keepset 0.23.0 → 0.24.0

package/README.md

@@ -63,7 +63,7 @@ The catalogue itself is a manifest (`catalogue/manifest.json`: id, version, repo
 
 ## Categories and normalisation targets
 
-A category names a normalisation target: the shape every entry in that category resolves to, whatever its native shape. `graph/v1` is `{ meta, nodes, edges }` with unique node ids and edges whose endpoints name nodes; its entries are graph datasets taken from upstream collections such as SuiteSparse, Netzschleuder, SNAP, KONECT and OGB. `or/v1` is `{ meta, nodes }`, an operations-research instance as the nodes it offers to visit, each flat with optional coordinates, prize, prize probability, duration, travel time, skill, territory, lifecycle status and times, and deadline, read from tabular sources by the `csv-table` parser (a delimited file, a member of an archive, a sheet of an Excel workbook, or an open data portal read a page at a time) and from ArcGIS feature layers by the `geojson` parser. Its entries come from a survey of open operations-research datasets, imported with a curation that records, for each source, what its download gives and how its columns map onto the node; a dataset whose source has no usable table is imported without a recipe and says why. A new category starts by defining its target schema in `src/targets/` and registering it in the category registry; an entry cannot name a category that does not exist. The statically generated site renders every entry with schema.org Dataset JSON-LD, so the catalogue surfaces in the dataset discovery layers.
+A category names a normalisation target: the shape every entry in that category resolves to, whatever its native shape. `graph/v1` is `{ meta, nodes, edges }` with unique node ids and edges whose endpoints name nodes; its entries are graph datasets taken from upstream collections such as SuiteSparse, Netzschleuder, SNAP, KONECT and OGB. `or/v1` is `{ meta, nodes }`, an operations-research instance as the nodes it offers to visit, each flat with optional coordinates, prize, prize probability, duration, travel time, skill, territory, lifecycle status and times, and deadline, read from tabular sources by the `csv-table` parser (a delimited file, a member of an archive, a sheet of an Excel workbook, or an open data portal read a page at a time) and from ArcGIS feature layers by the `geojson` parser. Its entries come from a survey of open operations-research datasets, imported with a curation that records, for each source, what its download gives and how its columns map onto the node; a dataset whose source has no usable table is imported without a recipe and says why. `or-set/v1` is `{ meta, instances }`, a collection of operations-research instances such as a benchmark suite, each instance an `or/v1` document under an id of its own (for a collection read from an archive, the path of the member it came from): a recipe selects every matching archive member with `members`, the parser reads each one, and a member that does not parse fails the fetch and is named. A new category starts by defining its target schema in `src/targets/` and registering it in the category registry; an entry cannot name a category that does not exist. The statically generated site renders every entry with schema.org Dataset JSON-LD, so the catalogue surfaces in the dataset discovery layers.
 
 ## Licensing
 

package/dist/{bundled-BKl0C7kF.js → bundled-BWINQKLQ.js}

@@ -113,14 +113,16 @@ const JsonScalarSchema = z.union([
 * Reports each node whose id an earlier node already has, as an issue on that node's id.
 * @param nodes - The nodes of a target document, in order.
 * @param ctx - The refinement context to report to.
+* @param prefix - The path of the object holding the `nodes` array, for a document that nests nodes inside another.
 * @returns The set of ids the nodes carry.
 */
-function addDuplicateIdIssues(nodes, ctx) {
+function addDuplicateIdIssues(nodes, ctx, prefix = []) {
 	const ids = /* @__PURE__ */ new Set();
 	for (const [index, node] of nodes.entries()) {
 		if (ids.has(node.id)) ctx.addIssue({
 			code: "custom",
 			path: [
+				...prefix,
 				"nodes",
 				index,
 				"id"
@@ -202,22 +204,65 @@ const OrV1Schema = define(z.object({
 	addDuplicateIdIssues(instance.nodes, ctx);
 }));
 //#endregion
+//#region src/targets/or-set-v1.ts
+/** One instance of a set: an or/v1 instance, `{ meta, nodes }`, with the id that names it within the set (for a collection read from an archive, the member's path). */
+const OrInstanceSchema = z.object({
+	id: z.string().min(1),
+	meta: recordOf(z.string().min(1), JsonScalarSchema),
+	nodes: z.array(OrNodeSchema)
+}).strict();
+/**
+* The or-set/v1 normalisation target: `{ meta, instances }`, a collection of operations-research instances such as a benchmark suite, each instance an or/v1 document (`{ meta, nodes }`) under an id of its own. Beyond the shapes it checks that instance ids are unique in the set and node ids unique within each instance. A node id need not be unique across instances, because every instance numbers its own nodes.
+*/
+const OrSetV1Schema = define(z.object({
+	meta: recordOf(z.string().min(1), JsonScalarSchema),
+	instances: z.array(OrInstanceSchema)
+}).strict().superRefine((set, ctx) => {
+	const ids = /* @__PURE__ */ new Set();
+	for (const [index, instance] of set.instances.entries()) {
+		if (ids.has(instance.id)) ctx.addIssue({
+			code: "custom",
+			path: [
+				"instances",
+				index,
+				"id"
+			],
+			message: `duplicate instance id ${JSON.stringify(instance.id)}`
+		});
+		ids.add(instance.id);
+		addDuplicateIdIssues(instance.nodes, ctx, ["instances", index]);
+	}
+}));
+//#endregion
 //#region src/categories.ts
 /** Every category id the registry knows. A category id is `<kind>/v<version>`: the kind of data, and the version of its normalisation target, so a breaking change to a target is a new category rather than an edit to an old one. */
-const CategoryIdSchema = z.enum(["graph/v1", "or/v1"]);
+const CategoryIdSchema = z.enum([
+	"graph/v1",
+	"or/v1",
+	"or-set/v1"
+]);
 /** The registry. Adding a category is adding its id to `CategoryIdSchema` and its entry here, with its target schema beside graph/v1's in src/targets/: the mapped type makes the compiler reject a registry that misses an id or has one too many. */
 const registry = {
 	"graph/v1": {
 		id: "graph/v1",
 		description: "A graph as { meta, nodes, edges }: a flat record of scalar metadata, nodes carrying at least an id, and edges carrying at least source and target node ids, with optional node and edge types.",
+		shape: "document",
 		target: GraphV1Schema,
 		counted: ["nodes", "edges"]
 	},
 	"or/v1": {
 		id: "or/v1",
 		description: "An operations-research instance as { meta, nodes }: a flat record of scalar metadata and nodes carrying at least an id, with optional coordinates, prize, prize probability, duration, travel time, skill, territory, lifecycle status and times, and deadline.",
+		shape: "document",
 		target: OrV1Schema,
 		counted: ["nodes"]
+	},
+	"or-set/v1": {
+		id: "or-set/v1",
+		description: "A collection of operations-research instances as { meta, instances }: a flat record of scalar metadata and instances, each an or/v1 document ({ meta, nodes }) under an id of its own, read from every matching member of an archive.",
+		shape: "collection",
+		target: OrSetV1Schema,
+		counted: ["instances"]
 	}
 };
 function getCategory(id) {
@@ -410,22 +455,26 @@ const CitationSchema = z.object({
 const RelativePathSchema = z.string().refine((path) => path.split("/").every((segment) => segment !== "" && segment !== "." && segment !== ".."), { message: "must be a relative path of non-empty segments with no '.' or '..'" });
 /** Which file of an archive to read: a member path, matched against a member's whole path or its trailing segments (so `net.gml` finds `net/net.gml`), or the member whose name ends with one of a list of extensions (a format can have several, such as Pajek's `.net` and `.paj`). Either must match exactly one member when the recipe runs; macOS resource-fork members (`__MACOSX/` and `._` files) never match. */
 const ArchiveMemberSchema = z.union([z.object({ path: RelativePathSchema }).strict(), z.object({ extensions: z.array(z.string().regex(/^\.[A-Za-z0-9][A-Za-z0-9.]*$/, { message: "must be a file extension such as .mtx" })).min(1) }).strict()]);
-/** How the downloaded bytes unpack into the one file the parser reads. A workbook (`xlsx`) is read as the CSV of its named sheet, with dates as ISO 8601 text, for the `csv-table` parser. */
-const ArchiveSchema = z.discriminatedUnion("kind", [
+/** The archive kinds that hold several files, from which a recipe selects one (`member`) or every one a selector matches (`members`). */
+const FileArchiveKindSchema = z.enum([
+	"zip",
+	"tar",
+	"tar.gz"
+]);
+/** An archive a collection is read from: every file a selector matches, at least one. */
+const CollectionArchiveSchema = z.object({
+	kind: FileArchiveKindSchema,
+	members: ArchiveMemberSchema
+}).strict();
+/** How the downloaded bytes unpack into the file the parser reads, or for a collection the files it reads one by one. A `members` selector gives every member it matches, at least one, in path order. A workbook (`xlsx`) is read as the CSV of its named sheet, with dates as ISO 8601 text, for the `csv-table` parser. */
+const ArchiveSchema = z.union([
 	z.object({ kind: z.literal("none") }).strict(),
 	z.object({ kind: z.literal("gzip") }).strict(),
 	z.object({
-		kind: z.literal("zip"),
-		member: ArchiveMemberSchema
-	}).strict(),
-	z.object({
-		kind: z.literal("tar"),
-		member: ArchiveMemberSchema
-	}).strict(),
-	z.object({
-		kind: z.literal("tar.gz"),
+		kind: FileArchiveKindSchema,
 		member: ArchiveMemberSchema
 	}).strict(),
+	CollectionArchiveSchema,
 	z.object({
 		kind: z.literal("xlsx"),
 		sheet: z.string().min(1)
@@ -468,6 +517,10 @@ const ParserSchema = z.discriminatedUnion("format", [
 		format: z.literal("pajek"),
 		encoding: EncodingSchema
 	}).strict(),
+	z.object({
+		format: z.literal("solomon"),
+		encoding: EncodingSchema
+	}).strict(),
 	z.object({
 		format: z.literal("csv-table"),
 		encoding: EncodingSchema,
@@ -587,6 +640,24 @@ const EntrySchema = define(z.object({
 		path: ["recipe", "source"],
 		message: `names source ${String(entry.recipe.source)} but the entry has ${String(entry.sources.length)}`
 	});
+	if (entry.recipe.status === "fetchable") {
+		const selectsMembers = "members" in entry.recipe.archive;
+		const { shape } = getCategory(entry.category);
+		if (shape === "collection" && !selectsMembers) ctx.addIssue({
+			code: "custom",
+			path: ["recipe", "archive"],
+			message: `${entry.category} is a collection, so the archive must select its members with \`members\``
+		});
+		if (shape === "document" && selectsMembers) ctx.addIssue({
+			code: "custom",
+			path: [
+				"recipe",
+				"archive",
+				"members"
+			],
+			message: `${entry.category} is one document, so the archive selects one file with \`member\``
+		});
+	}
 	if ("target" in entry.crosswalk && entry.crosswalk.target !== entry.category) ctx.addIssue({
 		code: "custom",
 		path: ["crosswalk", "target"],
@@ -909,7 +980,7 @@ var catalogue_default = {
 		"id": "keepset",
 		"name": "keepset",
 		"description": "The keepset catalogue: datasets with machine-readable recipes that fetch each one and normalise it into its category's target shape.",
-		"version": "1.11.0",
+		"version": "1.12.0",
 		"repository": "https://github.com/Mearman/keepset",
 		"namespaceAliases": {
 			"networkrepository": ["nr"],
@@ -7105,7 +7176,7 @@ var catalogue_default = {
 				"duration",
 				"travelTime"
 			],
-			"category": "or/v1",
+			"category": "or-set/v1",
 			"sources": [{
 				"url": "https://github.com/psxjpc/MOVRPTW-Generator",
 				"retrieved": "2026-10-19",
@@ -7121,12 +7192,95 @@ var catalogue_default = {
 			},
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is a collection of Solomon style instance files and distance matrices, and a collection is more than one or/v1 document, so no recipe reads it."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"members": { "extensions": [
+						".tw0Specs.dat",
+						".tw1Specs.dat",
+						".tw2Specs.dat",
+						".tw3Specs.dat",
+						".tw4Specs.dat"
+					] }
+				},
+				"parser": {
+					"format": "solomon",
+					"encoding": "utf-8"
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
-			"properties": { "surveyId": "or-9" }
+			"crosswalk": {
+				"source": "solomon",
+				"target": "or-set/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].id",
+						"to": "instances[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.name",
+						"to": "instances[].meta.name"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.vehicles",
+						"to": "instances[].meta.vehicles"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.capacity",
+						"to": "instances[].meta.capacity"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].id",
+						"to": "instances[].nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].due",
+						"to": "instances[].nodes[].deadline"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].service",
+						"to": "instances[].nodes[].duration"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].x",
+						"to": "instances[].nodes[].lat"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].y",
+						"to": "instances[].nodes[].lon"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].demand",
+						"to": "instances[].nodes[].demand"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].ready",
+						"to": "instances[].nodes[].ready"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "instances": 45 },
+			"properties": {
+				"surveyId": "or-9",
+				"timeUnit": "seconds"
+			}
 		},
 		{
 			"id": "github-zhangchenhaoseu-jd-cold-chain-nodes",
@@ -8586,21 +8740,102 @@ var catalogue_default = {
 				"duration",
 				"travelTime"
 			],
-			"category": "or/v1",
+			"category": "or-set/v1",
 			"sources": [{
 				"url": "https://neo.lcc.uma.es/vrp/vrp-instances/capacitated-vrp-with-time-windows-instances/",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://neo.lcc.uma.es/vrp/wp-content/data/instances/solomon/solomon_100.zip",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": { "kind": "unrecorded" },
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is a collection of Solomon instance files (planar coordinates, demands, service times and windows in abstract units, not seconds), and a collection is more than one or/v1 document, so no recipe reads it."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"members": { "extensions": [".txt"] }
+				},
+				"parser": {
+					"format": "solomon",
+					"encoding": "utf-8"
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
-			"properties": { "surveyId": "or-5" }
+			"crosswalk": {
+				"source": "solomon",
+				"target": "or-set/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].id",
+						"to": "instances[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.name",
+						"to": "instances[].meta.name"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.vehicles",
+						"to": "instances[].meta.vehicles"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.capacity",
+						"to": "instances[].meta.capacity"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].id",
+						"to": "instances[].nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].due",
+						"to": "instances[].nodes[].deadline"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].x",
+						"to": "instances[].nodes[].x"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].y",
+						"to": "instances[].nodes[].y"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].demand",
+						"to": "instances[].nodes[].demand"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].ready",
+						"to": "instances[].nodes[].ready"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].service",
+						"to": "instances[].nodes[].service"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "instances": 56 },
+			"properties": {
+				"surveyId": "or-5",
+				"timeUnit": "abstract units"
+			}
 		},
 		{
 			"id": "leeds-fixed-penalty-notices",

package/dist/{categories-BGjPWXpS.d.ts → categories-BfqmrAvf.d.ts}

@@ -70,28 +70,29 @@ declare const ArchiveMemberSchema: z.ZodUnion<readonly [z.ZodObject<{
   extensions: z.ZodArray<z.ZodString>;
 }, z.core.$strict>]>;
 type ArchiveMember = z.infer<typeof ArchiveMemberSchema>;
-/** How the downloaded bytes unpack into the one file the parser reads. A workbook (`xlsx`) is read as the CSV of its named sheet, with dates as ISO 8601 text, for the `csv-table` parser. */
-declare const ArchiveSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
+/** How the downloaded bytes unpack into the file the parser reads, or for a collection the files it reads one by one. A `members` selector gives every member it matches, at least one, in path order. A workbook (`xlsx`) is read as the CSV of its named sheet, with dates as ISO 8601 text, for the `csv-table` parser. */
+declare const ArchiveSchema: z.ZodUnion<readonly [z.ZodObject<{
   kind: z.ZodLiteral<"none">;
 }, z.core.$strict>, z.ZodObject<{
   kind: z.ZodLiteral<"gzip">;
 }, z.core.$strict>, z.ZodObject<{
-  kind: z.ZodLiteral<"zip">;
-  member: z.ZodUnion<readonly [z.ZodObject<{
-    path: z.ZodString;
-  }, z.core.$strict>, z.ZodObject<{
-    extensions: z.ZodArray<z.ZodString>;
-  }, z.core.$strict>]>;
-}, z.core.$strict>, z.ZodObject<{
-  kind: z.ZodLiteral<"tar">;
+  kind: z.ZodEnum<{
+    zip: "zip";
+    tar: "tar";
+    "tar.gz": "tar.gz";
+  }>;
   member: z.ZodUnion<readonly [z.ZodObject<{
     path: z.ZodString;
   }, z.core.$strict>, z.ZodObject<{
     extensions: z.ZodArray<z.ZodString>;
   }, z.core.$strict>]>;
 }, z.core.$strict>, z.ZodObject<{
-  kind: z.ZodLiteral<"tar.gz">;
-  member: z.ZodUnion<readonly [z.ZodObject<{
+  kind: z.ZodEnum<{
+    zip: "zip";
+    tar: "tar";
+    "tar.gz": "tar.gz";
+  }>;
+  members: z.ZodUnion<readonly [z.ZodObject<{
     path: z.ZodString;
   }, z.core.$strict>, z.ZodObject<{
     extensions: z.ZodArray<z.ZodString>;
@@ -99,7 +100,7 @@ declare const ArchiveSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
 }, z.core.$strict>, z.ZodObject<{
   kind: z.ZodLiteral<"xlsx">;
   sheet: z.ZodString;
-}, z.core.$strict>], "kind">;
+}, z.core.$strict>]>;
 type Archive = z.infer<typeof ArchiveSchema>;
 /** A column of a table as the table parser names it: the header with every run of characters outside letters, digits and underscore replaced by one underscore, so it is a valid segment of a twindex path. */
 declare const ColumnNameSchema: z.ZodString;
@@ -139,6 +140,12 @@ declare const ParserSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
     "utf-8": "utf-8";
     latin1: "latin1";
   }>;
+}, z.core.$strict>, z.ZodObject<{
+  format: z.ZodLiteral<"solomon">;
+  encoding: z.ZodEnum<{
+    "utf-8": "utf-8";
+    latin1: "latin1";
+  }>;
 }, z.core.$strict>, z.ZodObject<{
   format: z.ZodLiteral<"csv-table">;
   encoding: z.ZodEnum<{
@@ -187,27 +194,28 @@ declare const RecipeSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
     kind: z.ZodLiteral<"openfema">;
     pageSize: z.ZodInt;
   }, z.core.$strict>], "kind">>;
-  archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
+  archive: z.ZodUnion<readonly [z.ZodObject<{
     kind: z.ZodLiteral<"none">;
   }, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"gzip">;
   }, z.core.$strict>, z.ZodObject<{
-    kind: z.ZodLiteral<"zip">;
-    member: z.ZodUnion<readonly [z.ZodObject<{
-      path: z.ZodString;
-    }, z.core.$strict>, z.ZodObject<{
-      extensions: z.ZodArray<z.ZodString>;
-    }, z.core.$strict>]>;
-  }, z.core.$strict>, z.ZodObject<{
-    kind: z.ZodLiteral<"tar">;
+    kind: z.ZodEnum<{
+      zip: "zip";
+      tar: "tar";
+      "tar.gz": "tar.gz";
+    }>;
     member: z.ZodUnion<readonly [z.ZodObject<{
       path: z.ZodString;
     }, z.core.$strict>, z.ZodObject<{
       extensions: z.ZodArray<z.ZodString>;
     }, z.core.$strict>]>;
   }, z.core.$strict>, z.ZodObject<{
-    kind: z.ZodLiteral<"tar.gz">;
-    member: z.ZodUnion<readonly [z.ZodObject<{
+    kind: z.ZodEnum<{
+      zip: "zip";
+      tar: "tar";
+      "tar.gz": "tar.gz";
+    }>;
+    members: z.ZodUnion<readonly [z.ZodObject<{
       path: z.ZodString;
     }, z.core.$strict>, z.ZodObject<{
       extensions: z.ZodArray<z.ZodString>;
@@ -215,7 +223,7 @@ declare const RecipeSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
   }, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"xlsx">;
     sheet: z.ZodString;
-  }, z.core.$strict>], "kind">;
+  }, z.core.$strict>]>;
   parser: z.ZodDiscriminatedUnion<[z.ZodObject<{
     format: z.ZodLiteral<"edge-list">;
     encoding: z.ZodEnum<{
@@ -251,6 +259,12 @@ declare const RecipeSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
       "utf-8": "utf-8";
       latin1: "latin1";
     }>;
+  }, z.core.$strict>, z.ZodObject<{
+    format: z.ZodLiteral<"solomon">;
+    encoding: z.ZodEnum<{
+      "utf-8": "utf-8";
+      latin1: "latin1";
+    }>;
   }, z.core.$strict>, z.ZodObject<{
     format: z.ZodLiteral<"csv-table">;
     encoding: z.ZodEnum<{
@@ -309,6 +323,7 @@ declare const EntrySchema: z.ZodObject<{
   category: z.ZodEnum<{
     "graph/v1": "graph/v1";
     "or/v1": "or/v1";
+    "or-set/v1": "or-set/v1";
   }>;
   sources: z.ZodArray<z.ZodObject<{
     url: z.ZodURL;
@@ -353,27 +368,28 @@ declare const EntrySchema: z.ZodObject<{
       kind: z.ZodLiteral<"openfema">;
       pageSize: z.ZodInt;
     }, z.core.$strict>], "kind">>;
-    archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
+    archive: z.ZodUnion<readonly [z.ZodObject<{
       kind: z.ZodLiteral<"none">;
     }, z.core.$strict>, z.ZodObject<{
       kind: z.ZodLiteral<"gzip">;
     }, z.core.$strict>, z.ZodObject<{
-      kind: z.ZodLiteral<"zip">;
-      member: z.ZodUnion<readonly [z.ZodObject<{
-        path: z.ZodString;
-      }, z.core.$strict>, z.ZodObject<{
-        extensions: z.ZodArray<z.ZodString>;
-      }, z.core.$strict>]>;
-    }, z.core.$strict>, z.ZodObject<{
-      kind: z.ZodLiteral<"tar">;
+      kind: z.ZodEnum<{
+        zip: "zip";
+        tar: "tar";
+        "tar.gz": "tar.gz";
+      }>;
       member: z.ZodUnion<readonly [z.ZodObject<{
         path: z.ZodString;
       }, z.core.$strict>, z.ZodObject<{
         extensions: z.ZodArray<z.ZodString>;
       }, z.core.$strict>]>;
     }, z.core.$strict>, z.ZodObject<{
-      kind: z.ZodLiteral<"tar.gz">;
-      member: z.ZodUnion<readonly [z.ZodObject<{
+      kind: z.ZodEnum<{
+        zip: "zip";
+        tar: "tar";
+        "tar.gz": "tar.gz";
+      }>;
+      members: z.ZodUnion<readonly [z.ZodObject<{
         path: z.ZodString;
       }, z.core.$strict>, z.ZodObject<{
         extensions: z.ZodArray<z.ZodString>;
@@ -381,7 +397,7 @@ declare const EntrySchema: z.ZodObject<{
     }, z.core.$strict>, z.ZodObject<{
       kind: z.ZodLiteral<"xlsx">;
       sheet: z.ZodString;
-    }, z.core.$strict>], "kind">;
+    }, z.core.$strict>]>;
     parser: z.ZodDiscriminatedUnion<[z.ZodObject<{
       format: z.ZodLiteral<"edge-list">;
       encoding: z.ZodEnum<{
@@ -417,6 +433,12 @@ declare const EntrySchema: z.ZodObject<{
         "utf-8": "utf-8";
         latin1: "latin1";
       }>;
+    }, z.core.$strict>, z.ZodObject<{
+      format: z.ZodLiteral<"solomon">;
+      encoding: z.ZodEnum<{
+        "utf-8": "utf-8";
+        latin1: "latin1";
+      }>;
     }, z.core.$strict>, z.ZodObject<{
       format: z.ZodLiteral<"csv-table">;
       encoding: z.ZodEnum<{
@@ -463,7 +485,7 @@ declare const EntrySchema: z.ZodObject<{
     id: string;
     name: string;
     tags: string[];
-    category: "graph/v1" | "or/v1";
+    category: "graph/v1" | "or/v1" | "or-set/v1";
     sources: {
       url: string;
       retrieved: string;
@@ -496,22 +518,15 @@ declare const EntrySchema: z.ZodObject<{
       } | {
         kind: "gzip";
       } | {
-        kind: "zip";
-        member: {
-          path: string;
-        } | {
-          extensions: string[];
-        };
-      } | {
-        kind: "tar";
+        kind: "zip" | "tar" | "tar.gz";
         member: {
           path: string;
         } | {
           extensions: string[];
         };
       } | {
-        kind: "tar.gz";
-        member: {
+        kind: "zip" | "tar" | "tar.gz";
+        members: {
           path: string;
         } | {
           extensions: string[];
@@ -559,6 +574,9 @@ declare const EntrySchema: z.ZodObject<{
       } | {
         format: "pajek";
         encoding: "utf-8" | "latin1";
+      } | {
+        format: "solomon";
+        encoding: "utf-8" | "latin1";
       };
       paging?: {
         kind: "arcgis-query";
@@ -692,6 +710,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
     category: z.ZodEnum<{
       "graph/v1": "graph/v1";
       "or/v1": "or/v1";
+      "or-set/v1": "or-set/v1";
     }>;
     sources: z.ZodArray<z.ZodObject<{
       url: z.ZodURL;
@@ -736,27 +755,28 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         kind: z.ZodLiteral<"openfema">;
         pageSize: z.ZodInt;
       }, z.core.$strict>], "kind">>;
-      archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
+      archive: z.ZodUnion<readonly [z.ZodObject<{
         kind: z.ZodLiteral<"none">;
       }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"gzip">;
       }, z.core.$strict>, z.ZodObject<{
-        kind: z.ZodLiteral<"zip">;
-        member: z.ZodUnion<readonly [z.ZodObject<{
-          path: z.ZodString;
-        }, z.core.$strict>, z.ZodObject<{
-          extensions: z.ZodArray<z.ZodString>;
-        }, z.core.$strict>]>;
-      }, z.core.$strict>, z.ZodObject<{
-        kind: z.ZodLiteral<"tar">;
+        kind: z.ZodEnum<{
+          zip: "zip";
+          tar: "tar";
+          "tar.gz": "tar.gz";
+        }>;
         member: z.ZodUnion<readonly [z.ZodObject<{
           path: z.ZodString;
         }, z.core.$strict>, z.ZodObject<{
           extensions: z.ZodArray<z.ZodString>;
         }, z.core.$strict>]>;
       }, z.core.$strict>, z.ZodObject<{
-        kind: z.ZodLiteral<"tar.gz">;
-        member: z.ZodUnion<readonly [z.ZodObject<{
+        kind: z.ZodEnum<{
+          zip: "zip";
+          tar: "tar";
+          "tar.gz": "tar.gz";
+        }>;
+        members: z.ZodUnion<readonly [z.ZodObject<{
           path: z.ZodString;
         }, z.core.$strict>, z.ZodObject<{
           extensions: z.ZodArray<z.ZodString>;
@@ -764,7 +784,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
       }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"xlsx">;
         sheet: z.ZodString;
-      }, z.core.$strict>], "kind">;
+      }, z.core.$strict>]>;
       parser: z.ZodDiscriminatedUnion<[z.ZodObject<{
         format: z.ZodLiteral<"edge-list">;
         encoding: z.ZodEnum<{
@@ -800,6 +820,12 @@ declare const CatalogueIndexSchema: z.ZodObject<{
           "utf-8": "utf-8";
           latin1: "latin1";
         }>;
+      }, z.core.$strict>, z.ZodObject<{
+        format: z.ZodLiteral<"solomon">;
+        encoding: z.ZodEnum<{
+          "utf-8": "utf-8";
+          latin1: "latin1";
+        }>;
       }, z.core.$strict>, z.ZodObject<{
         format: z.ZodLiteral<"csv-table">;
         encoding: z.ZodEnum<{
@@ -846,7 +872,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
       id: string;
       name: string;
       tags: string[];
-      category: "graph/v1" | "or/v1";
+      category: "graph/v1" | "or/v1" | "or-set/v1";
       sources: {
         url: string;
         retrieved: string;
@@ -879,22 +905,15 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           kind: "gzip";
         } | {
-          kind: "zip";
+          kind: "zip" | "tar" | "tar.gz";
           member: {
             path: string;
           } | {
             extensions: string[];
           };
         } | {
-          kind: "tar";
-          member: {
-            path: string;
-          } | {
-            extensions: string[];
-          };
-        } | {
-          kind: "tar.gz";
-          member: {
+          kind: "zip" | "tar" | "tar.gz";
+          members: {
             path: string;
           } | {
             extensions: string[];
@@ -942,6 +961,9 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           format: "pajek";
           encoding: "utf-8" | "latin1";
+        } | {
+          format: "solomon";
+          encoding: "utf-8" | "latin1";
         };
         paging?: {
           kind: "arcgis-query";
@@ -991,7 +1013,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
       id: string;
       name: string;
       tags: string[];
-      category: "graph/v1" | "or/v1";
+      category: "graph/v1" | "or/v1" | "or-set/v1";
       sources: {
         url: string;
         retrieved: string;
@@ -1024,22 +1046,15 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           kind: "gzip";
         } | {
-          kind: "zip";
+          kind: "zip" | "tar" | "tar.gz";
           member: {
             path: string;
           } | {
             extensions: string[];
           };
         } | {
-          kind: "tar";
-          member: {
-            path: string;
-          } | {
-            extensions: string[];
-          };
-        } | {
-          kind: "tar.gz";
-          member: {
+          kind: "zip" | "tar" | "tar.gz";
+          members: {
             path: string;
           } | {
             extensions: string[];
@@ -1087,6 +1102,9 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           format: "pajek";
           encoding: "utf-8" | "latin1";
+        } | {
+          format: "solomon";
+          encoding: "utf-8" | "latin1";
         };
         paging?: {
           kind: "arcgis-query";
@@ -1222,6 +1240,62 @@ declare const GraphV1Schema: z.ZodObject<{
 };
 type GraphV1 = z.infer<typeof GraphV1Schema>;
 //#endregion
+//#region src/targets/or-set-v1.d.ts
+/**
+ * The or-set/v1 normalisation target: `{ meta, instances }`, a collection of operations-research instances such as a benchmark suite, each instance an or/v1 document (`{ meta, nodes }`) under an id of its own. Beyond the shapes it checks that instance ids are unique in the set and node ids unique within each instance. A node id need not be unique across instances, because every instance numbers its own nodes.
+ */
+declare const OrSetV1Schema: z.ZodObject<{
+  meta: z.ZodPreprocess<z.ZodRecord<z.ZodString, z.ZodUnion<readonly [z.ZodString, z.ZodNumber, z.ZodBoolean, z.ZodNull]>>, unknown>;
+  instances: z.ZodArray<z.ZodObject<{
+    id: z.ZodString;
+    meta: z.ZodPreprocess<z.ZodRecord<z.ZodString, z.ZodUnion<readonly [z.ZodString, z.ZodNumber, z.ZodBoolean, z.ZodNull]>>, unknown>;
+    nodes: z.ZodArray<z.ZodObject<{
+      id: z.ZodString;
+      lat: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      lon: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      x: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      y: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      prize: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      prizeProbability: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      duration: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      travelTime: z.ZodOptional<z.ZodNullable<z.ZodNumber>>;
+      skill: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+      territory: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+      status: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+      opened: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+      closed: z.ZodOptional<z.ZodNullable<z.ZodString>>;
+      deadline: z.ZodOptional<z.ZodNullable<z.ZodUnion<readonly [z.ZodString, z.ZodNumber]>>>;
+    }, z.core.$loose>>;
+  }, z.core.$strict>>;
+}, z.core.$strict> & {
+  is: (value: unknown) => value is {
+    meta: Record<string, string | number | boolean | null>;
+    instances: {
+      id: string;
+      meta: Record<string, string | number | boolean | null>;
+      nodes: {
+        [x: string]: unknown;
+        id: string;
+        lat?: number | null | undefined;
+        lon?: number | null | undefined;
+        x?: number | null | undefined;
+        y?: number | null | undefined;
+        prize?: number | null | undefined;
+        prizeProbability?: number | null | undefined;
+        duration?: number | null | undefined;
+        travelTime?: number | null | undefined;
+        skill?: string | null | undefined;
+        territory?: string | null | undefined;
+        status?: string | null | undefined;
+        opened?: string | null | undefined;
+        closed?: string | null | undefined;
+        deadline?: string | number | null | undefined;
+      }[];
+    }[];
+  };
+};
+type OrSetV1 = z.infer<typeof OrSetV1Schema>;
+//#endregion
 //#region src/targets/or-v1.d.ts
 /**
  * A node of an operations-research instance: a place or case a route may visit, flat so a twindex crosswalk can map a table's columns onto it (a wildcard path must keep the same shape on both sides). The target fixes only what the datasets share and lets a crosswalk map any further column through under its own name, as graph/v1 does. `lat` and `lon` are WGS84 degrees and `x` and `y` are planar coordinates in the dataset's own system; `duration` and `travelTime` are seconds, `prize` is the dataset's own reward measure, and `prizeProbability` is the chance the reward is collected. `skill` and `territory` name what is needed to serve the node and where it sits, `status` is where it is in its lifecycle, and `opened`, `closed` and `deadline` are its lifecycle times as the source writes them, a deadline being a number where the source gives a time of day.
@@ -1296,21 +1370,26 @@ type OrV1 = z.infer<typeof OrV1Schema>;
 declare const CategoryIdSchema: z.ZodEnum<{
   "graph/v1": "graph/v1";
   "or/v1": "or/v1";
+  "or-set/v1": "or-set/v1";
 }>;
 type CategoryId = z.infer<typeof CategoryIdSchema>;
 /** The normalisation target each category resolves to, keyed by category id. */
 interface CategoryTargets {
   readonly "graph/v1": GraphV1;
   readonly "or/v1": OrV1;
+  readonly "or-set/v1": OrSetV1;
 }
 /** The normalised shape of any category: what `fetchDataset` can resolve to. */
 type CategoryTarget = CategoryTargets[CategoryId];
 /** The keys of a target whose values are arrays: the collections an entry's `expectedCounts` may name. */
 type CountedKey<Target> = { readonly [Key in keyof Target]: Target[Key] extends readonly unknown[] ? Key : never; }[keyof Target] & string;
-/** One category: its id, what it holds, the schema of its normalisation target, and which of the target's arrays an entry may record expected counts for. */
+/** What one fetch of an entry in a category reads: a single document, or a set of documents, one per member of the archive the source unpacks to. */
+type CategoryShape = "document" | "collection";
+/** One category: its id, what it holds, whether an entry is one document or a collection of them, the schema of its normalisation target, and which of the target's arrays an entry may record expected counts for. */
 interface CategoryOf<Id extends CategoryId> {
   readonly id: Id;
   readonly description: string;
+  readonly shape: CategoryShape;
   readonly target: z.ZodType<CategoryTargets[Id]>;
   readonly counted: readonly CountedKey<CategoryTargets[Id]>[];
 }

package/dist/index.d.ts

@@ -1,4 +1,4 @@
-import { $ as Parser, A as LineageSchema, B as Citation, C as CatalogueIndex, D as NamespaceNode, E as DatasetQuery, F as AliasSchema, G as Entry, H as ColumnNameSchema, I as Archive, J as IdSchema, K as EntrySchema, L as ArchiveMember, M as ManifestSchema, N as SemVerSchema, O as createCatalogue, P as parseManifest, Q as NamespaceSchema, R as ArchiveMemberSchema, S as Catalogue, T as Dataset, U as CrosswalkReference, V as CitationSchema, W as CrosswalkReferenceSchema, X as LicenceSchema, Y as Licence, Z as Namespace, _ as GraphEdgeSchema, a as CategoryTargets, at as Snapshot, b as GraphV1, c as getCategory, ct as SourceSchema, d as listCategories, dt as waybackUrl, et as ParserSchema, f as OrNode, g as GraphEdge, h as OrV1Schema, i as CategoryTarget, it as RelativePathSchema, j as Manifest, k as Lineage, l as getTarget, lt as WaybackTimestampSchema, m as OrV1, n as CategoryId, nt as Recipe, o as CountedKey, ot as SnapshotSchema, p as OrNodeSchema, q as HttpUrlSchema, r as CategoryIdSchema, rt as RecipeSchema, s as countTarget, st as Source, t as Category, tt as RESERVED_META_KEYS, u as kindOf, ut as parseEntry, v as GraphNode, w as CatalogueIndexSchema, x as GraphV1Schema, y as GraphNodeSchema, z as ArchiveSchema } from "./categories-BGjPWXpS.js";
+import { $ as Parser, A as LineageSchema, B as Citation, C as CatalogueIndex, D as NamespaceNode, E as DatasetQuery, F as AliasSchema, G as Entry, H as ColumnNameSchema, I as Archive, J as IdSchema, K as EntrySchema, L as ArchiveMember, M as ManifestSchema, N as SemVerSchema, O as createCatalogue, P as parseManifest, Q as NamespaceSchema, R as ArchiveMemberSchema, S as Catalogue, T as Dataset, U as CrosswalkReference, V as CitationSchema, W as CrosswalkReferenceSchema, X as LicenceSchema, Y as Licence, Z as Namespace, _ as GraphEdgeSchema, a as CategoryTargets, at as Snapshot, b as GraphV1, c as getCategory, ct as SourceSchema, d as listCategories, dt as waybackUrl, et as ParserSchema, f as OrNode, g as GraphEdge, h as OrV1Schema, i as CategoryTarget, it as RelativePathSchema, j as Manifest, k as Lineage, l as getTarget, lt as WaybackTimestampSchema, m as OrV1, n as CategoryId, nt as Recipe, o as CountedKey, ot as SnapshotSchema, p as OrNodeSchema, q as HttpUrlSchema, r as CategoryIdSchema, rt as RecipeSchema, s as countTarget, st as Source, t as Category, tt as RESERVED_META_KEYS, u as kindOf, ut as parseEntry, v as GraphNode, w as CatalogueIndexSchema, x as GraphV1Schema, y as GraphNodeSchema, z as ArchiveSchema } from "./categories-BfqmrAvf.js";
 import { Crosswalk } from "twindex";
 import { z } from "zod";
 //#region src/errors.d.ts

package/dist/index.js

@@ -1,4 +1,4 @@
-import { $ as NativeParseError, A as toIssues, B as kindOf, C as RESERVED_META_KEYS, D as SourceSchema, E as SnapshotSchema, G as GraphNodeSchema, H as OrNodeSchema, I as CategoryIdSchema, J as ArchiveError, K as GraphV1Schema, L as countTarget, M as NAME_SEPARATOR, N as groupsOf, O as WaybackTimestampSchema, P as namesOf, Q as KeepsetError, R as getCategory, S as ParserSchema, T as RelativePathSchema, U as OrV1Schema, V as listCategories, W as GraphEdgeSchema, X as DatasetAmbiguousError, Y as CatalogueSyntaxError, Z as DatasetNotFoundError, _ as EntrySchema, a as CatalogueIndexSchema, b as LicenceSchema, c as ManifestSchema, d as AliasSchema, et as RecipeError, f as ArchiveMemberSchema, g as CrosswalkReferenceSchema, h as ColumnNameSchema, i as listNamespaces, j as waybackUrl, k as parseEntry, l as SemVerSchema, m as CitationSchema, n as getManifest, o as createCatalogue, p as ArchiveSchema, q as JsonScalarSchema, r as listDatasets, s as LineageSchema, t as getDataset, tt as UnknownCategoryError, u as parseManifest, v as HttpUrlSchema, w as RecipeSchema, x as NamespaceSchema, y as IdSchema, z as getTarget } from "./bundled-BKl0C7kF.js";
+import { $ as NativeParseError, A as toIssues, B as kindOf, C as RESERVED_META_KEYS, D as SourceSchema, E as SnapshotSchema, G as GraphNodeSchema, H as OrNodeSchema, I as CategoryIdSchema, J as ArchiveError, K as GraphV1Schema, L as countTarget, M as NAME_SEPARATOR, N as groupsOf, O as WaybackTimestampSchema, P as namesOf, Q as KeepsetError, R as getCategory, S as ParserSchema, T as RelativePathSchema, U as OrV1Schema, V as listCategories, W as GraphEdgeSchema, X as DatasetAmbiguousError, Y as CatalogueSyntaxError, Z as DatasetNotFoundError, _ as EntrySchema, a as CatalogueIndexSchema, b as LicenceSchema, c as ManifestSchema, d as AliasSchema, et as RecipeError, f as ArchiveMemberSchema, g as CrosswalkReferenceSchema, h as ColumnNameSchema, i as listNamespaces, j as waybackUrl, k as parseEntry, l as SemVerSchema, m as CitationSchema, n as getManifest, o as createCatalogue, p as ArchiveSchema, q as JsonScalarSchema, r as listDatasets, s as LineageSchema, t as getDataset, tt as UnknownCategoryError, u as parseManifest, v as HttpUrlSchema, w as RecipeSchema, x as NamespaceSchema, y as IdSchema, z as getTarget } from "./bundled-BWINQKLQ.js";
 import { validateCrosswalk } from "twindex";
 //#region src/serialise.ts
 /**

package/dist/node/fetch-dataset.d.ts

@@ -1,4 +1,4 @@
-import { G as Entry, T as Dataset, at as Snapshot, i as CategoryTarget, st as Source } from "../categories-BGjPWXpS.js";
+import { G as Entry, T as Dataset, at as Snapshot, i as CategoryTarget, st as Source } from "../categories-BfqmrAvf.js";
 //#region src/node/fetch-dataset.d.ts
 /** The network port a recipe downloads through: `downloadRaw` by default, which returns the bytes exactly as served, or a stand-in under test. */
 export type FetchPort = (url: string, init: {
@@ -24,7 +24,10 @@ export interface Provenance {
   readonly retrievedAt: string;
   readonly sha256: string;
   readonly byteLength: number;
+  /** The archive member a document was read from. */
   readonly member?: string;
+  /** The archive members a collection was read from, in the order they were read. */
+  readonly members?: readonly string[];
 }
 /** A fetched dataset: the resolved category target, the provenance of the bytes it came from, and any warnings, such as a count that differs from the one the entry records. */
 export interface FetchedDataset {

package/dist/node/fetch-dataset.js

@@ -1,4 +1,4 @@
-import { $ as NativeParseError, F as getOwn, J as ArchiveError, L as countTarget, Q as KeepsetError, et as RecipeError, t as getDataset, z as getTarget } from "../bundled-BKl0C7kF.js";
+import { $ as NativeParseError, F as getOwn, J as ArchiveError, L as countTarget, Q as KeepsetError, et as RecipeError, t as getDataset, z as getTarget } from "../bundled-BWINQKLQ.js";
 import { RoundTripMismatchError, assertRoundTrip, encode } from "twindex";
 import { z } from "zod";
 import { createHash } from "node:crypto";
@@ -104,7 +104,7 @@ function parseEdgeList(text, options) {
 //#endregion
 //#region src/parsers/gml.ts
 const KEY = /[A-Za-z_][A-Za-z0-9_]*/y;
-const NUMBER = /[+-]?(?:\d+\.?\d*|\.\d+)(?:[Ee][+-]?\d+)?/y;
+const NUMBER$1 = /[+-]?(?:\d+\.?\d*|\.\d+)(?:[Ee][+-]?\d+)?/y;
 /**
 * Matches a sticky pattern at a position.
 * @param pattern - A sticky regular expression.
@@ -155,7 +155,7 @@ function tokenise(text) {
 			position = end + 1;
 		} else {
 			const key = matchAt(KEY, text, position);
-			const number = key === void 0 ? matchAt(NUMBER, text, position) : void 0;
+			const number = key === void 0 ? matchAt(NUMBER$1, text, position) : void 0;
 			const match = key ?? number;
 			if (match === void 0) throw new NativeParseError("gml", line, `unexpected character ${JSON.stringify(char)}`);
 			tokens.push({
@@ -832,6 +832,89 @@ function parsePajek(text) {
 	};
 }
 //#endregion
+//#region src/parsers/solomon.ts
+const FORMAT = "solomon";
+/** A line of integers or decimals separated by any run of spaces and tabs, as the files write rows (Solomon's with padded spaces, the Tenerife set with tabs). */
+const NUMBER = /^-?\d+(?:\.\d+)?$/;
+/**
+* Reads a token as a finite number.
+* @param token - The token.
+* @param line - The 1-based line, for the error.
+* @param what - What the token should be, for the error.
+* @returns The number.
+* @throws NativeParseError when the token is not a decimal number.
+*/
+function numberOf(token, line, what) {
+	if (!NUMBER.test(token)) throw new NativeParseError(FORMAT, line, `expected ${what}, found ${JSON.stringify(token)}`);
+	return Number(token);
+}
+/**
+* Parses a Solomon style vehicle routing instance (R. M. Solomon, 1987): a name on the first line, a `VEHICLE` section holding the number of vehicles and their capacity, and a `CUSTOMER` section of one row per node, the depot first, with its id, position, demand, ready time, due date and service time. Columns are separated by any run of spaces or tabs, so the original files and the tab separated Tenerife set read alike, and blank or whitespace only lines are skipped. The header lines (`NUMBER CAPACITY` and the column titles) are not read: a row is a line of exactly seven numbers after the `CUSTOMER` keyword, and anything else after it is an error. Node ids are kept as the file writes them, so a file whose ids are not consecutive keeps its own. The time quantities are in the unit the source uses, which the format does not record.
+* @param text - The file text.
+* @returns The instance: its name, vehicle count and capacity as `meta`, and its nodes in file order.
+* @throws NativeParseError naming the line for a missing name, `VEHICLE` or `CUSTOMER` section, a vehicle line that is not two numbers, a customer row that is not seven numbers, a repeated node id, or a file with no customer.
+*/
+function parseSolomon(text) {
+	const all = lines(text);
+	const present = (index) => (all[index] ?? "").trim();
+	const name = present(0);
+	if (name === "") throw new NativeParseError(FORMAT, 1, "expected the instance name on the first line");
+	const meta = { name };
+	const nodes = [];
+	const seen = /* @__PURE__ */ new Set();
+	let section = "preamble";
+	let awaitingHeader = false;
+	for (const [index, raw] of all.entries()) {
+		const line = index + 1;
+		const trimmed = raw.trim();
+		if (index === 0 || trimmed === "") continue;
+		const keyword = trimmed.toUpperCase();
+		if (keyword === "VEHICLE" || keyword === "CUSTOMER") {
+			section = keyword === "VEHICLE" ? "vehicle" : "customer";
+			awaitingHeader = true;
+			continue;
+		}
+		if (section === "preamble") throw new NativeParseError(FORMAT, line, `expected VEHICLE or CUSTOMER, found ${JSON.stringify(trimmed)}`);
+		if (awaitingHeader) {
+			awaitingHeader = false;
+			continue;
+		}
+		const tokens = trimmed.split(/\s+/);
+		if (section === "vehicle") {
+			const [vehicles, capacity, ...extra] = tokens;
+			if (vehicles === void 0 || capacity === void 0 || extra.length > 0) throw new NativeParseError(FORMAT, line, `expected the vehicle count and capacity, found ${JSON.stringify(trimmed)}`);
+			meta.vehicles = numberOf(vehicles, line, "the vehicle count");
+			meta.capacity = numberOf(capacity, line, "the vehicle capacity");
+			continue;
+		}
+		const [id, x, y, demand, ready, due, service, ...extra] = tokens;
+		if (id === void 0 || x === void 0 || y === void 0 || demand === void 0 || ready === void 0 || due === void 0 || service === void 0 || extra.length > 0) throw new NativeParseError(FORMAT, line, `expected seven numbers (id, x, y, demand, ready, due, service), found ${JSON.stringify(trimmed)}`);
+		if (!/^\d+$/.test(id)) throw new NativeParseError(FORMAT, line, `expected a whole number node id, found ${JSON.stringify(id)}`);
+		if (seen.has(id)) throw new NativeParseError(FORMAT, line, `node ${id} appears twice`);
+		seen.add(id);
+		nodes.push({
+			id,
+			x: numberOf(x, line, "x"),
+			y: numberOf(y, line, "y"),
+			demand: numberOf(demand, line, "the demand"),
+			ready: numberOf(ready, line, "the ready time"),
+			due: numberOf(due, line, "the due date"),
+			service: numberOf(service, line, "the service time")
+		});
+	}
+	if (section !== "customer") throw new NativeParseError(FORMAT, all.length, "expected a CUSTOMER section");
+	if (nodes.length === 0) throw new NativeParseError(FORMAT, all.length, "the CUSTOMER section has no rows");
+	if (meta.vehicles === void 0 || meta.capacity === void 0) throw new NativeParseError(FORMAT, all.length, "expected a VEHICLE section with the vehicle count and capacity");
+	return {
+		meta: {
+			name: meta.name,
+			vehicles: meta.vehicles,
+			capacity: meta.capacity
+		},
+		nodes
+	};
+}
+//#endregion
 //#region src/parsers/parse-native.ts
 /** Bytes converted per String.fromCharCode call: spreading a whole file would exceed the engine's argument-count limit, and this many stays well inside every engine's. */
 const LATIN1_CHUNK = 32768;
@@ -887,6 +970,7 @@ function parseNative(bytes, parser) {
 	if (parser.format === "edge-list") return parseEdgeList(text, parser);
 	if (parser.format === "matrix-market") return parseMatrixMarket(text);
 	if (parser.format === "gml") return parseGml(text);
+	if (parser.format === "solomon") return parseSolomon(text);
 	return parsePajek(text);
 }
 /** The ustar header layout (POSIX.1-1988, pax(1) "ustar Interchange Format"): each field's byte offset and length. */
@@ -1285,6 +1369,18 @@ function selects(path, member) {
 	return member.extensions.some((extension) => path.toLowerCase().endsWith(extension.toLowerCase()));
 }
 /**
+* Picks every member a selector matches, in path order, so a collection reads in the same order wherever it is unpacked.
+* @param paths - Every member path.
+* @param member - The recipe's selector.
+* @returns The matching paths, sorted, at least one.
+* @throws ArchiveError when no member matches, listing the members.
+*/
+function pickAll(paths, member) {
+	const matched = paths.filter((path) => selects(path, member)).sort((left, right) => left < right ? -1 : left > right ? 1 : 0);
+	if (matched.length === 0) throw new ArchiveError(`expected at least one ${"path" in member ? `member ${member.path}` : `member ending ${member.extensions.join(" or ")}`}, found none among: ${paths.join(", ")}`);
+	return matched;
+}
+/**
 * Picks the one member a selector matches.
 * @param paths - Every member path.
 * @param member - The recipe's selector.
@@ -1338,6 +1434,42 @@ function unpack(bytes, archive) {
 	};
 }
 /**
+* Unpacks downloaded bytes into every file a collection's selector matches.
+* @param bytes - The downloaded bytes.
+* @param archive - The recipe's archive layout, which selects members with `members`.
+* @returns Each matching file, in path order.
+* @throws ArchiveError when the bytes are not a well-formed archive of the stated kind or the selector matches no member.
+*/
+function unpackMembers(bytes, archive) {
+	if (archive.kind === "zip") {
+		const names = [];
+		unzip(bytes, (file) => {
+			names.push(file.name);
+			return false;
+		});
+		const paths = pickAll(names, archive.members);
+		const chosen = new Set(paths);
+		const files = unzip(bytes, (candidate) => chosen.has(candidate.name));
+		return paths.map((path) => {
+			const file = getOwn(files, path);
+			if (file === void 0) throw new ArchiveError(`zip member ${path} could not be read`);
+			return {
+				path,
+				bytes: file
+			};
+		});
+	}
+	const members = readTar(archive.kind === "tar" ? bytes : gunzip(bytes));
+	return pickAll(members.map((member) => member.path), archive.members).map((path) => {
+		const found = members.find((member) => member.path === path);
+		if (found === void 0) throw new ArchiveError(`tar member ${path} could not be read`);
+		return {
+			path,
+			bytes: found.data
+		};
+	});
+}
+/**
 * Gunzips bytes, rethrowing zlib's failure in the package's taxonomy.
 * @param bytes - Gzip data.
 * @returns The decompressed bytes.
@@ -1808,28 +1940,54 @@ async function runToNative(dataset, options) {
 		if (!response.ok) throw new KeepsetError(`${fetchedUrl} answered ${String(response.status)} ${response.statusText}`);
 		return new Uint8Array(await response.arrayBuffer());
 	});
-	const unpacked = await step(entry.id, "extract", () => unpack(bytes, recipe.archive));
+	const provenanceOf = (extra) => ({
+		entryId: entry.id,
+		catalogue: dataset.catalogue,
+		category: entry.category,
+		source: {
+			url: source.url,
+			status: source.status,
+			...source.snapshot === void 0 ? {} : { snapshot: source.snapshot }
+		},
+		fetchedUrl,
+		retrievedAt,
+		sha256: createHash("sha256").update(bytes).digest("hex"),
+		byteLength: bytes.byteLength,
+		...extra
+	});
+	const { archive } = recipe;
+	if ("members" in archive) {
+		const files = await step(entry.id, "extract", () => unpackMembers(bytes, archive));
+		const members = await step(entry.id, "parse", () => files.map((file) => {
+			try {
+				return {
+					id: file.path,
+					meta: {},
+					...parseNative(file.bytes, recipe.parser)
+				};
+			} catch (error) {
+				throw new KeepsetError(`member ${file.path}: ${error instanceof Error ? error.message : String(error)}`, { cause: error });
+			}
+		}));
+		return {
+			native: {
+				meta: metaOf(entry),
+				members
+			},
+			provenance: provenanceOf({ members: files.map((file) => file.path) })
+		};
+	}
+	const unpacked = await step(entry.id, "extract", () => unpack(bytes, archive));
 	const parsed = await step(entry.id, "parse", () => parseNative(unpacked.bytes, recipe.parser));
 	return {
 		native: {
-			meta: metaOf(entry),
-			...parsed
+			...parsed,
+			meta: {
+				..."meta" in parsed ? parsed.meta : {},
+				...metaOf(entry)
+			}
 		},
-		provenance: {
-			entryId: entry.id,
-			catalogue: dataset.catalogue,
-			category: entry.category,
-			source: {
-				url: source.url,
-				status: source.status,
-				...source.snapshot === void 0 ? {} : { snapshot: source.snapshot }
-			},
-			fetchedUrl,
-			retrievedAt,
-			sha256: createHash("sha256").update(bytes).digest("hex"),
-			byteLength: bytes.byteLength,
-			...unpacked.member === void 0 ? {} : { member: unpacked.member }
-		}
+		provenance: provenanceOf(unpacked.member === void 0 ? {} : { member: unpacked.member })
 	};
 }
 /**

package/dist-cli/keepset.mjs

@@ -4,7 +4,7 @@ import { KeepsetError, NAME_SEPARATOR, countTarget, getDataset, listDatasets, li
 import { fetchDataset, roundTripDataset } from "keepset/node";
 import { parseArgs } from "node:util";
 //#region package.json
-var version = "0.23.0";
+var version = "0.24.0";
 const usage = `Usage: keepset <command> [options]
 
 Commands:

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "keepset",
   "description": "Library, CLI and statically generated catalogue for dataset fetch-and-normalise recipes",
-  "version": "0.23.0",
+  "version": "0.24.0",
   "bin": {
     "keepset": "./dist-cli/keepset.mjs"
   },

package/schemas/entry.schema.json

@@ -1 +1 @@
-{"$defs":{"Crosswalk":{"$ref":"https://cdn.jsdelivr.net/npm/twindex@0.1.4/schemas/twindex.schema.json"},"Entry":{"additionalProperties":false,"properties":{"aliases":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*(?:\\/[A-Za-z0-9][A-Za-z0-9._-]*)*$","type":"string"},"minItems":1,"type":"array"},"category":{"enum":["graph/v1","or/v1"],"type":"string"},"citations":{"items":{"additionalProperties":false,"properties":{"doi":{"pattern":"^10\\.\\d{4,9}\\/\\S+$","type":"string"},"text":{"minLength":1,"type":"string"},"url":{"format":"uri","type":"string"}},"required":["text"],"type":"object"},"type":"array"},"crosswalk":{"anyOf":[{"$ref":"#/$defs/Crosswalk"},{"additionalProperties":false,"properties":{"file":{"type":"string"}},"required":["file"],"type":"object"}]},"description":{"minLength":1,"type":"string"},"domain":{"minLength":1,"type":"string"},"expectedCounts":{"additionalProperties":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"propertyNames":{"pattern":"^[A-Za-z][A-Za-z0-9]*$","type":"string"},"type":"object"},"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"licence":{"oneOf":[{"additionalProperties":false,"properties":{"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9.+-]*$","type":"string"},"kind":{"const":"spdx","type":"string"}},"required":["kind","id"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"custom","type":"string"},"name":{"minLength":1,"type":"string"},"url":{"format":"uri","type":"string"}},"required":["kind","name","url"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"unrecorded","type":"string"}},"required":["kind"],"type":"object"}]},"name":{"minLength":1,"type":"string"},"namespace":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"minItems":1,"type":"array"},"properties":{"additionalProperties":{"type":["string","number","boolean","null"]},"propertyNames":{"pattern":"^[A-Za-z][A-Za-z0-9]*$","type":"string"},"type":"object"},"recipe":{"oneOf":[{"additionalProperties":false,"properties":{"archive":{"oneOf":[{"additionalProperties":false,"properties":{"kind":{"const":"none","type":"string"}},"required":["kind"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"gzip","type":"string"}},"required":["kind"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"zip","type":"string"},"member":{"anyOf":[{"additionalProperties":false,"properties":{"path":{"type":"string"}},"required":["path"],"type":"object"},{"additionalProperties":false,"properties":{"extensions":{"items":{"pattern":"^\\.[A-Za-z0-9][A-Za-z0-9.]*$","type":"string"},"minItems":1,"type":"array"}},"required":["extensions"],"type":"object"}]}},"required":["kind","member"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"tar","type":"string"},"member":{"anyOf":[{"additionalProperties":false,"properties":{"path":{"type":"string"}},"required":["path"],"type":"object"},{"additionalProperties":false,"properties":{"extensions":{"items":{"pattern":"^\\.[A-Za-z0-9][A-Za-z0-9.]*$","type":"string"},"minItems":1,"type":"array"}},"required":["extensions"],"type":"object"}]}},"required":["kind","member"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"tar.gz","type":"string"},"member":{"anyOf":[{"additionalProperties":false,"properties":{"path":{"type":"string"}},"required":["path"],"type":"object"},{"additionalProperties":false,"properties":{"extensions":{"items":{"pattern":"^\\.[A-Za-z0-9][A-Za-z0-9.]*$","type":"string"},"minItems":1,"type":"array"}},"required":["extensions"],"type":"object"}]}},"required":["kind","member"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"xlsx","type":"string"},"sheet":{"minLength":1,"type":"string"}},"required":["kind","sheet"],"type":"object"}]},"paging":{"oneOf":[{"additionalProperties":false,"properties":{"kind":{"const":"arcgis-query","type":"string"},"orderBy":{"pattern":"^[A-Za-z_][A-Za-z0-9_]*$","type":"string"}},"required":["kind","orderBy"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"socrata-resource","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"ckan-datastore","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"openfema","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"}]},"parser":{"oneOf":[{"additionalProperties":false,"properties":{"commentPrefixes":{"items":{"minLength":1,"type":"string"},"type":"array"},"delimiter":{"enum":["detect","whitespace","tab","comma","space"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"edge-list","type":"string"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"sourceColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"targetColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","commentPrefixes","sourceColumn","targetColumn"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"matrix-market","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"gml","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"pajek","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"type":"array"},"delimiter":{"enum":["comma","tab","semicolon"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"csv-table","type":"string"},"idColumn":{"anyOf":[{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},{"type":"null"}]},"nullValues":{"items":{"minLength":1,"type":"string"},"type":"array"},"numericColumns":{"items":{"additionalProperties":false,"properties":{"column":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"factor":{"exclusiveMinimum":0,"type":"number"}},"required":["column","factor"],"type":"object"},"type":"array"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","idColumn","numericColumns"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"minItems":1,"type":"array"},"dateColumns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"type":"array"},"encoding":{"const":"utf-8","type":"string"},"format":{"const":"geojson","type":"string"},"idColumn":{"anyOf":[{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},{"type":"null"}]},"numericColumns":{"items":{"additionalProperties":false,"properties":{"column":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"factor":{"exclusiveMinimum":0,"type":"number"}},"required":["column","factor"],"type":"object"},"type":"array"}},"required":["format","encoding","idColumn","numericColumns","dateColumns","columns"],"type":"object"}]},"source":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"status":{"const":"fetchable","type":"string"}},"required":["status","source","archive","parser"],"type":"object"},{"additionalProperties":false,"properties":{"reason":{"minLength":1,"type":"string"},"status":{"const":"not-fetchable","type":"string"}},"required":["status","reason"],"type":"object"}]},"sources":{"items":{"additionalProperties":false,"properties":{"retrieved":{"format":"date","pattern":"^(?:(?:\\d\\d[2468][048]|\\d\\d[13579][26]|\\d\\d0[48]|[02468][048]00|[13579][26]00)-02-29|\\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\\d|30)|(?:02)-(?:0[1-9]|1\\d|2[0-8])))$","type":"string"},"snapshot":{"additionalProperties":false,"properties":{"timestamp":{"pattern":"^\\d{8}(?:\\d{6})?$","type":"string"},"url":{"format":"uri","type":"string"}},"required":["url","timestamp"],"type":"object"},"status":{"enum":["live","dead"],"type":"string"},"url":{"format":"uri","type":"string"}},"required":["url","retrieved","status"],"type":"object"},"minItems":1,"type":"array"},"tags":{"items":{"minLength":1,"type":"string"},"type":"array"}},"required":["id","name","tags","category","sources","licence","citations","recipe","crosswalk","expectedCounts","properties"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.23.0/schemas/entry.schema.json","$ref":"#/$defs/Entry","$schema":"https://json-schema.org/draft/2020-12/schema"}
+{"$defs":{"Crosswalk":{"$ref":"https://cdn.jsdelivr.net/npm/twindex@0.1.4/schemas/twindex.schema.json"},"Entry":{"additionalProperties":false,"properties":{"aliases":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*(?:\\/[A-Za-z0-9][A-Za-z0-9._-]*)*$","type":"string"},"minItems":1,"type":"array"},"category":{"enum":["graph/v1","or/v1","or-set/v1"],"type":"string"},"citations":{"items":{"additionalProperties":false,"properties":{"doi":{"pattern":"^10\\.\\d{4,9}\\/\\S+$","type":"string"},"text":{"minLength":1,"type":"string"},"url":{"format":"uri","type":"string"}},"required":["text"],"type":"object"},"type":"array"},"crosswalk":{"anyOf":[{"$ref":"#/$defs/Crosswalk"},{"additionalProperties":false,"properties":{"file":{"type":"string"}},"required":["file"],"type":"object"}]},"description":{"minLength":1,"type":"string"},"domain":{"minLength":1,"type":"string"},"expectedCounts":{"additionalProperties":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"propertyNames":{"pattern":"^[A-Za-z][A-Za-z0-9]*$","type":"string"},"type":"object"},"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"licence":{"oneOf":[{"additionalProperties":false,"properties":{"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9.+-]*$","type":"string"},"kind":{"const":"spdx","type":"string"}},"required":["kind","id"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"custom","type":"string"},"name":{"minLength":1,"type":"string"},"url":{"format":"uri","type":"string"}},"required":["kind","name","url"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"unrecorded","type":"string"}},"required":["kind"],"type":"object"}]},"name":{"minLength":1,"type":"string"},"namespace":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"minItems":1,"type":"array"},"properties":{"additionalProperties":{"type":["string","number","boolean","null"]},"propertyNames":{"pattern":"^[A-Za-z][A-Za-z0-9]*$","type":"string"},"type":"object"},"recipe":{"oneOf":[{"additionalProperties":false,"properties":{"archive":{"anyOf":[{"additionalProperties":false,"properties":{"kind":{"const":"none","type":"string"}},"required":["kind"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"gzip","type":"string"}},"required":["kind"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"enum":["zip","tar","tar.gz"],"type":"string"},"member":{"anyOf":[{"additionalProperties":false,"properties":{"path":{"type":"string"}},"required":["path"],"type":"object"},{"additionalProperties":false,"properties":{"extensions":{"items":{"pattern":"^\\.[A-Za-z0-9][A-Za-z0-9.]*$","type":"string"},"minItems":1,"type":"array"}},"required":["extensions"],"type":"object"}]}},"required":["kind","member"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"enum":["zip","tar","tar.gz"],"type":"string"},"members":{"anyOf":[{"additionalProperties":false,"properties":{"path":{"type":"string"}},"required":["path"],"type":"object"},{"additionalProperties":false,"properties":{"extensions":{"items":{"pattern":"^\\.[A-Za-z0-9][A-Za-z0-9.]*$","type":"string"},"minItems":1,"type":"array"}},"required":["extensions"],"type":"object"}]}},"required":["kind","members"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"xlsx","type":"string"},"sheet":{"minLength":1,"type":"string"}},"required":["kind","sheet"],"type":"object"}]},"paging":{"oneOf":[{"additionalProperties":false,"properties":{"kind":{"const":"arcgis-query","type":"string"},"orderBy":{"pattern":"^[A-Za-z_][A-Za-z0-9_]*$","type":"string"}},"required":["kind","orderBy"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"socrata-resource","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"ckan-datastore","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"openfema","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"}]},"parser":{"oneOf":[{"additionalProperties":false,"properties":{"commentPrefixes":{"items":{"minLength":1,"type":"string"},"type":"array"},"delimiter":{"enum":["detect","whitespace","tab","comma","space"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"edge-list","type":"string"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"sourceColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"targetColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","commentPrefixes","sourceColumn","targetColumn"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"matrix-market","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"gml","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"pajek","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"solomon","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"type":"array"},"delimiter":{"enum":["comma","tab","semicolon"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"csv-table","type":"string"},"idColumn":{"anyOf":[{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},{"type":"null"}]},"nullValues":{"items":{"minLength":1,"type":"string"},"type":"array"},"numericColumns":{"items":{"additionalProperties":false,"properties":{"column":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"factor":{"exclusiveMinimum":0,"type":"number"}},"required":["column","factor"],"type":"object"},"type":"array"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","idColumn","numericColumns"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"minItems":1,"type":"array"},"dateColumns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"type":"array"},"encoding":{"const":"utf-8","type":"string"},"format":{"const":"geojson","type":"string"},"idColumn":{"anyOf":[{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},{"type":"null"}]},"numericColumns":{"items":{"additionalProperties":false,"properties":{"column":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","type":"string"},"factor":{"exclusiveMinimum":0,"type":"number"}},"required":["column","factor"],"type":"object"},"type":"array"}},"required":["format","encoding","idColumn","numericColumns","dateColumns","columns"],"type":"object"}]},"source":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"status":{"const":"fetchable","type":"string"}},"required":["status","source","archive","parser"],"type":"object"},{"additionalProperties":false,"properties":{"reason":{"minLength":1,"type":"string"},"status":{"const":"not-fetchable","type":"string"}},"required":["status","reason"],"type":"object"}]},"sources":{"items":{"additionalProperties":false,"properties":{"retrieved":{"format":"date","pattern":"^(?:(?:\\d\\d[2468][048]|\\d\\d[13579][26]|\\d\\d0[48]|[02468][048]00|[13579][26]00)-02-29|\\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\\d|30)|(?:02)-(?:0[1-9]|1\\d|2[0-8])))$","type":"string"},"snapshot":{"additionalProperties":false,"properties":{"timestamp":{"pattern":"^\\d{8}(?:\\d{6})?$","type":"string"},"url":{"format":"uri","type":"string"}},"required":["url","timestamp"],"type":"object"},"status":{"enum":["live","dead"],"type":"string"},"url":{"format":"uri","type":"string"}},"required":["url","retrieved","status"],"type":"object"},"minItems":1,"type":"array"},"tags":{"items":{"minLength":1,"type":"string"},"type":"array"}},"required":["id","name","tags","category","sources","licence","citations","recipe","crosswalk","expectedCounts","properties"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.24.0/schemas/entry.schema.json","$ref":"#/$defs/Entry","$schema":"https://json-schema.org/draft/2020-12/schema"}

package/schemas/manifest.schema.json

@@ -1 +1 @@
-{"$defs":{"Manifest":{"additionalProperties":false,"properties":{"description":{"minLength":1,"type":"string"},"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"lineage":{"additionalProperties":false,"properties":{"catalogue":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["catalogue","version","repository"],"type":"object"},"name":{"minLength":1,"type":"string"},"namespaceAliases":{"additionalProperties":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"minItems":1,"type":"array"},"propertyNames":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"type":"object"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["id","name","description","version","repository"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.23.0/schemas/manifest.schema.json","$ref":"#/$defs/Manifest","$schema":"https://json-schema.org/draft/2020-12/schema"}
+{"$defs":{"Manifest":{"additionalProperties":false,"properties":{"description":{"minLength":1,"type":"string"},"id":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"lineage":{"additionalProperties":false,"properties":{"catalogue":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["catalogue","version","repository"],"type":"object"},"name":{"minLength":1,"type":"string"},"namespaceAliases":{"additionalProperties":{"items":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"minItems":1,"type":"array"},"propertyNames":{"pattern":"^[A-Za-z0-9][A-Za-z0-9._-]*$","type":"string"},"type":"object"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["id","name","description","version","repository"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.24.0/schemas/manifest.schema.json","$ref":"#/$defs/Manifest","$schema":"https://json-schema.org/draft/2020-12/schema"}