- an addon's own texture files resolve from its folder, TGA included, and SetTexture answers 1 or nil the way the client does — custom icons and skins draw
- runtime SetParent re-levels a frame by the client's strata law, and the raise survives hide-and-show round-trips — AtlasLoot layers right
- an addon that unregisters an event mid-fire no longer robs the next addon of the same event — Bagnon's per-character database fills again
- the Model widget's clock truncates milliseconds like the client's, and the no-text cast-bar law is confirmed at the bytes and narrowed to the cast bar alone
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- `--surface` re-runs the comparison against the captured in-world global table on demand, and `--surface-dump` writes our raw side for any tool to join — a one-shot measurement stops drifting
- CreateFrame with an unresolvable template raises exactly as the client does, instead of handing back a bare frame; `inherits=` is one name, matched case-insensitively
- CreateFontString and CreateTexture honour their third argument — 49 corpus call sites were passing an inherit into the void
- the addon-message inbound byte map is corrected (a test had agreed with the wrong constants), and the group-loot roll template publishes under the reference's own name
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- the one shared region table splits into the client's own maps: a Texture no longer answers SetText, a FontString no longer answers SetTexture, and feature-detecting addons see the true surface
- CreateTitleRegion lands whole — object and drag — so an options window that is its own drag handle actually moves
- font-object names resolve whatever their case, and the CursorUpdate/CursorOnUpdate pair arrives for the container-item templates
- GetStringHeight is pruned (1.12 has no such method) and SetPortraitToTexture moves to the global it really is
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- GetObjectType and IsObjectType land from the binary's own carve, with the client's exact inheritance answers
- the survey seats money and a quest log — sixteen addons call GetMoney, five walk the quest log, and none of those paths had ever run
- `date("*t")` returns its table form; the missing shape was an infinite loop for any addon walking days backwards
- unit tokens fold case like the client's resolver, and a FontString template carries the font object it inherits through the whole chain
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- the justify verbs had been written three times and drifted three ways; one implementation now serves the region, font-object and message-frame tables, and a cleared axis paints centre, as the client's draw path really does
- MessageFrame and ScrollingMessageFrame gain the ten-verb font block they were missing entirely, so font calls on message frames stop raising
- EnableKeyboard and IsKeyboardEnabled land as real Frame methods — eight addons were dying on their absence
- the harness's METHODS ranking states its own limits instead of reading as a build queue
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- the corpus harness now seats a buff, a target and a ticking cooldown instead of an empty world, so "nothing to draw" and "failed to draw" stop sharing a row — and the richer seat immediately exposed two real gaps
- an addon button inheriting our action-button template now works even when it replaces the template's OnLoad: the handlers resolve their regions by name, the way the reference does on every access
- more of the reference's hidden-by-default frames exist by name — ReputationWatchBar, MiniMapBattlefieldFrame, MiniMapMeetingStoneFrame — for the addons that reposition them
- StartSizing lands beside its stop verb, and the sourced bag-anchor code runs against windows we actually build
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Lua 5.0's generic-for and nested long strings are restored in a forked VM — 60 corpus addons stopped dying on one deleted opcode
- a day of demand-driven carving takes the 218-addon corpus from 24 addons surviving a session start to 114, with 144 loading without a single error
- Bagnon draws a real bag and answers real gestures — hover tooltip, pickup, use, drag — because our frames wear the reference's names, and the reference's own ContainerFrame.lua runs off the player's install instead of being transcribed
- the physics broad phase leaves: nothing in the client reads a contact, and it was measured at up to 41% of main-thread samples in real sessions
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- benilla loads real 1.12 addons from Interface/AddOns — TOC, XML and Lua through one loader, with per-addon saved variables, Bindings.xml and the lifecycle events addons expect
- both AddOns screens are in: the character-select list with per-addon enable state, and the ESC menu's in-game list
- the API surface is measured, not guessed — a corpus harness runs hundreds of real 1.12 addons and ranks exactly which globals, templates and verbs they demand
- the Lua dialect is the client's own, read off the binary — 5.0 semantics with the 1.12 additions, and UI source files are bytes, not UTF-8
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- The UI stopped paying to discover that nothing changed. Layout used to hash
~1,900 frames' worth of inputs every frame just to decide "quiet"; it now
tracks a mutation epoch and answers with one integer compare, and a settled
frame reuses last frame's rendered quads instead of re-rasterizing every
glyph. ~0.9 ms off every idle frame — with a verify mode, forced on across
all 489 UI tests, that re-runs every skipped resolve in full and asserts it
would have produced identical results.
- Across this run of snapshots, the one number everything is graded by: CPU
work per frame at the crowded dungeon spot went from ~29 ms to ~16.5 ms.
- A lighting question got the full treatment and an honest answer. Units
briefly took a different day/night lighting law; it read wrong on screen the
same evening, and a build-by-build hunt a full week back confirmed the
world's look never regressed in the first place — the change is reverted,
nothing about the look changes in this snapshot, and the question stays
parked until the one remaining unknown in the reference client is pinned.
- The character pane's model booth grew a headless eye — a dump of exactly
what it draws, no live session needed — built for that hunt, kept for the
next one.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- The main menu bar's two empty recesses fill. The micro menu — the eight arched
buttons: character, spellbook, talents, quest log, socials, world map, main menu,
help — sits in the dark panel between the page arrows and the bags, transcribed
from the reference's own layout: the character button wears your portrait, a button
holds itself down while its panel is up, and the talent button appears at level 10.
The latency meter fills the one slot the bar art leaves transparent for exactly it,
tinted green/yellow/red on the reference's 300/600 ms thresholds and drawn UNDER
the bar art, seen through its window, the way 1.12 actually does it. The socials
and main-menu buttons find their panels one snapshot later; help still waits.
- The UI explains itself on hover: 1.12 ships its detailed two-line tooltips ON — a
FrameXML default, not the opt-in extra it is often remembered as — so the micro
buttons, your own portrait, the XP bar, the five bag buttons and the ping meter
all answer a hover with the reference's own explanations.
- The PvP flag, end to end: crossed swords on the player, target and party frames,
driven by the two descriptor bits the wire actually carries, and `/pvp` speaks —
the yellow toast plus the system chat line, the OFF sentence being the one that
warns the flag lingers for five minutes. It really does linger: turning it off
waits out the server's timer, which looks like a bug and is the reference.
- Quest markers stopped freezing. The grey `!` over a questgiver above your level
goes gold the moment you ding, an NPC re-entering view is asked afresh instead of
wearing a session-old answer, and the marker drops when the questgiver bit goes
off — the re-ask triggers mirror the reference's own set, plus one named addition
(your own level changing).
- Water and heights: swimming downhill no longer bounces between swim and fall ten
times a second — the surface line is a two-sided constraint, so the swimmer
settles with the river instead of being left behind by it. And collision height
is per-unit now, read from the unit's own model: a gnome used to swim with a
third of a yard of water over her head because five depth lines were quoting a
human's number.
- A round of pixel honesty in text and chat: the newest chat line no longer wears a
wider drop shadow than every line above it (shadows move in whole pixels now),
keeps its descenders, and hover boxes stop painting over the chat body. The chat
input's history browse survives programmatic edits, a failed script chunk warns
instead of vanishing, and recall deepens to 64 lines. Idle animation got honest
too: a model's idle is Stand resolved through its own lookup table — never
file order, which had the duel flag bobbing on its spawn clip — and an armed idle
is culled on the box its animation authors, not the bind pose, so the flag stops
vanishing at most camera angles.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
