@semiont/jobs 0.5.23 → 0.5.25

package/dist/index.d.ts

@@ -1,4 +1,4 @@
-import { JobId, UserId, ResourceId, EntityType, AnnotationId, GatheredContext, SupportedMediaType, TagSchema, Logger, EventBus, components } from '@semiont/core';
+import { JobId, UserId, ResourceId, EntityType, AnnotationId, GatheredContext, SupportedMediaType, TagSchema, Logger, EventBus, components, Annotation } from '@semiont/core';
 import { SemiontProject } from '@semiont/core/node';
 import { InferenceClient } from '@semiont/inference';
 
@@ -524,17 +524,6 @@ interface GenerationCitation {
     exact: string;
 }
 
-/**
- * Job Processors
- *
- * Pure functions that take content + inference client + params,
- * report progress via callback, and return annotations + results.
- *
- * No EventBus, no JobQueue, no side effects except calling inference.
- * Driven by the remote worker process (worker-process.ts), which claims
- * jobs over SSE and dispatches by jobType to these functions.
- */
-
 /** A detected span — offsets into the extracted `.text`, plus optional context. */
 type SpanMatch = {
     exact: string;
@@ -550,7 +539,7 @@ type SpanMatch = {
  * motivation-specific body. This is the single axis that varies by media type,
  * so the detection processors themselves stay media-agnostic.
  */
-type BuildAnnotation = (motivation: string, match: SpanMatch, body?: Record<string, unknown> | Record<string, unknown>[]) => Record<string, unknown>;
+type BuildAnnotation = (motivation: Motivation, match: SpanMatch, body?: Annotation['body']) => Annotation;
 /**
  * Progress callback. The three positional args satisfy the minimum
  * `JobProgress` required fields (`percentage`, `message`, `stage`).
@@ -560,8 +549,11 @@ type BuildAnnotation = (motivation: string, match: SpanMatch, body?: Record<stri
  */
 type OnProgress = (percentage: number, message: string, stage: string, extra?: Partial<JobProgress>) => void;
 type JobProgress = components['schemas']['JobProgress'];
+/** The five W3C motivations this system mints — a closed vocabulary, so it is
+ *  typed as one rather than as `string`. */
+type Motivation = Annotation['motivation'];
 interface ProcessorResult<R> {
-    annotations: Record<string, unknown>[];
+    annotations: Annotation[];
     result: R;
 }
 declare function processHighlightJob(content: string, inferenceClient: InferenceClient, params: HighlightDetectionParams, buildAnnotation: BuildAnnotation, onProgress: OnProgress): Promise<ProcessorResult<HighlightDetectionResult>>;
@@ -570,7 +562,7 @@ declare function processAssessmentJob(content: string, inferenceClient: Inferenc
 declare function processReferenceJob(content: string, inferenceClient: InferenceClient, params: DetectionParams, buildAnnotation: BuildAnnotation, onProgress: OnProgress, logger: Logger): Promise<ProcessorResult<DetectionResult>>;
 declare function processTagJob(content: string, inferenceClient: InferenceClient, params: TagDetectionParams, buildAnnotation: BuildAnnotation, onProgress: OnProgress): Promise<ProcessorResult<TagDetectionResult>>;
 declare function processGenerationJob(inferenceClient: InferenceClient, params: GenerationParams, onProgress: OnProgress, logger: Logger): Promise<{
-    content: string;
+    content: Uint8Array;
     title: string;
     format: SupportedMediaType;
     citations: GenerationCitation[];
@@ -652,7 +644,7 @@ declare class AnnotationDetection {
      * (source-resource locale). See `types.ts` "Locale conventions" for the
      * full discussion.
      */
-    static detectComments(content: string, client: InferenceClient, instructions?: string, tone?: string, density?: number, language?: string, sourceLanguage?: string): Promise<CommentMatch[]>;
+    static detectComments(content: string, client: InferenceClient, instructions?: string, tone?: string, density?: number, language?: string, sourceLanguage?: string, onChunk?: (completedChunks: number, totalChunks: number) => void): Promise<CommentMatch[]>;
     /**
      * Detect highlights in content.
      *
@@ -660,7 +652,7 @@ declare class AnnotationDetection {
      * applies, used in the prompt so the LLM analyzes non-English source
      * correctly.
      */
-    static detectHighlights(content: string, client: InferenceClient, instructions?: string, density?: number, sourceLanguage?: string): Promise<HighlightMatch[]>;
+    static detectHighlights(content: string, client: InferenceClient, instructions?: string, density?: number, sourceLanguage?: string, onChunk?: (completedChunks: number, totalChunks: number) => void): Promise<HighlightMatch[]>;
     /**
      * Detect assessments in content.
      *
@@ -668,7 +660,7 @@ declare class AnnotationDetection {
      * (annotation body locale). `sourceLanguage` is the locale of the content
      * being analyzed (source-resource locale).
      */
-    static detectAssessments(content: string, client: InferenceClient, instructions?: string, tone?: string, density?: number, language?: string, sourceLanguage?: string): Promise<AssessmentMatch[]>;
+    static detectAssessments(content: string, client: InferenceClient, instructions?: string, tone?: string, density?: number, language?: string, sourceLanguage?: string, onChunk?: (completedChunks: number, totalChunks: number) => void): Promise<AssessmentMatch[]>;
     /**
      * Detect tags in content for a specific category.
      *
@@ -681,7 +673,7 @@ declare class AnnotationDetection {
      * identifiers, not LLM-generated text — so it's consumed at the body-stamp
      * site, not here.
      */
-    static detectTags(content: string, client: InferenceClient, schema: TagSchema, category: string, sourceLanguage?: string): Promise<TagMatch[]>;
+    static detectTags(content: string, client: InferenceClient, schema: TagSchema, category: string, sourceLanguage?: string, onChunk?: (completedChunks: number, totalChunks: number) => void): Promise<TagMatch[]>;
 }
 
 /**
@@ -701,7 +693,10 @@ declare class AnnotationDetection {
  * generate German content from an English source resource. See
  * `types.ts` "Locale conventions" for the full discussion.
  */
-declare function generateResourceFromTopic(topic: string, entityTypes: string[], client: InferenceClient, logger: Logger, userPrompt?: string, locale?: string, context?: GatheredContext, temperature?: number, maxTokens?: number, sourceLanguage?: string, outputMediaType?: SupportedMediaType, task?: string, structure?: string, cite?: boolean): Promise<{
+declare function generateResourceFromTopic(topic: string, entityTypes: string[], client: InferenceClient, logger: Logger, userPrompt?: string, locale?: string, context?: GatheredContext, temperature?: number, maxTokens?: number, sourceLanguage?: string, outputMediaType?: SupportedMediaType, task?: string, structure?: string, cite?: boolean, repair?: {
+    source: string;
+    error: string;
+}): Promise<{
     title: string;
     content: string;
 }>;