wiki-graph-core 0.3.1 → 0.4.1

package/dist/retrieval/search-index/search/index.js

@@ -0,0 +1,3 @@
+export * from "./core.js";
+export * from "./tokenizer.js";
+//# sourceMappingURL=index.js.map

package/dist/retrieval/search-index/search/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/retrieval/search-index/search/index.ts"],"sourcesContent":["export * from \"./core.js\";\nexport * from \"./tokenizer.js\";\n"],"mappings":"AAAA,cAAc;AACd,cAAc;","names":[]}

package/dist/retrieval/search-index/search/match.d.ts

@@ -0,0 +1,34 @@
+import { SearchTokenPlan } from './tokenizer.js';
+import { ArchiveFindMatch } from '../../query/archive-view/types.js';
+import '../../../document/types.js';
+import '../../../text/source/meta.js';
+import 'zod';
+import '../../../graph/reading.js';
+import '../../../document/directory/types.js';
+import '../../../text/source/toc.js';
+import '../../../text/source/types.js';
+import '../../../document/database.js';
+import 'sqlite3';
+import '../../../document/text-streams/serial.js';
+import '../../../document/text-streams/draft.js';
+import '../../../document/text-streams/types.js';
+import '../../../document/stores/chunk.js';
+import '../../../document/stores/types.js';
+import '../../../document/stores/fragment-group.js';
+import '../../../document/stores/graph-build-parameter.js';
+import '../../../document/stores/mention-link.js';
+import '../../../document/stores/mention.js';
+import '../../../document/stores/object-metadata.js';
+import '../../../document/stores/reading-edge.js';
+import '../../../document/stores/serial.js';
+import '../../../document/stores/snake/chunk.js';
+import '../../../document/stores/snake/core.js';
+import '../../../document/stores/snake/edge.js';
+import '../../../document/chapter/types.js';
+
+interface TierQuery {
+    readonly matchExpression: string;
+}
+declare function createTierQueries(query: string, plan: SearchTokenPlan, match: ArchiveFindMatch): readonly TierQuery[];
+
+export { type TierQuery, createTierQueries };

package/dist/retrieval/search-index/search/match.js

@@ -0,0 +1,82 @@
+import {
+  createSearchTokenPlan,
+  hasSearchTokens,
+  normalizeSearchText
+} from "./tokenizer.js";
+function createTierQueries(query, plan, match) {
+  if (match === "all") {
+    return createAllMatchTierQueries(query);
+  }
+  return [
+    {
+      matchExpression: createMatchExpression([
+        { column: "tier1", tokens: plan.tier1.map((token) => token.encoded) }
+      ])
+    },
+    {
+      matchExpression: createMatchExpression([
+        { column: "tier1", tokens: plan.tier1.map((token) => token.encoded) },
+        { column: "tier2", tokens: plan.tier2.map((token) => token.encoded) }
+      ])
+    },
+    {
+      matchExpression: createMatchExpression([
+        { column: "tier1", tokens: plan.tier1.map((token) => token.encoded) },
+        { column: "tier2", tokens: plan.tier2.map((token) => token.encoded) },
+        { column: "tier3", tokens: plan.tier3.map((token) => token.encoded) }
+      ])
+    }
+  ];
+}
+function createAllMatchTierQueries(query) {
+  const termPlans = normalizeSearchText(query).split(/\s+/u).map((term) => createSearchTokenPlan(term)).filter(hasSearchTokens);
+  const createExpression = (selectTokens) => termPlans.map((termPlan) => {
+    const tokens = [...new Set(selectTokens(termPlan))];
+    return tokens.length === 0 ? "" : `(${tokens.join(" OR ")})`;
+  }).filter((term) => term !== "").join(" AND ");
+  return [
+    {
+      matchExpression: createExpression(
+        (termPlan) => termPlan.tier1.map((token) => `tier1:${escapeFtsToken(token.encoded)}`)
+      )
+    },
+    {
+      matchExpression: createExpression((termPlan) => [
+        ...termPlan.tier1.map(
+          (token) => `tier1:${escapeFtsToken(token.encoded)}`
+        ),
+        ...termPlan.tier2.map(
+          (token) => `tier2:${escapeFtsToken(token.encoded)}`
+        )
+      ])
+    },
+    {
+      matchExpression: createExpression((termPlan) => [
+        ...termPlan.tier1.map(
+          (token) => `tier1:${escapeFtsToken(token.encoded)}`
+        ),
+        ...termPlan.tier2.map(
+          (token) => `tier2:${escapeFtsToken(token.encoded)}`
+        ),
+        ...termPlan.tier3.map(
+          (token) => `tier3:${escapeFtsToken(token.encoded)}`
+        )
+      ])
+    }
+  ];
+}
+function createMatchExpression(groups) {
+  return groups.map((group) => ({
+    ...group,
+    tokens: [...new Set(group.tokens)]
+  })).filter((group) => group.tokens.length > 0).map(
+    (group) => `${group.column}:(${group.tokens.map(escapeFtsToken).join(" OR ")})`
+  ).join(" OR ");
+}
+function escapeFtsToken(token) {
+  return `"${token.replaceAll('"', '""')}"`;
+}
+export {
+  createTierQueries
+};
+//# sourceMappingURL=match.js.map

package/dist/retrieval/search-index/search/match.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/retrieval/search-index/search/match.ts"],"sourcesContent":["import {\n  createSearchTokenPlan,\n  hasSearchTokens,\n  normalizeSearchText,\n  type SearchTokenPlan,\n} from \"./tokenizer.js\";\nimport type { ArchiveFindMatch } from \"../../query/view.js\";\n\nexport interface TierQuery {\n  readonly matchExpression: string;\n}\n\nexport function createTierQueries(\n  query: string,\n  plan: SearchTokenPlan,\n  match: ArchiveFindMatch,\n): readonly TierQuery[] {\n  if (match === \"all\") {\n    return createAllMatchTierQueries(query);\n  }\n\n  return [\n    {\n      matchExpression: createMatchExpression([\n        { column: \"tier1\", tokens: plan.tier1.map((token) => token.encoded) },\n      ]),\n    },\n    {\n      matchExpression: createMatchExpression([\n        { column: \"tier1\", tokens: plan.tier1.map((token) => token.encoded) },\n        { column: \"tier2\", tokens: plan.tier2.map((token) => token.encoded) },\n      ]),\n    },\n    {\n      matchExpression: createMatchExpression([\n        { column: \"tier1\", tokens: plan.tier1.map((token) => token.encoded) },\n        { column: \"tier2\", tokens: plan.tier2.map((token) => token.encoded) },\n        { column: \"tier3\", tokens: plan.tier3.map((token) => token.encoded) },\n      ]),\n    },\n  ];\n}\n\nfunction createAllMatchTierQueries(query: string): readonly TierQuery[] {\n  const termPlans = normalizeSearchText(query)\n    .split(/\\s+/u)\n    .map((term) => createSearchTokenPlan(term))\n    .filter(hasSearchTokens);\n\n  const createExpression = (\n    selectTokens: (plan: SearchTokenPlan) => readonly string[],\n  ): string =>\n    termPlans\n      .map((termPlan) => {\n        const tokens = [...new Set(selectTokens(termPlan))];\n\n        return tokens.length === 0 ? \"\" : `(${tokens.join(\" OR \")})`;\n      })\n      .filter((term) => term !== \"\")\n      .join(\" AND \");\n\n  return [\n    {\n      matchExpression: createExpression((termPlan) =>\n        termPlan.tier1.map((token) => `tier1:${escapeFtsToken(token.encoded)}`),\n      ),\n    },\n    {\n      matchExpression: createExpression((termPlan) => [\n        ...termPlan.tier1.map(\n          (token) => `tier1:${escapeFtsToken(token.encoded)}`,\n        ),\n        ...termPlan.tier2.map(\n          (token) => `tier2:${escapeFtsToken(token.encoded)}`,\n        ),\n      ]),\n    },\n    {\n      matchExpression: createExpression((termPlan) => [\n        ...termPlan.tier1.map(\n          (token) => `tier1:${escapeFtsToken(token.encoded)}`,\n        ),\n        ...termPlan.tier2.map(\n          (token) => `tier2:${escapeFtsToken(token.encoded)}`,\n        ),\n        ...termPlan.tier3.map(\n          (token) => `tier3:${escapeFtsToken(token.encoded)}`,\n        ),\n      ]),\n    },\n  ];\n}\n\nfunction createMatchExpression(\n  groups: readonly {\n    readonly column: string;\n    readonly tokens: readonly string[];\n  }[],\n): string {\n  return groups\n    .map((group) => ({\n      ...group,\n      tokens: [...new Set(group.tokens)],\n    }))\n    .filter((group) => group.tokens.length > 0)\n    .map(\n      (group) =>\n        `${group.column}:(${group.tokens.map(escapeFtsToken).join(\" OR \")})`,\n    )\n    .join(\" OR \");\n}\n\nfunction escapeFtsToken(token: string): string {\n  return `\"${token.replaceAll('\"', '\"\"')}\"`;\n}\n"],"mappings":"AAAA;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,OAEK;AAOA,SAAS,kBACd,OACA,MACA,OACsB;AACtB,MAAI,UAAU,OAAO;AACnB,WAAO,0BAA0B,KAAK;AAAA,EACxC;AAEA,SAAO;AAAA,IACL;AAAA,MACE,iBAAiB,sBAAsB;AAAA,QACrC,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,MACtE,CAAC;AAAA,IACH;AAAA,IACA;AAAA,MACE,iBAAiB,sBAAsB;AAAA,QACrC,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,QACpE,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,MACtE,CAAC;AAAA,IACH;AAAA,IACA;AAAA,MACE,iBAAiB,sBAAsB;AAAA,QACrC,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,QACpE,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,QACpE,EAAE,QAAQ,SAAS,QAAQ,KAAK,MAAM,IAAI,CAAC,UAAU,MAAM,OAAO,EAAE;AAAA,MACtE,CAAC;AAAA,IACH;AAAA,EACF;AACF;AAEA,SAAS,0BAA0B,OAAqC;AACtE,QAAM,YAAY,oBAAoB,KAAK,EACxC,MAAM,MAAM,EACZ,IAAI,CAAC,SAAS,sBAAsB,IAAI,CAAC,EACzC,OAAO,eAAe;AAEzB,QAAM,mBAAmB,CACvB,iBAEA,UACG,IAAI,CAAC,aAAa;AACjB,UAAM,SAAS,CAAC,GAAG,IAAI,IAAI,aAAa,QAAQ,CAAC,CAAC;AAElD,WAAO,OAAO,WAAW,IAAI,KAAK,IAAI,OAAO,KAAK,MAAM,CAAC;AAAA,EAC3D,CAAC,EACA,OAAO,CAAC,SAAS,SAAS,EAAE,EAC5B,KAAK,OAAO;AAEjB,SAAO;AAAA,IACL;AAAA,MACE,iBAAiB;AAAA,QAAiB,CAAC,aACjC,SAAS,MAAM,IAAI,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC,EAAE;AAAA,MACxE;AAAA,IACF;AAAA,IACA;AAAA,MACE,iBAAiB,iBAAiB,CAAC,aAAa;AAAA,QAC9C,GAAG,SAAS,MAAM;AAAA,UAChB,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC;AAAA,QACnD;AAAA,QACA,GAAG,SAAS,MAAM;AAAA,UAChB,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC;AAAA,QACnD;AAAA,MACF,CAAC;AAAA,IACH;AAAA,IACA;AAAA,MACE,iBAAiB,iBAAiB,CAAC,aAAa;AAAA,QAC9C,GAAG,SAAS,MAAM;AAAA,UAChB,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC;AAAA,QACnD;AAAA,QACA,GAAG,SAAS,MAAM;AAAA,UAChB,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC;AAAA,QACnD;AAAA,QACA,GAAG,SAAS,MAAM;AAAA,UAChB,CAAC,UAAU,SAAS,eAAe,MAAM,OAAO,CAAC;AAAA,QACnD;AAAA,MACF,CAAC;AAAA,IACH;AAAA,EACF;AACF;AAEA,SAAS,sBACP,QAIQ;AACR,SAAO,OACJ,IAAI,CAAC,WAAW;AAAA,IACf,GAAG;AAAA,IACH,QAAQ,CAAC,GAAG,IAAI,IAAI,MAAM,MAAM,CAAC;AAAA,EACnC,EAAE,EACD,OAAO,CAAC,UAAU,MAAM,OAAO,SAAS,CAAC,EACzC;AAAA,IACC,CAAC,UACC,GAAG,MAAM,MAAM,KAAK,MAAM,OAAO,IAAI,cAAc,EAAE,KAAK,MAAM,CAAC;AAAA,EACrE,EACC,KAAK,MAAM;AAChB;AAEA,SAAS,eAAe,OAAuB;AAC7C,SAAO,IAAI,MAAM,WAAW,KAAK,IAAI,CAAC;AACxC;","names":[]}

package/dist/retrieval/search-index/search/query.d.ts

@@ -0,0 +1,44 @@
+import { ReadonlyDocument } from '../../../document/directory/types.js';
+import { ArchiveFindMatch, ArchiveFindObjectType } from '../../query/archive-view/types.js';
+import { TextSentenceKind, SearchIndexQueryResult } from './types.js';
+import '../../../text/source/meta.js';
+import 'zod';
+import '../../../text/source/toc.js';
+import '../../../text/source/types.js';
+import '../../../document/database.js';
+import 'sqlite3';
+import '../../../document/types.js';
+import '../../../document/text-streams/serial.js';
+import '../../../document/text-streams/draft.js';
+import '../../../document/text-streams/types.js';
+import '../../../document/stores/chunk.js';
+import '../../../document/stores/types.js';
+import '../../../document/stores/fragment-group.js';
+import '../../../document/stores/graph-build-parameter.js';
+import '../../../document/stores/mention-link.js';
+import '../../../document/stores/mention.js';
+import '../../../document/stores/object-metadata.js';
+import '../../../document/stores/reading-edge.js';
+import '../../../document/stores/serial.js';
+import '../../../document/stores/snake/chunk.js';
+import '../../../document/stores/snake/core.js';
+import '../../../document/stores/snake/edge.js';
+import '../../../graph/reading.js';
+import '../../../document/chapter/types.js';
+
+declare function querySearchIndex(document: ReadonlyDocument, query: string, options?: {
+    readonly chapters?: readonly number[];
+    readonly match?: ArchiveFindMatch;
+    readonly objectHitLimit?: number;
+    readonly textAfter?: {
+        readonly archiveId: number;
+        readonly chapterId: number;
+        readonly kind: TextSentenceKind;
+        readonly rank: number;
+        readonly sentenceIndex: number;
+    };
+    readonly textHitLimit?: number;
+    readonly types?: readonly ArchiveFindObjectType[] | null;
+}): Promise<SearchIndexQueryResult | undefined>;
+
+export { querySearchIndex };

package/dist/retrieval/search-index/search/query.js

@@ -0,0 +1,219 @@
+import { getNumber } from "../../../document/database.js";
+import {
+  createSearchTokenPlan,
+  hasSearchTokens,
+  listSearchPlanTerms
+} from "./tokenizer.js";
+import { createTierQueries } from "./match.js";
+import {
+  createChapterParams,
+  createChapterSql,
+  createLimitParams,
+  createLimitSql,
+  createObjectHitKey,
+  createObjectTypeParams,
+  createObjectTypeSql,
+  createTextHitKey,
+  createTextKindFilter,
+  rankToScore,
+  shouldQueryObjects
+} from "./helpers.js";
+import { SEARCH_INDEX_FTS_HIT_LIMIT, TIER_WEIGHTS } from "./types.js";
+import { assertSearchIndexNotDirty } from "./status.js";
+async function querySearchIndex(document, query, options = {}) {
+  const plan = createSearchTokenPlan(query);
+  if (!hasSearchTokens(plan)) {
+    return void 0;
+  }
+  const terms = listSearchPlanTerms(plan);
+  return await document.readSearchIndexDatabase(async (database) => {
+    await assertSearchIndexNotDirty(database);
+    const tierQueries = createTierQueries(query, plan, options.match ?? "any");
+    const objectHitLimit = options.objectHitLimit ?? SEARCH_INDEX_FTS_HIT_LIMIT;
+    const textHitLimit = options.textHitLimit ?? SEARCH_INDEX_FTS_HIT_LIMIT;
+    const queriesObjects = shouldQueryObjects(options.types);
+    const queriesText = createTextKindFilter(options.types).length > 0;
+    const objectHitsByKey = /* @__PURE__ */ new Map();
+    const textHitsByKey = /* @__PURE__ */ new Map();
+    for (const tierQuery of tierQueries) {
+      if (tierQuery.matchExpression === "") {
+        continue;
+      }
+      const objectHitRemaining = Math.max(
+        0,
+        objectHitLimit - objectHitsByKey.size
+      );
+      const textHitRemaining = Math.max(0, textHitLimit - textHitsByKey.size);
+      if ((!queriesObjects || objectHitRemaining <= 0) && (!queriesText || textHitRemaining <= 0)) {
+        break;
+      }
+      const [objectRows, textRows] = await Promise.all([
+        queryObjectRows(database, tierQuery.matchExpression, {
+          ...options,
+          objectHitLimit: objectHitRemaining
+        }),
+        queryTextRows(database, tierQuery.matchExpression, {
+          ...options,
+          textHitLimit: textHitRemaining
+        })
+      ]);
+      for (const hit of objectRows) {
+        const key = createObjectHitKey(hit);
+        if (!objectHitsByKey.has(key)) {
+          objectHitsByKey.set(key, hit);
+        }
+      }
+      for (const hit of textRows) {
+        const key = createTextHitKey(hit);
+        if (!textHitsByKey.has(key)) {
+          textHitsByKey.set(key, hit);
+        }
+      }
+      if ((!queriesObjects || objectHitsByKey.size >= objectHitLimit) && (!queriesText || textHitsByKey.size >= textHitLimit)) {
+        break;
+      }
+    }
+    return {
+      objectHits: [...objectHitsByKey.values()],
+      terms,
+      textHits: [...textHitsByKey.values()]
+    };
+  });
+}
+async function queryObjectRows(database, matchExpression, options) {
+  if (!shouldQueryObjects(options.types) || options.objectHitLimit === 0) {
+    return [];
+  }
+  return await database.queryAll(
+    `
+      SELECT
+        r.owner_kind AS owner_kind,
+        r.owner_id AS owner_id,
+        r.property_kind AS property_kind,
+        r.archive_id AS archive_id,
+        r.chapter_id AS chapter_id,
+        bm25(search_object_properties_fts, ?, ?, ?) AS rank
+      FROM search_object_properties_fts
+      JOIN search_object_properties_records AS r
+        ON r.id = search_object_properties_fts.rowid
+      WHERE search_object_properties_fts MATCH ?
+        ${createChapterSql(options.chapters)}
+        ${createObjectTypeSql(options.types)}
+      ORDER BY rank ASC, r.archive_id, r.chapter_id, r.owner_kind, r.owner_id, r.property_kind
+      ${createLimitSql(options.objectHitLimit)}
+    `,
+    [
+      ...TIER_WEIGHTS,
+      matchExpression,
+      ...createChapterParams(options.chapters),
+      ...createObjectTypeParams(options.types),
+      ...createLimitParams(options.objectHitLimit)
+    ],
+    (row) => ({
+      ownerId: String(row.owner_id),
+      archiveId: getNumber(row, "archive_id"),
+      ownerKind: getNumber(row, "owner_kind"),
+      propertyKind: getNumber(row, "property_kind"),
+      score: rankToScore(getNumber(row, "rank")),
+      ...row.chapter_id === null ? {} : { chapterId: getNumber(row, "chapter_id") }
+    })
+  );
+}
+async function queryTextRows(database, matchExpression, options) {
+  const kinds = createTextKindFilter(options.types);
+  if (kinds.length === 0) {
+    return [];
+  }
+  if (options.textHitLimit === 0) {
+    return [];
+  }
+  const after = options.textAfter;
+  return await database.queryAll(
+    `
+      SELECT
+        kind,
+        archive_id,
+        chapter_id,
+        sentence_index,
+        words_count,
+        rank
+      FROM (
+        SELECT
+          r.kind AS kind,
+          r.archive_id AS archive_id,
+          r.chapter_id AS chapter_id,
+          r.sentence_index AS sentence_index,
+          r.words_count AS words_count,
+          bm25(text_sentence_fts, ?, ?, ?) AS rank
+        FROM text_sentence_fts
+        JOIN text_sentence_records AS r
+          ON r.id = text_sentence_fts.rowid
+        WHERE text_sentence_fts MATCH ?
+          AND r.kind IN (${kinds.map(() => "?").join(", ")})
+          ${createChapterSql(options.chapters)}
+      )
+      ${after === void 0 ? "" : `
+            WHERE (
+              rank > ?
+              OR (rank = ? AND archive_id > ?)
+              OR (rank = ? AND archive_id = ? AND chapter_id > ?)
+              OR (
+                rank = ?
+                AND archive_id = ?
+                AND chapter_id = ?
+                AND sentence_index > ?
+              )
+              OR (
+                rank = ?
+                AND archive_id = ?
+                AND chapter_id = ?
+                AND sentence_index = ?
+                AND kind > ?
+              )
+            )
+          `}
+      ORDER BY rank ASC, archive_id, chapter_id, sentence_index, kind
+      ${createLimitSql(options.textHitLimit)}
+    `,
+    [
+      ...TIER_WEIGHTS,
+      matchExpression,
+      ...kinds,
+      ...createChapterParams(options.chapters),
+      ...after === void 0 ? [] : [
+        after.rank,
+        after.rank,
+        after.archiveId,
+        after.rank,
+        after.archiveId,
+        after.chapterId,
+        after.rank,
+        after.archiveId,
+        after.chapterId,
+        after.sentenceIndex,
+        after.rank,
+        after.archiveId,
+        after.chapterId,
+        after.sentenceIndex,
+        after.kind
+      ],
+      ...createLimitParams(options.textHitLimit)
+    ],
+    (row) => {
+      const rank = getNumber(row, "rank");
+      return {
+        chapterId: getNumber(row, "chapter_id"),
+        archiveId: getNumber(row, "archive_id"),
+        kind: getNumber(row, "kind"),
+        rank,
+        score: rankToScore(rank),
+        sentenceIndex: getNumber(row, "sentence_index"),
+        wordsCount: getNumber(row, "words_count")
+      };
+    }
+  );
+}
+export {
+  querySearchIndex
+};
+//# sourceMappingURL=query.js.map

package/dist/retrieval/search-index/search/query.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/retrieval/search-index/search/query.ts"],"sourcesContent":["import { getNumber, type Database } from \"../../../document/database.js\";\nimport type { ReadonlyDocument } from \"../../../document/index.js\";\nimport type {\n  ArchiveFindMatch,\n  ArchiveFindObjectType,\n} from \"../../query/view.js\";\nimport {\n  createSearchTokenPlan,\n  hasSearchTokens,\n  listSearchPlanTerms,\n} from \"./tokenizer.js\";\nimport { createTierQueries } from \"./match.js\";\nimport {\n  createChapterParams,\n  createChapterSql,\n  createLimitParams,\n  createLimitSql,\n  createObjectHitKey,\n  createObjectTypeParams,\n  createObjectTypeSql,\n  createTextHitKey,\n  createTextKindFilter,\n  rankToScore,\n  shouldQueryObjects,\n} from \"./helpers.js\";\nimport type {\n  SearchIndexObjectHit,\n  SearchIndexQueryResult,\n  SearchIndexTextHit,\n  SearchObjectPropertyKind,\n  SearchObjectPropertyOwnerKind,\n  TextSentenceKind,\n} from \"./types.js\";\nimport { SEARCH_INDEX_FTS_HIT_LIMIT, TIER_WEIGHTS } from \"./types.js\";\nimport { assertSearchIndexNotDirty } from \"./status.js\";\n\nexport async function querySearchIndex(\n  document: ReadonlyDocument,\n  query: string,\n  options: {\n    readonly chapters?: readonly number[];\n    readonly match?: ArchiveFindMatch;\n    readonly objectHitLimit?: number;\n    readonly textAfter?: {\n      readonly archiveId: number;\n      readonly chapterId: number;\n      readonly kind: TextSentenceKind;\n      readonly rank: number;\n      readonly sentenceIndex: number;\n    };\n    readonly textHitLimit?: number;\n    readonly types?: readonly ArchiveFindObjectType[] | null;\n  } = {},\n): Promise<SearchIndexQueryResult | undefined> {\n  const plan = createSearchTokenPlan(query);\n\n  if (!hasSearchTokens(plan)) {\n    return undefined;\n  }\n\n  const terms = listSearchPlanTerms(plan);\n\n  return await document.readSearchIndexDatabase(async (database) => {\n    await assertSearchIndexNotDirty(database);\n    const tierQueries = createTierQueries(query, plan, options.match ?? \"any\");\n    const objectHitLimit = options.objectHitLimit ?? SEARCH_INDEX_FTS_HIT_LIMIT;\n    const textHitLimit = options.textHitLimit ?? SEARCH_INDEX_FTS_HIT_LIMIT;\n    const queriesObjects = shouldQueryObjects(options.types);\n    const queriesText = createTextKindFilter(options.types).length > 0;\n    const objectHitsByKey = new Map<string, SearchIndexObjectHit>();\n    const textHitsByKey = new Map<string, SearchIndexTextHit>();\n\n    for (const tierQuery of tierQueries) {\n      if (tierQuery.matchExpression === \"\") {\n        continue;\n      }\n\n      const objectHitRemaining = Math.max(\n        0,\n        objectHitLimit - objectHitsByKey.size,\n      );\n      const textHitRemaining = Math.max(0, textHitLimit - textHitsByKey.size);\n\n      if (\n        (!queriesObjects || objectHitRemaining <= 0) &&\n        (!queriesText || textHitRemaining <= 0)\n      ) {\n        break;\n      }\n\n      const [objectRows, textRows] = await Promise.all([\n        queryObjectRows(database, tierQuery.matchExpression, {\n          ...options,\n          objectHitLimit: objectHitRemaining,\n        }),\n        queryTextRows(database, tierQuery.matchExpression, {\n          ...options,\n          textHitLimit: textHitRemaining,\n        }),\n      ]);\n\n      for (const hit of objectRows) {\n        const key = createObjectHitKey(hit);\n\n        if (!objectHitsByKey.has(key)) {\n          objectHitsByKey.set(key, hit);\n        }\n      }\n      for (const hit of textRows) {\n        const key = createTextHitKey(hit);\n\n        if (!textHitsByKey.has(key)) {\n          textHitsByKey.set(key, hit);\n        }\n      }\n      if (\n        (!queriesObjects || objectHitsByKey.size >= objectHitLimit) &&\n        (!queriesText || textHitsByKey.size >= textHitLimit)\n      ) {\n        break;\n      }\n    }\n\n    return {\n      objectHits: [...objectHitsByKey.values()],\n      terms,\n      textHits: [...textHitsByKey.values()],\n    };\n  });\n}\n\nasync function queryObjectRows(\n  database: Database,\n  matchExpression: string,\n  options: {\n    readonly chapters?: readonly number[];\n    readonly objectHitLimit?: number;\n    readonly types?: readonly ArchiveFindObjectType[] | null;\n  },\n): Promise<readonly SearchIndexObjectHit[]> {\n  if (!shouldQueryObjects(options.types) || options.objectHitLimit === 0) {\n    return [];\n  }\n\n  return await database.queryAll(\n    `\n      SELECT\n        r.owner_kind AS owner_kind,\n        r.owner_id AS owner_id,\n        r.property_kind AS property_kind,\n        r.archive_id AS archive_id,\n        r.chapter_id AS chapter_id,\n        bm25(search_object_properties_fts, ?, ?, ?) AS rank\n      FROM search_object_properties_fts\n      JOIN search_object_properties_records AS r\n        ON r.id = search_object_properties_fts.rowid\n      WHERE search_object_properties_fts MATCH ?\n        ${createChapterSql(options.chapters)}\n        ${createObjectTypeSql(options.types)}\n      ORDER BY rank ASC, r.archive_id, r.chapter_id, r.owner_kind, r.owner_id, r.property_kind\n      ${createLimitSql(options.objectHitLimit)}\n    `,\n    [\n      ...TIER_WEIGHTS,\n      matchExpression,\n      ...createChapterParams(options.chapters),\n      ...createObjectTypeParams(options.types),\n      ...createLimitParams(options.objectHitLimit),\n    ],\n    (row) => ({\n      ownerId: String(row.owner_id),\n      archiveId: getNumber(row, \"archive_id\"),\n      ownerKind: getNumber(row, \"owner_kind\") as SearchObjectPropertyOwnerKind,\n      propertyKind: getNumber(row, \"property_kind\") as SearchObjectPropertyKind,\n      score: rankToScore(getNumber(row, \"rank\")),\n      ...(row.chapter_id === null\n        ? {}\n        : { chapterId: getNumber(row, \"chapter_id\") }),\n    }),\n  );\n}\n\nasync function queryTextRows(\n  database: Database,\n  matchExpression: string,\n  options: {\n    readonly chapters?: readonly number[];\n    readonly textAfter?: {\n      readonly archiveId: number;\n      readonly chapterId: number;\n      readonly kind: TextSentenceKind;\n      readonly rank: number;\n      readonly sentenceIndex: number;\n    };\n    readonly textHitLimit?: number;\n    readonly types?: readonly ArchiveFindObjectType[] | null;\n  },\n): Promise<readonly SearchIndexTextHit[]> {\n  const kinds = createTextKindFilter(options.types);\n\n  if (kinds.length === 0) {\n    return [];\n  }\n  if (options.textHitLimit === 0) {\n    return [];\n  }\n\n  const after = options.textAfter;\n\n  return await database.queryAll(\n    `\n      SELECT\n        kind,\n        archive_id,\n        chapter_id,\n        sentence_index,\n        words_count,\n        rank\n      FROM (\n        SELECT\n          r.kind AS kind,\n          r.archive_id AS archive_id,\n          r.chapter_id AS chapter_id,\n          r.sentence_index AS sentence_index,\n          r.words_count AS words_count,\n          bm25(text_sentence_fts, ?, ?, ?) AS rank\n        FROM text_sentence_fts\n        JOIN text_sentence_records AS r\n          ON r.id = text_sentence_fts.rowid\n        WHERE text_sentence_fts MATCH ?\n          AND r.kind IN (${kinds.map(() => \"?\").join(\", \")})\n          ${createChapterSql(options.chapters)}\n      )\n      ${\n        after === undefined\n          ? \"\"\n          : `\n            WHERE (\n              rank > ?\n              OR (rank = ? AND archive_id > ?)\n              OR (rank = ? AND archive_id = ? AND chapter_id > ?)\n              OR (\n                rank = ?\n                AND archive_id = ?\n                AND chapter_id = ?\n                AND sentence_index > ?\n              )\n              OR (\n                rank = ?\n                AND archive_id = ?\n                AND chapter_id = ?\n                AND sentence_index = ?\n                AND kind > ?\n              )\n            )\n          `\n      }\n      ORDER BY rank ASC, archive_id, chapter_id, sentence_index, kind\n      ${createLimitSql(options.textHitLimit)}\n    `,\n    [\n      ...TIER_WEIGHTS,\n      matchExpression,\n      ...kinds,\n      ...createChapterParams(options.chapters),\n      ...(after === undefined\n        ? []\n        : [\n            after.rank,\n            after.rank,\n            after.archiveId,\n            after.rank,\n            after.archiveId,\n            after.chapterId,\n            after.rank,\n            after.archiveId,\n            after.chapterId,\n            after.sentenceIndex,\n            after.rank,\n            after.archiveId,\n            after.chapterId,\n            after.sentenceIndex,\n            after.kind,\n          ]),\n      ...createLimitParams(options.textHitLimit),\n    ],\n    (row) => {\n      const rank = getNumber(row, \"rank\");\n\n      return {\n        chapterId: getNumber(row, \"chapter_id\"),\n        archiveId: getNumber(row, \"archive_id\"),\n        kind: getNumber(row, \"kind\") as TextSentenceKind,\n        rank,\n        score: rankToScore(rank),\n        sentenceIndex: getNumber(row, \"sentence_index\"),\n        wordsCount: getNumber(row, \"words_count\"),\n      };\n    },\n  );\n}\n"],"mappings":"AAAA,SAAS,iBAAgC;AAMzC;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,OACK;AACP,SAAS,yBAAyB;AAClC;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AASP,SAAS,4BAA4B,oBAAoB;AACzD,SAAS,iCAAiC;AAE1C,eAAsB,iBACpB,UACA,OACA,UAaI,CAAC,GACwC;AAC7C,QAAM,OAAO,sBAAsB,KAAK;AAExC,MAAI,CAAC,gBAAgB,IAAI,GAAG;AAC1B,WAAO;AAAA,EACT;AAEA,QAAM,QAAQ,oBAAoB,IAAI;AAEtC,SAAO,MAAM,SAAS,wBAAwB,OAAO,aAAa;AAChE,UAAM,0BAA0B,QAAQ;AACxC,UAAM,cAAc,kBAAkB,OAAO,MAAM,QAAQ,SAAS,KAAK;AACzE,UAAM,iBAAiB,QAAQ,kBAAkB;AACjD,UAAM,eAAe,QAAQ,gBAAgB;AAC7C,UAAM,iBAAiB,mBAAmB,QAAQ,KAAK;AACvD,UAAM,cAAc,qBAAqB,QAAQ,KAAK,EAAE,SAAS;AACjE,UAAM,kBAAkB,oBAAI,IAAkC;AAC9D,UAAM,gBAAgB,oBAAI,IAAgC;AAE1D,eAAW,aAAa,aAAa;AACnC,UAAI,UAAU,oBAAoB,IAAI;AACpC;AAAA,MACF;AAEA,YAAM,qBAAqB,KAAK;AAAA,QAC9B;AAAA,QACA,iBAAiB,gBAAgB;AAAA,MACnC;AACA,YAAM,mBAAmB,KAAK,IAAI,GAAG,eAAe,cAAc,IAAI;AAEtE,WACG,CAAC,kBAAkB,sBAAsB,OACzC,CAAC,eAAe,oBAAoB,IACrC;AACA;AAAA,MACF;AAEA,YAAM,CAAC,YAAY,QAAQ,IAAI,MAAM,QAAQ,IAAI;AAAA,QAC/C,gBAAgB,UAAU,UAAU,iBAAiB;AAAA,UACnD,GAAG;AAAA,UACH,gBAAgB;AAAA,QAClB,CAAC;AAAA,QACD,cAAc,UAAU,UAAU,iBAAiB;AAAA,UACjD,GAAG;AAAA,UACH,cAAc;AAAA,QAChB,CAAC;AAAA,MACH,CAAC;AAED,iBAAW,OAAO,YAAY;AAC5B,cAAM,MAAM,mBAAmB,GAAG;AAElC,YAAI,CAAC,gBAAgB,IAAI,GAAG,GAAG;AAC7B,0BAAgB,IAAI,KAAK,GAAG;AAAA,QAC9B;AAAA,MACF;AACA,iBAAW,OAAO,UAAU;AAC1B,cAAM,MAAM,iBAAiB,GAAG;AAEhC,YAAI,CAAC,cAAc,IAAI,GAAG,GAAG;AAC3B,wBAAc,IAAI,KAAK,GAAG;AAAA,QAC5B;AAAA,MACF;AACA,WACG,CAAC,kBAAkB,gBAAgB,QAAQ,oBAC3C,CAAC,eAAe,cAAc,QAAQ,eACvC;AACA;AAAA,MACF;AAAA,IACF;AAEA,WAAO;AAAA,MACL,YAAY,CAAC,GAAG,gBAAgB,OAAO,CAAC;AAAA,MACxC;AAAA,MACA,UAAU,CAAC,GAAG,cAAc,OAAO,CAAC;AAAA,IACtC;AAAA,EACF,CAAC;AACH;AAEA,eAAe,gBACb,UACA,iBACA,SAK0C;AAC1C,MAAI,CAAC,mBAAmB,QAAQ,KAAK,KAAK,QAAQ,mBAAmB,GAAG;AACtE,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,MAAM,SAAS;AAAA,IACpB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,UAYM,iBAAiB,QAAQ,QAAQ,CAAC;AAAA,UAClC,oBAAoB,QAAQ,KAAK,CAAC;AAAA;AAAA,QAEpC,eAAe,QAAQ,cAAc,CAAC;AAAA;AAAA,IAE1C;AAAA,MACE,GAAG;AAAA,MACH;AAAA,MACA,GAAG,oBAAoB,QAAQ,QAAQ;AAAA,MACvC,GAAG,uBAAuB,QAAQ,KAAK;AAAA,MACvC,GAAG,kBAAkB,QAAQ,cAAc;AAAA,IAC7C;AAAA,IACA,CAAC,SAAS;AAAA,MACR,SAAS,OAAO,IAAI,QAAQ;AAAA,MAC5B,WAAW,UAAU,KAAK,YAAY;AAAA,MACtC,WAAW,UAAU,KAAK,YAAY;AAAA,MACtC,cAAc,UAAU,KAAK,eAAe;AAAA,MAC5C,OAAO,YAAY,UAAU,KAAK,MAAM,CAAC;AAAA,MACzC,GAAI,IAAI,eAAe,OACnB,CAAC,IACD,EAAE,WAAW,UAAU,KAAK,YAAY,EAAE;AAAA,IAChD;AAAA,EACF;AACF;AAEA,eAAe,cACb,UACA,iBACA,SAYwC;AACxC,QAAM,QAAQ,qBAAqB,QAAQ,KAAK;AAEhD,MAAI,MAAM,WAAW,GAAG;AACtB,WAAO,CAAC;AAAA,EACV;AACA,MAAI,QAAQ,iBAAiB,GAAG;AAC9B,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,QAAQ,QAAQ;AAEtB,SAAO,MAAM,SAAS;AAAA,IACpB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,2BAoBuB,MAAM,IAAI,MAAM,GAAG,EAAE,KAAK,IAAI,CAAC;AAAA,YAC9C,iBAAiB,QAAQ,QAAQ,CAAC;AAAA;AAAA,QAGtC,UAAU,SACN,KACA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,WAoBN;AAAA;AAAA,QAEE,eAAe,QAAQ,YAAY,CAAC;AAAA;AAAA,IAExC;AAAA,MACE,GAAG;AAAA,MACH;AAAA,MACA,GAAG;AAAA,MACH,GAAG,oBAAoB,QAAQ,QAAQ;AAAA,MACvC,GAAI,UAAU,SACV,CAAC,IACD;AAAA,QACE,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,QACN,MAAM;AAAA,MACR;AAAA,MACJ,GAAG,kBAAkB,QAAQ,YAAY;AAAA,IAC3C;AAAA,IACA,CAAC,QAAQ;AACP,YAAM,OAAO,UAAU,KAAK,MAAM;AAElC,aAAO;AAAA,QACL,WAAW,UAAU,KAAK,YAAY;AAAA,QACtC,WAAW,UAAU,KAAK,YAAY;AAAA,QACtC,MAAM,UAAU,KAAK,MAAM;AAAA,QAC3B;AAAA,QACA,OAAO,YAAY,IAAI;AAAA,QACvB,eAAe,UAAU,KAAK,gBAAgB;AAAA,QAC9C,YAAY,UAAU,KAAK,aAAa;AAAA,MAC1C;AAAA,IACF;AAAA,EACF;AACF;","names":[]}

package/dist/retrieval/search-index/search/settings.d.ts

@@ -0,0 +1,29 @@
+import { ReadonlyDocument, Document } from '../../../document/directory/types.js';
+import { ArchiveIndexSettings } from './types.js';
+import '../../../text/source/meta.js';
+import 'zod';
+import '../../../text/source/toc.js';
+import '../../../text/source/types.js';
+import '../../../document/database.js';
+import 'sqlite3';
+import '../../../document/types.js';
+import '../../../document/text-streams/serial.js';
+import '../../../document/text-streams/draft.js';
+import '../../../document/text-streams/types.js';
+import '../../../document/stores/chunk.js';
+import '../../../document/stores/types.js';
+import '../../../document/stores/fragment-group.js';
+import '../../../document/stores/graph-build-parameter.js';
+import '../../../document/stores/mention-link.js';
+import '../../../document/stores/mention.js';
+import '../../../document/stores/object-metadata.js';
+import '../../../document/stores/reading-edge.js';
+import '../../../document/stores/serial.js';
+import '../../../document/stores/snake/chunk.js';
+import '../../../document/stores/snake/core.js';
+import '../../../document/stores/snake/edge.js';
+
+declare function readArchiveIndexSettings(document: ReadonlyDocument): Promise<ArchiveIndexSettings>;
+declare function setFtsIndexEmbedded(document: Document, embedded: boolean): Promise<void>;
+
+export { readArchiveIndexSettings, setFtsIndexEmbedded };

package/dist/retrieval/search-index/search/settings.js

@@ -0,0 +1,35 @@
+import { getNumber } from "../../../document/database.js";
+async function readArchiveIndexSettings(document) {
+  return await document.readDatabase(async (database) => {
+    const row = await database.queryOne(
+      `
+        SELECT fts_embedded
+        FROM archive_index_settings
+        WHERE id = 1
+      `,
+      void 0,
+      (value) => ({
+        ftsEmbedded: getNumber(value, "fts_embedded") !== 0
+      })
+    );
+    return row ?? { ftsEmbedded: false };
+  });
+}
+async function setFtsIndexEmbedded(document, embedded) {
+  await document.readDatabase(async (database) => {
+    await database.run(
+      `
+        INSERT INTO archive_index_settings(id, fts_embedded)
+        VALUES (1, ?)
+        ON CONFLICT(id)
+        DO UPDATE SET fts_embedded = excluded.fts_embedded
+      `,
+      [embedded ? 1 : 0]
+    );
+  });
+}
+export {
+  readArchiveIndexSettings,
+  setFtsIndexEmbedded
+};
+//# sourceMappingURL=settings.js.map

package/dist/retrieval/search-index/search/settings.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/retrieval/search-index/search/settings.ts"],"sourcesContent":["import { getNumber } from \"../../../document/database.js\";\nimport type { Document, ReadonlyDocument } from \"../../../document/index.js\";\nimport type { ArchiveIndexSettings } from \"./types.js\";\n\nexport async function readArchiveIndexSettings(\n  document: ReadonlyDocument,\n): Promise<ArchiveIndexSettings> {\n  return await document.readDatabase(async (database) => {\n    const row = await database.queryOne(\n      `\n        SELECT fts_embedded\n        FROM archive_index_settings\n        WHERE id = 1\n      `,\n      undefined,\n      (value) => ({\n        ftsEmbedded: getNumber(value, \"fts_embedded\") !== 0,\n      }),\n    );\n\n    return row ?? { ftsEmbedded: false };\n  });\n}\n\nexport async function setFtsIndexEmbedded(\n  document: Document,\n  embedded: boolean,\n): Promise<void> {\n  await document.readDatabase(async (database) => {\n    await database.run(\n      `\n        INSERT INTO archive_index_settings(id, fts_embedded)\n        VALUES (1, ?)\n        ON CONFLICT(id)\n        DO UPDATE SET fts_embedded = excluded.fts_embedded\n      `,\n      [embedded ? 1 : 0],\n    );\n  });\n}\n"],"mappings":"AAAA,SAAS,iBAAiB;AAI1B,eAAsB,yBACpB,UAC+B;AAC/B,SAAO,MAAM,SAAS,aAAa,OAAO,aAAa;AACrD,UAAM,MAAM,MAAM,SAAS;AAAA,MACzB;AAAA;AAAA;AAAA;AAAA;AAAA,MAKA;AAAA,MACA,CAAC,WAAW;AAAA,QACV,aAAa,UAAU,OAAO,cAAc,MAAM;AAAA,MACpD;AAAA,IACF;AAEA,WAAO,OAAO,EAAE,aAAa,MAAM;AAAA,EACrC,CAAC;AACH;AAEA,eAAsB,oBACpB,UACA,UACe;AACf,QAAM,SAAS,aAAa,OAAO,aAAa;AAC9C,UAAM,SAAS;AAAA,MACb;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,MAMA,CAAC,WAAW,IAAI,CAAC;AAAA,IACnB;AAAA,EACF,CAAC;AACH;","names":[]}

package/dist/retrieval/search-index/search/status.d.ts

@@ -0,0 +1,31 @@
+import { Database } from '../../../document/database.js';
+import { ReadonlyDocument } from '../../../document/directory/types.js';
+import { SearchIndexInput, SearchIndexStatus } from './types.js';
+import 'sqlite3';
+import '../../../text/source/meta.js';
+import 'zod';
+import '../../../text/source/toc.js';
+import '../../../text/source/types.js';
+import '../../../document/types.js';
+import '../../../document/text-streams/serial.js';
+import '../../../document/text-streams/draft.js';
+import '../../../document/text-streams/types.js';
+import '../../../document/stores/chunk.js';
+import '../../../document/stores/types.js';
+import '../../../document/stores/fragment-group.js';
+import '../../../document/stores/graph-build-parameter.js';
+import '../../../document/stores/mention-link.js';
+import '../../../document/stores/mention.js';
+import '../../../document/stores/object-metadata.js';
+import '../../../document/stores/reading-edge.js';
+import '../../../document/stores/serial.js';
+import '../../../document/stores/snake/chunk.js';
+import '../../../document/stores/snake/core.js';
+import '../../../document/stores/snake/edge.js';
+
+declare function isSearchIndexCurrent(document: ReadonlyDocument, input?: SearchIndexInput): Promise<boolean>;
+declare function readSearchIndexStatus(document: ReadonlyDocument, input?: SearchIndexInput): Promise<SearchIndexStatus>;
+declare function hasDirtySearchIndexChapters(database: Database): Promise<boolean>;
+declare function assertSearchIndexNotDirty(database: Database): Promise<void>;
+
+export { assertSearchIndexNotDirty, hasDirtySearchIndexChapters, isSearchIndexCurrent, readSearchIndexStatus };

package/dist/retrieval/search-index/search/status.js

@@ -0,0 +1,70 @@
+import { getNumber } from "../../../document/database.js";
+import { isMissingSearchIndexError } from "./errors.js";
+import {
+  createSearchIndexFingerprint,
+  readSearchIndexFingerprintFromDatabase
+} from "./fingerprint.js";
+async function isSearchIndexCurrent(document, input) {
+  return await readSearchIndexStatus(document, input) === "current";
+}
+async function readSearchIndexStatus(document, input) {
+  const fingerprint = input === void 0 ? void 0 : createSearchIndexFingerprint(input);
+  try {
+    return await document.readSearchIndexDatabase(async (database) => {
+      if (await hasDirtySearchIndexChapters(database)) {
+        return "dirty";
+      }
+      const indexedFingerprint = await readSearchIndexFingerprintFromDatabase(database);
+      if (indexedFingerprint === void 0) {
+        return "dirty";
+      }
+      return fingerprint === void 0 || indexedFingerprint === fingerprint ? "current" : "dirty";
+    });
+  } catch (error) {
+    if (isMissingSearchIndexError(error)) {
+      return "missing";
+    }
+    throw error;
+  }
+}
+async function hasDirtySearchIndexChapters(database) {
+  if (!await hasTable(database, "index_dirty_chapters")) {
+    return false;
+  }
+  const dirty = await database.queryOne(
+    `
+      SELECT 1 AS found
+      FROM index_dirty_chapters
+      LIMIT 1
+    `,
+    void 0,
+    (row) => getNumber(row, "found")
+  );
+  return dirty === 1;
+}
+async function hasTable(database, table) {
+  const row = await database.queryOne(
+    `
+      SELECT 1 AS found
+      FROM sqlite_master
+      WHERE type = 'table' AND name = ?
+    `,
+    [table],
+    (value) => getNumber(value, "found")
+  );
+  return row === 1;
+}
+async function assertSearchIndexNotDirty(database) {
+  if (await hasDirtySearchIndexChapters(database)) {
+    throw new Error(
+      "Archive search index is dirty; rebuild the index before querying."
+    );
+  }
+}
+export {
+  assertSearchIndexNotDirty,
+  hasDirtySearchIndexChapters,
+  isSearchIndexCurrent,
+  readSearchIndexStatus
+};
+//# sourceMappingURL=status.js.map

package/dist/retrieval/search-index/search/status.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../src/retrieval/search-index/search/status.ts"],"sourcesContent":["import { getNumber, type Database } from \"../../../document/database.js\";\nimport type { ReadonlyDocument } from \"../../../document/index.js\";\nimport { isMissingSearchIndexError } from \"./errors.js\";\nimport {\n  createSearchIndexFingerprint,\n  readSearchIndexFingerprintFromDatabase,\n} from \"./fingerprint.js\";\nimport type { SearchIndexInput, SearchIndexStatus } from \"./types.js\";\n\nexport async function isSearchIndexCurrent(\n  document: ReadonlyDocument,\n  input?: SearchIndexInput,\n): Promise<boolean> {\n  return (await readSearchIndexStatus(document, input)) === \"current\";\n}\n\nexport async function readSearchIndexStatus(\n  document: ReadonlyDocument,\n  input?: SearchIndexInput,\n): Promise<SearchIndexStatus> {\n  const fingerprint =\n    input === undefined ? undefined : createSearchIndexFingerprint(input);\n\n  try {\n    return await document.readSearchIndexDatabase(async (database) => {\n      if (await hasDirtySearchIndexChapters(database)) {\n        return \"dirty\";\n      }\n\n      const indexedFingerprint =\n        await readSearchIndexFingerprintFromDatabase(database);\n\n      if (indexedFingerprint === undefined) {\n        return \"dirty\";\n      }\n\n      return fingerprint === undefined || indexedFingerprint === fingerprint\n        ? \"current\"\n        : \"dirty\";\n    });\n  } catch (error) {\n    if (isMissingSearchIndexError(error)) {\n      return \"missing\";\n    }\n\n    throw error;\n  }\n}\n\nexport async function hasDirtySearchIndexChapters(\n  database: Database,\n): Promise<boolean> {\n  if (!(await hasTable(database, \"index_dirty_chapters\"))) {\n    return false;\n  }\n\n  const dirty = await database.queryOne(\n    `\n      SELECT 1 AS found\n      FROM index_dirty_chapters\n      LIMIT 1\n    `,\n    undefined,\n    (row) => getNumber(row, \"found\"),\n  );\n\n  return dirty === 1;\n}\n\nasync function hasTable(database: Database, table: string): Promise<boolean> {\n  const row = await database.queryOne(\n    `\n      SELECT 1 AS found\n      FROM sqlite_master\n      WHERE type = 'table' AND name = ?\n    `,\n    [table],\n    (value) => getNumber(value, \"found\"),\n  );\n\n  return row === 1;\n}\n\nexport async function assertSearchIndexNotDirty(\n  database: Database,\n): Promise<void> {\n  if (await hasDirtySearchIndexChapters(database)) {\n    throw new Error(\n      \"Archive search index is dirty; rebuild the index before querying.\",\n    );\n  }\n}\n"],"mappings":"AAAA,SAAS,iBAAgC;AAEzC,SAAS,iCAAiC;AAC1C;AAAA,EACE;AAAA,EACA;AAAA,OACK;AAGP,eAAsB,qBACpB,UACA,OACkB;AAClB,SAAQ,MAAM,sBAAsB,UAAU,KAAK,MAAO;AAC5D;AAEA,eAAsB,sBACpB,UACA,OAC4B;AAC5B,QAAM,cACJ,UAAU,SAAY,SAAY,6BAA6B,KAAK;AAEtE,MAAI;AACF,WAAO,MAAM,SAAS,wBAAwB,OAAO,aAAa;AAChE,UAAI,MAAM,4BAA4B,QAAQ,GAAG;AAC/C,eAAO;AAAA,MACT;AAEA,YAAM,qBACJ,MAAM,uCAAuC,QAAQ;AAEvD,UAAI,uBAAuB,QAAW;AACpC,eAAO;AAAA,MACT;AAEA,aAAO,gBAAgB,UAAa,uBAAuB,cACvD,YACA;AAAA,IACN,CAAC;AAAA,EACH,SAAS,OAAO;AACd,QAAI,0BAA0B,KAAK,GAAG;AACpC,aAAO;AAAA,IACT;AAEA,UAAM;AAAA,EACR;AACF;AAEA,eAAsB,4BACpB,UACkB;AAClB,MAAI,CAAE,MAAM,SAAS,UAAU,sBAAsB,GAAI;AACvD,WAAO;AAAA,EACT;AAEA,QAAM,QAAQ,MAAM,SAAS;AAAA,IAC3B;AAAA;AAAA;AAAA;AAAA;AAAA,IAKA;AAAA,IACA,CAAC,QAAQ,UAAU,KAAK,OAAO;AAAA,EACjC;AAEA,SAAO,UAAU;AACnB;AAEA,eAAe,SAAS,UAAoB,OAAiC;AAC3E,QAAM,MAAM,MAAM,SAAS;AAAA,IACzB;AAAA;AAAA;AAAA;AAAA;AAAA,IAKA,CAAC,KAAK;AAAA,IACN,CAAC,UAAU,UAAU,OAAO,OAAO;AAAA,EACrC;AAEA,SAAO,QAAQ;AACjB;AAEA,eAAsB,0BACpB,UACe;AACf,MAAI,MAAM,4BAA4B,QAAQ,GAAG;AAC/C,UAAM,IAAI;AAAA,MACR;AAAA,IACF;AAAA,EACF;AACF;","names":[]}