@vanillaskyai/video 0.5.8 → 0.7.0

package/dist/{chunk-3OU7HIEB.js → chunk-RE4IMWJR.js}

@@ -3,23 +3,23 @@ import {
 } from "./chunk-E7CL7UPB.js";
 import {
   applyVideoEvent,
-  checksumVideo,
   createReplayableStream,
   createVideoState
-} from "./chunk-FZHMQFG3.js";
+} from "./chunk-GAIOHGNR.js";
 import {
   parseVideoPlanPart,
   safePublicDiagnostic
-} from "./chunk-G66Z5CWR.js";
+} from "./chunk-MMUXVA47.js";
 import {
   MAX_RETAINED_INSTRUCTIONS_LENGTH,
   MAX_RETAINED_MEDIA_URLS,
   MAX_RETAINED_MEDIA_URL_LENGTH,
   MAX_RETAINED_SOURCE_LENGTH,
+  checksumVideo,
+  createVideoEventFactory,
   parseVideo
-} from "./chunk-66MNRUCR.js";
+} from "./chunk-2PYO6VAC.js";
 import {
-  VIDEO_PROTOCOL_VERSION,
   VIDEO_SCHEMA_VERSION,
   createVideoRequest
 } from "./chunk-AOWSXU2K.js";
@@ -32,25 +32,6 @@ import {
   resolveVideoBrand
 } from "./chunk-XGRA2MUV.js";
 
-// src/protocol/events.ts
-function createVideoEventFactory(options) {
-  let sequence = options.initialSequence ?? 0;
-  return {
-    create(type, data) {
-      const currentSequence = sequence;
-      sequence += 1;
-      return {
-        protocolVersion: VIDEO_PROTOCOL_VERSION,
-        runId: options.runId,
-        sequence: currentSequence,
-        eventId: `${options.runId}:${currentSequence}`,
-        type,
-        data
-      };
-    }
-  };
-}
-
 // src/server/prompts/user-prompt.ts
 var SUPPLIED_MEDIA_REFERENCE_BASE = "https://vanillasky.invalid/supplied/";
 function suppliedMediaReferences(input) {
@@ -92,10 +73,12 @@ function buildVideoUserPrompt(input, openingDurationSec = 0) {
     "Compose a video response from the structured customer input below.",
     `Knowledge mode: ${input.knowledgeMode ?? "input-only"}.`,
     `Maximum duration: ${input.maxDurationSec ?? 30} seconds, including the supplied opening.`,
-    typeof input.opening === "string" && input.opening.trim() ? `The host has already added the opening scene, which consumes ${openingDurationSec} seconds. Continue after it and do not repeat or rewrite it.` : "Add the first grounded scene as soon as it is complete.",
-    "The first generated body scene must be fully playable without external media. Use a content-fit text, data, comparison, list, or device-free template with no media URL or keyword.",
-    "Never use media, ctaMedia, or reaction as the first generated body template, even with mediaType=gradient. Choose a non-media template first.",
-    "Add that scene before resolving any stock or supplied asset. Resolve media as part of each complete later body scene.",
+    input.opening === false ? "The host owns the opening wait. Add the first grounded scene as soon as it is complete; it may use host-resolved media." : typeof input.opening === "string" && input.opening.trim() ? `The host has already added the opening scene, which consumes ${openingDurationSec} seconds. Continue after it and do not repeat or rewrite it.` : "Add the first grounded scene as soon as it is complete.",
+    ...input.opening === false ? [] : [
+      "The first generated body scene must be fully playable without external media. Use a content-fit text, data, comparison, list, or device-free template with no media URL or keyword.",
+      "Never use media, ctaMedia, or reaction as the first generated body template, even with mediaType=gradient. Choose a non-media template first.",
+      "Add that scene before resolving any stock or supplied asset. Resolve media as part of each complete later body scene."
+    ],
     "Use only claims supported by the factual basis permitted by the trusted system prompt.",
     "Select the most decision-relevant grounded takeaways that fit the duration; represent each selected takeaway once before completing the response.",
     "For a long source, summarize instead of attempting to represent every fact, unless the creative instructions explicitly request complete fact coverage that fits the duration.",
@@ -107,7 +90,7 @@ function buildVideoUserPrompt(input, openingDurationSec = 0) {
     "After the first playable body scene, emit exactly one scene.add with placement closer. Generate a short grounded conclusion that answers the story's so-what; it must not repeat hook language or imply another scene follows.",
     "Use a different suitable template for each body scene when the catalog supports it.",
     "Never add filler to satisfy a count or diversity target.",
-    input.suppliedMedia?.length ? "Select zero or more relevant opaque supplied-media references for visible later scenes. Never invent or transform a reference. Only emit mediaKeyword when the trusted system catalog explicitly exposes it, and only on later scenes. Never invent mediaUrl or mediaPoster." : "No supplied media URL is available. Use asset-free templates unless the trusted system catalog explicitly permits host-resolved media intent. Only emit mediaKeyword when the trusted system catalog explicitly exposes it, and only on later scenes. Never invent mediaUrl or mediaPoster.",
+    input.suppliedMedia?.length ? input.opening === false ? "Select zero or more relevant opaque supplied-media references for visible scenes. Never invent or transform a reference. Only emit mediaKeyword when the trusted system catalog explicitly exposes it. Never invent mediaUrl or mediaPoster." : "Select zero or more relevant opaque supplied-media references for visible later scenes. Never invent or transform a reference. Only emit mediaKeyword when the trusted system catalog explicitly exposes it, and only on later scenes. Never invent mediaUrl or mediaPoster." : input.opening === false ? "No supplied media URL is available. Use asset-free templates unless the trusted system catalog explicitly permits host-resolved media intent, which may begin on the first scene. Only emit mediaKeyword when the trusted system catalog explicitly exposes it. Never invent mediaUrl or mediaPoster." : "No supplied media URL is available. Use asset-free templates unless the trusted system catalog explicitly permits host-resolved media intent. Only emit mediaKeyword when the trusted system catalog explicitly exposes it, and only on later scenes. Never invent mediaUrl or mediaPoster.",
     "",
     "RAW INPUT",
     input.input.trim(),
@@ -225,7 +208,8 @@ function buildInitialComposition(input, audio, snapshotRetention, closerReserveS
         motion: input.style?.motion ?? "normal",
         defaultBackgroundEffect: input.style?.backgroundEffect ?? "static",
         defaultTextArchetype: input.style?.textArchetype ?? "subtle",
-        defaultTransition: "crossfade"
+        defaultTransition: "crossfade",
+        ...input.style?.generatedLook ? { generatedLook: input.style.generatedLook } : {}
       },
       meta
     },
@@ -670,7 +654,7 @@ function createVideo(rawInput, options) {
           position
         }));
       }
-      const systemPrompt = options.systemPrompt ?? (await import("./system-prompt-4I6Z5HK3.js")).createVideoSystemPrompt(input.knowledgeMode);
+      const systemPrompt = options.systemPrompt ?? (await import("./system-prompt-RRXIWDDD.js")).createVideoSystemPrompt(input.knowledgeMode);
       const context = { request, systemPrompt, userPrompt, initialConfig, signal: controller.signal };
       attachGenerationLifecycleSink(context, lifecycle);
       for await (const untrustedPart of options.generate(context)) {

package/dist/{chunk-FVTMYS6U.js → chunk-RGF452LL.js}

@@ -1,6 +1,6 @@
 import {
   BUILTIN_TEMPLATE_CATALOG
-} from "./chunk-3RV4YKB3.js";
+} from "./chunk-JNN3EYVP.js";
 import {
   assertTemplateTransitionMetadata
 } from "./chunk-2E6T633S.js";

package/dist/{chunk-34O5BY6X.js → chunk-RXHW4EP4.js}

@@ -6,7 +6,7 @@ import {
 } from "./chunk-E7CL7UPB.js";
 import {
   parseVideoPlanPart
-} from "./chunk-G66Z5CWR.js";
+} from "./chunk-MMUXVA47.js";
 
 // src/server/model/truncated-json.ts
 function isConfidentlyTruncatedJson(source) {
@@ -48,11 +48,28 @@ function parsePlanLine(line) {
   if (value && typeof value === "object" && !Array.isArray(value)) {
     const object = value;
     const scene = object.scene;
-    if (object.type === "scene.add" && object.timing != null && scene && typeof scene === "object" && !Array.isArray(scene) && !("timing" in scene)) {
-      const { timing, ...planPart } = object;
+    if (object.type === "scene.add" && scene && typeof scene === "object" && !Array.isArray(scene)) {
+      const planPart = { ...object };
+      const repairedScene = { ...scene };
+      let repaired = false;
+      if (object.placement != null && object.placement !== "closer") {
+        delete planPart.placement;
+        repaired = true;
+      }
+      if (object.timing != null && !("timing" in scene)) {
+        repairedScene.timing = object.timing;
+        delete planPart.timing;
+        repaired = true;
+      }
+      if (object.narration != null && !("narration" in scene)) {
+        repairedScene.narration = object.narration;
+        delete planPart.narration;
+        repaired = true;
+      }
+      if (!repaired) return parseVideoPlanPart(value);
       return parseVideoPlanPart({
         ...planPart,
-        scene: { ...scene, timing }
+        scene: repairedScene
       });
     }
   }

package/dist/chunk-SKRGRKHY.js

@@ -0,0 +1,142 @@
+import {
+  checksumVideo,
+  createVideoEventFactory
+} from "./chunk-2PYO6VAC.js";
+import {
+  VIDEO_SCHEMA_VERSION
+} from "./chunk-AOWSXU2K.js";
+import {
+  getReadableSceneDuration
+} from "./chunk-3O7OMMMF.js";
+
+// src/protocol/scene-timeline.ts
+function createQueue() {
+  const buffered = [];
+  const waiting = [];
+  let closed = false;
+  return {
+    push(value) {
+      const next = waiting.shift();
+      if (next) next({ value, done: false });
+      else buffered.push(value);
+    },
+    close() {
+      closed = true;
+      let next = waiting.shift();
+      while (next) {
+        next({ value: void 0, done: true });
+        next = waiting.shift();
+      }
+    },
+    iterable: {
+      [Symbol.asyncIterator]() {
+        return {
+          next() {
+            const buffer = buffered.shift();
+            if (buffer !== void 0) return Promise.resolve({ value: buffer, done: false });
+            if (closed) return Promise.resolve({ value: void 0, done: true });
+            return new Promise((resolve) => waiting.push(resolve));
+          }
+        };
+      }
+    }
+  };
+}
+var runCounter = 0;
+function createSceneTimeline(options) {
+  const queue = createQueue();
+  const runId = options.runId ?? `composed-${runCounter += 1}`;
+  const events = createVideoEventFactory({ runId });
+  const orientation = options.orientation ?? "landscape";
+  const snapshot = {
+    schemaVersion: VIDEO_SCHEMA_VERSION,
+    orientation,
+    scenes: [],
+    style: options.style
+  };
+  let position = 0;
+  let finished = false;
+  const held = [];
+  let openForScenes = !options.awaitAudio;
+  const emit = (event) => {
+    if (finished) return;
+    queue.push(event);
+  };
+  emit(events.create("response.start", {
+    requestId: options.requestId ?? runId,
+    format: { orientation },
+    style: options.style
+  }));
+  if (options.audio) {
+    snapshot.audio = options.audio;
+    emit(events.create("audio.set", { audio: options.audio }));
+  }
+  const release = () => {
+    const pending = held.splice(0, held.length);
+    for (const scene of pending) {
+      snapshot.scenes.push(scene);
+      emit(events.create("scene.add", { scene, position: position++ }));
+    }
+  };
+  return {
+    stream: queue.iterable,
+    add(scene) {
+      if (finished) return;
+      if (!openForScenes) {
+        held.push(scene);
+        return;
+      }
+      snapshot.scenes.push(scene);
+      emit(events.create("scene.add", { scene, position: position++ }));
+    },
+    setAudio(audio) {
+      if (finished || openForScenes) return;
+      if (audio) {
+        snapshot.audio = audio;
+        emit(events.create("audio.set", { audio }));
+      }
+      openForScenes = true;
+      release();
+    },
+    complete() {
+      if (finished) return;
+      if (!openForScenes) {
+        openForScenes = true;
+        release();
+      }
+      emit(events.create("response.complete", {
+        finishReason: "stop",
+        snapshot,
+        checksum: checksumVideo(snapshot)
+      }));
+      finished = true;
+      queue.close();
+    }
+  };
+}
+
+// src/protocol/scene-duration.ts
+var NARRATION_WORDS_PER_SECOND = 2.6;
+var NARRATION_TAIL_SECONDS = 0.6;
+function getSceneDurationBounds(scene, metadata) {
+  return {
+    minimum: Math.max(1e-3, metadata?.minDuration ?? 1),
+    readable: getReadableSceneDuration(scene, metadata)
+  };
+}
+function getSpokenDuration(narration) {
+  const words = narration.trim().split(/\s+/u).filter(Boolean).length;
+  if (words === 0) return 0;
+  return words / NARRATION_WORDS_PER_SECOND + NARRATION_TAIL_SECONDS;
+}
+function getSceneDuration(scene, metadata) {
+  const { readable } = getSceneDurationBounds(scene, metadata);
+  return Math.max(readable, scene.narration ? getSpokenDuration(scene.narration) : 0);
+}
+
+export {
+  createSceneTimeline,
+  getSceneDurationBounds,
+  getSpokenDuration,
+  getSceneDuration
+};

package/dist/{chunk-TCEHK2BW.js → chunk-YJJC4N4D.js}

@@ -572,6 +572,7 @@ var TemplateText = ({
   } : baseTypo;
   const presetWeight = tt ? Math.min(900, Math.max(100, fontWeight + tt.weightDelta)) : fontWeight;
   const presetSize = tt ? fontSize * tt.sizeScale : fontSize;
+  const overMediaColor = overMedia && isLikelyDark(color) ? "#FFFFFF" : color;
   const textShadow = overMedia ? MEDIA_TEXT_SHADOW : dropShadowFor(color);
   const containerStyle = {
     position: "absolute",
@@ -580,7 +581,7 @@ var TemplateText = ({
     display: "flex",
     justifyContent: "center",
     padding: `0 ${safeZone.right}px 0 ${safeZone.left}px`,
-    color,
+    color: overMediaColor,
     fontFamily: font,
     fontWeight: presetWeight,
     fontSize: presetSize,