keepset 0.26.0 → 0.27.0

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. `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.
+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` (by path, by the extensions its name ends with, or by a regular expression its whole path matches), the parser reads each one, and a member that does not parse fails the fetch and is named. Instance files are read by one parser per format (Solomon and TOPTW time window files, Chao team orienteering files, TSPLIB style orienteering files, VRP-REP XML, VeRoLog 2019 files and customer point lists), which gives each node the columns the file has and each instance a header of its own; a curation says which of them fill the `or/v1` node's fields, which pass through under their own names and which are dropped, so what an entry leaves out is written down, and a source whose time quantities are not seconds keeps them under the source's names with the unit recorded as the entry's `timeUnit`. 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-V1EJ0k9e.js → bundled-qRweq3SB.js}

@@ -453,8 +453,21 @@ const CitationSchema = z.object({
 }).strict();
 /** A relative path inside an archive or a catalogue: `/`-separated segments, none empty, `.` or `..`, so it can never climb out of the directory it is resolved against. */
 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.]*$/, { message: "must be a file extension such as .mtx" })).min(1) }).strict()]);
+/** A regular expression a member path must match in full, in JavaScript syntax; it must compile, so a recipe with a malformed pattern is refused when the entry is parsed and not when it runs. */
+const MemberPatternSchema = z.string().min(1).refine((pattern) => {
+	try {
+		new RegExp(`^(?:${pattern})$`, "u");
+		return true;
+	} catch {
+		return false;
+	}
+}, { message: "must be a valid regular expression" });
+/** Which file of an archive to read: a member path, matched against a member's whole path or its trailing segments (so `net.gml` finds `net/net.gml`), the member whose name ends with one of a list of extensions (a format can have several, such as Pajek's `.net` and `.paj`), or the member whose whole path matches a regular expression (for an archive whose instance files cannot be told from its readmes and solutions by extension). A `member` selector must match exactly one member when the recipe runs, a `members` selector at least one; macOS resource-fork members (`__MACOSX/` and `._` files) never match. */
+const ArchiveMemberSchema = z.union([
+	z.object({ path: RelativePathSchema }).strict(),
+	z.object({ extensions: z.array(z.string().regex(/^\.[A-Za-z0-9][A-Za-z0-9.]*$/, { message: "must be a file extension such as .mtx" })).min(1) }).strict(),
+	z.object({ pattern: MemberPatternSchema }).strict()
+]);
 /** 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",
@@ -517,6 +530,10 @@ const ParserSchema = z.discriminatedUnion("format", [
 		format: z.literal("pajek"),
 		encoding: EncodingSchema
 	}).strict(),
+	z.object({
+		format: z.literal("customer-points"),
+		encoding: EncodingSchema
+	}).strict(),
 	z.object({
 		format: z.literal("solomon"),
 		encoding: EncodingSchema
@@ -533,6 +550,14 @@ const ParserSchema = z.discriminatedUnion("format", [
 		format: z.literal("tsplib"),
 		encoding: EncodingSchema
 	}).strict(),
+	z.object({
+		format: z.literal("verolog-2019"),
+		encoding: z.literal("utf-8")
+	}).strict(),
+	z.object({
+		format: z.literal("vrp-rep"),
+		encoding: z.literal("utf-8")
+	}).strict(),
 	z.object({
 		format: z.literal("csv-table"),
 		encoding: EncodingSchema,
@@ -992,7 +1017,7 @@ 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": "1.14.0",
+		"version": "1.17.0",
 		"repository": "https://github.com/Mearman/keepset",
 		"namespaceAliases": {
 			"networkrepository": ["nr"],
@@ -7081,7 +7106,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The instances are keyword format text files of matrices (a collection of many, indexed by dataset.csv), and a collection is more than one or/v1 document, so no recipe reads it."
+				"reason": "The instances are keyword format text files with node positions, time windows, service durations and a travel time matrix, but the repository holds them as a very large number of separate files in size directories and publishes no archive of the curated selection, so a recipe has no single download to read; the repository's own archive holds every instance and is several gigabytes."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -7099,20 +7124,153 @@ var catalogue_default = {
 				"skills",
 				"travelTime"
 			],
-			"category": "or/v1",
+			"category": "or-set/v1",
 			"sources": [{
 				"url": "https://github.com/BasCoding/VeRoLog-2019",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://codeload.github.com/BasCoding/VeRoLog-2019/zip/refs/heads/main",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": { "kind": "unrecorded" },
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The instances are keyword format text files in a collection mirrored by a student without a licence statement, and a collection is more than one or/v1 document, so no recipe reads it."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"members": { "pattern": "VeRoLog-2019-main/All Instances/VSC2019_(?:EX\\d\\d|ORTEC_early_\\d\\d)\\.txt" }
+				},
+				"parser": {
+					"format": "verolog-2019",
+					"encoding": "utf-8"
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
+			"crosswalk": {
+				"source": "verolog-2019",
+				"target": "or-set/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].id",
+						"to": "instances[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.dataset",
+						"to": "instances[].meta.dataset"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.name",
+						"to": "instances[].meta.name"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.days",
+						"to": "instances[].meta.days"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.truckCapacity",
+						"to": "instances[].meta.truckCapacity"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.truckMaxDistance",
+						"to": "instances[].meta.truckMaxDistance"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.truckDistanceCost",
+						"to": "instances[].meta.truckDistanceCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.truckDayCost",
+						"to": "instances[].meta.truckDayCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.truckCost",
+						"to": "instances[].meta.truckCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.technicianDistanceCost",
+						"to": "instances[].meta.technicianDistanceCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.technicianDayCost",
+						"to": "instances[].meta.technicianDayCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.technicianCost",
+						"to": "instances[].meta.technicianCost"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.technicians",
+						"to": "instances[].meta.technicians"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].id",
+						"to": "instances[].nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].machine",
+						"to": "instances[].nodes[].skill"
+					},
+					{
+						"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[].kind",
+						"to": "instances[].nodes[].kind"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].location",
+						"to": "instances[].nodes[].location"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].fromDay",
+						"to": "instances[].nodes[].fromDay"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].toDay",
+						"to": "instances[].nodes[].toDay"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].amount",
+						"to": "instances[].nodes[].amount"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "instances": 27 },
 			"properties": { "surveyId": "or-24" }
 		},
 		{
@@ -17178,7 +17336,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The instances are keyword format text files on a Wayback Machine snapshot, a collection of separate files, and a collection is more than one or/v1 document, so no recipe reads it."
+				"reason": "The instances are keyword format text files that the Wayback Machine serves one at a time, as the archive zips of each set were not captured, and a recipe reads one download, so no recipe reaches the collection."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -56436,7 +56594,7 @@ var catalogue_default = {
 			"description": "Park-and-loop routing instances built from real client locations in France, with coordinates and service times in minutes.",
 			"domain": "Vehicle routing with time windows",
 			"tags": ["location", "duration"],
-			"category": "or/v1",
+			"category": "or-set/v1",
 			"sources": [{
 				"url": "http://www.vrp-rep.org/datasets/download/cabrera-et-al-2021.zip",
 				"retrieved": "2026-10-19",
@@ -56449,12 +56607,86 @@ var catalogue_default = {
 			"licence": { "kind": "unrecorded" },
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is a collection of separate coordinate files plus a readme, and a collection is more than one or/v1 document, so no recipe reads it."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"members": { "pattern": "Coordinates_\\d+\\.txt" }
+				},
+				"parser": {
+					"format": "csv-table",
+					"encoding": "utf-8",
+					"delimiter": "comma",
+					"skipLines": 0,
+					"idColumn": "ID",
+					"numericColumns": [
+						{
+							"column": "Longitude",
+							"factor": 1
+						},
+						{
+							"column": "Latitude",
+							"factor": 1
+						},
+						{
+							"column": "ServiceTime_min",
+							"factor": 60
+						}
+					],
+					"columns": [
+						"ServiceTime_min",
+						"Latitude",
+						"Longitude"
+					]
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
-			"properties": { "surveyId": "or-10" }
+			"crosswalk": {
+				"source": "csv-table",
+				"target": "or-set/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].id",
+						"to": "instances[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta",
+						"to": "instances[].meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].id",
+						"to": "instances[].nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].ServiceTime_min",
+						"to": "instances[].nodes[].duration"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].Latitude",
+						"to": "instances[].nodes[].lat"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].Longitude",
+						"to": "instances[].nodes[].lon"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "instances": 60 },
+			"properties": {
+				"surveyId": "or-10",
+				"timeUnit": "seconds"
+			}
 		},
 		{
 			"id": "vrp-rep-probabilistic-orienteering",
@@ -56593,20 +56825,93 @@ var catalogue_default = {
 				"duration",
 				"travelTime"
 			],
-			"category": "or/v1",
+			"category": "or-set/v1",
 			"sources": [{
 				"url": "http://www.vrp-rep.org/datasets/download/pillac-et-al-2013.zip",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://www.vrp-rep.org/datasets/download/pillac-et-al-2013.zip",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": { "kind": "unrecorded" },
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is a collection of VRP-REP XML instance files holding coordinates and skills but no times, and a collection is more than one or/v1 document, so no recipe reads it."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"members": { "extensions": [".xml"] }
+				},
+				"parser": {
+					"format": "vrp-rep",
+					"encoding": "utf-8"
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
+			"crosswalk": {
+				"source": "vrp-rep",
+				"target": "or-set/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "members[].id",
+						"to": "instances[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.dataset",
+						"to": "instances[].meta.dataset"
+					},
+					{
+						"op": "move",
+						"from": "members[].meta.name",
+						"to": "instances[].meta.name"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].id",
+						"to": "instances[].nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].skill",
+						"to": "instances[].nodes[].skill"
+					},
+					{
+						"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[].type",
+						"to": "instances[].nodes[].type"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].skills",
+						"to": "instances[].nodes[].skills"
+					},
+					{
+						"op": "move",
+						"from": "members[].nodes[].resources",
+						"to": "instances[].nodes[].resources"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "instances": 57 },
 			"properties": { "surveyId": "or-21" }
 		},
 		{
@@ -57655,6 +57960,10 @@ var catalogue_default = {
 				"url": "https://zenodo.org/records/5823834",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://zenodo.org/records/5823834/files/MeterReading_Instance.zip?download=1",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": {
 				"kind": "spdx",
@@ -57662,11 +57971,45 @@ var catalogue_default = {
 			},
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is one arc routing instance in two tab separated text files with no column names, holding only planar coordinates of customers and a road network, so there is no field of the or/v1 node beyond a position and no table the parser reads."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"member": { "path": "customers.txt" }
+				},
+				"parser": {
+					"format": "customer-points",
+					"encoding": "utf-8"
+				}
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
-			"expectedCounts": {},
+			"crosswalk": {
+				"source": "customer-points",
+				"target": "or/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].id",
+						"to": "nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].x",
+						"to": "nodes[].x"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].y",
+						"to": "nodes[].y"
+					}
+				],
+				"oneWay": []
+			},
+			"expectedCounts": { "nodes": 6067 },
 			"properties": { "surveyId": "or-6" }
 		},
 		{
@@ -57734,7 +58077,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The Zenodo files are a custom sectioned text format (technicians, tasks and a travel time matrix of millions of lines) and not a table, and two of them are collections of instance files."
+				"reason": "The instances are a custom sectioned text format whose tasks have no positions, and the collection is hundreds of megabytes compressed, almost all of it a pairwise travel time matrix that or/v1 has no arcs to hold, so reading it would hold gigabytes in memory to keep a few fields per task."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},