wiki-graph-core 0.3.1 → 0.4.1

package/dist/retrieval/query/archive-view/search/core.d.ts

@@ -0,0 +1,31 @@
+import { ReadonlyDocument } from '../../../../document/directory/types.js';
+import { ArchiveFindOptions, ArchiveFindResult } 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 findArchiveObjects(document: ReadonlyDocument, query: string, options?: ArchiveFindOptions): Promise<ArchiveFindResult>;
+declare function grepArchiveObjects(document: ReadonlyDocument, query: string, options?: ArchiveFindOptions): Promise<ArchiveFindResult>;
+
+export { findArchiveObjects, grepArchiveObjects };

package/dist/retrieval/query/archive-view/search/core.js

@@ -0,0 +1,388 @@
+import {
+  createLexicalQuery,
+  listLexicalQueryCandidateTerms
+} from "../../lexical-search.js";
+import {
+  createEntitySearchSession,
+  createSearchSession,
+  decodeSearchSessionCursor,
+  readCachedEntitySearchSessionPage,
+  readCachedSearchSessionPage,
+  readEntitySearchSessionPage,
+  readSearchSessionDescriptor,
+  readSearchSessionPage
+} from "../../search-cache/index.js";
+import {
+  BROAD_FIND_LENS_HINT,
+  DEFAULT_FIND_LIMIT,
+  compareNumbers,
+  createFindResult,
+  createPhraseSearch,
+  createRankedFindResult,
+  isFindCursor,
+  parseFindLens,
+  parseFindMatch,
+  parseFindTypes
+} from "../helpers.js";
+import { isArchiveSearchIndexCurrent } from "../index-state.js";
+import {
+  assertSearchCursorTypesMatch,
+  createEntitySearchCacheInput,
+  createSentenceEvidenceSearchCacheInput,
+  isEntityOnlySearch,
+  isEntitySearchTypes
+} from "./cache-input.js";
+import {
+  filterLexicalHitsByMatch,
+  findChapters,
+  findEntities,
+  findMeta,
+  findNodes,
+  findTriples
+} from "../find.js";
+import {
+  createFindEvidenceHydrationOptions,
+  hydrateFindHitEvidence
+} from "../evidence.js";
+import { hydrateFindResultBacklinks } from "../backlinks.js";
+import {
+  readBucketedSearchResultPage,
+  tryDecodeBucketSearchSessionCursor
+} from "./buckets.js";
+import { findArchiveObjectsIndexed, isTextOnlySearch } from "./hydration.js";
+async function findArchiveObjects(document, query, options = {}) {
+  const limit = options.limit ?? DEFAULT_FIND_LIMIT;
+  const textOnlySearch = isTextOnlySearch(options);
+  if (options.cursor !== void 0 && (!textOnlySearch || !isFindCursor(options.cursor))) {
+    const bucketCursor = tryDecodeBucketSearchSessionCursor(options.cursor);
+    if (bucketCursor !== void 0) {
+      return await readBucketedSearchResultPage(document, bucketCursor, {
+        ...options,
+        limit
+      });
+    }
+    const cursor = decodeSearchSessionCursor(options.cursor);
+    const descriptor = await readSearchSessionDescriptor(
+      cursor.sessionId,
+      options.archiveKey ?? "archive"
+    );
+    assertSearchCursorTypesMatch(options.types, descriptor.types);
+    const page = isEntitySearchTypes(descriptor.types) ? await readEntitySearchSessionPage(
+      cursor.sessionId,
+      cursor.offset,
+      limit,
+      options.archiveKey ?? "archive",
+      cursor.createdAt
+    ) : await readSearchSessionPage(
+      cursor.sessionId,
+      cursor.offset,
+      limit,
+      options.archiveKey ?? "archive",
+      cursor.createdAt
+    );
+    return await hydrateFindResultBacklinks(
+      document,
+      {
+        chapters: page.chapters,
+        items: await hydrateFindHitEvidence(
+          document,
+          page.items,
+          createFindEvidenceHydrationOptions(options, cursor.sessionId)
+        ),
+        lens: parseFindLens(page.lens),
+        lensHint: page.lens === "broad" ? BROAD_FIND_LENS_HINT : null,
+        limit,
+        match: parseFindMatch(page.match),
+        nextCursor: page.nextCursor,
+        order: options.order ?? "doc-asc",
+        query: page.query,
+        terms: page.terms,
+        types: parseFindTypes(descriptor.types)
+      },
+      options
+    );
+  }
+  const requestedTypes = options.types ?? null;
+  const wantsStructuredSearch = requestedTypes === null || requestedTypes.includes("entity") || requestedTypes.includes("triple");
+  const search = createLexicalQuery(query);
+  if (search === void 0) {
+    return createFindResult(query, [], options);
+  }
+  if (textOnlySearch) {
+    return await findTextOnlyArchiveObjectsIndexed(
+      document,
+      query,
+      options,
+      search
+    );
+  }
+  const revisionScope = await createSearchRevisionScope(
+    document,
+    options.chapters
+  );
+  const cacheInput = {
+    archiveKey: options.archiveKey ?? "archive",
+    chapters: options.chapters ?? null,
+    lens: options.types === void 0 ? "broad" : "typed",
+    match: options.match ?? "any",
+    order: options.order ?? "doc-asc",
+    query,
+    revisionScope,
+    terms: search.terms,
+    types: options.types ?? null
+  };
+  const canReadSearchCache = options.triplePattern === void 0;
+  const usesBucketedSearch = options.types === void 0 && options.triplePattern === void 0;
+  if (canReadSearchCache && isEntityOnlySearch(options)) {
+    const cachedPage = await readCachedEntitySearchSessionPage(
+      cacheInput,
+      0,
+      limit
+    );
+    if (cachedPage !== void 0) {
+      return await hydrateFindResultBacklinks(
+        document,
+        {
+          chapters: cachedPage.chapters,
+          items: await hydrateFindHitEvidence(
+            document,
+            cachedPage.items,
+            createFindEvidenceHydrationOptions(options, cachedPage.sessionId)
+          ),
+          lens: parseFindLens(cachedPage.lens),
+          lensHint: cachedPage.lens === "broad" ? BROAD_FIND_LENS_HINT : null,
+          limit,
+          match: parseFindMatch(cachedPage.match),
+          nextCursor: cachedPage.nextCursor,
+          order: options.order ?? "doc-asc",
+          query: cachedPage.query,
+          terms: cachedPage.terms,
+          types: parseFindTypes(cachedPage.types)
+        },
+        options
+      );
+    }
+  } else if (canReadSearchCache && !usesBucketedSearch) {
+    const cachedPage = await readCachedSearchSessionPage(cacheInput, 0, limit);
+    if (cachedPage !== void 0) {
+      return await hydrateFindResultBacklinks(
+        document,
+        {
+          chapters: cachedPage.chapters,
+          items: await hydrateFindHitEvidence(
+            document,
+            cachedPage.items,
+            createFindEvidenceHydrationOptions(options)
+          ),
+          lens: parseFindLens(cachedPage.lens),
+          lensHint: cachedPage.lens === "broad" ? BROAD_FIND_LENS_HINT : null,
+          limit,
+          match: parseFindMatch(cachedPage.match),
+          nextCursor: cachedPage.nextCursor,
+          order: options.order ?? "doc-asc",
+          query: cachedPage.query,
+          terms: cachedPage.terms,
+          types: parseFindTypes(cachedPage.types)
+        },
+        options
+      );
+    }
+  }
+  if (usesBucketedSearch) {
+    if (!await isArchiveSearchIndexCurrent(document)) {
+      throw new Error(
+        "Wiki Graph search index is missing or outdated. Run `<archive-uri>/index enable` before searching."
+      );
+    }
+    const sessionId2 = await createSearchSession({
+      archiveKey: options.archiveKey ?? "archive",
+      chapters: options.chapters ?? null,
+      lens: "broad",
+      match: options.match ?? "any",
+      order: options.order ?? "doc-asc",
+      query,
+      revisionScope,
+      terms: search.terms,
+      types: null
+    });
+    const descriptor = await readSearchSessionDescriptor(
+      sessionId2,
+      options.archiveKey ?? "archive"
+    );
+    return await readBucketedSearchResultPage(
+      document,
+      {
+        createdAt: descriptor.createdAt,
+        cursor: { bucket: 0 },
+        sessionId: sessionId2
+      },
+      { ...options, limit }
+    );
+  }
+  const allMentions = wantsStructuredSearch ? await document.mentions.listBySurfaceTerms(
+    listLexicalQueryCandidateTerms(query)
+  ) : [];
+  const indexed = await findArchiveObjectsIndexed(document, query, options);
+  const structuredHits = wantsStructuredSearch ? [
+    ...findEntities(search, { mentions: allMentions }),
+    ...await findTriples(document, search, { mentions: allMentions })
+  ] : [];
+  const hits = [...structuredHits, ...indexed?.hits ?? []];
+  if (isEntityOnlySearch(options)) {
+    const ranked2 = createRankedFindResult(
+      query,
+      filterLexicalHitsByMatch(hits, search, options.match ?? "any"),
+      options,
+      search.terms
+    );
+    const entityCacheInput2 = createEntitySearchCacheInput(
+      ranked2.items,
+      indexed?.result
+    );
+    const sentenceCacheInput2 = await createSentenceEvidenceSearchCacheInput(
+      document,
+      indexed?.result,
+      options
+    );
+    const sessionId2 = await createEntitySearchSession({
+      archiveKey: options.archiveKey ?? "archive",
+      chapters: ranked2.chapters,
+      chunkHits: sentenceCacheInput2.chunkHits,
+      entityHits: [
+        ...entityCacheInput2.entityHits,
+        ...sentenceCacheInput2.entityHits
+      ],
+      evidenceEvents: [
+        ...entityCacheInput2.evidenceEvents,
+        ...sentenceCacheInput2.evidenceEvents
+      ],
+      lens: ranked2.lens,
+      match: ranked2.match,
+      order: ranked2.order,
+      query,
+      revisionScope,
+      terms: ranked2.terms,
+      tripleHits: sentenceCacheInput2.tripleHits,
+      types: ranked2.types
+    });
+    const firstPage2 = await readEntitySearchSessionPage(sessionId2, 0, limit);
+    return await hydrateFindResultBacklinks(
+      document,
+      {
+        ...ranked2,
+        items: await hydrateFindHitEvidence(
+          document,
+          firstPage2.items,
+          createFindEvidenceHydrationOptions(options, sessionId2)
+        ),
+        nextCursor: firstPage2.nextCursor
+      },
+      options
+    );
+  }
+  const ranked = createRankedFindResult(
+    query,
+    filterLexicalHitsByMatch(hits, search, options.match ?? "any"),
+    options,
+    search.terms
+  );
+  const entityCacheInput = createEntitySearchCacheInput(
+    ranked.items,
+    indexed?.result
+  );
+  const sentenceCacheInput = await createSentenceEvidenceSearchCacheInput(
+    document,
+    indexed?.result,
+    options
+  );
+  const sessionId = await createSearchSession({
+    archiveKey: options.archiveKey ?? "archive",
+    chapters: ranked.chapters,
+    chunkHits: sentenceCacheInput.chunkHits,
+    entityHits: [
+      ...entityCacheInput.entityHits,
+      ...sentenceCacheInput.entityHits
+    ],
+    evidenceEvents: [
+      ...entityCacheInput.evidenceEvents,
+      ...sentenceCacheInput.evidenceEvents
+    ],
+    items: ranked.items,
+    lens: ranked.lens,
+    match: ranked.match,
+    order: ranked.order,
+    query,
+    revisionScope,
+    terms: ranked.terms,
+    tripleHits: sentenceCacheInput.tripleHits,
+    types: ranked.types
+  });
+  const firstPage = await readSearchSessionPage(sessionId, 0, limit);
+  return await hydrateFindResultBacklinks(
+    document,
+    {
+      ...ranked,
+      items: await hydrateFindHitEvidence(
+        document,
+        firstPage.items,
+        createFindEvidenceHydrationOptions(options)
+      ),
+      nextCursor: firstPage.nextCursor
+    },
+    options
+  );
+}
+async function findTextOnlyArchiveObjectsIndexed(document, query, options, search) {
+  const indexed = await findArchiveObjectsIndexed(document, query, options);
+  const hits = indexed?.hits ?? [];
+  return createFindResult(
+    query,
+    filterLexicalHitsByMatch(hits, search, options.match ?? "any"),
+    options,
+    indexed?.result.terms ?? search.terms
+  );
+}
+async function createSearchRevisionScope(document, chapters) {
+  if (chapters === void 0 || chapters.length === 0) {
+    return JSON.stringify({
+      chaptersRevision: await document.serials.getChaptersRevision(),
+      scope: "all"
+    });
+  }
+  const uniqueChapters = [...new Set(chapters)].sort(compareNumbers);
+  const revisions = await document.serials.getRevisions(uniqueChapters);
+  return JSON.stringify({
+    chapters: uniqueChapters.map(
+      (chapterId) => [chapterId, revisions.get(chapterId) ?? 0]
+    ),
+    scope: "chapters"
+  });
+}
+async function grepArchiveObjects(document, query, options = {}) {
+  const search = createPhraseSearch(query);
+  if (search === void 0) {
+    return createFindResult(
+      query,
+      [],
+      { ...options, match: "all" },
+      [],
+      "exact"
+    );
+  }
+  const hits = [];
+  hits.push(...findMeta(await document.readBookMeta(), search));
+  hits.push(...await findChapters(document, search));
+  hits.push(...await findNodes(document, search));
+  return createFindResult(
+    query,
+    hits,
+    { ...options, match: "all" },
+    [query.trim().toLowerCase()],
+    "exact"
+  );
+}
+export {
+  findArchiveObjects,
+  grepArchiveObjects
+};
+//# sourceMappingURL=core.js.map

package/dist/retrieval/query/archive-view/search/core.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../../../../src/retrieval/query/archive-view/search/core.ts"],"sourcesContent":["import type { ReadonlyDocument } from \"../../../../document/index.js\";\nimport {\n  createLexicalQuery,\n  listLexicalQueryCandidateTerms,\n  type LexicalQuery,\n} from \"../../lexical-search.js\";\nimport {\n  createEntitySearchSession,\n  createSearchSession,\n  decodeSearchSessionCursor,\n  readCachedEntitySearchSessionPage,\n  readCachedSearchSessionPage,\n  readEntitySearchSessionPage,\n  readSearchSessionDescriptor,\n  readSearchSessionPage,\n} from \"../../search-cache/index.js\";\n\nimport {\n  BROAD_FIND_LENS_HINT,\n  DEFAULT_FIND_LIMIT,\n  compareNumbers,\n  createFindResult,\n  createPhraseSearch,\n  createRankedFindResult,\n  isFindCursor,\n  parseFindLens,\n  parseFindMatch,\n  parseFindTypes,\n} from \"../helpers.js\";\nimport { isArchiveSearchIndexCurrent } from \"../index-state.js\";\nimport {\n  assertSearchCursorTypesMatch,\n  createEntitySearchCacheInput,\n  createSentenceEvidenceSearchCacheInput,\n  isEntityOnlySearch,\n  isEntitySearchTypes,\n} from \"./cache-input.js\";\nimport {\n  filterLexicalHitsByMatch,\n  findChapters,\n  findEntities,\n  findMeta,\n  findNodes,\n  findTriples,\n} from \"../find.js\";\nimport {\n  createFindEvidenceHydrationOptions,\n  hydrateFindHitEvidence,\n} from \"../evidence.js\";\nimport { hydrateFindResultBacklinks } from \"../backlinks.js\";\nimport {\n  readBucketedSearchResultPage,\n  tryDecodeBucketSearchSessionCursor,\n} from \"./buckets.js\";\nimport { findArchiveObjectsIndexed, isTextOnlySearch } from \"./hydration.js\";\nimport type {\n  ArchiveFindHit,\n  ArchiveFindOptions,\n  ArchiveFindResult,\n} from \"../types.js\";\n\nexport async function findArchiveObjects(\n  document: ReadonlyDocument,\n  query: string,\n  options: ArchiveFindOptions = {},\n): Promise<ArchiveFindResult> {\n  const limit = options.limit ?? DEFAULT_FIND_LIMIT;\n  const textOnlySearch = isTextOnlySearch(options);\n\n  if (\n    options.cursor !== undefined &&\n    (!textOnlySearch || !isFindCursor(options.cursor))\n  ) {\n    const bucketCursor = tryDecodeBucketSearchSessionCursor(options.cursor);\n\n    if (bucketCursor !== undefined) {\n      return await readBucketedSearchResultPage(document, bucketCursor, {\n        ...options,\n        limit,\n      });\n    }\n\n    const cursor = decodeSearchSessionCursor(options.cursor);\n    const descriptor = await readSearchSessionDescriptor(\n      cursor.sessionId,\n      options.archiveKey ?? \"archive\",\n    );\n\n    assertSearchCursorTypesMatch(options.types, descriptor.types);\n\n    const page = isEntitySearchTypes(descriptor.types)\n      ? await readEntitySearchSessionPage(\n          cursor.sessionId,\n          cursor.offset,\n          limit,\n          options.archiveKey ?? \"archive\",\n          cursor.createdAt,\n        )\n      : await readSearchSessionPage(\n          cursor.sessionId,\n          cursor.offset,\n          limit,\n          options.archiveKey ?? \"archive\",\n          cursor.createdAt,\n        );\n\n    return await hydrateFindResultBacklinks(\n      document,\n      {\n        chapters: page.chapters,\n        items: await hydrateFindHitEvidence(\n          document,\n          page.items,\n          createFindEvidenceHydrationOptions(options, cursor.sessionId),\n        ),\n        lens: parseFindLens(page.lens),\n        lensHint: page.lens === \"broad\" ? BROAD_FIND_LENS_HINT : null,\n        limit,\n        match: parseFindMatch(page.match),\n        nextCursor: page.nextCursor,\n        order: options.order ?? \"doc-asc\",\n        query: page.query,\n        terms: page.terms,\n        types: parseFindTypes(descriptor.types),\n      },\n      options,\n    );\n  }\n\n  const requestedTypes = options.types ?? null;\n  const wantsStructuredSearch =\n    requestedTypes === null ||\n    requestedTypes.includes(\"entity\") ||\n    requestedTypes.includes(\"triple\");\n  const search = createLexicalQuery(query);\n\n  if (search === undefined) {\n    return createFindResult(query, [], options);\n  }\n\n  if (textOnlySearch) {\n    return await findTextOnlyArchiveObjectsIndexed(\n      document,\n      query,\n      options,\n      search,\n    );\n  }\n\n  const revisionScope = await createSearchRevisionScope(\n    document,\n    options.chapters,\n  );\n  const cacheInput = {\n    archiveKey: options.archiveKey ?? \"archive\",\n    chapters: options.chapters ?? null,\n    lens: options.types === undefined ? \"broad\" : \"typed\",\n    match: options.match ?? \"any\",\n    order: options.order ?? \"doc-asc\",\n    query,\n    revisionScope,\n    terms: search.terms,\n    types: options.types ?? null,\n  };\n  const canReadSearchCache = options.triplePattern === undefined;\n  const usesBucketedSearch =\n    options.types === undefined && options.triplePattern === undefined;\n\n  if (canReadSearchCache && isEntityOnlySearch(options)) {\n    const cachedPage = await readCachedEntitySearchSessionPage(\n      cacheInput,\n      0,\n      limit,\n    );\n\n    if (cachedPage !== undefined) {\n      return await hydrateFindResultBacklinks(\n        document,\n        {\n          chapters: cachedPage.chapters,\n          items: await hydrateFindHitEvidence(\n            document,\n            cachedPage.items,\n            createFindEvidenceHydrationOptions(options, cachedPage.sessionId),\n          ),\n          lens: parseFindLens(cachedPage.lens),\n          lensHint: cachedPage.lens === \"broad\" ? BROAD_FIND_LENS_HINT : null,\n          limit,\n          match: parseFindMatch(cachedPage.match),\n          nextCursor: cachedPage.nextCursor,\n          order: options.order ?? \"doc-asc\",\n          query: cachedPage.query,\n          terms: cachedPage.terms,\n          types: parseFindTypes(cachedPage.types),\n        },\n        options,\n      );\n    }\n  } else if (canReadSearchCache && !usesBucketedSearch) {\n    const cachedPage = await readCachedSearchSessionPage(cacheInput, 0, limit);\n\n    if (cachedPage !== undefined) {\n      return await hydrateFindResultBacklinks(\n        document,\n        {\n          chapters: cachedPage.chapters,\n          items: await hydrateFindHitEvidence(\n            document,\n            cachedPage.items,\n            createFindEvidenceHydrationOptions(options),\n          ),\n          lens: parseFindLens(cachedPage.lens),\n          lensHint: cachedPage.lens === \"broad\" ? BROAD_FIND_LENS_HINT : null,\n          limit,\n          match: parseFindMatch(cachedPage.match),\n          nextCursor: cachedPage.nextCursor,\n          order: options.order ?? \"doc-asc\",\n          query: cachedPage.query,\n          terms: cachedPage.terms,\n          types: parseFindTypes(cachedPage.types),\n        },\n        options,\n      );\n    }\n  }\n\n  if (usesBucketedSearch) {\n    if (!(await isArchiveSearchIndexCurrent(document))) {\n      throw new Error(\n        \"Wiki Graph search index is missing or outdated. Run `<archive-uri>/index enable` before searching.\",\n      );\n    }\n    const sessionId = await createSearchSession({\n      archiveKey: options.archiveKey ?? \"archive\",\n      chapters: options.chapters ?? null,\n      lens: \"broad\",\n      match: options.match ?? \"any\",\n      order: options.order ?? \"doc-asc\",\n      query,\n      revisionScope,\n      terms: search.terms,\n      types: null,\n    });\n    const descriptor = await readSearchSessionDescriptor(\n      sessionId,\n      options.archiveKey ?? \"archive\",\n    );\n\n    return await readBucketedSearchResultPage(\n      document,\n      {\n        createdAt: descriptor.createdAt,\n        cursor: { bucket: 0 },\n        sessionId,\n      },\n      { ...options, limit },\n    );\n  }\n\n  const allMentions = wantsStructuredSearch\n    ? await document.mentions.listBySurfaceTerms(\n        listLexicalQueryCandidateTerms(query),\n      )\n    : [];\n  const indexed = await findArchiveObjectsIndexed(document, query, options);\n  const structuredHits = wantsStructuredSearch\n    ? [\n        ...findEntities(search, { mentions: allMentions }),\n        ...(await findTriples(document, search, { mentions: allMentions })),\n      ]\n    : [];\n  const hits = [...structuredHits, ...(indexed?.hits ?? [])];\n  if (isEntityOnlySearch(options)) {\n    const ranked = createRankedFindResult(\n      query,\n      filterLexicalHitsByMatch(hits, search, options.match ?? \"any\"),\n      options,\n      search.terms,\n    );\n    const entityCacheInput = createEntitySearchCacheInput(\n      ranked.items,\n      indexed?.result,\n    );\n    const sentenceCacheInput = await createSentenceEvidenceSearchCacheInput(\n      document,\n      indexed?.result,\n      options,\n    );\n    const sessionId = await createEntitySearchSession({\n      archiveKey: options.archiveKey ?? \"archive\",\n      chapters: ranked.chapters,\n      chunkHits: sentenceCacheInput.chunkHits,\n      entityHits: [\n        ...entityCacheInput.entityHits,\n        ...sentenceCacheInput.entityHits,\n      ],\n      evidenceEvents: [\n        ...entityCacheInput.evidenceEvents,\n        ...sentenceCacheInput.evidenceEvents,\n      ],\n      lens: ranked.lens,\n      match: ranked.match,\n      order: ranked.order,\n      query,\n      revisionScope,\n      terms: ranked.terms,\n      tripleHits: sentenceCacheInput.tripleHits,\n      types: ranked.types,\n    });\n    const firstPage = await readEntitySearchSessionPage(sessionId, 0, limit);\n\n    return await hydrateFindResultBacklinks(\n      document,\n      {\n        ...ranked,\n        items: await hydrateFindHitEvidence(\n          document,\n          firstPage.items,\n          createFindEvidenceHydrationOptions(options, sessionId),\n        ),\n        nextCursor: firstPage.nextCursor,\n      },\n      options,\n    );\n  }\n\n  const ranked = createRankedFindResult(\n    query,\n    filterLexicalHitsByMatch(hits, search, options.match ?? \"any\"),\n    options,\n    search.terms,\n  );\n  const entityCacheInput = createEntitySearchCacheInput(\n    ranked.items,\n    indexed?.result,\n  );\n  const sentenceCacheInput = await createSentenceEvidenceSearchCacheInput(\n    document,\n    indexed?.result,\n    options,\n  );\n  const sessionId = await createSearchSession({\n    archiveKey: options.archiveKey ?? \"archive\",\n    chapters: ranked.chapters,\n    chunkHits: sentenceCacheInput.chunkHits,\n    entityHits: [\n      ...entityCacheInput.entityHits,\n      ...sentenceCacheInput.entityHits,\n    ],\n    evidenceEvents: [\n      ...entityCacheInput.evidenceEvents,\n      ...sentenceCacheInput.evidenceEvents,\n    ],\n    items: ranked.items,\n    lens: ranked.lens,\n    match: ranked.match,\n    order: ranked.order,\n    query,\n    revisionScope,\n    terms: ranked.terms,\n    tripleHits: sentenceCacheInput.tripleHits,\n    types: ranked.types,\n  });\n  const firstPage = await readSearchSessionPage(sessionId, 0, limit);\n\n  return await hydrateFindResultBacklinks(\n    document,\n    {\n      ...ranked,\n      items: await hydrateFindHitEvidence(\n        document,\n        firstPage.items,\n        createFindEvidenceHydrationOptions(options),\n      ),\n      nextCursor: firstPage.nextCursor,\n    },\n    options,\n  );\n}\n\nasync function findTextOnlyArchiveObjectsIndexed(\n  document: ReadonlyDocument,\n  query: string,\n  options: ArchiveFindOptions,\n  search: LexicalQuery,\n): Promise<ArchiveFindResult> {\n  const indexed = await findArchiveObjectsIndexed(document, query, options);\n  const hits = indexed?.hits ?? [];\n\n  return createFindResult(\n    query,\n    filterLexicalHitsByMatch(hits, search, options.match ?? \"any\"),\n    options,\n    indexed?.result.terms ?? search.terms,\n  );\n}\n\nasync function createSearchRevisionScope(\n  document: ReadonlyDocument,\n  chapters: readonly number[] | undefined,\n): Promise<string> {\n  if (chapters === undefined || chapters.length === 0) {\n    return JSON.stringify({\n      chaptersRevision: await document.serials.getChaptersRevision(),\n      scope: \"all\",\n    });\n  }\n\n  const uniqueChapters = [...new Set(chapters)].sort(compareNumbers);\n  const revisions = await document.serials.getRevisions(uniqueChapters);\n\n  return JSON.stringify({\n    chapters: uniqueChapters.map(\n      (chapterId) => [chapterId, revisions.get(chapterId) ?? 0] as const,\n    ),\n    scope: \"chapters\",\n  });\n}\n\nexport async function grepArchiveObjects(\n  document: ReadonlyDocument,\n  query: string,\n  options: ArchiveFindOptions = {},\n): Promise<ArchiveFindResult> {\n  const search = createPhraseSearch(query);\n\n  if (search === undefined) {\n    return createFindResult(\n      query,\n      [],\n      { ...options, match: \"all\" },\n      [],\n      \"exact\",\n    );\n  }\n\n  const hits: ArchiveFindHit[] = [];\n\n  hits.push(...findMeta(await document.readBookMeta(), search));\n  hits.push(...(await findChapters(document, search)));\n  hits.push(...(await findNodes(document, search)));\n\n  return createFindResult(\n    query,\n    hits,\n    { ...options, match: \"all\" },\n    [query.trim().toLowerCase()],\n    \"exact\",\n  );\n}\n"],"mappings":"AACA;AAAA,EACE;AAAA,EACA;AAAA,OAEK;AACP;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AAEP;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AACP,SAAS,mCAAmC;AAC5C;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AACP;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AACP;AAAA,EACE;AAAA,EACA;AAAA,OACK;AACP,SAAS,kCAAkC;AAC3C;AAAA,EACE;AAAA,EACA;AAAA,OACK;AACP,SAAS,2BAA2B,wBAAwB;AAO5D,eAAsB,mBACpB,UACA,OACA,UAA8B,CAAC,GACH;AAC5B,QAAM,QAAQ,QAAQ,SAAS;AAC/B,QAAM,iBAAiB,iBAAiB,OAAO;AAE/C,MACE,QAAQ,WAAW,WAClB,CAAC,kBAAkB,CAAC,aAAa,QAAQ,MAAM,IAChD;AACA,UAAM,eAAe,mCAAmC,QAAQ,MAAM;AAEtE,QAAI,iBAAiB,QAAW;AAC9B,aAAO,MAAM,6BAA6B,UAAU,cAAc;AAAA,QAChE,GAAG;AAAA,QACH;AAAA,MACF,CAAC;AAAA,IACH;AAEA,UAAM,SAAS,0BAA0B,QAAQ,MAAM;AACvD,UAAM,aAAa,MAAM;AAAA,MACvB,OAAO;AAAA,MACP,QAAQ,cAAc;AAAA,IACxB;AAEA,iCAA6B,QAAQ,OAAO,WAAW,KAAK;AAE5D,UAAM,OAAO,oBAAoB,WAAW,KAAK,IAC7C,MAAM;AAAA,MACJ,OAAO;AAAA,MACP,OAAO;AAAA,MACP;AAAA,MACA,QAAQ,cAAc;AAAA,MACtB,OAAO;AAAA,IACT,IACA,MAAM;AAAA,MACJ,OAAO;AAAA,MACP,OAAO;AAAA,MACP;AAAA,MACA,QAAQ,cAAc;AAAA,MACtB,OAAO;AAAA,IACT;AAEJ,WAAO,MAAM;AAAA,MACX;AAAA,MACA;AAAA,QACE,UAAU,KAAK;AAAA,QACf,OAAO,MAAM;AAAA,UACX;AAAA,UACA,KAAK;AAAA,UACL,mCAAmC,SAAS,OAAO,SAAS;AAAA,QAC9D;AAAA,QACA,MAAM,cAAc,KAAK,IAAI;AAAA,QAC7B,UAAU,KAAK,SAAS,UAAU,uBAAuB;AAAA,QACzD;AAAA,QACA,OAAO,eAAe,KAAK,KAAK;AAAA,QAChC,YAAY,KAAK;AAAA,QACjB,OAAO,QAAQ,SAAS;AAAA,QACxB,OAAO,KAAK;AAAA,QACZ,OAAO,KAAK;AAAA,QACZ,OAAO,eAAe,WAAW,KAAK;AAAA,MACxC;AAAA,MACA;AAAA,IACF;AAAA,EACF;AAEA,QAAM,iBAAiB,QAAQ,SAAS;AACxC,QAAM,wBACJ,mBAAmB,QACnB,eAAe,SAAS,QAAQ,KAChC,eAAe,SAAS,QAAQ;AAClC,QAAM,SAAS,mBAAmB,KAAK;AAEvC,MAAI,WAAW,QAAW;AACxB,WAAO,iBAAiB,OAAO,CAAC,GAAG,OAAO;AAAA,EAC5C;AAEA,MAAI,gBAAgB;AAClB,WAAO,MAAM;AAAA,MACX;AAAA,MACA;AAAA,MACA;AAAA,MACA;AAAA,IACF;AAAA,EACF;AAEA,QAAM,gBAAgB,MAAM;AAAA,IAC1B;AAAA,IACA,QAAQ;AAAA,EACV;AACA,QAAM,aAAa;AAAA,IACjB,YAAY,QAAQ,cAAc;AAAA,IAClC,UAAU,QAAQ,YAAY;AAAA,IAC9B,MAAM,QAAQ,UAAU,SAAY,UAAU;AAAA,IAC9C,OAAO,QAAQ,SAAS;AAAA,IACxB,OAAO,QAAQ,SAAS;AAAA,IACxB;AAAA,IACA;AAAA,IACA,OAAO,OAAO;AAAA,IACd,OAAO,QAAQ,SAAS;AAAA,EAC1B;AACA,QAAM,qBAAqB,QAAQ,kBAAkB;AACrD,QAAM,qBACJ,QAAQ,UAAU,UAAa,QAAQ,kBAAkB;AAE3D,MAAI,sBAAsB,mBAAmB,OAAO,GAAG;AACrD,UAAM,aAAa,MAAM;AAAA,MACvB;AAAA,MACA;AAAA,MACA;AAAA,IACF;AAEA,QAAI,eAAe,QAAW;AAC5B,aAAO,MAAM;AAAA,QACX;AAAA,QACA;AAAA,UACE,UAAU,WAAW;AAAA,UACrB,OAAO,MAAM;AAAA,YACX;AAAA,YACA,WAAW;AAAA,YACX,mCAAmC,SAAS,WAAW,SAAS;AAAA,UAClE;AAAA,UACA,MAAM,cAAc,WAAW,IAAI;AAAA,UACnC,UAAU,WAAW,SAAS,UAAU,uBAAuB;AAAA,UAC/D;AAAA,UACA,OAAO,eAAe,WAAW,KAAK;AAAA,UACtC,YAAY,WAAW;AAAA,UACvB,OAAO,QAAQ,SAAS;AAAA,UACxB,OAAO,WAAW;AAAA,UAClB,OAAO,WAAW;AAAA,UAClB,OAAO,eAAe,WAAW,KAAK;AAAA,QACxC;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAAA,EACF,WAAW,sBAAsB,CAAC,oBAAoB;AACpD,UAAM,aAAa,MAAM,4BAA4B,YAAY,GAAG,KAAK;AAEzE,QAAI,eAAe,QAAW;AAC5B,aAAO,MAAM;AAAA,QACX;AAAA,QACA;AAAA,UACE,UAAU,WAAW;AAAA,UACrB,OAAO,MAAM;AAAA,YACX;AAAA,YACA,WAAW;AAAA,YACX,mCAAmC,OAAO;AAAA,UAC5C;AAAA,UACA,MAAM,cAAc,WAAW,IAAI;AAAA,UACnC,UAAU,WAAW,SAAS,UAAU,uBAAuB;AAAA,UAC/D;AAAA,UACA,OAAO,eAAe,WAAW,KAAK;AAAA,UACtC,YAAY,WAAW;AAAA,UACvB,OAAO,QAAQ,SAAS;AAAA,UACxB,OAAO,WAAW;AAAA,UAClB,OAAO,WAAW;AAAA,UAClB,OAAO,eAAe,WAAW,KAAK;AAAA,QACxC;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAAA,EACF;AAEA,MAAI,oBAAoB;AACtB,QAAI,CAAE,MAAM,4BAA4B,QAAQ,GAAI;AAClD,YAAM,IAAI;AAAA,QACR;AAAA,MACF;AAAA,IACF;AACA,UAAMA,aAAY,MAAM,oBAAoB;AAAA,MAC1C,YAAY,QAAQ,cAAc;AAAA,MAClC,UAAU,QAAQ,YAAY;AAAA,MAC9B,MAAM;AAAA,MACN,OAAO,QAAQ,SAAS;AAAA,MACxB,OAAO,QAAQ,SAAS;AAAA,MACxB;AAAA,MACA;AAAA,MACA,OAAO,OAAO;AAAA,MACd,OAAO;AAAA,IACT,CAAC;AACD,UAAM,aAAa,MAAM;AAAA,MACvBA;AAAA,MACA,QAAQ,cAAc;AAAA,IACxB;AAEA,WAAO,MAAM;AAAA,MACX;AAAA,MACA;AAAA,QACE,WAAW,WAAW;AAAA,QACtB,QAAQ,EAAE,QAAQ,EAAE;AAAA,QACpB,WAAAA;AAAA,MACF;AAAA,MACA,EAAE,GAAG,SAAS,MAAM;AAAA,IACtB;AAAA,EACF;AAEA,QAAM,cAAc,wBAChB,MAAM,SAAS,SAAS;AAAA,IACtB,+BAA+B,KAAK;AAAA,EACtC,IACA,CAAC;AACL,QAAM,UAAU,MAAM,0BAA0B,UAAU,OAAO,OAAO;AACxE,QAAM,iBAAiB,wBACnB;AAAA,IACE,GAAG,aAAa,QAAQ,EAAE,UAAU,YAAY,CAAC;AAAA,IACjD,GAAI,MAAM,YAAY,UAAU,QAAQ,EAAE,UAAU,YAAY,CAAC;AAAA,EACnE,IACA,CAAC;AACL,QAAM,OAAO,CAAC,GAAG,gBAAgB,GAAI,SAAS,QAAQ,CAAC,CAAE;AACzD,MAAI,mBAAmB,OAAO,GAAG;AAC/B,UAAMC,UAAS;AAAA,MACb;AAAA,MACA,yBAAyB,MAAM,QAAQ,QAAQ,SAAS,KAAK;AAAA,MAC7D;AAAA,MACA,OAAO;AAAA,IACT;AACA,UAAMC,oBAAmB;AAAA,MACvBD,QAAO;AAAA,MACP,SAAS;AAAA,IACX;AACA,UAAME,sBAAqB,MAAM;AAAA,MAC/B;AAAA,MACA,SAAS;AAAA,MACT;AAAA,IACF;AACA,UAAMH,aAAY,MAAM,0BAA0B;AAAA,MAChD,YAAY,QAAQ,cAAc;AAAA,MAClC,UAAUC,QAAO;AAAA,MACjB,WAAWE,oBAAmB;AAAA,MAC9B,YAAY;AAAA,QACV,GAAGD,kBAAiB;AAAA,QACpB,GAAGC,oBAAmB;AAAA,MACxB;AAAA,MACA,gBAAgB;AAAA,QACd,GAAGD,kBAAiB;AAAA,QACpB,GAAGC,oBAAmB;AAAA,MACxB;AAAA,MACA,MAAMF,QAAO;AAAA,MACb,OAAOA,QAAO;AAAA,MACd,OAAOA,QAAO;AAAA,MACd;AAAA,MACA;AAAA,MACA,OAAOA,QAAO;AAAA,MACd,YAAYE,oBAAmB;AAAA,MAC/B,OAAOF,QAAO;AAAA,IAChB,CAAC;AACD,UAAMG,aAAY,MAAM,4BAA4BJ,YAAW,GAAG,KAAK;AAEvE,WAAO,MAAM;AAAA,MACX;AAAA,MACA;AAAA,QACE,GAAGC;AAAA,QACH,OAAO,MAAM;AAAA,UACX;AAAA,UACAG,WAAU;AAAA,UACV,mCAAmC,SAASJ,UAAS;AAAA,QACvD;AAAA,QACA,YAAYI,WAAU;AAAA,MACxB;AAAA,MACA;AAAA,IACF;AAAA,EACF;AAEA,QAAM,SAAS;AAAA,IACb;AAAA,IACA,yBAAyB,MAAM,QAAQ,QAAQ,SAAS,KAAK;AAAA,IAC7D;AAAA,IACA,OAAO;AAAA,EACT;AACA,QAAM,mBAAmB;AAAA,IACvB,OAAO;AAAA,IACP,SAAS;AAAA,EACX;AACA,QAAM,qBAAqB,MAAM;AAAA,IAC/B;AAAA,IACA,SAAS;AAAA,IACT;AAAA,EACF;AACA,QAAM,YAAY,MAAM,oBAAoB;AAAA,IAC1C,YAAY,QAAQ,cAAc;AAAA,IAClC,UAAU,OAAO;AAAA,IACjB,WAAW,mBAAmB;AAAA,IAC9B,YAAY;AAAA,MACV,GAAG,iBAAiB;AAAA,MACpB,GAAG,mBAAmB;AAAA,IACxB;AAAA,IACA,gBAAgB;AAAA,MACd,GAAG,iBAAiB;AAAA,MACpB,GAAG,mBAAmB;AAAA,IACxB;AAAA,IACA,OAAO,OAAO;AAAA,IACd,MAAM,OAAO;AAAA,IACb,OAAO,OAAO;AAAA,IACd,OAAO,OAAO;AAAA,IACd;AAAA,IACA;AAAA,IACA,OAAO,OAAO;AAAA,IACd,YAAY,mBAAmB;AAAA,IAC/B,OAAO,OAAO;AAAA,EAChB,CAAC;AACD,QAAM,YAAY,MAAM,sBAAsB,WAAW,GAAG,KAAK;AAEjE,SAAO,MAAM;AAAA,IACX;AAAA,IACA;AAAA,MACE,GAAG;AAAA,MACH,OAAO,MAAM;AAAA,QACX;AAAA,QACA,UAAU;AAAA,QACV,mCAAmC,OAAO;AAAA,MAC5C;AAAA,MACA,YAAY,UAAU;AAAA,IACxB;AAAA,IACA;AAAA,EACF;AACF;AAEA,eAAe,kCACb,UACA,OACA,SACA,QAC4B;AAC5B,QAAM,UAAU,MAAM,0BAA0B,UAAU,OAAO,OAAO;AACxE,QAAM,OAAO,SAAS,QAAQ,CAAC;AAE/B,SAAO;AAAA,IACL;AAAA,IACA,yBAAyB,MAAM,QAAQ,QAAQ,SAAS,KAAK;AAAA,IAC7D;AAAA,IACA,SAAS,OAAO,SAAS,OAAO;AAAA,EAClC;AACF;AAEA,eAAe,0BACb,UACA,UACiB;AACjB,MAAI,aAAa,UAAa,SAAS,WAAW,GAAG;AACnD,WAAO,KAAK,UAAU;AAAA,MACpB,kBAAkB,MAAM,SAAS,QAAQ,oBAAoB;AAAA,MAC7D,OAAO;AAAA,IACT,CAAC;AAAA,EACH;AAEA,QAAM,iBAAiB,CAAC,GAAG,IAAI,IAAI,QAAQ,CAAC,EAAE,KAAK,cAAc;AACjE,QAAM,YAAY,MAAM,SAAS,QAAQ,aAAa,cAAc;AAEpE,SAAO,KAAK,UAAU;AAAA,IACpB,UAAU,eAAe;AAAA,MACvB,CAAC,cAAc,CAAC,WAAW,UAAU,IAAI,SAAS,KAAK,CAAC;AAAA,IAC1D;AAAA,IACA,OAAO;AAAA,EACT,CAAC;AACH;AAEA,eAAsB,mBACpB,UACA,OACA,UAA8B,CAAC,GACH;AAC5B,QAAM,SAAS,mBAAmB,KAAK;AAEvC,MAAI,WAAW,QAAW;AACxB,WAAO;AAAA,MACL;AAAA,MACA,CAAC;AAAA,MACD,EAAE,GAAG,SAAS,OAAO,MAAM;AAAA,MAC3B,CAAC;AAAA,MACD;AAAA,IACF;AAAA,EACF;AAEA,QAAM,OAAyB,CAAC;AAEhC,OAAK,KAAK,GAAG,SAAS,MAAM,SAAS,aAAa,GAAG,MAAM,CAAC;AAC5D,OAAK,KAAK,GAAI,MAAM,aAAa,UAAU,MAAM,CAAE;AACnD,OAAK,KAAK,GAAI,MAAM,UAAU,UAAU,MAAM,CAAE;AAEhD,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA,EAAE,GAAG,SAAS,OAAO,MAAM;AAAA,IAC3B,CAAC,MAAM,KAAK,EAAE,YAAY,CAAC;AAAA,IAC3B;AAAA,EACF;AACF;","names":["sessionId","ranked","entityCacheInput","sentenceCacheInput","firstPage"]}

package/dist/retrieval/query/archive-view/search/hydration.d.ts

@@ -0,0 +1,51 @@
+import { ReadonlyDocument } from '../../../../document/directory/types.js';
+import { ChapterEntry } from '../../../../document/chapter/types.js';
+import { SearchIndexQueryResult, SearchIndexObjectHit, SearchIndexTextHit } from '../../../search-index/search/types.js';
+import { querySearchIndex } from '../../../search-index/search/query.js';
+import { ArchiveFindOptions, ArchiveFindHit, EvidenceReadContext } 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';
+
+declare function findArchiveObjectsIndexed(document: ReadonlyDocument, query: string, options: ArchiveFindOptions): Promise<{
+    readonly hits: readonly ArchiveFindHit[];
+    readonly result: SearchIndexQueryResult;
+} | undefined>;
+declare function queryRequiredSearchIndex(document: ReadonlyDocument, query: string, options: Parameters<typeof querySearchIndex>[2]): Promise<SearchIndexQueryResult | undefined>;
+declare function hydrateSearchIndexHits(document: ReadonlyDocument, result: SearchIndexQueryResult, options?: {
+    readonly textHitLimit?: number;
+}): Promise<readonly ArchiveFindHit[]>;
+declare function createSearchIndexHydrationOptions(options: ArchiveFindOptions): {
+    readonly textHitLimit?: number;
+};
+declare function createSearchIndexQueryLimitOptions(options: ArchiveFindOptions): {
+    readonly textHitLimit?: number;
+};
+declare function createTextOnlySearchCacheWindow(limit: number): number;
+declare function isTextOnlySearch(options: ArchiveFindOptions): boolean;
+declare function withSearchTerms(hit: ArchiveFindHit, terms: readonly string[]): ArchiveFindHit;
+declare function hydrateSearchObjectHit(document: ReadonlyDocument, chapters: ReadonlyMap<number, ChapterEntry>, hit: SearchIndexObjectHit): Promise<ArchiveFindHit | undefined>;
+declare function parseSearchPropertyIntegerOwnerId(ownerId: string): number;
+declare function hydrateSearchTextHit(document: ReadonlyDocument, chapters: ReadonlyMap<number, ChapterEntry>, hit: SearchIndexTextHit, context?: EvidenceReadContext): Promise<ArchiveFindHit | undefined>;
+
+export { createSearchIndexHydrationOptions, createSearchIndexQueryLimitOptions, createTextOnlySearchCacheWindow, findArchiveObjectsIndexed, hydrateSearchIndexHits, hydrateSearchObjectHit, hydrateSearchTextHit, isTextOnlySearch, parseSearchPropertyIntegerOwnerId, queryRequiredSearchIndex, withSearchTerms };