (REPEAT), with minimal diagonal side caps at the ramp's own cmin/cmax edges — the original scene walls are left untouched. Verified live (headless, Kakariko scene 0x52): the Death-Mountain-trail staircase and the main entrance staircase render as clean coursed-stone steps, watertight (no cyan between steps), no flat slab, brick wall preserved with a clean seam. The cyan UNDER the main staircase in a side view is inherent to the OoT3D scene (the ramp is a thin cantilevered ledge above dropped-away terrain) — confirmed identical with stairs off, so not a regression. Evidence on HarbourMasters#1. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01FkowFhfE3A1KxK1VdQVndP
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 19, 2026
…oT3D terrain The N64 title demo (gHyruleFieldIntroCs) and scripted cutscenes author the camera eye at low/sub-ground heights (spot00 intro holds eye.y=-1 while the ground is ~+10). On N64 the double-sided terrain hid a buried eye; SoH3D culls terrain backfaces (HarbourMasters#1), so the same eye now sees THROUGH the ground (void/skybox seam across the lower frame). Fix: for a cinematic camera (active cutscene OR non-MAIN subcamera), if the eye is below the visible OoT3D room mesh at its XZ, translate the whole camera (eye + lookAt by the same delta, preserving look direction) up until the eye clears the mesh by an 18u near-plane margin. Only ever lifts; the correction -> 0 continuously as the eye rises above ground, so no pop at shot boundaries. Gameplay cams (csState idle + MAIN_CAM) are left alone. Applied in SoH3D_ReplPoll after Play_Update, before Play_Draw. Gate: SoH3D_Enabled() + env SOH3D_CAMLIFT (default ON) + REPL `camlift`. Verified live on the title demo: first shot eye.y=-1 -> lift 29.3 -> proper field vista; shots already above ground get lift 0; gameplay unaffected. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QJTwduyac1qsfD2dR7L2A4
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 20, 2026
…mp (HarbourMasters#1) The generated stairs were anchored at the step's lower y, tucking the whole flight at/under the original kaidan ramp diagonal so the ramp touched only the step nosings (the steps' max height). The walking surface therefore sat a half-step low. Raise treads/risers by dy/2 (render + collision, kept in lockstep) so the treads average the original ramp plane, as intended. Extend the side caps to a quad spanning from the ORIGINAL ramp diagonal (the terrain-meet line) up to the raised step silhouette, so the surrounding terrain still occludes the sides and the half-step skirt the raise opens never shows through. Verified A/B in Kakariko: watertight, no new see-through. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MnbTTwyKrEkgwBEGpSSbJQ
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 20, 2026
…ng (HarbourMasters#1) Render + collision treads raised a FULL step (yr = f.dy) above the original ramp diagonal (was half-step) so the top tread reaches the upper ground and the flight reads as real steps, not sitting below the walked surface. Side caps are now ONE outward-facing triangle per step on each c-edge (hypotenuse on the diagonal) — a stepped outer silhouette with no smooth-slope strip, closing the open side so nothing shows through to the channel wall behind. Step color uses the original kaidan ramp's averaged baked vertex color (baseCol) instead of flat white, multiplied by a per-face shade (tread 1.0 / riser 0.72 / side 0.55) so the steps sit in the same scene tone as the ramp they replace and read as 3D. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MnbTTwyKrEkgwBEGpSSbJQ
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 21, 2026
…ters#21: port HUD texture identity to PC HarbourMasters#1 (stairs): generateStairsGroup emitted only treads/risers + single diagonal side triangles, leaving the stepped volume open (side ends, underside, bottom-front, top back-wall) so the OoT3D sky dome bled through as a cyan halo, plus a stray protruding triangle anchored to the old ramp diagonal. Re-emit each patch as a closed solid: full-rectangle side walls from a flat bottom plane, bottom-front riser, top back wall, underside quad; stray diagonal removed. Verified live (Kakariko s01_kaidan): solid steps, no cyan bleed, no stray face. HarbourMasters#21 (HUD corruption): Fast3D's texture cache keys textures by raw source address with MANUAL invalidation (the N64 model). SoH3D's PC heap-allocated HUD buffers (hearts, rupee/counter icons, button disc, digits, glyphs) can land on an address a prior texture was cached under and freed without Gfx_TextureCacheDelete -> the first HUD draw hits the stale entry and renders garbage that persists the session (clears on restart). Each getter now evicts any stale cache entry for its buffer address on first use (SoH3D_HudTexClaim -> Gfx_TextureCacheDelete). Verified no HUD regression live. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GQ4xk7K8GenPY2vbxicMau
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 21, 2026
…urying the flanking wall) The prior HarbourMasters#1 closure emitted side walls as full rectangles down to the patch floor (yBot=f.ymin) plus bottom/back/underside planes, which blanketed the stone-brick wall flanking the staircase ("these faces should only be between the wall and the stairs, not cover the whole wall"). Replace with per-step wedge triangles bounded above by the tread/riser and below by the original kaidan ramp diagonal (yRamp(a)=ymin+(a-amin)*dy/da; tread top yk lands on the ramp at the step's back edge). The ramp diagonal closes the underside, so no separate bottom/back/underside planes -> the wedge occupies only the steps-vs-ramp envelope, the flanking wall stays visible, and no sky bleeds through. Verified live (Kakariko s01_kaidan): walls visible, steps solid, no cyan bleed. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GQ4xk7K8GenPY2vbxicMau
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 21, 2026
…elf-calibrated scale HarbourMasters#1 (stairs): the per-step riser is emitted at the BACK of each tread, so the first step had no downhill front face -> sky bled through a cyan wedge at the base of the flight ("first step missing its Y vertex"). Replace the (wrong, top-end) seal with a bottom-front wall on the first step only: a=amin, full width, from the first tread down to the ramp base ymin (one step tall, flush with the ground, nothing below ymin so the flanking wall stays visible). Verified live (Kakariko village staircase): steps meet the grass solid, no cyan. windmill: the rotor blades (Bg_Spot01_Fusya / c_s01fusya) rendered way too small because they used the shared SOH3D_SPOT01_WORLD_SCALE (0.01268, mis-derived). Self-calibrate the windmill's own scale = N64 blade height / OoT3D CMB height (the HarbourMasters#77 well-arch pattern, new sentinel SOH3D_MEASKEY_WINDMILL). Log-confirmed scale=0.0983 (n64h=1008, modelh=10255), ~7.7x larger, matching the tower. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GQ4xk7K8GenPY2vbxicMau
SomeoneIsWorking
added a commit
to SomeoneIsWorking/Shipwright
that referenced
this pull request
Jun 22, 2026
HarbourMasters#1, HarbourMasters#93) Skeletal NPCs (Kokiri En_Ko, Saria En_Sa) turn their head/torso toward the player. OoT3D's actor OverrideLimbDraw applies this to the head/torso limb matrix (MTXMODE_APPLY) from interactInfo; SoH3D's CSAB auto-draw path samples a raw (anim, frame) pose and never runs that callback, so they stared straight ahead. Port the override onto the OoT3D rig directly (not by retargeting N64 limb math): new module soh3d_anim_override.{h,cpp} reads the live interactInfo the faithful actor logic already computes and post-multiplies the OoT3D-native rotation onto the OoT3D head/torso BONE. The OoT3D head bone's origin sits at the neck, so a local rotation pivots there naturally — no N64<->OoT3D axis derivation, no pivot translate. New CSAB skinner channel: per-bone post-rotation matrix (SoH3D_SetBonePostRot), post-multiplied onto the bone's animated local rotation (R*Rpost) — matching the actor's MTXMODE_APPLY and propagating to child bones. Threaded through skinMatrices and skinMatricesMorph; distinct from the euler-add boneRotDelta channel (cucco flap), which an earlier attempt wrongly reused (euler addition composes in the wrong frame and pitched the head down instead of yawing). Override table (km1/kw1 head bone 10 / torso bone 9, En_Ko RotateX(-ty)/RotateZ(tx); zelda_sa torso RotateY/RotateX). Gated by REPL `track 0|1` / env SOH3D_TRACK. Verified headless (Kokiri Forest): track 1 turns the OoT3D Kokiri's head to match the faithful reference pose at the same interactInfo (head/yaw + slight pitch); track 0 = straight ahead. Live "head follows the player" is limited by the headless actor-update throttle (idle kids aren't in an active-track state), so it reads the same interactInfo the faithful logic produces — confirm on a non-throttled run. Saria's bone ids assume the km1 layout; re-confirm in the Meadow. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GcSA214RFwTHKziZZmAu6b
serprex
pushed a commit
that referenced
this pull request
Aug 1, 2026
* Add Torch migration plan Plan for replacing ZAPDTR + OTRExporter with Torch as SoH's only asset pipeline, following HarbourMasters/Torch#219 ("oot support"). Co-Authored-By: Claude <noreply@anthropic.com> * Plan: yml repo is source of truth, drop the move-to-assets-repo steps - Name the actual repo (briaguya0/soh-asset-yml); it holds yml only, and is edited directly rather than regenerated. zapd_to_torch.py is a one-shot conversion, not an ongoing workflow, since nobody edits the ZAPD XMLs once ZAPD is gone. - Phase 5 is pure deletion. Generator inputs are already preserved by the harness's shipwright@95d8f7e submodule pin and by git history, so nothing needs snapshotting or moving out of Shipwright. - Clarify soh/CMakeLists.txt:107-109 as a configure-soh-directly fallback with nothing left to fall back to. - Downgrade the extraction wall-clock risk; perf work landed in Torch#219. Co-Authored-By: Claude <noreply@anthropic.com> * Phase 0: flesh out the verification gates Adds torch-migration/PHASE0.md with the full gate detail: what each gate's variable is, the driver source, exact commands, cost, and exit criteria. Two gates the plan was missing: - Gate C (Release build) - the harness 14/14 is a Debug measurement; CI and releases build -O3. - Gate A2 (two extractions in one process) - free once the driver exists, retires the re-entrancy risk offline instead of by hand in the game. Also: gates run full-tree + check.sh rather than test_assets.py (which hardcodes the torch path and copies a filtered yml subset), 19 ROM dumps not 17, and libgfxd being USE_STANDALONE-only is called out as the real hazard in Gate A. * Phase 0: correct the test_assets.py claim The "filtered subset" was a misread. The log line prints the pre-recursion yml count (1320); setup_scratch's external_files recursion then pulls in the remaining 130, so an unfiltered run copies all 1450 - verified by replaying the logic. So the gates use test_assets.py, the same tool that produced the existing 14/14, with a TORCH_BIN env override as the only harness patch. Gate A2 keeps the check.sh path, since it needs two extractions in one process and test_assets.py execs the binary once per ROM. * Phase 0: preflight green, 19/19 Baseline Torch build (USE_STANDALONE=ON, all games, Debug, PORT_VERSION_ ENDIANNESS=ON) reproduces every reference archive: 19 ROM dumps, all 0 failed / 0 not generated / 0 not in reference. Counts match the harness README's 14-target table, and the duplicate dumps agree with each other, which the 14-target matrix never showed. Reference manifests also spot-checked by re-derivation from two o2r files -- the baseline the gates measure against hasn't drifted. * Phase 0: Gate A green, 19/19 USE_STANDALONE=OFF plus driving Companion by hand produces byte-identical archives for all 19 ROM dumps. The real question was libgfxd, which is fetched only if(USE_STANDALONE) -- _deps/ confirms it is absent, so the Binary export path is genuinely gfxd-free. Reading said so; now it's run. Progress denominator confirmed exact: SetPhaseCallback fires once per yml file, matching the on-disk count for all three classes (1449/1450/1480). Phase 3's bar can be monotonic with no change to OTRGlobals. Also observed for Phase 2: zlib is fetched even with BUILD_STORMLIB=OFF and no zlib.h use in torch/src, confirming risk #1; spdlog is not fetched, so only the global-logger stomping applies there. * Phase 0: Gate A2 green, and the duplicate-entry finding Two extractions in one process, vanilla then MQ, both byte-identical to their references. The second run's phases=1450 matches pal_mq's on-disk yml count exactly, which is the direct refutation of the gProcessedFiles risk -- a leak would have skipped files and truncated the archive. A fresh Companion per extraction is sufficient; PLAN.md risk #6 retired. Also records what the gate turned up on the way: torch writes 25 duplicate archive entries, benign because libultraship indexes by CRC64 of the path. Filed as Torch#233 item 5. Corrects the Gate A note that attributed the whole ~343 KB size delta to compression -- 76 KB of it is these duplicates. * Phase 0: Gates B and C green OoT-only (eight BUILD_<game> flags OFF) and Release (-O3) each 19/19 identical, one variable moved apiece -- verified against the option() defaults so the other flags genuinely match the baseline cache. Gate C mattered: every parity measurement to date, including the original 14/14, was Debug, while CI and releases build -O3. Also records extraction wall-clock from torch's own timing line: Release averages 11.4s per ROM against Debug's 26.7s, and the OoT-only flags cost nothing. SoH ships Release, so PLAN.md risk #5 looks much smaller than the Debug numbers suggested -- though ZAPD's time on the same machine is still unmeasured. * Phase 0: Gate A' green -- the shipping configuration is the spec Static lib + OoT-only + Release together: 19/19 identical plus the vanilla-then-MQ pair. Flags verified from the generated cache rather than assumed, so Phase 2's CMake block and TorchExtract.cpp can be transcriptions of this build and RunOnce respectively. Archive bytes turn out to be configuration-independent -- the pair produced exactly the sizes the Debug static-lib run did, phases=1450 on both runs. Extraction 11.8s mean, so the static-lib wrapper costs nothing over Gate C. Confirms both Phase 2 dependency risks in the configuration that actually ships: zlib fetched with BUILD_STORMLIB=OFF and no zlib.h includes, tinyxml2 fetched with BUILD_NAUDIO=OFF removing its only consumers. Both FetchContent declarations are unconditional. * Copy soh.o2r from a checked-in prebuilt archive (TEMPORARY) Unblocks the rest of the migration. soh.o2r is produced by ZAPD via OTRExporter/extract_assets.py, both of which this migration deletes; its replacement is a small in-tree packer that has to reproduce ZAPD's texture quantisation exactly. Rather than block Phases 2/3/5 on that packer, check the archive in and make GenerateSohOtr a copy. The archive doubles as Phase 4's acceptance oracle -- the packer is correct when its output matches this entry-for-entry -- which is strictly more useful than Gate E's planned manifest of hashes of it. Generation turns out to be content-deterministic but NOT byte- deterministic: two runs give identical payloads for all 1,042 entries but different file hashes, because zip stores per-entry timestamps. So the Phase 4 comparison must be entry-wise; whole-file would fail even on a correct archive. Recorded in prebuilt/README.md. Adds a configure-time guard on the baked-in portVersion. OTRGlobals.cpp :283 requires exact major.minor.patch equality and RunExtract exit(1)s with "soh.o2r is outdated", which is an opaque symptom for a stale checked-in file; verified the guard fires by building against 9.2.4. ExtractAssets, CI and copy-existing-otrs.cmake are deliberately untouched -- ExtractAssets still generates a correct archive while ZAPD exists, and the CI job keeps working since it just builds this target. * Extract ROMs with Torch instead of ZAPD Adds soh/assets/yml as a submodule (briaguya0/soh-asset-yml, 20,353 asset definitions) and FetchContents torch at 4cae4416, configured exactly as Gate A' measured it: static lib, OoT only, no UI, no StormLib. find_package(ZLIB) has to run before torch is declared. Torch declares zlib with OVERRIDE_FIND_PACKAGE, so it takes over the whole build's find_package(ZLIB), and its copy provides no ZLIB::ZLIB -- which both StormLib and CMake's FindPNG link by name. Filed upstream as Torch#233. CallZapd becomes CallTorch: no chdir, no symlinked assets dir, no 22-entry argv. Everything that needed the working directory is now a parameter. TorchExtract.cpp is the only TU that includes Companion.h, since torch exports its whole lib/ as PUBLIC includes. Progress uses torch's phase callback, which fires once per yml file, with the denominator counted off disk. Gate A confirmed the two match exactly for all three yml counts, so the bar runs 0-100 without the sawtooth the old per-file counter would have produced. GetZapdVerStr becomes GetTorchVersionDir, returning the version directory under the yml tree rather than a ZAPD xml directory. soh.elf links torch and no longer references zapd_report. * Record the two-ROM session passing in-game Vanilla then MQ extracted in one process, both loaded and played. Closes PLAN.md risk #6 in the game rather than only against the harness driver, and confirms config.yml-driven archive naming end to end. * Add soh-torch, rebuild ExtractAssets on it soh links torch as a static library, which compiles out torch's own CLI, so build-time extraction needs an entry point. soh-torch supplies one around the same SohTorch::Extract the game calls -- one implementation, so the CLI and the in-game path can't drift. ExtractAssets keeps its old contract: SOH_ROM_PATH accepts roms and/or directories of roms and defaults to roms/, so dropping a vanilla and a master quest rom in produces oot.o2r and oot-mq.o2r in one run, and chaining GenerateSohOtr keeps soh.o2r coming out of it too. Extract now returns the archive name torch chose rather than deriving it from IsMasterQuest(), so config.yml is the only thing naming archives. The name comes from Companion::GetOutputPath(): scanning the destination for *.o2r picks up an unrelated soh.o2r when extracting into the build directory, which is only invisible in the game because it extracts into an empty temp dir. ExtractAssetHeaders stays as a target that explains why it can't run -- torch registers no OoT header exporter and emits no #define d<sym>, which soh source references ~24k times. Documented in BUILDING.md alongside it. This removes the last ZAPD dependency from the build. * Add the roms directory ExtractAssets defaults to BUILDING.md points people at roms/, but git can't track an empty directory, so a fresh clone didn't have one. Its .gitignore keeps everything but itself and the readme out. * Delete ZAPDTR and OTRExporter Nothing drives them any more: ExtractAssets runs soh-torch, the in-game extractor calls Companion directly, GenerateSohOtr copies a prebuilt archive, and soh links torch instead of ZAPDLib. Verified with a build from a fresh tree with both submodules absent from disk. Removes the submodules, the 7,680 ZAPD xmls under soh/assets/xml, the extractor inputs (Config_*.xml, TexturePool.xml, filelists, symbols) and copy-existing-otrs.cmake. The xmls remain in this repo's history and in the test harness's pinned shipwright submodule, which is where the yaml was generated from. Packaging changes shape as well as content: the extractor assets used to install into the 'extractor' component, which Packaging-2.cmake only includes for the AppImage generator, so linux zips shipped without them. torch is linked into soh rather than being a separate binary, so the yml tree installs into 'ship' on every platform and every package can extract a rom. * Keep CI providing tinyxml2 itself Torch's fetched copy satisfies libultraship today, so the install steps are redundant -- but that rests on torch declaring tinyxml2 with an unconditional OVERRIDE_FIND_PACKAGE, which is an implementation detail, and one we've asked upstream to reconsider in Torch#233. If the fetch gets gated, find_package needs a system copy again and CI would break because we removed the steps providing it. Also notes that the generate-soh-otr slim-down was written before the prebuilt archive, which makes that job build nothing for now. * Generate soh.o2r again, replacing the prebuilt archive soh-o2r-packer builds soh.o2r from soh/assets/custom, reproducing the ZAPD/OTRExporter archive exactly: 1,042 entries, 0 missing, 0 extra, 0 content mismatch, checked entry-wise because zip stores per-entry timestamps and whole-file hashes never match between runs. Most of it is torch's: Companion::Pack walks the directory, zips it and writes portVersion; BaseExporter::WriteHeader writes the resource header; TextureType and CalculateTextureSize come from TextureUtils. The packer stages the assets into the shape the archive should have and hands that over. Encoding a PNG into an N64 texture is the only piece nothing else provides -- torch decodes rom data that is already N64 format, never the reverse -- so PngTexture.cpp is the whole of what had to be written, and it follows ZAPD's quantisation rather than n64graphics', which scales where ZAPD shifts. GenerateSohOtr copies the libultraship shaders into assets/custom again before packing; assets/custom/shaders is gitignored and nothing else populates it, so a fresh clone would otherwise pack three files short. Removes prebuilt/ and its configure-time portVersion guard. * Move the asset tools under soh/assets soh-torch and soh-o2r-packer both exist to turn assets into archives, so they sit better beside the assets they read than at the repo root. Nothing tangled: both targets are declared in the root CMakeLists with explicit paths rather than add_subdirectory, and soh/CMakeLists.txt globs only include/, soh/ and src/, so sources under assets/ aren't swept into the soh target. * Build soh.o2r in CI without libultraship or soh SOH_TOOLS_ONLY returns from the root CMakeLists once the asset tools are declared, before libultraship and soh are added, so a configure that only needs soh-o2r-packer never reaches LUS's find_package(SDL2 REQUIRED). That lets generate-soh-otr drop building SDL2 from source, SDL2_net, and the deps cache that existed to hold the SDL2 tarball -- eleven steps down to six. Verified a tools-only build produces the archive with all 1,042 entries matching. The tool targets move above the sub-projects so the early return can sit between them; they only ever needed torch. The packer creates its output directory, which previously came for free from soh's build directory. * Fix macos and windows builds, keep the tools out of the game build macos: libultraship declares spdlog with OVERRIDE_FIND_PACKAGE so its own find_package(spdlog REQUIRED) resolves, but torch declares spdlog without it, and FetchContent_Declare is first-wins. Configuring torch first made LUS's declaration a no-op and left find_package with no config to find. It only showed on macos because that's the platform where neither project finds an installed spdlog -- it's kept out of macports.yml because its fmt dependency breaks the universal build -- so both fall through to fetching. Declaring it ahead of both keeps macos on the same source-built v1.16.0 it used before. Torch already uses OVERRIDE_FIND_PACKAGE for tinyxml2 and zlib, so the real fix is upstream. windows: soh builds with /WX, and TorchExtract.cpp is the one translation unit reaching yaml-cpp through torch's headers, which trips the dll-interface warnings. Suppress those two on that file. EXCLUDE_FROM_ALL on soh-torch and soh-o2r-packer: they are build-time tools, so building the game shouldn't compile them. ExtractAssets and GenerateSohOtr still pull them in through DEPENDS. Drops the .gitattributes rule for the prebuilt archive, which is gone. * Drop the gitignore negation for the prebuilt archive Left over with the .gitattributes rule when prebuilt/ was removed. * TEMPORARY: put soh/assets/xml back so the branch diff is reviewable Deleting 7,680 xmls puts the diff past what GitHub will render, and they are 99.8% of it. Restoring them here leaves the ~46 files that actually changed, which reviews in the compare view. Revert this before opening the PR. * Remove the migration planning docs Scaffolding for the migration itself rather than anything the repo needs to carry afterwards. * Fix NTSC master quest roms being rejected as unsupported verMap's two master quest rows keyed off the GameCube constants instead of the MQ ones, so with duplicate keys dropped the map never contained OOT_NTSC_US_MQ or OOT_NTSC_JP_MQ and both roms were filtered out before extraction. IsMasterQuest already handles both, and the asset yml covers ntsc_u_mq and ntsc_j_mq, so keying them correctly is all that's needed. Predates this migration; the torch cli was unaffected because it hashes the rom rather than consulting verMap. * Revert "TEMPORARY: put soh/assets/xml back so the branch diff is reviewable" This reverts commit 28f4113. * Run clang-format Renaming CallZapd to CallTorch made the call sites one character longer and left their continuation lines under-indented. Co-Authored-By: Claude <noreply@anthropic.com> * Temporarily pin torch to the fork carrying the Windows fixes HarbourMasters/Torch#234 fixes three Windows-only defects in the OoT factories: a 64 KiB over-read past the code segment that crashes extraction, backslashes in exported scene resource names that crash at scene load, and external_files being keyed differently from the directory walk, which processed every dependency twice. Without them a Windows build of this branch cannot extract a rom. Goes back to a HarbourMasters sha once that PR merges. Co-Authored-By: Claude <noreply@anthropic.com> * Point torch back at upstream HarbourMasters/Torch#234 landed as 65eb11c, squashing the three Windows fixes this branch was pinned to a fork for. The squashed tree is identical to the fork commit that was tested, and extraction output is unchanged: 35,386 entries, same CRCs. Co-Authored-By: Claude <noreply@anthropic.com> * Vendor the asset yml instead of submoduling it The submodule existed only so this PR's diff stayed reviewable; squash-merging it would have put a submodule on develop. This replaces it with the 20,353 files themselves, from briaguya0/soh-asset-yml@523be1d. The submodule's own README is dropped rather than vendored, since it describes that repo and was showing up in the assets folder of release downloads. Co-Authored-By: Claude <noreply@anthropic.com> * Bump torch to pick up the yaml-cpp 0.9.0 bump HarbourMasters/Torch#235 landed as e92c210. The pinned yaml-cpp predated jbeder/yaml-cpp@4fe2fb8, so every scalar conversion constructed a named std::locale; cheap on glibc, expensive on MSVC. Windows extraction of a PAL GC rom drops from 49.0s to 13.9s on the same machine, which also puts it ahead of ZAPDTR's 48.4s rather than fractionally behind. The merged commit is tree-identical to the fork commit those numbers were measured on. Extraction output is unchanged: 35,386 entries, same CRCs. Co-Authored-By: Claude <noreply@anthropic.com> * Consume torch as a submodule instead of FetchContent Reviewer feedback on #6989: FetchContent makes iterating on torch itself awkward, since testing a change means pushing it somewhere the pin can reach. A submodule is editable in place, which is how libultraship already works. Nothing else changes -- FetchContent_MakeAvailable was already doing add_subdirectory, so the include topology and every build option are the same. The pin moves from GIT_TAG into the gitlink, at the same sha. Co-Authored-By: Claude <noreply@anthropic.com> --------- Co-authored-by: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.