Number the stars the archive places after their host's name, so a refresh keeps each one's id

4c8e4a0 numbered each star it adds from the Exoplanet Archive by its place among the unmatched
hosts, in pl_name order, and the refresh workflow re-queries the archive every Monday. One host
added or dropped ahead of another renumbers it: in the review, removing a single planet row
renamed 3 276 of the 3 277 ids, and a bookmark kept on Kepler-186 (1070001620) opened Kepler-1860
under Kepler-186's stored name. HYG's and Gaia's ids do not move between refreshes; the merge's own
comment says ids are meant to hold.

archiveStarId (host-star-matching.ts, beside the matcher the ETL already imports from there) hashes
the host name with FNV-1a into the 3.7 million ids between 1 070 000 000 and 2^30, and moves a name
whose id is taken to the next free one. The added stars are sorted by id before they are appended,
so the published list stays in id order. Measured: all 4 237 archive-hosted planets change host id
once (Kepler-186 is now 1073671518), none of the others; one of the 3 277 names was probed past a
collision, and one new id falls in the old 1070000000-1070003276 range, so a bookmark saved on that
old id would open the wrong star once. Dropping the same row from a copy of the cached answer and
rerunning the exoplanet step in a scratch copy of the ETL now leaves 3 276 of 3 276 ids unchanged.
The validators pass: no duplicate id, all under 2^30.

Controls, each failing its named test: the id taken from the order of arrival, and no probing past
a taken id (1 of 801 each).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

src/app/shared/astro/host-star-matching.spec.ts

@@ -1,6 +1,6 @@
 import { describe, expect, it } from 'vitest';
 
-import { buildStarNameIndex, normalizeStarName, resolveHostStarId } from './host-star-matching';
+import { ARCHIVE_ID_BASE, archiveStarId, buildStarNameIndex, normalizeStarName, resolveHostStarId } from './host-star-matching';
 import { propagateProperMotion, raDegDecDistanceToXyz } from './coordinates';
 import { StarRecord } from '../models/star.model';
 
@@ -214,3 +214,19 @@ describe('resolveHostStarId', () => {
     });
   });
 });
+
+describe('archiveStarId', () => {
+  it('gives a host the same id whatever else the archive holds, inside the range left for it', () => {
+    const kepler186 = archiveStarId('Kepler-186', new Set());
+    const others = new Set(['1RXS J160929.1-210524', 'Kepler-1860', 'Kepler-452', 'TOI-700'].map((name) => archiveStarId(name, new Set())));
+    expect(archiveStarId('Kepler-186', others)).toBe(kepler186);
+    expect(others.has(kepler186)).toBe(false);
+    expect(kepler186).toBeGreaterThanOrEqual(ARCHIVE_ID_BASE);
+    expect(kepler186).toBeLessThan(2 ** 30);
+  });
+
+  it('moves a host whose id is taken to the next free one', () => {
+    const kepler186 = archiveStarId('Kepler-186', new Set());
+    expect(archiveStarId('Kepler-186', new Set([kepler186, kepler186 + 1]))).toBe(kepler186 + 2);
+  });
+});

src/app/shared/astro/host-star-matching.ts

@@ -171,3 +171,30 @@ export function resolveHostStarId(
 
   return best ? best.id : null;
 }
+
+/**
+ * Where the ids of the stars only the archive places begin: past Gaia's two ranges, and under
+ * the 2^30 `validateStars` holds every id to — which leaves 3.7 million.
+ */
+export const ARCHIVE_ID_BASE = 1_070_000_000;
+const ARCHIVE_ID_RANGE = 2 ** 30 - ARCHIVE_ID_BASE;
+
+/**
+ * The id of a star the ETL places from the archive, from its host's name rather than from its
+ * place in the answer. Numbered in pl_name order, a refresh that added or dropped one host
+ * renumbered every host after it — one row dropped renamed 3 276 of 3 277 ids, and a bookmark kept
+ * on Kepler-186 opened Kepler-1860 — while HYG's and Gaia's ids hold. FNV-1a over the name, into
+ * the range above; a name whose id is taken takes the next free one, the one case a refresh can
+ * still move, and only between the two names that collided.
+ */
+export function archiveStarId(hostname: string, taken: ReadonlySet<number>): number {
+  let hash = 0x811c9dc5;
+  for (let i = 0; i < hostname.length; i++) {
+    hash = Math.imul(hash ^ hostname.charCodeAt(i), 0x01000193) >>> 0;
+  }
+  let offset = hash % ARCHIVE_ID_RANGE;
+  while (taken.has(ARCHIVE_ID_BASE + offset)) {
+    offset = (offset + 1) % ARCHIVE_ID_RANGE;
+  }
+  return ARCHIVE_ID_BASE + offset;
+}

tools/etl/fetchExoplanets.ts

@@ -2,7 +2,7 @@ import { createHash } from 'node:crypto';
 import { writeFileSync } from 'node:fs';
 
 import { propagateProperMotion, raDegDecDistanceToXyz } from '../../src/app/shared/astro/coordinates';
-import { buildStarNameIndex, resolveHostStarId } from '../../src/app/shared/astro/host-star-matching';
+import { archiveStarId, buildStarNameIndex, resolveHostStarId } from '../../src/app/shared/astro/host-star-matching';
 import { temperatureToColorIndex } from '../../src/app/shared/astro/spectral';
 import { ExoplanetRecord } from '../../src/app/shared/models/exoplanet.model';
 import { isDesignation } from '../../src/app/shared/models/star-catalog';
@@ -85,11 +85,6 @@ const COMPOSITE_CACHE_FILE = `exoplanet-archive-pscomppars-${createHash('sha1').
 const DISTANCE_ERRORS_URL = `${TAP_BASE_URL}?query=select+pl_name,sy_disterr1,sy_disterr2+from+pscomppars+order+by+pl_name&format=csv`;
 const DISTANCE_ERRORS_CACHE_FILE = `exoplanet-archive-disterr-${createHash('sha1').update(DISTANCE_ERRORS_URL).digest('hex').slice(0, 8)}.csv`;
 
-/**
- * Where the ids of the stars only the archive places begin: past Gaia's two ranges, and under
- * the 2^30 `validateStars` holds every id to.
- */
-const ARCHIVE_ID_BASE = 1_070_000_000;
 const ARCHIVE_SOURCE = 'exoplanet-archive';
 /**
  * The archive's positions are at Gaia's epoch, J2016, not the catalogue's J2000: of the 746
@@ -123,6 +118,7 @@ export async function fetchExoplanets(stars?: StarRecord[]): Promise<{ exoplanet
   // the only way "TRAPPIST-1" or "Teegarden's Star" can be found by search.
   const renamed = new Map<number, string>();
   const archiveStars = new Map<string, StarRecord>();
+  const archiveIds = new Set<number>();
   const bands = { V: 0, G: 0, none: 0 };
   const exoplanets: ExoplanetRecord[] = rows.map((row, index) => {
     const compositeRow = composite.get(row['pl_name']);
@@ -166,7 +162,7 @@ export async function fetchExoplanets(stars?: StarRecord[]): Promise<{ exoplanet
         const errors = distanceErrors.get(row['pl_name']);
         const [above, below] = [parseOptionalNumber(errors?.['sy_disterr1']), parseOptionalNumber(errors?.['sy_disterr2'])];
         archiveStar = {
-          id: ARCHIVE_ID_BASE + archiveStars.size,
+          id: archiveStarId(row['hostname'], archiveIds),
           name: row['hostname'],
           ...raDegDecDistanceToXyz(j2000.raDeg, j2000.decDeg, distancePc),
           magnitude: v ?? g ?? UNKNOWN_MAGNITUDE,
@@ -178,6 +174,7 @@ export async function fetchExoplanets(stars?: StarRecord[]): Promise<{ exoplanet
           source: ARCHIVE_SOURCE
         };
         archiveStars.set(row['hostname'], archiveStar);
+        archiveIds.add(archiveStar.id);
       }
       hostStarId = archiveStar.id;
     }
@@ -215,9 +212,9 @@ export async function fetchExoplanets(stars?: StarRecord[]): Promise<{ exoplanet
     };
   });
 
-  // Appended after the catalogue, whose ids all sit below ARCHIVE_ID_BASE, so the list stays in
-  // the id order `fetchStars` sorted it into.
-  const added = [...archiveStars.values()];
+  // Appended after the catalogue, whose ids all sit below ARCHIVE_ID_BASE, and sorted, so the list
+  // stays in the id order `fetchStars` sorted it into.
+  const added = [...archiveStars.values()].sort((a, b) => a.id - b.id);
   const allStars = [...knownStars.map((star) => (renamed.has(star.id) ? { ...star, name: renamed.get(star.id)! } : star)), ...added];
   writeStarAssets(allStars);