keepset 0.6.1 → 0.6.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +1 -1
- package/dist/{bundled-P_132GDn.js → bundled-BIBLzetw.js} +1 -1
- package/dist/index.js +1 -1
- package/dist/node/fetch-dataset.js +2 -2
- package/dist-cli/keepset.mjs +1 -1
- package/package.json +2 -2
- package/schemas/entry.schema.json +1 -1
- package/schemas/manifest.schema.json +1 -1
package/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# keepset
|
|
2
2
|
|
|
3
|
-
[](https://github.com/
|
|
3
|
+
[](https://github.com/Mearman/keepset) [](https://www.npmjs.com/package/keepset) [](https://github.com/Mearman/keepset/releases/latest) [](https://github.com/Mearman/keepset/actions)
|
|
4
4
|
|
|
5
5
|
A catalogue of datasets, each with a machine-readable recipe that says how to fetch it and how to normalise it into the shape its category names. The `keepset` package is the consumable library over that catalogue; the CLI and the statically generated site are thin surfaces on top of it. Entries are expressed with [twindex](https://github.com/ExaDev/twindex) crosswalks, so fetching a dataset yields a normalised shape directly in user code.
|
|
6
6
|
|
|
@@ -557,7 +557,7 @@ var catalogue_default = {
|
|
|
557
557
|
"name": "keepset",
|
|
558
558
|
"description": "The keepset catalogue: datasets with machine-readable recipes that fetch each one and normalise it into its category's target shape.",
|
|
559
559
|
"version": "1.0.0",
|
|
560
|
-
"repository": "https://github.com/
|
|
560
|
+
"repository": "https://github.com/Mearman/keepset"
|
|
561
561
|
},
|
|
562
562
|
crosswalks: { "crosswalks/graph-v1/topology.json": {
|
|
563
563
|
"source": "keepset/topology",
|
package/dist/index.js
CHANGED
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
import { A as countTarget, B as CatalogueSyntaxError, C as SnapshotSchema, D as toIssues, E as parseEntry, F as GraphEdgeSchema, G as UnknownCategoryError, H as KeepsetError, I as GraphNodeSchema, L as GraphV1Schema, M as getTarget, N as kindOf, O as waybackUrl, P as listCategories, R as JsonScalarSchema, S as RelativePathSchema, T as WaybackTimestampSchema, U as NativeParseError, V as DatasetNotFoundError, W as RecipeError, _ as IdSchema, a as createCatalogue, b as RESERVED_META_KEYS, c as ManifestSchema, d as ArchiveMemberSchema, f as ArchiveSchema, g as HttpUrlSchema, h as EntrySchema, i as CatalogueIndexSchema, j as getCategory, k as CategoryIdSchema, l as SemVerSchema, m as CrosswalkReferenceSchema, n as getManifest, p as CitationSchema, r as listDatasets, s as LineageSchema, t as getDataset, u as parseManifest, v as LicenceSchema, w as SourceSchema, x as RecipeSchema, y as ParserSchema, z as ArchiveError } from "./bundled-
|
|
1
|
+
import { A as countTarget, B as CatalogueSyntaxError, C as SnapshotSchema, D as toIssues, E as parseEntry, F as GraphEdgeSchema, G as UnknownCategoryError, H as KeepsetError, I as GraphNodeSchema, L as GraphV1Schema, M as getTarget, N as kindOf, O as waybackUrl, P as listCategories, R as JsonScalarSchema, S as RelativePathSchema, T as WaybackTimestampSchema, U as NativeParseError, V as DatasetNotFoundError, W as RecipeError, _ as IdSchema, a as createCatalogue, b as RESERVED_META_KEYS, c as ManifestSchema, d as ArchiveMemberSchema, f as ArchiveSchema, g as HttpUrlSchema, h as EntrySchema, i as CatalogueIndexSchema, j as getCategory, k as CategoryIdSchema, l as SemVerSchema, m as CrosswalkReferenceSchema, n as getManifest, p as CitationSchema, r as listDatasets, s as LineageSchema, t as getDataset, u as parseManifest, v as LicenceSchema, w as SourceSchema, x as RecipeSchema, y as ParserSchema, z as ArchiveError } from "./bundled-BIBLzetw.js";
|
|
2
2
|
import { validateCrosswalk } from "twindex";
|
|
3
3
|
//#region src/serialise.ts
|
|
4
4
|
/**
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
import { A as countTarget, H as KeepsetError, M as getTarget, U as NativeParseError, W as RecipeError, o as getOwn, t as getDataset, z as ArchiveError } from "../bundled-
|
|
1
|
+
import { A as countTarget, H as KeepsetError, M as getTarget, U as NativeParseError, W as RecipeError, o as getOwn, t as getDataset, z as ArchiveError } from "../bundled-BIBLzetw.js";
|
|
2
2
|
import { RoundTripMismatchError, assertRoundTrip, encode } from "twindex";
|
|
3
3
|
import { createHash } from "node:crypto";
|
|
4
4
|
import { gunzipSync } from "node:zlib";
|
|
@@ -726,7 +726,7 @@ const REDIRECTS = new Set(Object.values({
|
|
|
726
726
|
permanentRedirect: 308
|
|
727
727
|
}));
|
|
728
728
|
/** What the downloader says it is, so a server operator can tell where the traffic comes from. */
|
|
729
|
-
const USER_AGENT = "keepset (+https://github.com/
|
|
729
|
+
const USER_AGENT = "keepset (+https://github.com/Mearman/keepset)";
|
|
730
730
|
/**
|
|
731
731
|
* Sends one GET and resolves with the response head, its body still unread.
|
|
732
732
|
* @param url - The URL.
|
package/dist-cli/keepset.mjs
CHANGED
|
@@ -4,7 +4,7 @@ import { KeepsetError, countTarget, getDataset, listDatasets, validateDataset }
|
|
|
4
4
|
import { fetchDataset, roundTripDataset } from "keepset/node";
|
|
5
5
|
import { parseArgs } from "node:util";
|
|
6
6
|
//#region package.json
|
|
7
|
-
var version = "0.6.
|
|
7
|
+
var version = "0.6.2";
|
|
8
8
|
const usage = `Usage: keepset <command> [options]
|
|
9
9
|
|
|
10
10
|
Commands:
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "keepset",
|
|
3
3
|
"description": "Library, CLI and statically generated catalogue for dataset fetch-and-normalise recipes",
|
|
4
|
-
"version": "0.6.
|
|
4
|
+
"version": "0.6.2",
|
|
5
5
|
"bin": {
|
|
6
6
|
"keepset": "./dist-cli/keepset.mjs"
|
|
7
7
|
},
|
|
@@ -74,7 +74,7 @@
|
|
|
74
74
|
},
|
|
75
75
|
"repository": {
|
|
76
76
|
"type": "git",
|
|
77
|
-
"url": "git+https://github.com/
|
|
77
|
+
"url": "git+https://github.com/Mearman/keepset.git"
|
|
78
78
|
},
|
|
79
79
|
"scripts": {
|
|
80
80
|
"build": "pnpm run catalogue && tsdown && tsx scripts/generate-json-schema.ts",
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"$defs":{"Crosswalk":{"$ref":"https://cdn.jsdelivr.net/npm/twindex@0.1.1/schemas/twindex.schema.json"},"Entry":{"additionalProperties":false,"properties":{"category":{"enum":["graph/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"},"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"}]},"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"}]},"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"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.6.
|
|
1
|
+
{"$defs":{"Crosswalk":{"$ref":"https://cdn.jsdelivr.net/npm/twindex@0.1.1/schemas/twindex.schema.json"},"Entry":{"additionalProperties":false,"properties":{"category":{"enum":["graph/v1"],"type":"string"},"citations":{"items":{"additionalProperties":false,"properties":{"doi":{"pattern":"^10\\.\\d{4,9}\\/\\S+$","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.+-]*$","type":"string"},"kind":{"const":"spdx","type":"string"}},"required":["kind","id"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"custom","type":"string"},"name":{"minLength":1,"type":"string"},"url":{"format":"uri","type":"string"}},"required":["kind","name","url"],"type":"object"},{"additionalProperties":false,"properties":{"kind":{"const":"unrecorded","type":"string"}},"required":["kind"],"type":"object"}]},"name":{"minLength":1,"type":"string"},"properties":{"additionalProperties":{"type":["string","number","boolean","null"]},"propertyNames":{"pattern":"^[A-Za-z][A-Za-z0-9]*$","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"}]},"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"}]},"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@0.6.2/schemas/entry.schema.json","$ref":"#/$defs/Entry","$schema":"https://json-schema.org/draft/2020-12/schema"}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"$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-]+)*)?$","type":"string"}},"required":["catalogue","version","repository"],"type":"object"},"name":{"minLength":1,"type":"string"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["id","name","description","version","repository"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.6.
|
|
1
|
+
{"$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-]+)*)?$","type":"string"}},"required":["catalogue","version","repository"],"type":"object"},"name":{"minLength":1,"type":"string"},"repository":{"format":"uri","type":"string"},"version":{"pattern":"^(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)\\.(?:0|[1-9]\\d*)(?:-[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?(?:\\+[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*)?$","type":"string"}},"required":["id","name","description","version","repository"],"type":"object"}},"$id":"https://cdn.jsdelivr.net/npm/keepset@0.6.2/schemas/manifest.schema.json","$ref":"#/$defs/Manifest","$schema":"https://json-schema.org/draft/2020-12/schema"}
|