050b8251a73cbc67ef27b02f6971c5fbb08d8371
12 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 claude
|
050b8251a7 |
Stage 3: inventory, ground loot, and two loot visibilities
ci / verify (push) Successful in 48s
Four always-on-screen slots, items as data, and loot tables on enemies and bosses. Health potions drop rarely from trash and always from the Warden; the Warden also drops a Warden's Ration, one per living player, which does nothing at all. The ration is not filler. Player-instanced loot is a separate code path from shared loot -- a distinct entity per owner, filtered per peer in the snapshot encoder -- and the cheapest way to keep that path honest is to have something in the game that exercises it on every boss kill. Item actions ride the input frame rather than becoming new client messages. InputFrame gained BTN_USE, BTN_DROP and a slot byte, which buys the packet-loss redundancy, the replay guard on last_input_tick, ordering against movement on the same tick, and a rate limit of one action per tick -- all of which a separate RPC would have needed bolted back on. The cost is that anything in the frame which must not repeat has to be edge-triggered, since frames are resent and a starved server coasts on the last one it holds. Instanced loot is enforced in NetCodec.encode_snapshot, beside the actor interest radius: a peer is never told another player's copy exists. Hiding it client-side would have been the same mistake as relying on fog to hide enemies. Inventories live on the character and are written to the store on every transaction, so a crash between "picked it up" and "wrote it down" cannot lose or duplicate an item. Anything dropped becomes world-shared whatever it was before, and a potion used at full health is refused rather than spent. tools/diag_loot.tscn covers drop -> snapshot -> pick up -> persist -> use -> drop plus both visibilities on the wire, for the same reason diag_progression exists: bots are poor shots and almost never produce a drop. It asserts each input frame was actually consumed, after an early version silently dropped its first press and every later check passed for the wrong reason. check.sh clean, 266 tests, SMOKE PASS, all three diagnostics green. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> |
||
|
claude
|
ded7bf96d5 |
Documentation pass before context compaction
ci / verify (push) Successful in 47s
The roadmap was actively misleading: an earlier stage renumbering left Stage 2's completed work sitting under a "Stage 3 -- todo" heading, and the inventory table orphaned with no heading at all. Since that file is the primary handoff document, a fresh session would have started by re-implementing accounts and characters. Rewritten. Captured in full, from the original brief rather than from memory, the two stages not yet built: - Stage 3 (inventory and loot): 4 slots, potions rare from trash and guaranteed from bosses, world-shared loot, and the player-instanced food item -- with a note that two loot visibilities must exist from the start, because proving the instanced path works is the food item's entire purpose. - Stage 4 (upgrades): every upgrade with its exact stated effect, plus the two constraints it will collide with -- sniper's 2x bullet speed against the tunnelling threshold, and per-player bullet travel against the interest radius that test_interest.gd currently derives from static content. Ten open questions are listed as explicitly do-not-guess, seven of them blocking Stage 4. ARCHITECTURE.md still claimed "the arena is a rectangle" and "no tilemap collision", both untrue since Stage 1. Rewritten around MapGrid, with what the grid costs (axis-aligned, 32px-quantised, bullets under a tile per tick) rather than only what it buys. Added a "verification traps" section to WORKFLOW.md recording six mistakes made during this work, each of which cost a round trip of reporting something fixed that was not: verifying the artefact rather than the behaviour, dumping the wrong channel, measuring a configuration where the bug cannot exist, a test whose setup silently invalidated it, git checkout reverting real work alongside a probe, and a pattern edit matching in two files. They are specific enough to be actionable. Also documented the diagnostics in CLAUDE.md -- they were undiscoverable -- and recorded the current verification surface so "everything passes" has a stated meaning. 206 tests, 15 smoke assertions, both diagnostics pass. |
||
|
claude
|
d8197885ca |
Hide dead characters, suggest names, make the XP bar live
ci / verify (push) Successful in 46s
The roster sent to clients now contains living characters only. Retirement stays server-side bookkeeping for archival; from the player's side a dead character is simply gone, and listing it offers a choice that cannot be taken. The XP bar only moved on a level-up or a character swap because it read the character roster, which is re-sent only when the SET of characters changes. Experience now rides the snapshot -- four bytes on a message already going out at 20Hz -- and the server mirrors each grant into the world immediately rather than only when a level is crossed. The create field starts with a suggested name instead of blank, and offers another after each creation. Two bugs found while testing, both mine: The first was a bad patch of my own: a change meant for the snapshot decoder also matched inside decode_characters, which then read a four-byte field its encoder never wrote and ran off the end of every packet. This is precisely the "encodes but decodes wrong" failure the codec tests exist to catch, and it was caught within a minute of the test being written. Chasing that exposed a real robustness gap: StreamPeerBuffer.get_utf8_string() pushes an engine error and returns garbage when the buffer is short, so a truncated or hostile character/roster packet produced error spam instead of degrading. Both decoders now bounds-check every field, with tests that slice each packet at many lengths and assert it degrades rather than inventing entries -- the same guarantee the input decoder already had. 206 tests. check.sh, test.sh, smoke.sh and both diagnostics pass. |
||
|
claude
|
7ef972e3b3 |
Hub character swapping, XP percentage, passive health regeneration
ci / verify (push) Successful in 47s
Character swapping is hub-only, and refused by the SERVER rather than merely greyed out in the menu. Allowing it inside a dungeon would be an instant, uninterruptible exit from danger -- strictly better than the one-second escape channel, which would make that channel pointless. Creating a character in a dungeon is refused for the same reason. Both are covered by diag_progression. Health regenerates at 0.5% of MAXIMUM per second. A percentage rather than a flat rate so it does not become irrelevant at level 15: a capped character regains 1.2 hp/s against a level 1's 0.5, and both take about 200 seconds to heal from nothing. No out-of-combat gate -- at this rate it cannot out-heal anything actually shooting at you, and a trickle that never stops is easier to reason about than a timer players have to learn. A fractional carry is needed because a tick heals well under one hit point, so truncating each tick would heal exactly nothing; there is a test for that specifically. The XP bar now states the percentage and the level it leads to, since a bar answers "how far" vaguely and a number answers it exactly. Recorded the two Stage 3 answers: 4 inventory slots, and the boss's food item is player-instanced now so the mechanism gets exercised rather than deferred. Writing the swap guard's test caught my own mistake: the first version created its spare character through the store, which has no opinion about where you are, so it bypassed the guard it was meant to prove and left a stray character behind that broke a later assertion. 200 tests. check.sh, test.sh, smoke.sh and both diagnostics pass. |
||
|
claude
|
4765bbce28 |
Stage 2: accounts, characters, permadeath, levels and experience
ci / verify (push) Successful in 47s
Identity is shaped like Steamworks so swapping to it is one subclass and no schema change: the client presents an opaque ticket, the server validates it into a stable 64-bit account id, and nothing downstream sees anything else. LocalAuthProvider takes any ticket at face value -- insecure on purpose, and labelled as such everywhere, because the point is the shape rather than the security. Do not ship it. Characters persist as JSON keyed by account. Account ids are written as decimal strings because they are 64-bit and JSON numbers are doubles, which would silently round them. A corrupt store aborts the server rather than starting empty: starting empty looks like it worked and then saves over every character on the first level-up. Levels 1-15, +10 max health each, level DERIVED from lifetime experience rather than stored beside it, so a hand-edited save cannot produce a level 12 character with a level 3's experience. Experience is shared undivided across everyone alive in the instance -- splitting it would make bringing a friend cost you progress. A level-up heals by what it added, so gaining one mid-fight is relief rather than a bar that moved further from full. Death is permanent and unbinds the character entirely: no "return to the hub as the character who just died", because the run is over. The record is retired, never deleted. The five-character cap counts LIVING characters only -- counting the dead would lock a player out of their own account after five deaths. Verified by tools/diag_progression.tscn, which drives the real server through kill -> xp -> level -> health and death -> retire -> roster. The bot smoke test cannot cover that: bots are poor shots and rarely kill anything. Writing it caught two real ordering bugs -- the death event was dispatched before the payload that tells the player they died, and the dead character stayed bound to the peer. Also added --account and --store so several clients and test runs can coexist on one machine. The smoke test now uses a scratch store; without it a rerun resumed the previous run's characters and "a character was created" quietly stopped being true. 193 tests. check.sh, test.sh, smoke.sh, diag_progression and diag_prediction all pass. |
||
|
claude
|
3586555aea |
Remove non-redistributable art from the repo; record licence terms
Researched the licence at each of the four itch.io sources. Two of the four
forbid redistribution, and publishing a repository containing them is exactly
that, so they are out of the tree.
0x72 DungeonTileset II CC0 1.0 commercial ok, no credit required
Helton Yan Pixel Combat CC BY 4.0 commercial ok, CREDIT REQUIRED
Fire Pixel Bullet 16x16 custom non-commercial, NO REDISTRIBUTION
750+ Effect and FX Pixel custom non-commercial, NO REDISTRIBUTION
Both bdragon1727 packs state verbatim: "You cannot do: Resell / redistribute
this asset." The FX pack's paid tier is no better for this purpose -- "You can
NOT re-distribute the file, no matter how much you modify it" -- so paying does
not unlock it and neither does editing the art.
bullets.png and impact.png therefore move to assets/local/, gitignored.
Art.bullets_texture() / impact_texture() load them at runtime and return null
when absent; the bullet renderer falls back to the generated dot it used before,
tinted per kind so bullet types stay distinguishable. Nothing there may be
preload()ed -- preload resolves at compile time and would fail the build on
every machine without the files, which is every clone.
Verified both ways rather than assumed: with the files present the game uses
them and no fallback triggers; with them moved aside, 156/156 tests pass and the
game runs clean on the generated dots.
Added CREDITS.md, because Helton Yan's CC BY 4.0 makes attribution a licence
obligation rather than a courtesy. A file in the source tree does not discharge
it once the game ships -- an in-game credits screen is now a tracked todo, and
flagged as a licence requirement rather than polish.
NOT addressed here, and it needs a decision: these files remain in git history
at
|
||
|
claude
|
9fa2e260f2 |
Replace placeholder rendering with real sprites and audio
Terrain, actors, boss and bullets are sprites now, and four sounds play off server events. 548KB total across seven files, out of 3.9GB of source packs. - src/view/art.gd holds every atlas rect and sound path in one table, copied from the 0x72 pack's own tile_list. Scattering coordinates through draw calls would make re-cutting the atlas a hunt, and a wrong rect invisible. - Bullets get one MultiMesh per kind rather than one overall: a MultiMeshInstance2D carries a single texture and each kind needs a different region of the sheet. Four draw calls, no custom shader passing UVs through per-instance data. - src/view/sfx.gd is a 16-voice round-robin pool. Bullet-hell fire rates mean sounds overlap constantly, and identical sounds landing within two frames are collapsed so a ring hitting eight bullets is a bang rather than clipping. Every sound is triggered by a server event, never a local guess, so what you hear matches what happened. Audio was converted, not copied: the pack ships 24-bit/96kHz masters averaging 2MB. Downsampled to 16-bit/44.1kHz mono, silence-trimmed, and the shoot sound hard-capped to 0.30s -- it came out at 2.03s, against a fire cooldown of 0.12s, which would have smeared held fire into noise. tests/unit/test_art.gd checks what cannot be eyeballed here: every atlas rect lands inside its texture, every animation frame of a strip fits (the last frame is what runs off the sheet, not the first), there is a sprite per bullet kind and per enemy visual, and no sound is long enough to stack badly. A wrong atlas coordinate does not error -- it silently draws the wrong pixels. Licence sources recorded in docs/ASSETS.md now that they are known. All four are free versions and four rows still say "confirm at source": free itch packs vary on credit and commercial use, and there is no credits screen yet. 155 tests (was 146). check.sh, test.sh and smoke.sh pass. |
||
|
claude
|
7ee6c8e761 |
Close Stage 1: per-peer actor interest, boss room confinement
ci / verify (push) Successful in 46s
The gap Stage 1 left open: snapshots were encoded once and broadcast to every peer in an instance, so clients were handed enemies the fog then hid. Fog is a rendering rule -- a modified client draws whatever it holds -- so that was no defence at all. Snapshots are now encoded per peer, and actors beyond ACTOR_INTEREST_RADIUS (800u) are never sent. Measured cost of four filtered encodes against one shared: 95.6us vs 24.9us, or 0.032 ms/tick amortised over the snapshot interval. Two asymmetries worth keeping: - The observer's own player record is never filtered, however far out the arithmetic puts it. The client reconciles its prediction against that record, so dropping it breaks the player's own movement rather than hiding someone. - Only bullet SPAWNS are filtered, never despawns. A client told about a bullet must always be told it died, or it keeps a phantom. Bullet spawns use a much wider radius (2200u) than actors, deliberately. An enemy appearing at the edge of sight is cosmetic; a bullet withheld at spawn that later flies into view is invisible damage. The floor is longest bullet travel + fog radius -- 1500 for the Warden's Collapse snipe -- and test_interest.gd recomputes that from live content, so adding a faster or longer-lived bullet fails a test instead of producing bullets that wink into existence. Also clamped bosses to their room. A no-op today since every boss is stationary, which is exactly when the invariant is cheap to establish: boss rooms have no door that locks, so the only thing keeping a fight in the boss room is the boss, and Stage 5's movement work would otherwise break it quietly. 146 tests (was 137). check.sh, test.sh and smoke.sh pass. Stage 1 has no partials left; docs/ROADMAP.md now records the three interest radii and the floor each must respect. |
||
|
claude
|
20dd6bc502 |
Keep asset packs out of git and out of Godot's importer
ci / verify (push) Successful in 46s
The packs total 3.9GB, almost all of it Helton Yan's SFX shipped as 24-bit / 96kHz studio masters -- ~2MB for a single gunshot. Two guards, because they solve different problems: - .gitignore excludes the contents. The directory and its guard files stay tracked, so a fresh clone has the .gdignore in place before anyone drops packs back in. - assetpacks/.gdignore stops Godot descending into the tree at all. This is the urgent half: tools/check.sh runs --import, which walks all of res://, so the next check would have imported several thousand PNGs and 2,100 of those WAVs, and every run after it. Verified by controlled test rather than assumed -- an unguarded PNG under res:// imports and gets a .import file beside it; with .gdignore neither happens. check.sh stays at 3.3s and .godot/imported at 1.2MB. The trade the .gdignore buys is that nothing in assetpacks/ can be referenced from a scene or script. That is the intended shape: raw source on one side, game-ready files copied into res://assets/ on the other, so adding a huge pack costs the repo and the import step nothing. docs/ASSETS.md records attribution, because gitignoring the packs also gitignores their licence files. Worth flagging: four of the six packs shipped with no licence text at all, and two of the undecided ones are demo versions, which are often more restricted than the paid packs. Only RF Catacombs has explicit terms (public domain, no resale). Those rows need a source URL before anything ships. Also removed __MACOSX and .DS_Store extraction artefacts, and left considering_dont_use_yet/ untouched as the user's own "not chosen" marker. |
||
|
claude
|
de48afcbd9 |
Fix aiming under a scrolling camera; add status and decisions docs
ci / verify (push) Successful in 46s
The aim bug was collateral from the camera work. "Mouse relative to the centre of the screen" WAS the cursor's world position while the world was drawn fixed at the origin, so subtracting the player position gave the right vector. Once the camera scrolled, that expression became the aim vector itself, and subtracting the player position again made the ship aim at a fixed world location -- walking around swung the crosshair with the mouse held still. Fixed by inverting the transform the view actually draws with (world = screen - world_view.position, published by the game scene each frame) rather than assuming the player is centred, so it still holds if the camera later clamps at map edges or gets shake or look-ahead. tests/unit/test_aim.gd pins it, including the regression directly: moving the player must not move the crosshair. Documentation, for other sessions picking this up cold: - docs/ROADMAP.md rewritten as the status map -- every feature in the brief against its state and the file implementing it, the known gaps called out (actor interest management is the notable one), and the ten design questions that are genuinely unspecified and should not be guessed at. - docs/DECISIONS.md, new: settled decisions with their reasoning, so a session does not re-litigate or re-ask. Several are not the obvious default -- no i-frames, no contact damage, non-interruptible escape, and never sending the map seed. - CLAUDE.md and README point at both. 137 tests; check.sh, test.sh and smoke.sh pass. |
||
|
claude
|
005679f1b5 |
Fix bullet/ship desync; rework death, escape, arrival and hub awareness
(1) Bullets appeared to trail the ship. Two independent causes, measured with
the new tools/diag_prediction.gd rather than guessed at:
- ServerRuntime ticked before ClientRuntime, so input sampled on frame N was
not consumed until frame N+1, leaving the drawn ship a constant one tick
(4.00px at 240 u/s) ahead of the authoritative one that bullets spawn from.
ClientRuntime now sets process_physics_priority = -10. Gap on a listen
server: 4.00px -> 0.10px mean, 0.30px worst.
- PLAYER_MUZZLE_OFFSET was PLAYER_RADIUS + 6 = 12px against a 13px drawn
ship, so bullets were born inside the sprite. Regression from the previous
commit's hitbox shrink; it now derives from PLAYER_VISUAL_RADIUS.
(2) No more timed respawn. A downed player stays down until they ask for the
hub (E), which is an ordinary input -- the server has no "revive me" message.
(3) Escape channel 3s -> 1s, and damage no longer cancels it. An interruptible
channel makes killing the process strictly better than using the button, so a
dropped connection now runs the same channel: the player stays in the world as
linkdead, still killable, and is only released once it completes. Instances
refuse to close while a linkdead body is resolving, or a solo drop would delete
it on the next tick and hand the exploit straight back.
(4) Escape opens an in-game menu: return to hub (routed through the same held-
escape channel, not a new message), disconnect, quit.
(5) Server pushes a roster so the hub shows who is online and which dungeon
they are in. Entering a dungeon grants 2s arrival protection -- invulnerable
AND weapons-cold, since invulnerability alone would make the spawn a free
firing position -- flagged in the snapshot and drawn on every protected ship.
(6) Cleared dungeons hold the party 30s (was 5s) with a visible countdown.
(7) The hub's grey circle was a 100k-HP target dummy that read as scenery. Now
drawn as a bullseye so its purpose is legible.
Protocol version 1 -> 2. 91 tests (was 78); smoke.sh gains a bot that is
SIGKILLed mid-dungeon to prove the disconnect path end to end. check.sh,
test.sh and smoke.sh all pass.
|
||
|
claude
|
c4beeae38f |
Initial commit: Transcience MVP
Top-down twin-stick bullet-hell, Godot 4.7, server-authoritative dedicated server with client-side prediction. Clients send input only; the server resolves every hit for both players and enemies (no PvP). - SimWorld: whole simulation as plain RefCounted objects (no nodes, no physics server), ~0.24ms/tick at peak load -- runs headless for free and drives 78 tests in under a second - BulletPool: struct-of-arrays bullet storage, replicated as spawn/despawn events rather than per-tick state - Emitter framework (Ring/AimedSpread/WallGap/ArcSweep) shared by trash enemies and bosses -- a new boss is data in src/content/content.gd, no simulation changes - The Warden of the Fold: stationary 4-phase boss built entirely on that format - Lobby hub with a portal into on-demand dungeon instances; one process hosts the hub plus every concurrent dungeon - Emergency escape: 3s server-owned channel, cancelled by damage - tools/check.sh, test.sh (GUT), smoke.sh (real server + bot clients over ENet), bench.gd; git hooks wired to the same scripts - docs/ARCHITECTURE.md, NETCODE.md, WORKFLOW.md, ROADMAP.md |