@lde/dataset 0.7.5 → 0.7.7

package/dist/distribution.d.ts

@@ -18,6 +18,15 @@ export declare class Distribution {
      * Returns `undefined` when no compression format is declared.
      */
     get compressMimeType(): string | undefined;
+    /**
+     * The full content type a server is expected to send for the compressed
+     * download — {@link mimeType} with the compression suffix derived from
+     * {@link compressFormat} appended, e.g. `application/n-quads+gzip`. Returns
+     * `undefined` when no media type or no recognised compression format is
+     * declared. Lets a format check accept either the bare media type (the server
+     * decompressed the body) or the declared compressed form.
+     */
+    get compressedMimeType(): string | undefined;
     /**
      * @param accessUrl  Distribution access URL.
      * @param mediaType  IANA media type URI per DCAT-AP 3.0

package/dist/distribution.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"distribution.d.ts","sourceRoot":"","sources":["../src/distribution.ts"],"names":[],"mappings":"AAEA,eAAO,MAAM,sBAAsB,kDACc,CAAC;AAElD,qBAAa,YAAY;aAgCL,SAAS,EAAE,GAAG;aACd,SAAS,CAAC,EAAE,MAAM;aAClB,UAAU,CAAC,EAAE,GAAG;IAjC3B,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,YAAY,CAAC,EAAE,IAAI,CAAC;IACpB,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,aAAa,CAAC,EAAE,MAAM,CAAC;IAE9B;;;OAGG;IACH,SAAgB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAElC;;;OAGG;IACH,IAAW,gBAAgB,IAAI,MAAM,GAAG,SAAS,CAKhD;IAED;;;;;;OAMG;gBAEe,SAAS,EAAE,GAAG,EACd,SAAS,CAAC,EAAE,MAAM,YAAA,EAClB,UAAU,CAAC,EAAE,GAAG,YAAA;IAO3B,QAAQ;WASD,MAAM,CAAC,QAAQ,EAAE,GAAG,EAAE,UAAU,CAAC,EAAE,MAAM;CAUxD;AAED,oBAAY,SAAS;IACnB,WAAW,0BAA0B;IACrC,SAAS,wBAAwB;IACjC,MAAM,gBAAgB;CACvB;AAED,wBAAgB,wBAAwB,CAAC,SAAS,EAAE,SAAS,GAAG,MAAM,CAWrE"}
+{"version":3,"file":"distribution.d.ts","sourceRoot":"","sources":["../src/distribution.ts"],"names":[],"mappings":"AAEA,eAAO,MAAM,sBAAsB,kDACc,CAAC;AAYlD,qBAAa,YAAY;aA8CL,SAAS,EAAE,GAAG;aACd,SAAS,CAAC,EAAE,MAAM;aAClB,UAAU,CAAC,EAAE,GAAG;IA/C3B,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,YAAY,CAAC,EAAE,IAAI,CAAC;IACpB,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,aAAa,CAAC,EAAE,MAAM,CAAC;IAE9B;;;OAGG;IACH,SAAgB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAElC;;;OAGG;IACH,IAAW,gBAAgB,IAAI,MAAM,GAAG,SAAS,CAKhD;IAED;;;;;;;OAOG;IACH,IAAW,kBAAkB,IAAI,MAAM,GAAG,SAAS,CAIlD;IAED;;;;;;OAMG;gBAEe,SAAS,EAAE,GAAG,EACd,SAAS,CAAC,EAAE,MAAM,YAAA,EAClB,UAAU,CAAC,EAAE,GAAG,YAAA;IAO3B,QAAQ;WASD,MAAM,CAAC,QAAQ,EAAE,GAAG,EAAE,UAAU,CAAC,EAAE,MAAM;CAUxD;AAED,oBAAY,SAAS;IACnB,WAAW,0BAA0B;IACrC,SAAS,wBAAwB;IACjC,MAAM,gBAAgB;CACvB;AAED,wBAAgB,wBAAwB,CAAC,SAAS,EAAE,SAAS,GAAG,MAAM,CAWrE"}

package/dist/distribution.js

@@ -1,5 +1,14 @@
 const SPARQL_URI = 'https://www.w3.org/TR/sparql11-protocol/';
 export const IANA_MEDIA_TYPE_PREFIX = 'https://www.iana.org/assignments/media-types/';
+// Maps a compression content type to the structured-syntax suffix a server
+// appends to the RDF media type, e.g. `application/gzip` yields the `gzip` in
+// `application/n-quads+gzip`. The inverse of how a registry strips that suffix
+// into a separate compress format on ingest.
+const compressionSuffixesByMimeType = {
+    'application/gzip': 'gzip',
+    'application/x-gzip': 'gzip',
+    'application/zip': 'zip',
+};
 export class Distribution {
     accessUrl;
     mediaType;
@@ -25,6 +34,20 @@ export class Distribution {
             ? this.compressFormat.slice(IANA_MEDIA_TYPE_PREFIX.length)
             : this.compressFormat;
     }
+    /**
+     * The full content type a server is expected to send for the compressed
+     * download — {@link mimeType} with the compression suffix derived from
+     * {@link compressFormat} appended, e.g. `application/n-quads+gzip`. Returns
+     * `undefined` when no media type or no recognised compression format is
+     * declared. Lets a format check accept either the bare media type (the server
+     * decompressed the body) or the declared compressed form.
+     */
+    get compressedMimeType() {
+        if (this.mimeType === undefined)
+            return undefined;
+        const suffix = compressionSuffixesByMimeType[this.compressMimeType ?? ''];
+        return suffix === undefined ? undefined : `${this.mimeType}+${suffix}`;
+    }
     /**
      * @param accessUrl  Distribution access URL.
      * @param mediaType  IANA media type URI per DCAT-AP 3.0

package/dist/index.d.ts

@@ -1,5 +1,6 @@
 export * from './dataset.js';
 export * from './distribution.js';
 export * from './mediaType.js';
+export * from './skolem.js';
 export * from './validation.js';
 //# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,cAAc,CAAC;AAC7B,cAAc,mBAAmB,CAAC;AAClC,cAAc,gBAAgB,CAAC;AAC/B,cAAc,iBAAiB,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,cAAc,CAAC;AAC7B,cAAc,mBAAmB,CAAC;AAClC,cAAc,gBAAgB,CAAC;AAC/B,cAAc,aAAa,CAAC;AAC5B,cAAc,iBAAiB,CAAC"}

package/dist/index.js

@@ -1,4 +1,5 @@
 export * from './dataset.js';
 export * from './distribution.js';
 export * from './mediaType.js';
+export * from './skolem.js';
 export * from './validation.js';

package/dist/skolem.d.ts

@@ -0,0 +1,28 @@
+/**
+ * Mint a deterministic IRI for an otherwise-anonymous structural node — a PROV
+ * `Activity`/`Usage`, a DQV measurement, a `void:Linkset`, a `void:subset` or a
+ * `void:*Partition` — by extending a `base` IRI that is already unique within
+ * the dataset's graph (typically the dataset IRI or one of its subsets) with
+ * `-`-joined `suffixes`.
+ *
+ * Use this instead of a blank node for any such node. A dataset's graph is
+ * assembled from the output of several independent pipeline stages, and
+ * blank-node labels are not preserved across separately serialised documents:
+ * two stages' `_:b0` collapse into one node when the documents are merged into
+ * one graph, silently fusing unrelated provenance, measurements and linksets
+ * (see dataset-knowledge-graph issue #352). IRIs are never relabelled, so
+ * deriving every structural node from a unique base keeps distinct nodes
+ * distinct across stages and makes a re-run idempotent rather than additive —
+ * and lets a later stage address (and extend) a subset or partition another
+ * stage emitted.
+ *
+ * Pass opaque, possibly non-IRI-safe `suffixes` (e.g. a URL) through
+ * {@link hashSuffix} first.
+ */
+export declare function skolemIri(base: string, ...suffixes: string[]): string;
+/**
+ * An md5 hex digest, for embedding an opaque value (e.g. a URL) as a stable,
+ * IRI-safe segment in a {@link skolemIri}.
+ */
+export declare function hashSuffix(value: string): string;
+//# sourceMappingURL=skolem.d.ts.map

package/dist/skolem.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"skolem.d.ts","sourceRoot":"","sources":["../src/skolem.ts"],"names":[],"mappings":"AAEA;;;;;;;;;;;;;;;;;;;;GAoBG;AACH,wBAAgB,SAAS,CAAC,IAAI,EAAE,MAAM,EAAE,GAAG,QAAQ,EAAE,MAAM,EAAE,GAAG,MAAM,CAErE;AAED;;;GAGG;AACH,wBAAgB,UAAU,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAEhD"}

package/dist/skolem.js

@@ -0,0 +1,32 @@
+import { createHash } from 'node:crypto';
+/**
+ * Mint a deterministic IRI for an otherwise-anonymous structural node — a PROV
+ * `Activity`/`Usage`, a DQV measurement, a `void:Linkset`, a `void:subset` or a
+ * `void:*Partition` — by extending a `base` IRI that is already unique within
+ * the dataset's graph (typically the dataset IRI or one of its subsets) with
+ * `-`-joined `suffixes`.
+ *
+ * Use this instead of a blank node for any such node. A dataset's graph is
+ * assembled from the output of several independent pipeline stages, and
+ * blank-node labels are not preserved across separately serialised documents:
+ * two stages' `_:b0` collapse into one node when the documents are merged into
+ * one graph, silently fusing unrelated provenance, measurements and linksets
+ * (see dataset-knowledge-graph issue #352). IRIs are never relabelled, so
+ * deriving every structural node from a unique base keeps distinct nodes
+ * distinct across stages and makes a re-run idempotent rather than additive —
+ * and lets a later stage address (and extend) a subset or partition another
+ * stage emitted.
+ *
+ * Pass opaque, possibly non-IRI-safe `suffixes` (e.g. a URL) through
+ * {@link hashSuffix} first.
+ */
+export function skolemIri(base, ...suffixes) {
+    return [base, ...suffixes].join('-');
+}
+/**
+ * An md5 hex digest, for embedding an opaque value (e.g. a URL) as a stable,
+ * IRI-safe segment in a {@link skolemIri}.
+ */
+export function hashSuffix(value) {
+    return createHash('md5').update(value).digest('hex');
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lde/dataset",
-  "version": "0.7.5",
+  "version": "0.7.7",
   "repository": {
     "url": "git+https://github.com/ldelements/lde.git",
     "directory": "packages/dataset"