polite-crawler 0.1.3 → 0.1.4

package/build/index.d.mts

@@ -1,7 +1,7 @@
 import { DAYS, Day, Hours, HoursComparison, Interval, Week, compareHours, describeWeek, disagreements, fromCompact, fromSchemaOrg, fromText, toHoursWeek } from "./hours.mjs";
 import { t as ScopeOptions } from "./urls-BTDSgDLC.mjs";
 import { a as typesOf, c as decodeBody, n as PageLink, r as extractPage, s as bodyEncoding, t as PageExtract } from "./html-BACljlkf.mjs";
-import { $ as RECIPE_LIMITS, $n as FetchOutcome, $r as ReadOptions, $t as checkFields, A as RuntimeOptions, Ai as PlaywrightRenderer, An as DiffHunk, Ar as Found, At as VectorWrite, B as WatchEvent, Bn as VersionChoice, Br as Archived, Bt as ExtractOptions, C as FetchOptions, Ci as ScanOptions, Cn as SitemapEntry, Cr as DocDigest, Ct as createHybridSearch, D as OpenedCrawler, Di as Duration, Dn as feedDate, Dr as ExtractedDoc, Dt as VectorHit, E as OpenOptions, Ei as makeRecord, En as SitemapRead, Er as DocVersion, Et as VectorCounts, F as isMailbox, Fi as Renderer, Fn as DiffResult, Fr as ReadingIndex, Ft as CrawlRunOptions, G as WatchRun, Gn as CacheEntry, Gr as CatchUpResult, Gt as FieldCandidate, H as WatchList, Hn as diffDocument, Hr as archiveExtractor, Ht as ExtractSource, I as openCrawler, Ii as playwrightRenderer, In as DiffSide, Ir as SearchRank, It as CrawlStop, J as WatchStop, Jn as CrawlEvent, Jr as Extractors, Jt as FieldKind, K as WatchRunOptions, Kn as CacheIndex, Kr as CrawlRuntime, Kt as FieldCheck, L as WATCH_LIMITS, Li as BlobStore, Ln as DiffSummary, Lr as Variant, Lt as CrawlSummary, M as StoreRuntime, Mi as RenderOptions, Mn as DiffLimits, Mr as NO_EMBEDDER, Mt as CrawlOptions, N as createRuntime, Ni as RenderRefused, Nn as DiffLine, Nr as Passage, Nt as CrawlPage, O as PoliteCrawler, Oi as durationMs, On as readSitemap, Or as FAST_RANK_SHARE, Ot as VectorStore, P as defaultUserAgent, Pi as Rendered, Pn as DiffOptions, Pr as RankMode, Pt as CrawlRun, Q as runWatch, Qn as DEFAULT_SECRET_PARAMS, Qr as PageResult, Qt as FieldsChecked, R as WATCH_SCHEMA, Ri as FileBlobStore, Rn as MetaChange, Rr as VersionAt, Rt as UrlPattern, S as startDaemon, Si as DamagedLine, Sn as SITEMAP_LIMITS, Sr as Doc, St as Ranked, T as HybridCrawler, Ti as RecordInput, Tn as SitemapLimits, Tr as DocInput, Tt as VECTORS_SCHEMA, U as WatchOptions, Un as diffTexts, Ur as archivedDocId, Ut as ExtractSummary, V as WatchKind, Vn as WordPart, Vr as WebArchive, Vt as ExtractResult, W as WatchPending, Wn as CAPTURE_SCHEMA, Wr as CapturePayload, Wt as FIELD_KINDS, X as WatchStoreOptions, Xn as CrawlerOptions, Xr as Indexed, Xt as FieldSpan, Y as WatchStore, Yn as Crawler, Yr as HELD_BACK, Yt as FieldResult, Z as WatchSummary, Zn as DEADLINE_NOTE, Zr as PageOptions, Zt as FieldStatus, _ as WatchSchedule, _i as verifyQuote, _n as DiscoverSource, _r as verifySignature, _t as EmbedStop, a as DaemonDue, ai as fetchAndIndex, an as FusedRank, ar as KEY_DIRECTORY_PATH, at as RecipeExtract, b as readDaemonState, bi as CaptureLog, bn as discoverSite, br as egressLookup, bt as HybridSearch, c as DaemonMcpOptions, ci as heldBackMeta, cn as chunkText, cr as KeyInput, ct as RecipeResult, d as DaemonState, di as pageDocId, dn as fuseRanks, dr as SignatureCheck, dt as defineRecipe, ei as ReadResult, en as extractDocument, er as FetchRequest, et as RESERVED_RECIPES, f as DaemonStatus, fi as pageExtractor, fn as unitVector, fr as SignedRequest, ft as readJson, g as ScheduleResult, gi as renderAndIndex, gn as DiscoverRunOptions, gr as keyDirectory, gt as EmbedResult, h as ScheduleOptions, hi as reindex, hn as DiscoverResult, hr as createWebBotAuthSigner, ht as EmbedOptions, i as Daemon, ii as catchUpIndex, in as Embedder, ir as DirectoryKey, it as RecipeDoc, j as SearchOptions, ji as RefusedReason, jn as DiffInput, jr as Latest, jt as CRAWL_LIMITS, k as RankOptions, ki as NO_PLAYWRIGHT, kn as DIFF_LIMITS, kr as FindIndexOptions, kt as VectorStoreOptions, l as DaemonMcpService, li as isHeldBack, ln as chunkTokensOf, lr as RequestSigner, lt as RecipeRunOptions, m as DaemonTick, mi as recordOverride, mn as DiscoverOptions, mr as WebBotAuthSigner, mt as runRecipe, n as DAEMON_DEFAULTS, ni as VerifyOptions, nn as ChunkOptions, nr as RenderRequest, nt as RecipeCapture, o as DaemonMcpHooks, oi as fetchPage, on as HYBRID_DEFAULTS, or as KEY_DIRECTORY_TYPE, ot as RecipeRequest, p as DaemonTask, pi as readDocument, pn as DISCOVER_LIMITS, pr as WebBotAuthOptions, pt as registerRecipe, q as WatchRunnerOptions, qn as CaptureWriter, qr as Extractor, qt as FieldClaim, r as DAEMON_STATE_SCHEMA, ri as VerifyResult, rn as EmbedCallOptions, rr as renderedKey, rt as RecipeDefinition, s as DaemonMcpHost, si as findDocuments, sn as TextChunk, sr as KeyDirectory, st as RecipeRequestOptions, t as ChangeCommandOptions, ti as ReindexResult, tn as CHUNK_LIMITS, tr as Fetched, tt as Recipe, u as DaemonOptions, ui as isoTime, un as estimateTokens, ur as SignatureAgentType, ut as RecipeTarget, v as changeCommand, vi as versionAtTime, vn as DiscoverStop, vr as EgressOptions, vt as HybridFindOptions, w as FindOptions, wi as CaptureRecord, wn as SitemapKind, wr as DocIndex, wt as StoredChunk, x as scheduleWatches, xi as CaptureReader, xn as Changefreq, xr as DigestOptions, xt as HybridSearchOptions, y as daemonStateFile, yi as CaptureFilter, yn as DiscoveredUrl, yr as EgressRefused, yt as HybridOptions, z as WatchEntry, zn as TextDiff, zr as VersionText, zt as crawlSite } from "./daemon-DP4EqjFL.mjs";
+import { $ as RECIPE_LIMITS, $n as FetchOutcome, $r as ReadOptions, $t as checkFields, A as RuntimeOptions, Ai as PlaywrightRenderer, An as DiffHunk, Ar as Found, At as VectorWrite, B as WatchEvent, Bn as VersionChoice, Br as Archived, Bt as ExtractOptions, C as FetchOptions, Ci as ScanOptions, Cn as SitemapEntry, Cr as DocDigest, Ct as createHybridSearch, D as OpenedCrawler, Di as Duration, Dn as feedDate, Dr as ExtractedDoc, Dt as VectorHit, E as OpenOptions, Ei as makeRecord, En as SitemapRead, Er as DocVersion, Et as VectorCounts, F as isMailbox, Fi as Renderer, Fn as DiffResult, Fr as ReadingIndex, Ft as CrawlRunOptions, G as WatchRun, Gn as CacheEntry, Gr as CatchUpResult, Gt as FieldCandidate, H as WatchList, Hn as diffDocument, Hr as archiveExtractor, Ht as ExtractSource, I as openCrawler, Ii as playwrightRenderer, In as DiffSide, Ir as SearchRank, It as CrawlStop, J as WatchStop, Jn as CrawlEvent, Jr as Extractors, Jt as FieldKind, K as WatchRunOptions, Kn as CacheIndex, Kr as CrawlRuntime, Kt as FieldCheck, L as WATCH_LIMITS, Li as BlobStore, Ln as DiffSummary, Lr as Variant, Lt as CrawlSummary, M as StoreRuntime, Mi as RenderOptions, Mn as DiffLimits, Mr as NO_EMBEDDER, Mt as CrawlOptions, N as createRuntime, Ni as RenderRefused, Nn as DiffLine, Nr as Passage, Nt as CrawlPage, O as PoliteCrawler, Oi as durationMs, On as readSitemap, Or as FAST_RANK_SHARE, Ot as VectorStore, P as defaultUserAgent, Pi as Rendered, Pn as DiffOptions, Pr as RankMode, Pt as CrawlRun, Q as runWatch, Qn as DEFAULT_SECRET_PARAMS, Qr as PageResult, Qt as FieldsChecked, R as WATCH_SCHEMA, Ri as FileBlobStore, Rn as MetaChange, Rr as VersionAt, Rt as UrlPattern, S as startDaemon, Si as DamagedLine, Sn as SITEMAP_LIMITS, Sr as Doc, St as Ranked, T as HybridCrawler, Ti as RecordInput, Tn as SitemapLimits, Tr as DocInput, Tt as VECTORS_SCHEMA, U as WatchOptions, Un as diffTexts, Ur as archivedDocId, Ut as ExtractSummary, V as WatchKind, Vn as WordPart, Vr as WebArchive, Vt as ExtractResult, W as WatchPending, Wn as CAPTURE_SCHEMA, Wr as CapturePayload, Wt as FIELD_KINDS, X as WatchStoreOptions, Xn as CrawlerOptions, Xr as Indexed, Xt as FieldSpan, Y as WatchStore, Yn as Crawler, Yr as HELD_BACK, Yt as FieldResult, Z as WatchSummary, Zn as DEADLINE_NOTE, Zr as PageOptions, Zt as FieldStatus, _ as WatchSchedule, _i as verifyQuote, _n as DiscoverSource, _r as verifySignature, _t as EmbedStop, a as DaemonDue, ai as fetchAndIndex, an as FusedRank, ar as KEY_DIRECTORY_PATH, at as RecipeExtract, b as readDaemonState, bi as CaptureLog, bn as discoverSite, br as egressLookup, bt as HybridSearch, c as DaemonMcpOptions, ci as heldBackMeta, cn as chunkText, cr as KeyInput, ct as RecipeResult, d as DaemonState, di as pageDocId, dn as fuseRanks, dr as SignatureCheck, dt as defineRecipe, ei as ReadResult, en as extractDocument, er as FetchRequest, et as RESERVED_RECIPES, f as DaemonStatus, fi as pageExtractor, fn as unitVector, fr as SignedRequest, ft as readJson, g as ScheduleResult, gi as renderAndIndex, gn as DiscoverRunOptions, gr as keyDirectory, gt as EmbedResult, h as ScheduleOptions, hi as reindex, hn as DiscoverResult, hr as createWebBotAuthSigner, ht as EmbedOptions, i as Daemon, ii as catchUpIndex, in as Embedder, ir as DirectoryKey, it as RecipeDoc, j as SearchOptions, ji as RefusedReason, jn as DiffInput, jr as Latest, jt as CRAWL_LIMITS, k as RankOptions, ki as NO_PLAYWRIGHT, kn as DIFF_LIMITS, kr as FindIndexOptions, kt as VectorStoreOptions, l as DaemonMcpService, li as isHeldBack, ln as chunkTokensOf, lr as RequestSigner, lt as RecipeRunOptions, m as DaemonTick, mi as recordOverride, mn as DiscoverOptions, mr as WebBotAuthSigner, mt as runRecipe, n as DAEMON_DEFAULTS, ni as VerifyOptions, nn as ChunkOptions, nr as RenderRequest, nt as RecipeCapture, o as DaemonMcpHooks, oi as fetchPage, on as HYBRID_DEFAULTS, or as KEY_DIRECTORY_TYPE, ot as RecipeRequest, p as DaemonTask, pi as readDocument, pn as DISCOVER_LIMITS, pr as WebBotAuthOptions, pt as registerRecipe, q as WatchRunnerOptions, qn as CaptureWriter, qr as Extractor, qt as FieldClaim, r as DAEMON_STATE_SCHEMA, ri as VerifyResult, rn as EmbedCallOptions, rr as renderedKey, rt as RecipeDefinition, s as DaemonMcpHost, si as findDocuments, sn as TextChunk, sr as KeyDirectory, st as RecipeRequestOptions, t as ChangeCommandOptions, ti as ReindexResult, tn as CHUNK_LIMITS, tr as Fetched, tt as Recipe, u as DaemonOptions, ui as isoTime, un as estimateTokens, ur as SignatureAgentType, ut as RecipeTarget, v as changeCommand, vi as versionAtTime, vn as DiscoverStop, vr as EgressOptions, vt as HybridFindOptions, w as FindOptions, wi as CaptureRecord, wn as SitemapKind, wr as DocIndex, wt as StoredChunk, x as scheduleWatches, xi as CaptureReader, xn as Changefreq, xr as DigestOptions, xt as HybridSearchOptions, y as daemonStateFile, yi as CaptureFilter, yn as DiscoveredUrl, yr as EgressRefused, yt as HybridOptions, z as WatchEntry, zn as TextDiff, zr as VersionText, zt as crawlSite } from "./daemon-wwIWsUmU.mjs";
 import { r as Signals, t as AiInput } from "./signals-BHpy2xb2.mjs";
 import { E as nextValidatorMemo, O as realClock, S as ValidatorVerdict, _ as VALIDATOR_MEMO, b as ValidatorMemo, f as PolicyOptions, i as HostResult, n as DEFAULT_POLICY, o as HostStateStore, p as PolitenessPolicy, r as HostCounts, s as HostStop, t as Clock, u as MemoryHostStateStore, v as ValidatorAnswer, x as ValidatorReport, y as ValidatorLimits } from "./politeness-yI3xBEx9.mjs";
 import { a as Robots, c as RobotsRule, h as parseRobots, l as RobotsSignal, o as RobotsDecision, s as RobotsGroup, t as ALLOW_ALL, u as decide } from "./robots-R-qaEBOb.mjs";
@@ -34,7 +34,7 @@ export declare const locate: (text: string, quote: string) => Verified;
 //#endregion
 //#region src/version.d.ts
 /** This is the version of the package. A test makes sure that it agrees with `package.json`. */
-export declare const VERSION = "0.1.3";
+export declare const VERSION = "0.1.4";
 /** This is the product token that robots.txt rules match when an application gives no token. */
 export declare const DEFAULT_TOKEN = "PoliteCrawler";
 /** This page tells a site owner what the crawler is and how to control it. */