@lde/pipeline 0.6.25 → 0.6.26

package/dist/sparql/executor.d.ts

@@ -45,12 +45,16 @@ export interface SparqlConstructExecutorOptions {
     fetcher?: SparqlEndpointFetcher;
 }
 /**
- * A streaming SPARQL CONSTRUCT executor that parses the query once (in the
- * constructor) and operates on the AST for graph and VALUES injection.
+ * A streaming SPARQL CONSTRUCT executor.
+ *
+ * Queries **without** `#subjectFilter#` are parsed once in the constructor
+ * (fast path). Queries that contain the template are stored as raw strings
+ * and parsed at {@link execute} time after substitution.
  *
  * Template substitution (applied in order):
- * 1. `FROM <graph>` — set via `withDefaultGraph` if the distribution has a named graph
- * 2. `?dataset` — replaced with the dataset IRI (string substitution on the serialised query)
+ * 1. `#subjectFilter#` — replaced with `distribution.subjectFilter` (deferred to execute)
+ * 2. `FROM <graph>` — set via `withDefaultGraph` if the distribution has a named graph
+ * 3. `?dataset` — replaced with the dataset IRI (string substitution on the serialised query)
  *
  * @example
  * ```typescript
@@ -68,7 +72,8 @@ export interface SparqlConstructExecutorOptions {
  * ```
  */
 export declare class SparqlConstructExecutor implements Executor {
-    private readonly query;
+    private readonly rawQuery;
+    private readonly preParsed?;
     private readonly fetcher;
     private readonly generator;
     constructor(options: SparqlConstructExecutorOptions);
@@ -89,14 +94,6 @@ export declare class SparqlConstructExecutor implements Executor {
      */
     static fromFile(filename: string, options?: Omit<SparqlConstructExecutorOptions, 'query'>): Promise<SparqlConstructExecutor>;
 }
-/**
- * Substitute template variables in a SPARQL query.
- *
- * - `#subjectFilter#` — replaced with the distribution's subject filter
- * - `#namedGraph#` — replaced with `FROM <graph>` clause if the distribution has a named graph
- * - `?dataset` — replaced with the dataset IRI
- */
-export declare function substituteQueryTemplates(query: string, distribution: Distribution | null, dataset: Dataset): string;
 /**
  * Read a SPARQL query from a file.
  */

package/dist/sparql/executor.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"executor.d.ts","sourceRoot":"","sources":["../../src/sparql/executor.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,MAAM,cAAc,CAAC;AACrD,OAAO,EAAE,qBAAqB,EAAE,MAAM,uBAAuB,CAAC;AAC9D,OAAO,KAAK,EAAE,SAAS,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,cAAc,CAAC;AAC5D,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,aAAa,CAAC;AAO5C;;GAEG;AACH,qBAAa,YAAY;aACK,OAAO,EAAE,MAAM;gBAAf,OAAO,EAAE,MAAM;CAC5C;AAED,qEAAqE;AACrE,MAAM,MAAM,gBAAgB,GAAG,MAAM,CAAC,MAAM,EAAE,SAAS,CAAC,CAAC;AAEzD,MAAM,WAAW,cAAc;IAC7B;;;OAGG;IACH,QAAQ,CAAC,EAAE,gBAAgB,EAAE,CAAC;CAC/B;AAED,MAAM,WAAW,QAAQ;IACvB,OAAO,CACL,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,aAAa,CAAC,IAAI,CAAC,GAAG,YAAY,CAAC,CAAC;CAChD;AAED;;;GAGG;AACH,MAAM,MAAM,UAAU,GAAG,QAAQ,GAAG,MAAM,CAAC,IAAI,CAAC,CAAC;AAEjD;;GAEG;AACH,MAAM,WAAW,8BAA8B;IAC7C;;OAEG;IACH,KAAK,EAAE,MAAM,CAAC;IAEd;;;OAGG;IACH,OAAO,CAAC,EAAE,MAAM,CAAC;IAEjB;;OAEG;IACH,OAAO,CAAC,EAAE,qBAAqB,CAAC;CACjC;AAED;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,qBAAa,uBAAwB,YAAW,QAAQ;IACtD,OAAO,CAAC,QAAQ,CAAC,KAAK,CAAiB;IACvC,OAAO,CAAC,QAAQ,CAAC,OAAO,CAAwB;IAChD,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAmB;gBAEjC,OAAO,EAAE,8BAA8B;IAcnD;;;;;;;OAOG;IACG,OAAO,CACX,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,UAAU,CAAC;IAoBtB;;;;;OAKG;WACiB,QAAQ,CAC1B,QAAQ,EAAE,MAAM,EAChB,OAAO,CAAC,EAAE,IAAI,CAAC,8BAA8B,EAAE,OAAO,CAAC,GACtD,OAAO,CAAC,uBAAuB,CAAC;CAIpC;AAED;;;;;;GAMG;AACH,wBAAgB,wBAAwB,CACtC,KAAK,EAAE,MAAM,EACb,YAAY,EAAE,YAAY,GAAG,IAAI,EACjC,OAAO,EAAE,OAAO,GACf,MAAM,CAWR;AAED;;GAEG;AACH,wBAAsB,aAAa,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAErE"}
+{"version":3,"file":"executor.d.ts","sourceRoot":"","sources":["../../src/sparql/executor.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,MAAM,cAAc,CAAC;AACrD,OAAO,EAAE,qBAAqB,EAAE,MAAM,uBAAuB,CAAC;AAC9D,OAAO,KAAK,EAAE,SAAS,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,cAAc,CAAC;AAC5D,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,aAAa,CAAC;AAO5C;;GAEG;AACH,qBAAa,YAAY;aACK,OAAO,EAAE,MAAM;gBAAf,OAAO,EAAE,MAAM;CAC5C;AAED,qEAAqE;AACrE,MAAM,MAAM,gBAAgB,GAAG,MAAM,CAAC,MAAM,EAAE,SAAS,CAAC,CAAC;AAEzD,MAAM,WAAW,cAAc;IAC7B;;;OAGG;IACH,QAAQ,CAAC,EAAE,gBAAgB,EAAE,CAAC;CAC/B;AAED,MAAM,WAAW,QAAQ;IACvB,OAAO,CACL,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,aAAa,CAAC,IAAI,CAAC,GAAG,YAAY,CAAC,CAAC;CAChD;AAED;;;GAGG;AACH,MAAM,MAAM,UAAU,GAAG,QAAQ,GAAG,MAAM,CAAC,IAAI,CAAC,CAAC;AAEjD;;GAEG;AACH,MAAM,WAAW,8BAA8B;IAC7C;;OAEG;IACH,KAAK,EAAE,MAAM,CAAC;IAEd;;;OAGG;IACH,OAAO,CAAC,EAAE,MAAM,CAAC;IAEjB;;OAEG;IACH,OAAO,CAAC,EAAE,qBAAqB,CAAC;CACjC;AAED;;;;;;;;;;;;;;;;;;;;;;;;;;GA0BG;AACH,qBAAa,uBAAwB,YAAW,QAAQ;IACtD,OAAO,CAAC,QAAQ,CAAC,QAAQ,CAAS;IAClC,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAC,CAAiB;IAC5C,OAAO,CAAC,QAAQ,CAAC,OAAO,CAAwB;IAChD,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAmB;gBAEjC,OAAO,EAAE,8BAA8B;IAkBnD;;;;;;;OAOG;IACG,OAAO,CACX,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,OAAO,CAAC,EAAE,cAAc,GACvB,OAAO,CAAC,UAAU,CAAC;IAiCtB;;;;;OAKG;WACiB,QAAQ,CAC1B,QAAQ,EAAE,MAAM,EAChB,OAAO,CAAC,EAAE,IAAI,CAAC,8BAA8B,EAAE,OAAO,CAAC,GACtD,OAAO,CAAC,uBAAuB,CAAC;CAIpC;AAED;;GAEG;AACH,wBAAsB,aAAa,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAErE"}

package/dist/sparql/executor.js

@@ -14,12 +14,16 @@ export class NotSupported {
     }
 }
 /**
- * A streaming SPARQL CONSTRUCT executor that parses the query once (in the
- * constructor) and operates on the AST for graph and VALUES injection.
+ * A streaming SPARQL CONSTRUCT executor.
+ *
+ * Queries **without** `#subjectFilter#` are parsed once in the constructor
+ * (fast path). Queries that contain the template are stored as raw strings
+ * and parsed at {@link execute} time after substitution.
  *
  * Template substitution (applied in order):
- * 1. `FROM <graph>` — set via `withDefaultGraph` if the distribution has a named graph
- * 2. `?dataset` — replaced with the dataset IRI (string substitution on the serialised query)
+ * 1. `#subjectFilter#` — replaced with `distribution.subjectFilter` (deferred to execute)
+ * 2. `FROM <graph>` — set via `withDefaultGraph` if the distribution has a named graph
+ * 3. `?dataset` — replaced with the dataset IRI (string substitution on the serialised query)
  *
  * @example
  * ```typescript
@@ -37,16 +41,19 @@ export class NotSupported {
  * ```
  */
 export class SparqlConstructExecutor {
-    query;
+    rawQuery;
+    preParsed;
     fetcher;
     generator = new Generator();
     constructor(options) {
-        const parser = new Parser();
-        const parsed = parser.parse(options.query);
-        if (parsed.type !== 'query' || parsed.queryType !== 'CONSTRUCT') {
-            throw new Error('Query must be a CONSTRUCT query');
+        this.rawQuery = options.query;
+        if (!options.query.includes('#subjectFilter#')) {
+            const parsed = new Parser().parse(options.query);
+            if (parsed.type !== 'query' || parsed.queryType !== 'CONSTRUCT') {
+                throw new Error('Query must be a CONSTRUCT query');
+            }
+            this.preParsed = parsed;
         }
-        this.query = parsed;
         this.fetcher =
             options.fetcher ??
                 new SparqlEndpointFetcher({
@@ -63,7 +70,18 @@ export class SparqlConstructExecutor {
      */
     async execute(dataset, distribution, options) {
         const endpoint = distribution.accessUrl;
-        let ast = structuredClone(this.query);
+        let ast;
+        if (this.preParsed) {
+            ast = structuredClone(this.preParsed);
+        }
+        else {
+            const substituted = this.rawQuery.replace('#subjectFilter#', distribution.subjectFilter ?? '');
+            const parsed = new Parser().parse(substituted);
+            if (parsed.type !== 'query' || parsed.queryType !== 'CONSTRUCT') {
+                throw new Error('Query must be a CONSTRUCT query');
+            }
+            ast = parsed;
+        }
         if (distribution.namedGraph) {
             withDefaultGraph(ast, distribution.namedGraph);
         }
@@ -86,23 +104,6 @@ export class SparqlConstructExecutor {
         return new SparqlConstructExecutor({ ...options, query });
     }
 }
-/**
- * Substitute template variables in a SPARQL query.
- *
- * - `#subjectFilter#` — replaced with the distribution's subject filter
- * - `#namedGraph#` — replaced with `FROM <graph>` clause if the distribution has a named graph
- * - `?dataset` — replaced with the dataset IRI
- */
-export function substituteQueryTemplates(query, distribution, dataset) {
-    const subjectFilter = distribution?.subjectFilter ?? '';
-    const namedGraph = distribution?.namedGraph
-        ? `FROM <${distribution.namedGraph}>`
-        : '';
-    return query
-        .replace('#subjectFilter#', subjectFilter)
-        .replaceAll('?dataset', `<${dataset.iri}>`)
-        .replace('#namedGraph#', namedGraph);
-}
 /**
  * Read a SPARQL query from a file.
  */

package/dist/sparql/index.d.ts

@@ -1,4 +1,4 @@
-export { SparqlConstructExecutor, substituteQueryTemplates, NotSupported, readQueryFile, type ExecuteOptions, type Executor, type SparqlConstructExecutorOptions, type QuadStream, type VariableBindings, } from './executor.js';
+export { SparqlConstructExecutor, NotSupported, readQueryFile, type ExecuteOptions, type Executor, type SparqlConstructExecutorOptions, type QuadStream, type VariableBindings, } from './executor.js';
 export { SparqlSelector, type SparqlSelectorOptions } from './selector.js';
 export { injectValues } from './values.js';
 export { withDefaultGraph } from './graph.js';

package/dist/sparql/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/sparql/index.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,uBAAuB,EACvB,wBAAwB,EACxB,YAAY,EACZ,aAAa,EACb,KAAK,cAAc,EACnB,KAAK,QAAQ,EACb,KAAK,8BAA8B,EACnC,KAAK,UAAU,EACf,KAAK,gBAAgB,GACtB,MAAM,eAAe,CAAC;AACvB,OAAO,EAAE,cAAc,EAAE,KAAK,qBAAqB,EAAE,MAAM,eAAe,CAAC;AAE3E,OAAO,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAE3C,OAAO,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/sparql/index.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,uBAAuB,EACvB,YAAY,EACZ,aAAa,EACb,KAAK,cAAc,EACnB,KAAK,QAAQ,EACb,KAAK,8BAA8B,EACnC,KAAK,UAAU,EACf,KAAK,gBAAgB,GACtB,MAAM,eAAe,CAAC;AACvB,OAAO,EAAE,cAAc,EAAE,KAAK,qBAAqB,EAAE,MAAM,eAAe,CAAC;AAE3E,OAAO,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAE3C,OAAO,EAAE,gBAAgB,EAAE,MAAM,YAAY,CAAC"}

package/dist/sparql/index.js

@@ -1,4 +1,4 @@
-export { SparqlConstructExecutor, substituteQueryTemplates, NotSupported, readQueryFile, } from './executor.js';
+export { SparqlConstructExecutor, NotSupported, readQueryFile, } from './executor.js';
 export { SparqlSelector } from './selector.js';
 export { injectValues } from './values.js';
 export { withDefaultGraph } from './graph.js';

package/dist/stage.d.ts

@@ -2,10 +2,12 @@ import { Dataset, Distribution } from '@lde/dataset';
 import type { Executor, VariableBindings } from './sparql/executor.js';
 import { NotSupported } from './sparql/executor.js';
 import type { Writer } from './writer/writer.js';
+/** A selector, or a factory that receives the runtime distribution. */
+export type StageSelectorInput = StageSelector | ((distribution: Distribution) => StageSelector);
 export interface StageOptions {
     name: string;
     executors: Executor | Executor[];
-    selector?: StageSelector;
+    selector?: StageSelectorInput;
     /** Maximum number of bindings per executor call. @default 10 */
     batchSize?: number;
     /** Maximum concurrent in-flight executor batches. @default 10 */
@@ -20,7 +22,7 @@ export declare class Stage {
     readonly name: string;
     readonly stages: readonly Stage[];
     private readonly executors;
-    private readonly selector?;
+    private readonly selectorInput?;
     private readonly batchSize;
     private readonly maxConcurrency;
     constructor(options: StageOptions);

package/dist/stage.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"stage.d.ts","sourceRoot":"","sources":["../src/stage.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,MAAM,cAAc,CAAC;AAErD,OAAO,KAAK,EAAE,QAAQ,EAAE,gBAAgB,EAAE,MAAM,sBAAsB,CAAC;AACvE,OAAO,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAEpD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAC;AAGjD,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,MAAM,CAAC;IACb,SAAS,EAAE,QAAQ,GAAG,QAAQ,EAAE,CAAC;IACjC,QAAQ,CAAC,EAAE,aAAa,CAAC;IACzB,gEAAgE;IAChE,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,iEAAiE;IACjE,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,uDAAuD;IACvD,MAAM,CAAC,EAAE,KAAK,EAAE,CAAC;CAClB;AAED,MAAM,WAAW,UAAU;IACzB,UAAU,CAAC,EAAE,CAAC,iBAAiB,EAAE,MAAM,EAAE,cAAc,EAAE,MAAM,KAAK,IAAI,CAAC;CAC1E;AAED,qBAAa,KAAK;IAChB,QAAQ,CAAC,IAAI,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,MAAM,EAAE,SAAS,KAAK,EAAE,CAAC;IAClC,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAa;IACvC,OAAO,CAAC,QAAQ,CAAC,QAAQ,CAAC,CAAgB;IAC1C,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAS;IACnC,OAAO,CAAC,QAAQ,CAAC,cAAc,CAAS;gBAE5B,OAAO,EAAE,YAAY;IAW3B,GAAG,CACP,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,MAAM,EAAE,MAAM,EACd,OAAO,CAAC,EAAE,UAAU,GACnB,OAAO,CAAC,YAAY,GAAG,IAAI,CAAC;YAajB,eAAe;YA8Gf,UAAU;CAqBzB;AAUD,8HAA8H;AAE9H,MAAM,WAAW,aAAc,SAAQ,aAAa,CAAC,gBAAgB,CAAC;CAAG"}
+{"version":3,"file":"stage.d.ts","sourceRoot":"","sources":["../src/stage.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,YAAY,EAAE,MAAM,cAAc,CAAC;AAErD,OAAO,KAAK,EAAE,QAAQ,EAAE,gBAAgB,EAAE,MAAM,sBAAsB,CAAC;AACvE,OAAO,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAEpD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAC;AAGjD,uEAAuE;AACvE,MAAM,MAAM,kBAAkB,GAC1B,aAAa,GACb,CAAC,CAAC,YAAY,EAAE,YAAY,KAAK,aAAa,CAAC,CAAC;AAEpD,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,MAAM,CAAC;IACb,SAAS,EAAE,QAAQ,GAAG,QAAQ,EAAE,CAAC;IACjC,QAAQ,CAAC,EAAE,kBAAkB,CAAC;IAC9B,gEAAgE;IAChE,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,iEAAiE;IACjE,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,uDAAuD;IACvD,MAAM,CAAC,EAAE,KAAK,EAAE,CAAC;CAClB;AAED,MAAM,WAAW,UAAU;IACzB,UAAU,CAAC,EAAE,CAAC,iBAAiB,EAAE,MAAM,EAAE,cAAc,EAAE,MAAM,KAAK,IAAI,CAAC;CAC1E;AAED,qBAAa,KAAK;IAChB,QAAQ,CAAC,IAAI,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,MAAM,EAAE,SAAS,KAAK,EAAE,CAAC;IAClC,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAa;IACvC,OAAO,CAAC,QAAQ,CAAC,aAAa,CAAC,CAAqB;IACpD,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAS;IACnC,OAAO,CAAC,QAAQ,CAAC,cAAc,CAAS;gBAE5B,OAAO,EAAE,YAAY;IAW3B,GAAG,CACP,OAAO,EAAE,OAAO,EAChB,YAAY,EAAE,YAAY,EAC1B,MAAM,EAAE,MAAM,EACd,OAAO,CAAC,EAAE,UAAU,GACnB,OAAO,CAAC,YAAY,GAAG,IAAI,CAAC;YAuBjB,eAAe;YA+Gf,UAAU;CAqBzB;AAUD,8HAA8H;AAE9H,MAAM,WAAW,aAAc,SAAQ,aAAa,CAAC,gBAAgB,CAAC;CAAG"}

package/dist/stage.js

@@ -5,7 +5,7 @@ export class Stage {
     name;
     stages;
     executors;
-    selector;
+    selectorInput;
     batchSize;
     maxConcurrency;
     constructor(options) {
@@ -14,13 +14,16 @@ export class Stage {
         this.executors = Array.isArray(options.executors)
             ? options.executors
             : [options.executors];
-        this.selector = options.selector;
+        this.selectorInput = options.selector;
         this.batchSize = options.batchSize ?? 10;
         this.maxConcurrency = options.maxConcurrency ?? 10;
     }
     async run(dataset, distribution, writer, options) {
-        if (this.selector) {
-            return this.runWithSelector(dataset, distribution, writer, options);
+        if (this.selectorInput) {
+            const selector = typeof this.selectorInput === 'function'
+                ? this.selectorInput(distribution)
+                : this.selectorInput;
+            return this.runWithSelector(selector, dataset, distribution, writer, options);
         }
         const streams = await this.executeAll(dataset, distribution);
         if (streams instanceof NotSupported) {
@@ -28,10 +31,10 @@ export class Stage {
         }
         await writer.write(dataset, mergeStreams(streams));
     }
-    async runWithSelector(dataset, distribution, writer, options) {
+    async runWithSelector(selector, dataset, distribution, writer, options) {
         // Peek the first batch to detect an empty selector before starting the
         // writer (important because e.g. SparqlUpdateWriter does CLEAR GRAPH).
-        const batches = batch(this.selector, this.batchSize);
+        const batches = batch(selector, this.batchSize);
         const iter = batches[Symbol.asyncIterator]();
         const first = await iter.next();
         if (first.done) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lde/pipeline",
-  "version": "0.6.25",
+  "version": "0.6.26",
   "repository": {
     "url": "https://github.com/ldengine/lde",
     "directory": "packages/pipeline"