keepset 0.21.0 → 0.22.0

package/dist/{bundled-NjZZ73vT.js → bundled-CAsUNCDb.js}

@@ -482,7 +482,9 @@ const ParserSchema = z.discriminatedUnion("format", [
 			column: ColumnNameSchema,
 			factor: z.number().positive()
 		}).strict()),
-		columns: z.array(ColumnNameSchema).optional()
+		columns: z.array(ColumnNameSchema).optional(),
+		/** Cell texts, besides the empty cell, that stand for a missing value (`NA`): such a cell is null in every column, numeric or not. */
+		nullValues: z.array(z.string().min(1)).optional()
 	}).strict().refine((parser) => new Set(parser.numericColumns.map((numeric) => numeric.column)).size === parser.numericColumns.length, {
 		message: "a column may be declared numeric once",
 		path: ["numericColumns"]
@@ -514,14 +516,21 @@ const ParserSchema = z.discriminatedUnion("format", [
 		path: ["dateColumns"]
 	})
 ]);
-/** Where a source is a paged query and not one file. An ArcGIS feature layer returns a limited number of features per request: the source URL is the layer's, the download reads every page in `orderBy` order and the parser reads the one GeoJSON collection they make. A Socrata resource endpoint (`/resource/<id>.csv`) is read `pageSize` rows at a time in the order of its row ids, each page a CSV with its own header, and the download joins them into the one CSV the parser reads, so no single response has to survive a long transfer. */
-const PagingSchema = z.discriminatedUnion("kind", [z.object({
-	kind: z.literal("arcgis-query"),
-	orderBy: z.string().regex(/^[A-Za-z_][A-Za-z0-9_]*$/, { message: "must be a field name of the layer" })
-}).strict(), z.object({
-	kind: z.literal("socrata-resource"),
-	pageSize: z.int().min(1)
-}).strict()]);
+/** Where a source is a paged query and not one file. An ArcGIS feature layer returns a limited number of features per request: the source URL is the layer's, the download reads every page in `orderBy` order and the parser reads the one GeoJSON collection they make. A Socrata resource endpoint (`/resource/<id>.csv`) is read `pageSize` rows at a time in the order of its row ids, and a CKAN datastore dump (`/datastore/dump/<id>`) `pageSize` rows at a time, each page a CSV with its own header; the download joins them into the one CSV the parser reads, so no single response has to survive a long transfer. */
+const PagingSchema = z.discriminatedUnion("kind", [
+	z.object({
+		kind: z.literal("arcgis-query"),
+		orderBy: z.string().regex(/^[A-Za-z_][A-Za-z0-9_]*$/, { message: "must be a field name of the layer" })
+	}).strict(),
+	z.object({
+		kind: z.literal("socrata-resource"),
+		pageSize: z.int().min(1)
+	}).strict(),
+	z.object({
+		kind: z.literal("ckan-datastore"),
+		pageSize: z.int().min(1)
+	}).strict()
+]);
 /** How to obtain the dataset: a fetchable recipe naming the source (by index into `sources`), the archive layout and the parser, or a statement that no declarative recipe exists yet and why. */
 const RecipeSchema = z.discriminatedUnion("status", [z.object({
 	status: z.literal("fetchable"),
@@ -896,7 +905,7 @@ 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": "1.9.0",
+		"version": "1.10.0",
 		"repository": "https://github.com/Mearman/keepset",
 		"namespaceAliases": {
 			"networkrepository": ["nr"],
@@ -1319,6 +1328,10 @@ var catalogue_default = {
 			"recipe": {
 				"status": "fetchable",
 				"source": 1,
+				"paging": {
+					"kind": "ckan-datastore",
+					"pageSize": 5e4
+				},
 				"archive": { "kind": "none" },
 				"parser": {
 					"format": "csv-table",
@@ -3430,7 +3443,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The bulk download is a zip of 355,930,383 bytes covering every product, which is larger than a recipe can read, and the debt collection subset is only reachable through a search API."
+				"reason": "The bulk download is a zip of 355,930,383 bytes covering every product, and the debt collection subset is only reachable through a search API that keepset cannot page."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -4824,7 +4837,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The CSV starts with a title line and a production note before any column header, so it is not a rectangular table with the header on its first line."
+				"reason": "The CSV has a title and a production note above its header and its address contains the month it was produced, so it changes every month and has no stable download URL."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -5469,7 +5482,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The survey URL is a landing page and no direct download could be verified, because the files sit on a host whose TLS connection failed."
+				"reason": "The yearly download is a zip holding a ten sheet xlsx workbook with a title row above each header, durations as Excel day fractions and outlet positions only as National Grid references, so a workbook inside a zip with no coordinates in degrees has no recipe; the host the landing page names has no address, so the files are fetched through the catalogue's download API."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -5499,7 +5512,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The survey URL is a landing page; the zip it names is hosted on api.agrimetrics.co.uk, whose TLS connection failed during the survey, so no direct download or table could be verified, and the page lists the format as ODS (a spreadsheet) which keepset has no parser for yet."
+				"reason": "The download is a zip holding one OpenDocument spreadsheet whose member name changes every quarter, and keepset reads xlsx and not ods."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6406,7 +6419,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The data is served only as JSON pages by the OpenFEMA API (the survey page itself refuses automated requests), keepset has no JSON paging parser yet, and the successor v3 table holds 2725989 records, more than a recipe can read."
+				"reason": "The OpenFEMA v3 API pages claims anonymously as CSV, but the table has millions of rows whose locations are rounded to a tenth of a degree, the terms of use limit it to statistical use without reidentification, and keepset has no recipe paging for this API."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6657,7 +6670,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of shapefile layers, and keepset has no shapefile parser yet."
+				"reason": "The download is a per-request zip of a point shapefile (the attributes are in a dbf and the positions in a separate shp, in a projected system) with no CSV or GeoJSON route, and keepset has no shapefile parser."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6905,7 +6918,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The instances are keyword-format text files that keepset has no parser for yet, and dataset.csv only indexes the selected instances."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6933,7 +6946,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The instances are keyword-format text files (and semicolon-delimited copies with the same content) that keepset has no parser for yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6965,7 +6978,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The plain CSV probed holds only an index, date, dispatch time and order id with no mappable field, and the coordinate data sits in a zip that keepset cannot read."
+				"reason": "The data is released under CC BY-NC 4.0 with a bar on passing it to third parties without permission, and its coordinates are shifted by the publisher, so they are not WGS84 positions."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -6993,7 +7006,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The only data file is an xls spreadsheet and keepset has no spreadsheet parser yet."
+				"reason": "The only data file is an old binary xls workbook, and keepset reads only xlsx."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -7029,7 +7042,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of Solomon-style instance text files and tab-separated matrices, and keepset has no parser for that format yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -7928,6 +7941,10 @@ var catalogue_default = {
 				"url": "https://www.kaggle.com/datasets/dharun4772/doordash-eta-prediction",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://www.kaggle.com/api/v1/datasets/download/dharun4772/doordash-eta-prediction",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": {
 				"kind": "spdx",
@@ -7935,10 +7952,83 @@ var catalogue_default = {
 			},
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The download is a zip archive holding one CSV and keepset has no archive handling yet, so the table cannot be read directly from the download."
+				"status": "fetchable",
+				"source": 1,
+				"archive": {
+					"kind": "zip",
+					"member": { "path": "historical_data.csv" }
+				},
+				"parser": {
+					"format": "csv-table",
+					"encoding": "utf-8",
+					"delimiter": "comma",
+					"skipLines": 0,
+					"idColumn": null,
+					"numericColumns": [{
+						"column": "estimated_order_place_duration",
+						"factor": 1
+					}, {
+						"column": "estimated_store_to_consumer_driving_duration",
+						"factor": 1
+					}],
+					"columns": [
+						"actual_delivery_time",
+						"estimated_order_place_duration",
+						"created_at",
+						"store_primary_category",
+						"market_id",
+						"estimated_store_to_consumer_driving_duration"
+					],
+					"nullValues": ["NA"]
+				}
+			},
+			"crosswalk": {
+				"source": "csv-table",
+				"target": "or/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].id",
+						"to": "nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].actual_delivery_time",
+						"to": "nodes[].closed"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].estimated_order_place_duration",
+						"to": "nodes[].duration"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].created_at",
+						"to": "nodes[].opened"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].store_primary_category",
+						"to": "nodes[].skill"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].market_id",
+						"to": "nodes[].territory"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].estimated_store_to_consumer_driving_duration",
+						"to": "nodes[].travelTime"
+					}
+				],
+				"oneWay": []
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
 			"properties": { "surveyId": "or-31" }
 		},
@@ -7967,7 +8057,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip archive holding one CSV and keepset has no archive handling yet, so the table cannot be read directly from the download."
+				"reason": "The records are patient level health data under CC BY-NC-SA 4.0 with no location column."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -8325,7 +8415,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of instance files in a plain text format with a small header and no column names, and keepset has no parser for it yet."
+				"reason": "The download is a collection of 60 text instance files of one 100 node point set under different budgets (planar coordinates and a prize, with no time quantities), and a collection is more than one or/v1 document, so no recipe reads it."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -8359,7 +8449,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of instance files in a Solomon style text format with no column names, and keepset has no parser for it yet."
+				"reason": "The download is a collection of Solomon style time window instance files (planar coordinates, service durations, prizes and windows in abstract units, not seconds), and a collection is more than one or/v1 document, so no recipe reads it."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -8426,7 +8516,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of plain-text Solomon instance files and keepset has no parser for that format yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -8544,7 +8634,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The CSV has a title line (FPN Detail) above the real header, so it does not have its header on the first line and is not a plain table."
+				"reason": "Each quarter is its own file at its own tokenised address, the later quarters are xlsx, and each CSV has a title row above its header and a summary block and blank rows below its data."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -9540,7 +9630,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The downloads are MATLAB data files and keepset has no parser for them yet."
+				"reason": "The downloads are MATLAB data files and keepset has no parser for them."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -9579,7 +9669,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a spreadsheet (xlsx) with task, team and distance matrix sheets and no coordinates, and keepset has no spreadsheet parser yet."
+				"reason": "The download is hundreds of small Excel workbooks whose task sheets have no coordinates (places exist only as distance matrices) and whose file names repeat across folders, so no recipe can name one table."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -9610,7 +9700,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The only file is an Excel workbook of results and keepset has no spreadsheet parser yet."
+				"reason": "The only file is an Excel workbook of aggregate results per solved instance, with no row for a place or a task, so there are no nodes to map."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -9674,7 +9764,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a RAR archive of macro-enabled Excel workbooks and keepset has no archive or spreadsheet parser yet."
+				"reason": "The download is a RAR archive of macro enabled Excel workbooks, and keepset reads neither RAR archives nor macro workbooks."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -9709,7 +9799,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The only file is an xlsx workbook of about 19.7 MB, and keepset has no spreadsheet parser yet."
+				"reason": "The only file is an Excel workbook with a row per stop visit of a route that holds minute offsets and distances between stops and no location, so the only node field it could fill is a time window and a fetch would carry no place."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -16620,7 +16710,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The instances are keyword-format text files on a Wayback Machine snapshot and keepset has no parser for them yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -16647,7 +16737,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The source is an SRU search endpoint that returns XML notices in a query language, not a table, and keepset has no parser for it."
+				"reason": "The source is an SRU search endpoint that returns XML notices in a query language and not a table, and keepset has no parser for it."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -20723,6 +20813,10 @@ var catalogue_default = {
 				"url": "https://www.data.qld.gov.au/dataset/unclaimed-monies",
 				"retrieved": "2026-10-19",
 				"status": "live"
+			}, {
+				"url": "https://www.data.qld.gov.au/datastore/dump/872065ae-ddfd-4b5f-ad15-e1935dadd883",
+				"retrieved": "2026-10-19",
+				"status": "live"
 			}],
 			"licence": {
 				"kind": "spdx",
@@ -20730,10 +20824,62 @@ var catalogue_default = {
 			},
 			"citations": [],
 			"recipe": {
-				"status": "not-fetchable",
-				"reason": "The CSV download answers with an AWS WAF challenge and an empty body, the CKAN datastore API is paged JSON that keepset cannot read yet, and the table has 2553652 records, more than a recipe can read."
+				"status": "fetchable",
+				"source": 1,
+				"paging": {
+					"kind": "ckan-datastore",
+					"pageSize": 5e4
+				},
+				"archive": { "kind": "none" },
+				"parser": {
+					"format": "csv-table",
+					"encoding": "utf-8",
+					"delimiter": "comma",
+					"skipLines": 0,
+					"idColumn": "id",
+					"numericColumns": [{
+						"column": "Amount",
+						"factor": 1
+					}],
+					"columns": [
+						"DateRec",
+						"Amount",
+						"PCode"
+					]
+				}
+			},
+			"crosswalk": {
+				"source": "csv-table",
+				"target": "or/v1",
+				"rules": [
+					{
+						"op": "move",
+						"from": "meta",
+						"to": "meta"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].id",
+						"to": "nodes[].id"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].DateRec",
+						"to": "nodes[].opened"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].Amount",
+						"to": "nodes[].prize"
+					},
+					{
+						"op": "move",
+						"from": "nodes[].PCode",
+						"to": "nodes[].territory"
+					}
+				],
+				"oneWay": []
 			},
-			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
 			"properties": { "surveyId": "or-124" }
 		},
@@ -21321,7 +21467,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a set of zip archives of CSV files, one per financial year and report type with differing columns, and keepset has no archive parser yet."
+				"reason": "The portal answered a bot challenge to automated requests for the table dumps when probed, and the data is a file per financial year and report type with different columns before and after 2022, so no single recipe covers the dataset."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -55709,7 +55855,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of TSPLIB-style .cop instance files plus a readme and an xlsx workbook, and keepset has no parser for that format yet."
+				"reason": "The download is a collection of TSPLIB style instance files whose prizes belong to overlapping clusters of nodes and not to nodes, and a collection is more than one or/v1 document, so no recipe reads it."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -55737,7 +55883,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip holding many separate coordinate files plus a readme, and keepset has no zip handling for a recipe yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -55765,7 +55911,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of TSPLIB-style text files with a POP section and keepset has no parser for it yet."
+				"reason": "The download is a collection of TSPLIB style instance files with a prize and probability section, half of which carry no coordinates and some of which use the degrees and minutes form, and a collection is more than one or/v1 document, so no recipe reads it."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -55794,7 +55940,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of instance files in the VRP-REP XML format and keepset has no parser for it yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -56854,7 +57000,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of plain-text arc-routing instance files (customers, network and links) described by a readme, and keepset has no parser for that format yet."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -56890,7 +57036,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "The download is a zip of Excel (xlsx) files holding distance and time matrices, and keepset has no spreadsheet parser yet."
+				"reason": "The download is a zip of Excel workbooks that hold distance and time matrices only, so there are no nodes to map, and keepset reads a sheet of a workbook, not a workbook inside a zip."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},
@@ -56925,7 +57071,7 @@ var catalogue_default = {
 			"citations": [],
 			"recipe": {
 				"status": "not-fetchable",
-				"reason": "Zenodo refused automated requests, so the file list and formats could not be seen and no direct download could be verified."
+				"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."
 			},
 			"crosswalk": { "file": "crosswalks/or-v1/nodes.json" },
 			"expectedCounts": {},

package/dist/{categories-frr8n_SI.d.ts → categories-CJZ2LiJh.d.ts}

@@ -157,6 +157,7 @@ declare const ParserSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
     factor: z.ZodNumber;
   }, z.core.$strict>>;
   columns: z.ZodOptional<z.ZodArray<z.ZodString>>;
+  nullValues: z.ZodOptional<z.ZodArray<z.ZodString>>;
 }, z.core.$strict>, z.ZodObject<{
   format: z.ZodLiteral<"geojson">;
   encoding: z.ZodLiteral<"utf-8">;
@@ -179,6 +180,9 @@ declare const RecipeSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
   }, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"socrata-resource">;
     pageSize: z.ZodInt;
+  }, z.core.$strict>, z.ZodObject<{
+    kind: z.ZodLiteral<"ckan-datastore">;
+    pageSize: z.ZodInt;
   }, z.core.$strict>], "kind">>;
   archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
     kind: z.ZodLiteral<"none">;
@@ -262,6 +266,7 @@ declare const RecipeSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
       factor: z.ZodNumber;
     }, z.core.$strict>>;
     columns: z.ZodOptional<z.ZodArray<z.ZodString>>;
+    nullValues: z.ZodOptional<z.ZodArray<z.ZodString>>;
   }, z.core.$strict>, z.ZodObject<{
     format: z.ZodLiteral<"geojson">;
     encoding: z.ZodLiteral<"utf-8">;
@@ -338,6 +343,9 @@ declare const EntrySchema: z.ZodObject<{
     }, z.core.$strict>, z.ZodObject<{
       kind: z.ZodLiteral<"socrata-resource">;
       pageSize: z.ZodInt;
+    }, z.core.$strict>, z.ZodObject<{
+      kind: z.ZodLiteral<"ckan-datastore">;
+      pageSize: z.ZodInt;
     }, z.core.$strict>], "kind">>;
     archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
       kind: z.ZodLiteral<"none">;
@@ -421,6 +429,7 @@ declare const EntrySchema: z.ZodObject<{
         factor: z.ZodNumber;
       }, z.core.$strict>>;
       columns: z.ZodOptional<z.ZodArray<z.ZodString>>;
+      nullValues: z.ZodOptional<z.ZodArray<z.ZodString>>;
     }, z.core.$strict>, z.ZodObject<{
       format: z.ZodLiteral<"geojson">;
       encoding: z.ZodLiteral<"utf-8">;
@@ -524,6 +533,7 @@ declare const EntrySchema: z.ZodObject<{
           factor: number;
         }[];
         columns?: string[] | undefined;
+        nullValues?: string[] | undefined;
       } | {
         format: "geojson";
         encoding: "utf-8";
@@ -550,6 +560,9 @@ declare const EntrySchema: z.ZodObject<{
       } | {
         kind: "socrata-resource";
         pageSize: number;
+      } | {
+        kind: "ckan-datastore";
+        pageSize: number;
       } | undefined;
     } | {
       status: "not-fetchable";
@@ -707,6 +720,9 @@ declare const CatalogueIndexSchema: z.ZodObject<{
       }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"socrata-resource">;
         pageSize: z.ZodInt;
+      }, z.core.$strict>, z.ZodObject<{
+        kind: z.ZodLiteral<"ckan-datastore">;
+        pageSize: z.ZodInt;
       }, z.core.$strict>], "kind">>;
       archive: z.ZodDiscriminatedUnion<[z.ZodObject<{
         kind: z.ZodLiteral<"none">;
@@ -790,6 +806,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
           factor: z.ZodNumber;
         }, z.core.$strict>>;
         columns: z.ZodOptional<z.ZodArray<z.ZodString>>;
+        nullValues: z.ZodOptional<z.ZodArray<z.ZodString>>;
       }, z.core.$strict>, z.ZodObject<{
         format: z.ZodLiteral<"geojson">;
         encoding: z.ZodLiteral<"utf-8">;
@@ -893,6 +910,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
             factor: number;
           }[];
           columns?: string[] | undefined;
+          nullValues?: string[] | undefined;
         } | {
           format: "geojson";
           encoding: "utf-8";
@@ -919,6 +937,9 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           kind: "socrata-resource";
           pageSize: number;
+        } | {
+          kind: "ckan-datastore";
+          pageSize: number;
         } | undefined;
       } | {
         status: "not-fetchable";
@@ -1031,6 +1052,7 @@ declare const CatalogueIndexSchema: z.ZodObject<{
             factor: number;
           }[];
           columns?: string[] | undefined;
+          nullValues?: string[] | undefined;
         } | {
           format: "geojson";
           encoding: "utf-8";
@@ -1057,6 +1079,9 @@ declare const CatalogueIndexSchema: z.ZodObject<{
         } | {
           kind: "socrata-resource";
           pageSize: number;
+        } | {
+          kind: "ckan-datastore";
+          pageSize: number;
         } | undefined;
       } | {
         status: "not-fetchable";

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-frr8n_SI.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-CJZ2LiJh.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-NjZZ73vT.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-CAsUNCDb.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-frr8n_SI.js";
+import { G as Entry, T as Dataset, at as Snapshot, i as CategoryTarget, st as Source } from "../categories-CJZ2LiJh.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: {

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-NjZZ73vT.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-CAsUNCDb.js";
 import { RoundTripMismatchError, assertRoundTrip, encode } from "twindex";
 import { z } from "zod";
 import { createHash } from "node:crypto";
@@ -574,7 +574,7 @@ function* tableRecords(text, options) {
 	for (const record of records(body, DELIMITERS[options.delimiter], options.skipLines + 1)) if (record.fields.length > 1 || record.fields[0]?.trim() !== "") yield record;
 }
 /**
-* Parses a delimited table into rows: the first record after `skipLines` lines is the header, each later record is a row, and every row carries every column but the id column (or, when the recipe lists `columns`, exactly those, so a wide table is not held cell for cell), an empty cell being null. Cells are trimmed. A column the recipe declares numeric is read as a decimal number times its factor and any other column stays a string. `id` is the row's id, so a column of that name which is not the id column is named by its position. A row's id is the cell of the recipe's id column, which is then not also a column, or its 1-based row number when the recipe names none. Blank lines are skipped; a row with a different number of fields than the header, two columns with the same name, and an id cell that is empty are errors.
+* Parses a delimited table into rows (a cell whose text is one of the recipe's `nullValues` is empty, so null): the first record after `skipLines` lines is the header, each later record is a row, and every row carries every column but the id column (or, when the recipe lists `columns`, exactly those, so a wide table is not held cell for cell), an empty cell being null. Cells are trimmed. A column the recipe declares numeric is read as a decimal number times its factor and any other column stays a string. `id` is the row's id, so a column of that name which is not the id column is named by its position. A row's id is the cell of the recipe's id column, which is then not also a column, or its 1-based row number when the recipe names none. Blank lines are skipped; a row with a different number of fields than the header, two columns with the same name, and an id cell that is empty are errors.
 * @param text - The file text, or its pieces in order.
 * @param options - The recipe's table parser.
 * @returns The rows, in file order.
@@ -595,6 +595,7 @@ function parseCsvTable(text, options) {
 	if (idColumn !== null && wanted.has(idColumn)) throw new NativeParseError("csv-table", header.line, `the id column ${idColumn} is the row's id, so it is not listed among the columns`);
 	for (const { column } of numericColumns) if (!wanted.has(column)) throw new NativeParseError("csv-table", header.line, `the column ${column} is declared numeric but is not among the columns kept`);
 	const factors = new Map(numericColumns.map((entry) => [entry.column, entry.factor]));
+	const missing = new Set(options.nullValues ?? []);
 	const idPosition = idColumn === null ? -1 : names.indexOf(idColumn);
 	const kept = names.flatMap((name, position) => name !== idColumn && wanted.has(name) ? [{
 		name,
@@ -607,7 +608,8 @@ function parseCsvTable(text, options) {
 		const id = idColumn === null ? String(nodes.length + 1) : (record.fields[idPosition] ?? "").trim();
 		if (id === "") throw new NativeParseError("csv-table", record.line, `the id column ${idColumn ?? ""} is empty`);
 		const cells = kept.map(({ name, position, factor }) => {
-			const cell = (record.fields[position] ?? "").trim();
+			const trimmed = (record.fields[position] ?? "").trim();
+			const cell = missing.has(trimmed) ? "" : trimmed;
 			return [name, factor === void 0 ? cell === "" ? null : cell : numeric$1(cell, factor, name, record.line)];
 		});
 		nodes.push({
@@ -1596,8 +1598,8 @@ async function getBytes(port, url, signal) {
 	return new Uint8Array(await response.arrayBuffer());
 }
 /**
-* Reads a resource's pages, each starting where the rows held so far end, and checks each holds exactly the rows it should, so a response cut short is an error and not a table with rows missing.
-* @param read - The resource read.
+* Reads a source's pages, each starting where the rows held so far end, and checks each holds exactly the rows it should, so a response cut short is an error and not a table with rows missing.
+* @param read - The source read.
 * @param held - The pages read so far, each after the first without its header record.
 * @param offset - How many rows the pages hold.
 * @returns Every page.
@@ -1607,7 +1609,7 @@ async function readResourcePages(read, held, offset) {
 	if (offset >= read.count) return held;
 	const expected = Math.min(read.pageSize, read.count - offset);
 	const page = await withAttempts(async () => {
-		const bytes = await getBytes(read.port, `${read.resource}?%24order=%3Aid&%24limit=${String(read.pageSize)}&%24offset=${String(offset)}`, read.signal);
+		const bytes = await getBytes(read.port, read.pageUrl(offset), read.signal);
 		const rows = recordsIn(bytes) - 1;
 		if (rows !== expected) throw new KeepsetError(`${read.resource} gave ${String(rows)} rows at offset ${String(offset)} where ${String(expected)} of its ${String(read.count)} were due, so a response was cut short or the table changed during the download`);
 		return bytes;
@@ -1618,6 +1620,17 @@ async function readResourcePages(read, held, offset) {
 	return readResourcePages(read, [...held, first === void 0 ? page : page.subarray(headerEnd)], offset + expected);
 }
 /**
+* Downloads a CSV source a page at a time and joins the pages under the first header.
+* @param read - The source read, with its count.
+* @returns The bytes of one CSV, which is what the parser reads and what the provenance hashes.
+* @throws KeepsetError when the source has no rows or a page is not what it should be.
+*/
+async function downloadPages(read) {
+	const pages = await readResourcePages(read, [], 0);
+	if (pages.length === 0) throw new KeepsetError(`${read.resource} has no rows`);
+	return Buffer.concat(pages);
+}
+/**
 * Downloads a whole Socrata resource as one CSV: the row count is read first, then pages of `pageSize` rows in the order of the rows' ids, each checked to hold the rows it should, and joined under the first page's header.
 * @param resource - The resource's URL, such as `https://data.example.org/resource/abcd-1234.csv`, without a query.
 * @param pageSize - How many rows a page holds.
@@ -1629,15 +1642,45 @@ async function readResourcePages(read, held, offset) {
 async function downloadSocrataResource(resource, pageSize, port, signal) {
 	const reported = /^"count"\r?\n"(\d+)"/.exec(new TextDecoder("utf-8").decode(await getBytes(port, `${resource}?%24select=count(*)`, signal)));
 	if (reported?.[1] === void 0) throw new KeepsetError(`${resource} did not report a row count`);
-	const pages = await readResourcePages({
+	return downloadPages({
 		resource,
+		pageUrl: (offset) => `${resource}?%24order=%3Aid&%24limit=${String(pageSize)}&%24offset=${String(offset)}`,
 		pageSize,
 		count: Number(reported[1]),
 		port,
 		signal
-	}, [], 0);
-	if (pages.length === 0) throw new KeepsetError(`${resource} has no rows`);
-	return Buffer.concat(pages);
+	});
+}
+const CountResultSchema = z.object({ result: z.object({ total: z.int().min(0) }) });
+/**
+* Downloads a whole CKAN datastore table as one CSV: the row count is read from the portal's datastore API, then pages of `pageSize` rows of the table's CSV dump, each checked to hold the rows it should, and joined under the first page's header.
+* @param dump - The dump's URL, such as `https://data.example.org/datastore/dump/<resource id>`, without a query; a portal that serves CKAN under a path prefix keeps it before `/datastore/dump/`.
+* @param pageSize - How many rows a page holds.
+* @param port - The network port.
+* @param signal - Aborts the download.
+* @returns The bytes of one CSV, which is what the parser reads and what the provenance hashes.
+* @throws KeepsetError when the URL is not a datastore dump, the table does not report a count, or a page is not what it should be.
+*/
+async function downloadCkanDatastore(dump, pageSize, port, signal) {
+	const parsed = /^(.*)\/datastore\/dump\/([^/?#]+)$/.exec(dump);
+	if (parsed?.[1] === void 0 || parsed[2] === void 0) throw new KeepsetError(`${dump} is not a CKAN datastore dump URL`);
+	const url = `${parsed[1]}/api/3/action/datastore_search?resource_id=${parsed[2]}&limit=0`;
+	let body;
+	try {
+		body = JSON.parse(new TextDecoder("utf-8").decode(await getBytes(port, url, signal)));
+	} catch (error) {
+		throw new KeepsetError(`${url} did not answer JSON (${error instanceof Error ? error.message : String(error)})`);
+	}
+	const count = CountResultSchema.safeParse(body);
+	if (!count.success) throw new KeepsetError(`${dump} did not report a row count`);
+	return downloadPages({
+		resource: dump,
+		pageUrl: (offset) => `${dump}?format=csv&limit=${String(pageSize)}&offset=${String(offset)}`,
+		pageSize,
+		count: count.data.result.total,
+		port,
+		signal
+	});
 }
 /**
 * Downloads a paged source as the one file its pages make, in the way its recipe's paging says.
@@ -1649,7 +1692,8 @@ async function downloadSocrataResource(resource, pageSize, port, signal) {
 * @throws KeepsetError when a page cannot be read or the pages do not add up to what the source reports.
 */
 async function downloadPaged(url, paging, port, signal) {
-	return paging.kind === "arcgis-query" ? downloadArcgisLayer(url, paging.orderBy, port, signal) : downloadSocrataResource(url, paging.pageSize, port, signal);
+	if (paging.kind === "arcgis-query") return downloadArcgisLayer(url, paging.orderBy, port, signal);
+	return paging.kind === "socrata-resource" ? downloadSocrataResource(url, paging.pageSize, port, signal) : downloadCkanDatastore(url, paging.pageSize, port, signal);
 }
 //#endregion
 //#region src/node/fetch-dataset.ts

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 = "0.21.0";
+var version = "0.22.0";
 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": "0.21.0",
+  "version": "0.22.0",
   "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_]*$","type":"string"}},"required":["kind","orderBy"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"socrata-resource","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"}]},"parser":{"oneOf":[{"additionalProperties":false,"properties":{"commentPrefixes":{"items":{"minLength":1,"type":"string"},"type":"array"},"delimiter":{"enum":["detect","whitespace","tab","comma","space"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"edge-list","type":"string"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"sourceColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"targetColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","commentPrefixes","sourceColumn","targetColumn"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"matrix-market","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"gml","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"pajek","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","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])?$","type":"string"},{"type":"null"}]},"numericColumns":{"items":{"additionalProperties":false,"properties":{"column":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","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"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.21.0/schemas/entry.schema.json","$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._-]*)*$","type":"string"},"minItems":1,"type":"array"},"category":{"enum":["graph/v1","or/v1"],"type":"string"},"citations":{"items":{"additionalProperties":false,"properties":{"doi":{"pattern":"^10\\.\\d{4,9}\\/\\S+$","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_]*$","type":"string"}},"required":["kind","orderBy"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"socrata-resource","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"ckan-datastore","type":"string"},"pageSize":{"maximum":9007199254740991,"minimum":1,"type":"integer"}},"required":["kind","pageSize"],"type":"object"}]},"parser":{"oneOf":[{"additionalProperties":false,"properties":{"commentPrefixes":{"items":{"minLength":1,"type":"string"},"type":"array"},"delimiter":{"enum":["detect","whitespace","tab","comma","space"],"type":"string"},"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"edge-list","type":"string"},"skipLines":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"sourceColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"},"targetColumn":{"maximum":9007199254740991,"minimum":0,"type":"integer"}},"required":["format","encoding","delimiter","skipLines","commentPrefixes","sourceColumn","targetColumn"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"matrix-market","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"gml","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"encoding":{"enum":["utf-8","latin1"],"type":"string"},"format":{"const":"pajek","type":"string"}},"required":["format","encoding"],"type":"object"},{"additionalProperties":false,"properties":{"columns":{"items":{"pattern":"^[A-Za-z](?:[A-Za-z0-9_]*[A-Za-z0-9])?$","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"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.22.0/schemas/entry.schema.json","$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@0.21.0/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"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.22.0/schemas/manifest.schema.json","$ref":"#/$defs/Manifest","$schema":"https://json-schema.org/draft/2020-12/schema"}