UFL-1.0.txt is the upstream Ubuntu Font Licence 1.0 text (google/fonts ufl/ubuntu/UFL.txt).
Montserrat-OFL-1.1.txt is the OFL 1.1 body, byte-for-byte the text already shipped with
Liberation Sans, headed by the copyright the Montserrat font embeds (Copyright(c) 2014 by
Julieta Ulanovsky) rather than the project's later authors line. The Ubuntu and
Montserrat credits now name these files as their LicenceFile; the test checks each
carries its licence, sits beside its fonts, and (OFL) the embedded holder.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Commit 16cb82d09 said every third-party work whose own text states holder and licence
was credited. That was false: header-only notices were missed. Now credited, holder
and licence taken from each file's own text: Academy of Motion Picture Arts and
Sciences (ACES.hlsl, 2015), Unity Technologies 2016 MIT (the built-in shader sources
Particle Standard Unlit, StretchedSkybox, MaskedWorldSpaceUI, MaskedWorldSpaceAdditiveUI
and the three AdditiveUI shaders), Supyrb 2017 (AdditiveUI shaders; the header states
no licence for that portion), and Stefan Gustavson 2011 MIT (Celestial Body
SimplexNoise2.cginc).
The discovery test now also scans every Copyright line in .shader/.hlsl/.cginc/.cs/.compute
under Assets; its holder must appear in a credit notice or in ReviewedHeaders with the
reason. Microsoft Open Technologies (UniRx headers, no licence stated, points at a
License.txt that is not in the repo) is the one reviewed entry. The Unity headers point
at a license.txt that is likewise not in the repo.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Prefabs reference the TMP SDF asset made from a font, not the .ttf, so matching the
font file guid alone could never fail. The unreferenced-font check now follows the
font guid and the guid of every Assets/Fonts file that references a guid already
followed (the font's own SDF asset, any SDF asset listing it as a fallback); a file
outside Assets/Fonts naming any of them is a use.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
ShowCredits renders the selected credit as property rows instead of one multi-line label, refuses with a dialog when
the attribution TextAsset is missing or malformed, takes list rows out of automatic navigation, steps with Up/Down on
the focused Back button and leaves on Cancel. ScenarioPicker's focus hold moves to FocusHold so both screens share it.
The menu wiring test now strips comments before matching.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds Liberation Sans (OFL), UniRx, KDTree, webgl-noise simplex noise and ambientCG to the list. CreditsTests now
matches LICEN[CS]E, COPYING, NOTICE, README, SOURCES, OFL files under Assets and requires each to be credited or
listed as reviewed-not-credited with a reason; credit notices are checked against what their files state. The
attribution heading is checked rather than skipped, the test states independent expectations, and the split Hurricane
line is joined.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Twemoji, Playdead and the Noun Project icon credits, held against the licence files under Assets by CreditsTests.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
AetherDb ship-authoring regenerate rewrites a package under the current schema and requires it to read back identical. ElementIdUpgradeTests gains content preservation and a shipped-packages-writable check; ShipModCatalogTests compares mod records unblinded where the same packages are composed twice.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
f458206c7 authored the five per-manufacturer station reactor products with bare
ProductRole entries (the .5 / .15 default). The command copies each source product's
Mean and StandardDeviation per role to a role still at the default, and was applied
to the catalog. TwoAiDigest repinned for the catalog move.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The public ItemFits(item, coord) the inventory drag asks counted the dragged equipped
item's own cells as occupied, so a nudge overlapping its current cells was refused.
The private rule now takes the item being moved as an excluded occupant. TryEquip
and TryFindSpace pass none and stay strict, so equipping an equipped item twice is
still refused; the UI keeps TryUnequip then TryEquip (cargo bay emptiness rule).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The previous commit deleted the GPU Noise includes; this one carries the call-site
edits it described. Release map notes Background's remaining 2D gradient copy.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
CPU snoise (CultMath C#) and the GPU shaders now share one implementation, so
fBm agrees. Deletes the GPU Noise copies nothing includes any more (Classic 2D/3D,
Simplex 3D, SimplexGrad 3D). Background.shader keeps SimplexNoise2D/SimplexNoiseGrad2D:
CultMath has no 2D gradient (cultmath_snoise_grad is float3 -> float4 only).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The Arcs fighter had no hold, so its Autocannon and Cell Blasters went silent after one magazine: it now fits the one free 2x2 bay. The estimated crate counts are replaced by measured draws (ten minutes of play or death by heat) plus margin, authored for every ammunition-drawing shooter; ScenarioAmmunitionRunTests drives both scenarios and fails if any shooter is stocked below its measured draw.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Uranium, Hydrogen (and Bacteria, Fungi, Animal Products) were named by mining, volatiles, agriculture, habitat-support and power descriptions with no recipe behind them. Dropped the claims (smaller than authoring a power chain); batch StationKindDescriptions restates the five kinds in the catalog; test pins the rule.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Reads only the ship's velocity after the zone steps; reintroducing drag below top speed fails both cases (50 -> 18.39 in 10 s).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The test counted aetherdb-economy-* directories in the machine's temp folder, which every parallel test running the command also writes to.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A plain Dictionary written by parallel catalog loads corrupted itself ("Operations that change non-concurrent collections must have exclusive
access") and failed unrelated tests at random, through ValidateStatModifiers.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
ChunkInfoIsBitIdenticalToItsValueBeforeTheExtraction never passed on Linux: at c86f28e93 (the commit that added it) it already measures 1039274088 for the second observer where the recording says 1039274086, and base 12dfcef5, the code the recording claims to be taken from, measures the same ...088 on Linux (CultMath 0.4.1 and 0.5.0 alike). The sensor's bearing term (acos, curve) is libm arithmetic, which rounds differently on Windows, where the recording was made. It is a host-class recording read as universal, not a regression: no commit moved it.
The pin is now per host class like the two-AI digest (ruling replays-per-host-class): windows-x64 keeps the original recording, linux-x64 is base 12dfcef5 run on Linux, and the current code equals it. A host with no recording fails and prints its bits. RunStartTests.HostClass is now internal, the one host-class name.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
845f6c3aa (aetheria-gravity:ruling:no-drag-below-top-speed) removed hull drag below a hull's top speed. The crosser was staged at the Djinni's 50 m/s cap on the premise that drag slowed it to ~25 m/s at the bow line; it now holds 50 m/s and crosses at 2.57 s, 50 m/s relative, 1.1 rad/s, outside the tuned encounter (3.6 s, 25 m/s, 0.56 rad/s) that the Brief, the pass window and CrossingHoldsAcrossTheStartWindow all describe. CrossingConditions and CrossingHoldsAcrossTheStartWindow failed on that stale tuning, not on a defect.
The encounter is restaged to reproduce its tuning under the ruled physics: launched at 25 m/s from 90 m out (probed over speed x start; constant-speed passes are dx / v seconds, 45 m ahead). The pass lands at 3.63 s, 44.7 m, 24.9 m/s relative, 0.56 rad/s. The test's launch-speed assertions now pin the scenario's 25 m/s instead of "most of the cap"; the time, distance, relative-speed and angular-rate windows are unchanged. If drag below top speed returns, the crosser decays and the pass leaves the window.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Arcs' hostile turret fits two 100-round Light Machine Guns and had no hold, so it fell silent after one magazine; it now fits a bay and carries 30
crates of Ammo (about 8 minutes of both guns firing without pause). Pursuit's authored cargo had no Mines, capping the player at one magazine of six;
it now carries 10 crates (sixty-six mines). ScenarioStage.Cargo admits a commodity name as one unit, so authored cargo can state crates.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
ConstantWeapon names an AmmoType but has no magazine (a crate per AmmoInterval straight from the cargo), so the label printed its unfilled
magazine ("1/N"). Weapon.AmmoReadout owns the label's text, HasMagazine says whether there is a magazine to show, and SchematicDisplay reads it.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Every commodity a kind extracts (a recipe with no inputs) is named in that kind's description: checked over the shipped catalog, only these
two kinds broke it. The batch applies in the catalog commit that follows; until then the catalog tests for it fail.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The two tests capture Console.Out to prove the commands write to the writer they are given. Console is process-global, so in a full-suite run
other tests' diagnostic prints landed in the capture and failed the author dry-run test. ConsoleProbe keeps only writes made on the thread
running the command, and serialises probes so two cannot restore each other's writer.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Pins the default path the boss and the console fly use: a fed gun leaves the
ship with AmmunitionCrates, a gun that draws nothing leaves none. Removing the
default StockAmmunition call fails the fed case.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Materialize takes stockAmmunition (default true: boss, console ship). ScenarioStage
passes false and stocks at the end of staging only the fitted entities whose cargo
the scenario did not author with Cargo, so a scenario that names what a hold
carries gets exactly that. A packed (saved) fit is never stocked.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Loadouts.StockAmmunition is the single owner of crate stocking. The generator
calls it for generated ships and turrets, and Loadouts.Materialize calls it at the
end, which covers every Loadout-built entity: scenario stage.Fit/Preset, the boss
in ZoneGenerator, and the console-built ship. Paths left alone: Unpack of a pack
or save (an entity's cargo is its own and a saved game's cargo is never touched on
load), scenario stage.Cargo (adds on top of the stock), the LoadoutCommit
validation build (discarded). The new-run player ship is stage.Generated and so
already stocked by the generator. A hull with no bay gets no crates and a log line.
InstantWeapon: a weapon that names an AmmoType is magazine-fed with an effective
magazine of max(1, MagazineSize): every round leaves the magazine and the reload
takes a crate. A weapon with no AmmoType and a magazine of 0 or 1 stays free.
HUD shows rounds/reserve for every AmmoType weapon. ConstantWeapon keeps its own
per-interval crate draw.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
