A WMO with a NaN doodad quaternion would produce NaN euler angles
through glm::eulerAngles() and persist them into the WoB. Identity
quaternion fallback for a non-finite source, plus NaN scrub on
position/rotation/scale separately so the converted WOB is always
load-safe.
The previous --validate output told you whether *some* file of each
type existed, which was hard to act on for partially-valid zones.
Now reports the per-format file count and how many failed magic
validation, e.g. 'WOM (12 invalid: 2)' so a zone author can spot
missing or corrupted models without grepping through file listings.
readInfo previously trusted fileCount/infoSize blindly, so a malicious
or corrupted WCP could allocate a 4GB string just to print metadata
via --info-wcp. Same 1M file / 16MB info caps now applied. Also
categorizes .woc collision files (was bucketed under 'other').
WoW liquid types are 0=water/1=ocean/2=magma/3=slime. A user-edited
WOT could carry an out-of-range value that the editor renderer
silently maps to plain water but the server treats as undefined.
A WOT water entry with non-finite height would push NaN through
the water mesh builder and produce a degenerate Vulkan draw
(invisible water at best, GPU hang at worst).
Same scrub now applied symmetrically on the save side so a
corrupted in-memory doodad transform can't be persisted into a
WOB and then have to be cleaned up on every subsequent load.
Already had a guard for scale; extending to position/rotation too.
A WoB with non-finite doodad transforms produces NaN model matrices
that propagate into the M2 instance SSBO and crash the GPU.
Mirrors the load-side sanitize. nlohmann throws on NaN/inf, so
a corrupted in-memory baseHeight or volume would abort the manifest
save and lose all zone-level settings (flags, audio config, etc.).
Same json-serialization-safety fix as for NPC and WOT saves.
Object position/rotation NaN would abort the entire objects.json
save and lose all placement work in the session.
nlohmann::json throws on non-finite serialization, which would
abort the entire creatures.json save and lose every spawn change
in the session. Scrub position/orientation/scale/radii/patrol
floats on the way out so a single corrupt NPC can't kill the
batch save.
nlohmann::json serialization throws on NaN/inf floats, which would
abort the entire WOT save and leave the user with an unsaved zone
state. Scrub on the way out so a single corrupted placement can't
take down the whole save.
map_dbc.MapName / area_table_dbc.AreaName are varchar(100). Edited
zone.json could carry longer strings that would either fail the
INSERT or silently truncate. Cap mapName/displayName at 100,
biome at 64, description at 4096.
creature_template.name is varchar(100) in AzerothCore. Edited
creature JSON could carry longer names that would either fail
the INSERT or silently truncate. Cap name at 100 and modelPath
at 1024 on load.
quest_template.LogTitle is varchar(200) in AzerothCore. Edited
quest JSON could carry longer strings that would either fail the
INSERT or silently truncate at the server. Cap title at 200 chars
and the longer text fields at 8KB on load.
Mirrors the editor-side ZoneManifest sanitize on the discovery
scanner used by the launcher and asset manager. A custom_zones/
zone with bad mapId or out-of-range tile coords would otherwise
appear in the picker and silently fail when the user selects it.
Mirrors the existing --info-wom/--info-wob/--info-woc/--info-wcp
inspectors. Reports the tile coord, populated chunks, layer count,
water count, texture/doodad/WMO counts, and computed height range
across all chunks. Useful for triaging zones from the command line
without opening the GUI.
Same defensive guards as scatter: cap per-asset count at 50k to
prevent editor freeze under a high-density biome, and ensure scale
range preconditions (a<b, both positive) so distScale construction
doesn't undermine validity.
Same defensive guards now applied to NPC scatter: reject NaN/inf
center/radius, cap count at 100k (prevents editor freeze on huge
inputs), and ensure minScale < maxScale so uniform_real_distribution
preconditions hold.
A radius of 0 would either throw from the uniform distribution
constructor (uniform_real_distribution requires a < b) or divide
by zero in the sqrt-based area-uniform sampler. Also reject NaN
center, non-positive radius, and absurdly large counts (>10k)
which would freeze the editor on placement.
A creature_template row with modelid1=0 spawns an invisible NPC
in-game. Fall back to 11707 (a generic humanoid) on export so
"Creature"-named placeholder spawns are at least usable; the user
can edit the displayId after.
A WOT JSON could carry tile coords outside 0..63 (would compute
chunk world positions tens of thousands of units off-grid) or NaN
position/rotation values on doodad/WMO placements (would propagate
into rendering matrices and produce invisible geometry).
ADTWriter::writeFloat now coerces non-finite values to 0 before
emitting bits, so a stray NaN in a chunk position, MDDF rotation,
or MODF extent can't leak into the saved ADT and produce invisible
terrain or off-map placements after reload.
WoW's tile grid is 64x64. A tile (200,200) entry would generate an
ADT filename the loader rejects and silently leave the zone with no
terrain. Drop bad entries instead.
A zone named with spaces or punctuation (e.g. "My Zone!") used to
produce a module directory path "mod_wowee_My Zone!/" and conf
keys like "Wowee.My Zone!.Enabled" which AzerothCore's config
parser rejects. The slug is now stripped to [A-Za-z0-9_-], with
spaces/slashes mapped to underscores. SQL VALUES still use the
raw display text via SQLExporter::escape.
The map_dbc, area_table_dbc, and game_tele INSERTs previously
embedded mapName/displayName/manifest.mapName as raw strings — a
zone called "King's Land" or anything containing a single quote
would emit malformed SQL that AzerothCore would reject. Promotes
the existing escapeSql helper to a public SQLExporter::escape and
uses it in all three INSERTs.
A malformed stamp JSON could carry millions of entries (would OOM)
or NaN dx/dy/height (would propagate through brush blends and leave
permanent holes in the heightmap).
A hand-edited project.json could carry a 0 or massively-OOR mapId
(would break DBC indexing) or tile coords outside 0..63 (would
produce garbage ADT filenames). Defaults restore safe values.
A user-edited zone.json could carry NaN/inf or out-of-range volumes
which would silently corrupt terrain elevation or produce silent /
clipping audio. Now baseHeight defaults back to 100 if non-finite,
and music/ambience volumes clamp to [0,1] with NaN fallbacks.
Same float-NaN scrub for WoB save matching the WHM/WOC/WOM saves.
Building boundRadius defaults to 1.0 if non-finite; per-group bounds
zero out non-finite components (would otherwise corrupt the cull
frustum).
Mirrors the WHM and WOC save sanitize. boundRadius defaults to 1.0 if
non-finite (matches load-time default); boundMin/boundMax components
zero out non-finite values. Prevents an in-memory model with a NaN
spike (e.g. mid-edit) from being persisted into the WOM and requiring
load-time cleanup forever after.
In-memory collision can be polluted by addMesh on bad input (e.g. a
WMO with NaN vertex positions). Without this, the save would persist
that NaN into the WOC and the load-time guards would have to clean
it up forever. Now scrubs vertices and bounds at write time, matching
the WHM save sanitize.
Sanitize at write time too, not just on load. A mid-edit NaN spike
(e.g. brush operation that produced NaN before being committed) would
otherwise be persisted into the WHM and require the load-time guard
to clean it up forever after. AlphaSize is also capped at 64KB to
match the loader cap.
Three new sanity checks before stbi_write_png:
- dimensions <=0 or >8K rejected (matches PNG override loader cap)
- data buffer must be >= width * height * 4 bytes (corrupt BLP could
have mismatched dimensions vs data length, and stbi reads off the
end of the buffer otherwise)
Skips with warning rather than crashing the exporter mid-zone.
Two write-side guards mirroring the unpack-side ones:
- Path length truncated to 1KB (matches unpack cap; long paths would
silently wrap u16 and corrupt the pack)
- Files >4GB on disk skipped with a zero-length entry rather than
silently producing a truncated dataSize that overflows uint32
Same fix as WoB save just got. Without truncation a model name or
texture path over 65535 chars would silently get a wrap-around length
and corrupt the file.
Without truncation, names over 65535 chars would silently get a wrap-
around length value and produce a corrupt file (the actual string would
be longer than the saved length, shifting everything after it). Single
shared writeStr lambda replaces five copy-pasted (uint16 length + bytes)
write blocks.
JSON DBC values are mostly small integers, but a malicious file could
stuff 100MB strings into every cell to OOM the stringBlock (which has
no per-string or total cap). Added 4KB per-string + 64MB total caps —
fields exceeding either are zeroed. Float fields also get NaN scrub
(same pattern as the earlier vertex/keyframe guards).
Same defensive check as the WoB doodad path guard. Texture paths from
hostile WOM/WoB are passed to the asset manager; '..' or absolute paths
could probe outside the assets/ tree. Now cleared on detection — slot
survives but loads no texture (renderer falls back to white).
Single shared rejectTraversal lambda in WoB to avoid copy-paste.
Doodad model paths from a WoB are passed to the asset manager via
outModel.doodadNames. The asset manager only reads files, but '..' or
absolute paths from a hostile WoB could probe for files outside the
expected assets/ tree. Now clears the modelPath on traversal — the
doodad slot survives but loads no model.
Texture paths come from M2/WMO files which a malicious zone author
could craft to include '..' or absolute paths. Without this check,
exporting such a zone would write PNGs outside outputDir/textures/
and clobber sibling export files.
stbi_load happily decodes any PNG up to 32K x 32K — at 4 bytes/pixel
that's 4GB which OOMs the editor before the override even returns.
WoW textures top out at 4K; 8K cap leaves headroom for HD upgrades
without enabling abuse. Also widens the wxh multiplication to size_t
to defeat int overflow on 8K x 8K images.
Real DBCs cap at ~250 fields and a few million records (Spell.dbc is
the biggest at ~50K rows). A malicious JSON DBC declaring fieldCount=
1G or recordCount * recordSize > 256MB would OOM the recordData
allocation. Now rejects upfront — JSON DBCs are user-shareable so a
zone export downloaded from a forum should not be able to OOM the
client by including a bad data table.
ADT placement positions and rotations are loaded as raw floats from the
binary chunks. A corrupted MDDF/MODF entry could feed NaN into the
M2/WMO instance transform and crash render. Now position/rotation are
scrubbed for both MDDF doodads and MODF buildings; MODF also scrubs the
extentLower/extentUpper bounding box used for cull tests.
Three guards on quest objective loading from JSON:
- type out of QuestObjectiveType range (0..5) -> defaults to 0 (Kill)
- targetCount of 0 -> 1 (no-op objectives are nonsense)
- targetCount > 1000 -> 1000 (typo guard, biggest legit WoW quest is ~100)
- >10 objectives per quest -> dropped (matches SQL slot capacity, also
bounds per-quest memory)
Path length cap (1KB) — uint16 can hold 64KB but no real zone path
should exceed 256 chars. Path traversal check extended to also catch:
- Windows backslash absolute paths ('\' at start)
- Windows drive-prefixed paths ('C:\...')
A WCP downloaded from a forum and unpacked on Windows would otherwise
have these vectors open.
Three more per-record sanity bounds that the per-group sweep didn't
cover:
- material.texturePath length cap (1KB) — was unbounded
- doodad.modelPath length cap (1KB) — was unbounded
- portal.vertexCount cap (4096) — real portals are 4-12 verts;
>4K is corrupt and would OOM the resize
The WoB top-level header sanity bounds catch obviously-bad totals, but
each group's vc/ic/tc was still unbounded. A corrupted group could
declare 4G vertices and OOM the resize before the next group even
started. Now per-group: vc<=1M, ic<=4M, tc<=1K, name<=1KB.
Per-bone-anim cap of 10K keyframes still let a malicious file allocate
up to 1024 anims × 512 bones × 10K keys = 5.24B keyframes — multi-GB
pre-OOM allocation. Now tracks total across the whole model and stops
allocating after 10M (real models stay well under 100K). When the cap
is hit we still seek past the remaining payload to keep file alignment
intact for whatever follows.
