v0.4.24: sync anchor config in vanilla BE update packet

Bug: /videoPlace (and any path that places an anchor) reported success
on the server but the panel was invisible on the client — no quad, no
video. Walking around did not help.

Root cause: VideoAnchorBlockEntity did not override getUpdateTag() or
getUpdatePacket(). The base implementations return an empty CompoundTag
and null respectively, so url/facing/width/height were never carried in
vanilla BE sync. There is also a packet-ordering race between
level.setBlock() (queues a deferred chunk broadcast) and
ServerPlayNetworking.send(SyncAnchorPayload) (writes immediately): if
the payload arrives first, the client drops it because the BE does not
exist yet, then the chunk packet creates the BE with defaults
(url="") and the renderer silently no-ops.

Fix: override getUpdateTag(HolderLookup.Provider) → toNbt(), and
getUpdatePacket() → ClientboundBlockEntityDataPacket.create(this). NBT
key names already line up between toNbt() and loadAdditional(), so
vanilla wraps the CompoundTag in TagValueInput and existing load logic
reads it. Also fixes the 'walk far away, come back' case — that path
has no SyncAnchorPayload, just vanilla chunk re-sync.

gradle.properties

@@ -5,7 +5,7 @@ org.gradle.configuration-cache=false
 
 # Mod
 mod_id=video_player
-mod_version=0.4.20
+mod_version=0.4.24
 maven_group=com.ejclaw.videoplayer
 archives_base_name=video_player
 

src/main/java/com/ejclaw/videoplayer/VideoPlayerClient.java

@@ -20,6 +20,7 @@ import net.fabricmc.fabric.api.event.player.AttackBlockCallback;
 import net.minecraft.client.Minecraft;
 import net.minecraft.client.player.LocalPlayer;
 import net.minecraft.core.BlockPos;
+import net.minecraft.sounds.SoundSource;
 import net.minecraft.world.InteractionResult;
 import net.minecraft.world.phys.Vec3;
 
@@ -89,17 +90,26 @@ public class VideoPlayerClient implements ClientModInitializer {
      * Distance is measured from the player's eye to the <em>panel center</em>, not the anchor
      * block corner — for a 4×4 panel the corner is ~2 blocks off from where the screen visually
      * sits, which made the audio feel like it was off to the side.
+     *
+     * <p>Gain is also gated by the Minecraft sound options so the in-game sliders work as
+     * expected: vanilla {@code SoundEngine.calculateVolume} multiplies by master × category, so
+     * we do the same with {@link SoundSource#PLAYERS} as the category. Result: dragging the
+     * "Players" slider in Options → Music & Sounds attenuates video audio just like other
+     * player sounds, and "Master" still gates everything.
      */
     private static void updateDistanceGains(Minecraft client) {
         LocalPlayer p = client.player;
         if (p == null || client.level == null) return;
         Vec3 eye = p.getEyePosition();
+        float masterVol = client.options.getSoundSourceVolume(SoundSource.MASTER);
+        float playersVol = client.options.getSoundSourceVolume(SoundSource.PLAYERS);
+        float categoryScale = masterVol * playersVol;
         for (BlockPos pos : VideoPlayback.activePositions()) {
             if (!(client.level.getBlockEntity(pos) instanceof VideoAnchorBlockEntity be)) continue;
             Vec3 center = be.panelCenter();
             double d = center.distanceTo(eye);
             float attenuation = (float) Math.max(0.0, Math.min(1.0, 1.0 - d / 16.0));
-            float gain = be.isMuted() ? 0F : be.getVolume() * attenuation;
+            float gain = be.isMuted() ? 0F : be.getVolume() * attenuation * categoryScale;
             VideoPlayback.setGain(pos, gain);
         }
     }

src/main/java/com/ejclaw/videoplayer/block/VideoAnchorBlockEntity.java

@@ -3,7 +3,11 @@ package com.ejclaw.videoplayer.block;
 import com.ejclaw.videoplayer.registry.VideoPlayerBlockEntities;
 import net.minecraft.core.BlockPos;
 import net.minecraft.core.Direction;
+import net.minecraft.core.HolderLookup;
 import net.minecraft.nbt.CompoundTag;
+import net.minecraft.network.protocol.Packet;
+import net.minecraft.network.protocol.game.ClientGamePacketListener;
+import net.minecraft.network.protocol.game.ClientboundBlockEntityDataPacket;
 import net.minecraft.world.level.block.entity.BlockEntity;
 import net.minecraft.world.level.block.state.BlockState;
 import net.minecraft.world.level.storage.ValueInput;
@@ -123,6 +127,35 @@ public class VideoAnchorBlockEntity extends BlockEntity {
         out.putBoolean("autoplay", autoplay);
     }
 
+    /**
+     * Vanilla chunk-load BE sync. The base implementation returns an empty tag, which means
+     * when a client first sees this BE (chunk loads or player walks into range) it gets default
+     * values — url="" in particular makes the renderer no-op and the panel appears invisible.
+     *
+     * <p>Returning {@link #toNbt()} here carries the custom fields in the vanilla packet, so
+     * we don't depend on the {@code SyncAnchorPayload} arriving before the chunk's block-update
+     * packet (there's a race: {@code level.setBlock} queues a deferred chunk broadcast while
+     * {@code ServerPlayNetworking.send} writes immediately; if the payload wins, the client
+     * drops it because the BE doesn't exist yet, then the chunk packet creates the BE with
+     * defaults). It also fixes "player walks far away and comes back" — that path has no
+     * SyncAnchorPayload at all, just vanilla chunk re-sync.
+     */
+    @Override
+    public CompoundTag getUpdateTag(HolderLookup.Provider provider) {
+        return toNbt();
+    }
+
+    /**
+     * Triggers a {@link ClientboundBlockEntityDataPacket} whenever the chunk tracker decides
+     * this BE needs to push an update. Default implementation returns {@code null} (no packet
+     * sent on BE change). Combined with {@link #getUpdateTag} above, every BE-state change a
+     * client sees carries the full config.
+     */
+    @Override
+    public Packet<ClientGamePacketListener> getUpdatePacket() {
+        return ClientboundBlockEntityDataPacket.create(this);
+    }
+
     @Override
     protected void loadAdditional(ValueInput in) {
         super.loadAdditional(in);

src/main/java/com/ejclaw/videoplayer/client/playback/VideoCache.java

@@ -233,6 +233,33 @@ public final class VideoCache {
 
     // -- internals -----------------------------------------------------------------------
 
+    /**
+     * Publish {@code path} into {@link #READY} for {@code url}, but only if the cache
+     * epoch hasn't moved since {@code startEpoch} was captured. Wraps the
+     * pre-check / put / post-check / rollback dance so every publish site uses the
+     * same race-free pattern instead of re-implementing it.
+     *
+     * <p>Race coverage: if {@link #clearAll()} runs at any point between the
+     * pre-check and the post-check, the post-check sees the epoch bump and we
+     * roll back — {@link Map#remove(Object, Object)} is a compareAndRemove so we
+     * don't clobber a legitimate concurrent put under the same key, and we delete
+     * the on-disk file because clearAll's directory scan may have missed it (the
+     * file may not have existed yet when clearAll ran).
+     *
+     * @return {@code true} if the entry is now published; {@code false} if the
+     *         caller's work was cancelled and {@code path} has been cleaned up.
+     */
+    private static boolean publishIfNotCancelled(String url, Path path, long startEpoch) {
+        if (CACHE_EPOCH.get() != startEpoch) return false;
+        READY.put(url, path);
+        if (CACHE_EPOCH.get() != startEpoch) {
+            READY.remove(url, path);
+            try { Files.deleteIfExists(path); } catch (Throwable ignored) {}
+            return false;
+        }
+        return true;
+    }
+
     private static void download(String url, long startEpoch) {
         Path cacheDir = cacheDir();
         Path partPath = null;
@@ -259,18 +286,7 @@ public final class VideoCache {
             // Resume-friendly: if the file's already on disk from an earlier session, just
             // index it without re-downloading.
             if (Files.exists(finalPath) && Files.size(finalPath) > 0) {
-                // Pre-publish check — bail if clearAll has run since submit.
+                if (!publishIfNotCancelled(url, finalPath, startEpoch)) {
-                if (CACHE_EPOCH.get() != startEpoch) return;
-                READY.put(url, finalPath);
-                // Post-publish re-check: same window as the post-move check on the download
-                // path. If clearAll landed between the epoch read above and the READY.put,
-                // it would have wiped READY + the file, and our put() resurrected a stale
-                // entry pointing at a now-deleted path. Roll it back: remove our entry
-                // (compareAndRemove via remove(key, value) to avoid clobbering a concurrent
-                // legitimate re-put) and best-effort delete the file.
-                if (CACHE_EPOCH.get() != startEpoch) {
-                    READY.remove(url, finalPath);
-                    try { Files.deleteIfExists(finalPath); } catch (Throwable ignored) {}
                     VideoPlayerMod.LOG.info("[{}] preload: reindex cancelled (clearAll ran) — {}",
                         VideoPlayerMod.MOD_ID, url);
                     return;
@@ -370,21 +386,16 @@ public final class VideoCache {
             Files.move(partPath, finalPath, StandardCopyOption.REPLACE_EXISTING,
                 StandardCopyOption.ATOMIC_MOVE);
 
-            // Post-move cancellation check. clearAll() may have run between the pre-move
+            // Race-safe publish: this covers the window between Files.move and READY.put
-            // check above and the move itself — in that case clearAll's directory scan
+            // (clearAll deletes the freshly-moved file then we resurrect it in READY) AND
-            // either missed our file (it didn't exist yet) or saw the .part and skipped /
+            // the window between READY.put and the "완료" notification (clearAll wipes
-            // failed to delete it. Either way, finalPath now exists on disk but the
+            // READY then we re-publish + emit stale completion chat). publishIfNotCancelled
-            // user-visible cache state is "cleared", so we must delete the file and skip
+            // rolls back both READY and the on-disk file if the epoch moved at any time.
-            // both the READY.put and the "완료" chat. Without this check, a clear right at
+            if (!publishIfNotCancelled(url, finalPath, startEpoch)) {
-            // this window leaves a resurrected file in READY and a stale "완료" message.
+                VideoPlayerMod.LOG.info("[{}] preload: cancelled at publish (clearAll ran) — {}",
-            if (CACHE_EPOCH.get() != startEpoch) {
-                VideoPlayerMod.LOG.info("[{}] preload: cancelled after move (clearAll ran) — {}",
                     VideoPlayerMod.MOD_ID, url);
-                try { Files.deleteIfExists(finalPath); } catch (Throwable ignored) {}
                 return;
             }
-
-            READY.put(url, finalPath);
             VideoPlayerMod.LOG.info("[{}] preload: cached {} ({} bytes) -> {}",
                 VideoPlayerMod.MOD_ID, url, total, finalPath.getFileName());
             long mb = Math.max(1, total / (1024 * 1024));

src/main/java/com/ejclaw/videoplayer/client/render/VideoAnchorRenderer.java

@@ -31,8 +31,19 @@ import org.joml.Matrix4f;
 @Environment(EnvType.CLIENT)
 public class VideoAnchorRenderer implements BlockEntityRenderer<VideoAnchorBlockEntity, VideoAnchorRenderer.State> {
 
-    /** Tiny outward offset so the quad doesn't z-fight with the wall. */
+    /**
-    private static final float SURFACE_EPSILON = 0.001F;
+     * Outward offset so the quad doesn't z-fight with the wall it sits on.
+     *
+     * <p>Depth-buffer precision drops with the square of view distance: with near=0.05 and
+     * 24-bit depth, the smallest resolvable step at 30 blocks is ~1mm, and ~12mm at 100
+     * blocks. The old 0.001 (1mm) offset was right at the precision boundary at ~30 blocks
+     * — which is exactly the distance users started seeing the wall texture flicker through
+     * the video panel once {@code render_distance_blocks} was raised past the default.
+     *
+     * <p>2cm gives ~20× margin at 30 blocks and is still visually unnoticeable up close
+     * (~3% of a block thickness).
+     */
+    private static final float SURFACE_EPSILON = 0.02F;
 
     public VideoAnchorRenderer(BlockEntityRendererProvider.Context ctx) {
         // no-op
@@ -75,7 +86,12 @@ public class VideoAnchorRenderer implements BlockEntityRenderer<VideoAnchorBlock
         pose.translate(-0.5F, -0.5F, -0.5F + SURFACE_EPSILON);
 
         final Matrix4f mat = new Matrix4f(pose.last().pose());
-        RenderType rt = RenderTypes.entityCutout(tex);
+        // entitySolid (not entityCutout): video frames come from swscale → AV_PIX_FMT_RGBA with
+        // alpha hard-set to 255, so there is no alpha-tested cutout. Cutout's alpha-discard step
+        // adds nothing here and makes distant sampling unstable — without mipmaps on a dynamic
+        // texture, neighbouring texels can shimmer above/below the discard threshold at sub-pixel
+        // sampling rates, contributing to the flicker users see once render distance is raised.
+        RenderType rt = RenderTypes.entitySolid(tex);
         collector.submitCustomGeometry(pose, rt, (poseUnused, vc) -> {
             // Single-sided: the back of the anchor is by design pressed against the wall the
             // player clicked, so a back face is pure GPU waste. Halves the fragment shader work