polite-crawler 0.2.1 → 0.2.3

package/build/index.d.mts

@@ -3302,6 +3302,26 @@ interface WatchOptions {
   readonly onBehalfOf?: readonly string[] | undefined;
   /** This value is the number of checks that must see a change before it is confirmed. The default is 2. */
   readonly confirm?: 1 | 2 | undefined;
+  /**
+   * The value `true` keeps the time of the next check of a page that the list has. The default
+   * `false` sets that time one `every` after the last check. A new entry is due at once in the two cases.
+   *
+   * @beta
+   */
+  readonly keepSchedule?: boolean | undefined;
+}
+/**
+ * These options change the time of the next check of an entry (`reschedule`).
+ *
+ * @beta
+ */
+interface WatchRescheduleOptions {
+  /**
+   * This time is the next check of the entry: an ISO 8601 text or a `Date`. A run checks the due
+   * entries in the sequence of these times, so an earlier time also gives an earlier check. The
+   * time is not before one floor (`watchFloor`) after the last check, so a watch cannot hammer a site.
+   */
+  readonly at: string | Date;
 }
 /** A change that one check saw, and that the next check can confirm. */
 interface WatchPending {
@@ -3370,6 +3390,8 @@ interface WatchRunOptions {
 interface WatchRunnerOptions extends WatchRunOptions {
   /** These agents get their robots.txt rules applied to each check, before the agents of the entry. */
   readonly onBehalfOf?: readonly string[] | undefined;
+  /** This name connects the captures of each check to the job that asked for them, as `PageOptions.job` does. */
+  readonly job?: string | undefined;
   /** This function operates before the first check. `openCrawler` waits there for the open of the store. */
   readonly ready?: (() => Promise<void>) | undefined;
 }
@@ -3439,6 +3461,13 @@ interface WatchList {
   list(): Promise<WatchEntry[]>;
   /** This method checks the entries that are due, to a budget. A `for await` loop gets each event. */
   run(options?: WatchRunOptions): WatchRun;
+  /**
+   * This optional method changes the time of the next check of a page (`WatchStore.reschedule`). It
+   * gives the changed entry, or undefined when the list does not have the page. `openCrawler` gives it.
+   *
+   * @beta
+   */
+  reschedule?(url: string, options: WatchRescheduleOptions): Promise<WatchEntry | undefined>;
 }
 /**
  * These options open a watch store.
@@ -3540,10 +3569,17 @@ export declare class WatchStore {
   private row;
   /**
    * This method adds a page, or changes the options of its watch and keeps its state. A new entry
-   * is due at once. A changed entry is due one `every` after its last check. A bad option throws
-   * a `RangeError`.
+   * is due at once. A changed entry is due one `every` after its last check. With `keepSchedule`,
+   * it keeps the time of its next check. A bad option throws a `RangeError`.
    */
   add(url: string, options: WatchOptions, now: number): WatchEntry;
+  /**
+   * This method changes the time of the next check of a page, and nothing else. It gives the
+   * changed entry, or undefined when the list does not have the page. A bad time, or a time before
+   * one floor after the last check, throws a `RangeError`. A check in progress sets the time again
+   * when it ends.
+   */
+  reschedule(url: string, options: WatchRescheduleOptions, now: number): WatchEntry | undefined;
   /** This method removes a page. It gives false when the list does not have the page. */
   remove(url: string): boolean;
   /** This method gives the entry of a page, or undefined. */
@@ -3579,7 +3615,7 @@ export declare const runWatch: (runtime: CrawlRuntime, source: WatchStore | (()
 //#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 = "0.2.1";
+export declare const VERSION = "0.2.3";
 /** 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. */
@@ -3858,6 +3894,8 @@ interface OpenedCrawler extends PoliteCrawler {
 interface HybridCrawler extends OpenedCrawler {
   /** This method searches the stored documents. It also takes `rank: "hybrid"`, when the crawler has an embedder. */
   find(q: string, options?: SearchOptions): Promise<Found[]>;
+  /** This list is the watch list of `OpenedCrawler`, with each optional method, for example `reschedule`. */
+  readonly watch: Required<WatchList>;
   /** This search ranks with the vectors of the embedder, or it is undefined when the crawler has no embedder. */
   readonly hybrid: HybridSearch | undefined;
   /** This method gives vectors to the stored documents that have none, within a budget. It rejects when the crawler has no embedder. */
@@ -4920,5 +4958,5 @@ interface WaczResult extends ArchiveCounts {
  */
 export declare const exportWacz: (source: ArchiveSource, options: ArchiveOptions) => Promise<WaczResult>;
 //#endregion
-export { ALLOW_ALL, type AiInput, type ArchiveCounts, type ArchiveFilter, type ArchiveOptions, type ArchiveSource, type Archived, type BlobStore, type CacheEntry, type CacheIndex, type CaptureFilter, type CaptureLogOptions, type CapturePayload, type CaptureReader, type CaptureRecord, type CaptureWriter, type CatchUpResult, type ChangeCommandOptions, type Changefreq, type ChunkOptions, type Clock, type CrawlEvent, type CrawlOptions, type CrawlPage, type CrawlRun, type CrawlRunOptions, type CrawlRuntime, type CrawlStop, type CrawlSummary, type CrawlerOptions, DAYS, DEFAULT_POLICY, type Daemon, type DaemonDue, type DaemonMcpHooks, type DaemonMcpHost, type DaemonMcpOptions, type DaemonMcpService, type DaemonOptions, type DaemonState, type DaemonStatus, type DaemonTask, type DaemonTick, type DamagedLine, type Day, type DiffHunk, type DiffInput, type DiffLimits, type DiffLine, type DiffOptions, type DiffResult, type DiffSide, type DiffSummary, type DigestOptions, type DirectoryKey, type DiscoverOptions, type DiscoverResult, type DiscoverRunOptions, type DiscoverSource, type DiscoverStop, type DiscoveredUrl, type Doc, type DocDigest, type DocIndexOptions, type DocInput, type DocVersion, type Duration, type EgressOptions, type EgressResolve, type EmbedCallOptions, type EmbedOptions, type EmbedResult, type EmbedStop, type Embedder, type EmbeddingModel, type ExtractOptions, type ExtractResult, type ExtractSource, type ExtractSummary, type ExtractedDoc, type Extractor, type FeatureExtractor, type FetchOptions, type FetchOutcome, type FetchRequest, type Fetched, type FieldCandidate, type FieldCheck, type FieldClaim, type FieldKind, type FieldResult, type FieldSpan, type FieldStatus, type FieldsChecked, type FileBlobStoreOptions, type FindIndexOptions, type FindOptions, type Found, type FusedRank, type HostCounts, type HostResult, type HostStateStore, type HostStop, type Hours, type HoursComparison, type HybridCrawler, type HybridFindOptions, type HybridOptions, type HybridSearch, type HybridSearchOptions, type IndexBuild, type IndexCounts, type Indexed, type Interval, type JsonRecord, type JwkInput, type KeyDirectory, type KeyInput, type Latest, MemoryHostStateStore, type MetaChange, type Money, type OllamaEmbedderOptions, type OpenOptions, type OpenedCrawler, type Outcome, type PageExtract, type PageLink, type PageOptions, type PageResult, type Passage, type PdfOptions, type PdfText, type PlaywrightRenderer, type PolicyOptions, type PoliteCrawler, type PolitenessPolicy, type Pooling, type ProductOptions, type ProductRead, type ProductResult, type ProductVariant, type RankMode, type RankOptions, type Ranked, type ReadOptions, type ReadResult, type ReadingIndex, type Recipe, type RecipeCapture, type RecipeDefinition, type RecipeDoc, type RecipeExtract, type RecipeRequest, type RecipeRequestOptions, type RecipeResult, type RecipeRunOptions, type RecipeTarget, type RecordInput, type RefusedReason, type ReindexResult, type RenderOptions, type RenderRequest, type Rendered, type Renderer, type RequestSigner, type Robots, type RobotsDecision, type RobotsGroup, type RobotsRule, type RobotsSignal, type RuntimeOptions, type ScanOptions, type ScheduleOptions, type ScheduleResult, type ScopeOptions, type SearchOptions, type SearchRank, type Signals, type SignatureAgentType, type SignatureCheck, type SignedRequest, type SitemapEntry, type SitemapKind, type SitemapLimits, type SitemapRead, type SqliteHostStateOptions, type StoreRuntime, type StoredChunk, type TextChunk, type TextDiff, type TransformersEmbedderOptions, type TransformersModule, type UrlPattern, VALIDATOR_MEMO, type ValidatorAnswer, type ValidatorLimits, type ValidatorMemo, type ValidatorReport, type ValidatorVerdict, type Variant, type VectorCounts, type VectorHit, type VectorStoreOptions, type VectorWrite, type VenueLink, type VenueLocation, type VenueOptions, type VenueOsm, type VenuePage, type VenueRead, type VenueReading, type VenueResult, type VenueSource, type Verified, type VerifyOptions, type VerifyResult, type VersionAt, type VersionChoice, type VersionText, type WaczResult, type WarcResult, type WatchEntry, type WatchEvent, type WatchKind, type WatchList, type WatchOptions, type WatchPending, type WatchRow, type WatchRun, type WatchRunOptions, type WatchRunnerOptions, type WatchSchedule, type WatchState, type WatchStop, type WatchStoreOptions, type WatchSummary, type WatchUpdate, type WebArchive, type WebBotAuthOptions, type WebBotAuthSigner, type Week, type WordPart, bodyEncoding, compareHours, decide, decodeBody, describeWeek, disagreements, extractPage, fromCompact, fromSchemaOrg, fromText, nextValidatorMemo, parseRobots, realClock, toHoursWeek, typesOf };
+export { ALLOW_ALL, type AiInput, type ArchiveCounts, type ArchiveFilter, type ArchiveOptions, type ArchiveSource, type Archived, type BlobStore, type CacheEntry, type CacheIndex, type CaptureFilter, type CaptureLogOptions, type CapturePayload, type CaptureReader, type CaptureRecord, type CaptureWriter, type CatchUpResult, type ChangeCommandOptions, type Changefreq, type ChunkOptions, type Clock, type CrawlEvent, type CrawlOptions, type CrawlPage, type CrawlRun, type CrawlRunOptions, type CrawlRuntime, type CrawlStop, type CrawlSummary, type CrawlerOptions, DAYS, DEFAULT_POLICY, type Daemon, type DaemonDue, type DaemonMcpHooks, type DaemonMcpHost, type DaemonMcpOptions, type DaemonMcpService, type DaemonOptions, type DaemonState, type DaemonStatus, type DaemonTask, type DaemonTick, type DamagedLine, type Day, type DiffHunk, type DiffInput, type DiffLimits, type DiffLine, type DiffOptions, type DiffResult, type DiffSide, type DiffSummary, type DigestOptions, type DirectoryKey, type DiscoverOptions, type DiscoverResult, type DiscoverRunOptions, type DiscoverSource, type DiscoverStop, type DiscoveredUrl, type Doc, type DocDigest, type DocIndexOptions, type DocInput, type DocVersion, type Duration, type EgressOptions, type EgressResolve, type EmbedCallOptions, type EmbedOptions, type EmbedResult, type EmbedStop, type Embedder, type EmbeddingModel, type ExtractOptions, type ExtractResult, type ExtractSource, type ExtractSummary, type ExtractedDoc, type Extractor, type FeatureExtractor, type FetchOptions, type FetchOutcome, type FetchRequest, type Fetched, type FieldCandidate, type FieldCheck, type FieldClaim, type FieldKind, type FieldResult, type FieldSpan, type FieldStatus, type FieldsChecked, type FileBlobStoreOptions, type FindIndexOptions, type FindOptions, type Found, type FusedRank, type HostCounts, type HostResult, type HostStateStore, type HostStop, type Hours, type HoursComparison, type HybridCrawler, type HybridFindOptions, type HybridOptions, type HybridSearch, type HybridSearchOptions, type IndexBuild, type IndexCounts, type Indexed, type Interval, type JsonRecord, type JwkInput, type KeyDirectory, type KeyInput, type Latest, MemoryHostStateStore, type MetaChange, type Money, type OllamaEmbedderOptions, type OpenOptions, type OpenedCrawler, type Outcome, type PageExtract, type PageLink, type PageOptions, type PageResult, type Passage, type PdfOptions, type PdfText, type PlaywrightRenderer, type PolicyOptions, type PoliteCrawler, type PolitenessPolicy, type Pooling, type ProductOptions, type ProductRead, type ProductResult, type ProductVariant, type RankMode, type RankOptions, type Ranked, type ReadOptions, type ReadResult, type ReadingIndex, type Recipe, type RecipeCapture, type RecipeDefinition, type RecipeDoc, type RecipeExtract, type RecipeRequest, type RecipeRequestOptions, type RecipeResult, type RecipeRunOptions, type RecipeTarget, type RecordInput, type RefusedReason, type ReindexResult, type RenderOptions, type RenderRequest, type Rendered, type Renderer, type RequestSigner, type Robots, type RobotsDecision, type RobotsGroup, type RobotsRule, type RobotsSignal, type RuntimeOptions, type ScanOptions, type ScheduleOptions, type ScheduleResult, type ScopeOptions, type SearchOptions, type SearchRank, type Signals, type SignatureAgentType, type SignatureCheck, type SignedRequest, type SitemapEntry, type SitemapKind, type SitemapLimits, type SitemapRead, type SqliteHostStateOptions, type StoreRuntime, type StoredChunk, type TextChunk, type TextDiff, type TransformersEmbedderOptions, type TransformersModule, type UrlPattern, VALIDATOR_MEMO, type ValidatorAnswer, type ValidatorLimits, type ValidatorMemo, type ValidatorReport, type ValidatorVerdict, type Variant, type VectorCounts, type VectorHit, type VectorStoreOptions, type VectorWrite, type VenueLink, type VenueLocation, type VenueOptions, type VenueOsm, type VenuePage, type VenueRead, type VenueReading, type VenueResult, type VenueSource, type Verified, type VerifyOptions, type VerifyResult, type VersionAt, type VersionChoice, type VersionText, type WaczResult, type WarcResult, type WatchEntry, type WatchEvent, type WatchKind, type WatchList, type WatchOptions, type WatchPending, type WatchRescheduleOptions, type WatchRow, type WatchRun, type WatchRunOptions, type WatchRunnerOptions, type WatchSchedule, type WatchState, type WatchStop, type WatchStoreOptions, type WatchSummary, type WatchUpdate, type WebArchive, type WebBotAuthOptions, type WebBotAuthSigner, type Week, type WordPart, bodyEncoding, compareHours, decide, decodeBody, describeWeek, disagreements, extractPage, fromCompact, fromSchemaOrg, fromText, nextValidatorMemo, parseRobots, realClock, toHoursWeek, typesOf };
 //# sourceMappingURL=index.d.mts.map

package/build/index.mjs

@@ -1,21 +1,21 @@
 import { n as FileBlobStore, t as CaptureLog } from "./log-CmBWcyQm.mjs";
 import { n as RenderRefused, r as playwrightRenderer, t as NO_PLAYWRIGHT } from "./render-CKE2LILH.mjs";
 import { t as browserMissing } from "./browser-56dBBLec.mjs";
-import { i as durationMs, n as DEFAULT_TOKEN, r as VERSION, t as BOT_PAGE } from "./version-cPnAph6k.mjs";
-import { A as DEADLINE_NOTE, F as EgressRefused, I as egressLookup, M as makeRecord, O as CAPTURE_SCHEMA, S as readPdf, T as locate, _ as archivedDocId, a as fetchPage, c as isHeldBack, d as recordOverride, f as reindex, h as archiveExtractor, i as fetchAndIndex, j as DEFAULT_SECRET_PARAMS, k as Crawler, l as isoTime, m as versionAtTime, n as HELD_BACK, o as findDocuments, p as verifyQuote, r as catchUpIndex, s as heldBackMeta, t as Extractors, u as readDocument, x as isPdf } from "./runtime-z6x0pWSN.mjs";
+import { i as durationMs, n as DEFAULT_TOKEN, r as VERSION, t as BOT_PAGE } from "./version-BBVugNVm.mjs";
+import { A as DEADLINE_NOTE, F as EgressRefused, I as egressLookup, M as makeRecord, O as CAPTURE_SCHEMA, S as readPdf, T as locate, _ as archivedDocId, a as fetchPage, c as isHeldBack, d as recordOverride, f as reindex, h as archiveExtractor, i as fetchAndIndex, j as DEFAULT_SECRET_PARAMS, k as Crawler, l as isoTime, m as versionAtTime, n as HELD_BACK, o as findDocuments, p as verifyQuote, r as catchUpIndex, s as heldBackMeta, t as Extractors, u as readDocument, x as isPdf } from "./runtime-CmAOsSPn.mjs";
 import { DEFAULT_POLICY, MemoryHostStateStore, VALIDATOR_MEMO, nextValidatorMemo, realClock } from "./politeness.mjs";
 import { ALLOW_ALL, decide, parseRobots } from "./robots.mjs";
 import { i as keyDirectory, n as KEY_DIRECTORY_TYPE, o as verifySignature, r as createWebBotAuthSigner, t as KEY_DIRECTORY_PATH } from "./webbotauth-Dq_dA_u3.mjs";
-import { A as NO_EMBEDDER, C as DISCOVER_LIMITS, D as readSitemap, E as feedDate, O as DocIndex, S as crawlSite, T as SITEMAP_LIMITS, _ as estimateTokens, b as SqliteHostStateStore, c as WatchStore, d as VectorStore, f as createHybridSearch, g as chunkTokensOf, h as chunkText, i as openCrawler, k as FAST_RANK_SHARE, l as runWatch, m as HYBRID_DEFAULTS, n as defaultUserAgent, o as WATCH_LIMITS, p as CHUNK_LIMITS, r as isMailbox, s as WATCH_SCHEMA, t as createRuntime, u as VECTORS_SCHEMA, v as fuseRanks, w as discoverSite, x as CRAWL_LIMITS, y as unitVector } from "./open-9oUXhGCK.mjs";
+import { A as NO_EMBEDDER, C as DISCOVER_LIMITS, D as readSitemap, E as feedDate, O as DocIndex, S as crawlSite, T as SITEMAP_LIMITS, _ as estimateTokens, b as SqliteHostStateStore, c as WatchStore, d as VectorStore, f as createHybridSearch, g as chunkTokensOf, h as chunkText, i as openCrawler, k as FAST_RANK_SHARE, l as runWatch, m as HYBRID_DEFAULTS, n as defaultUserAgent, o as WATCH_LIMITS, p as CHUNK_LIMITS, r as isMailbox, s as WATCH_SCHEMA, t as createRuntime, u as VECTORS_SCHEMA, v as fuseRanks, w as discoverSite, x as CRAWL_LIMITS, y as unitVector } from "./open-DWzixXcq.mjs";
 import { a as bodyEncoding, o as decodeBody, r as typesOf, t as extractPage } from "./html-DkQUAE_d.mjs";
-import { a as formatDiff, n as diffDocument, r as diffTexts, t as DIFF_LIMITS } from "./diff-ByOuHmq-.mjs";
-import { n as checkFields, r as extractDocument, t as FIELD_KINDS } from "./extract-Ccgs1ERp.mjs";
-import { a as readJson, i as defineRecipe, n as RESERVED_RECIPES, o as registerRecipe, s as runRecipe, t as RECIPE_LIMITS } from "./recipe-BODo4plI.mjs";
-import { c as startDaemon, i as daemonStateFile, n as DAEMON_STATE_SCHEMA, o as readDaemonState, r as changeCommand, s as scheduleWatches, t as DAEMON_DEFAULTS } from "./daemon-Cm6Mb8EE.mjs";
+import { a as formatDiff, n as diffDocument, r as diffTexts, t as DIFF_LIMITS } from "./diff-DxhmN3bo.mjs";
+import { n as checkFields, r as extractDocument, t as FIELD_KINDS } from "./extract-DmNeGMHX.mjs";
+import { a as readJson, i as defineRecipe, n as RESERVED_RECIPES, o as registerRecipe, s as runRecipe, t as RECIPE_LIMITS } from "./recipe-BQX7GI1z.mjs";
+import { c as startDaemon, i as daemonStateFile, n as DAEMON_STATE_SCHEMA, o as readDaemonState, r as changeCommand, s as scheduleWatches, t as DAEMON_DEFAULTS } from "./daemon-CSge_vif.mjs";
 import { i as ollamaEmbedder, n as OLLAMA_MODELS, r as isLoopback, t as NO_OLLAMA } from "./ollama-D1xRh3hY.mjs";
 import { n as TRANSFORMERS_MODELS, r as transformersEmbedder, t as NO_TRANSFORMERS } from "./transformers-BNiuiLxT.mjs";
 import { DAYS, compareHours, describeWeek, disagreements, fromCompact, fromSchemaOrg, fromText, toHoursWeek } from "./hours.mjs";
-import { a as readAmount, c as locationMatches, d as readVenue, f as venueReadingsOf, i as productOf, l as namesVenue, n as pickProduct, o as readProduct, r as pickVariant, s as VENUE_LIMITS, t as PRODUCT_LIMITS, u as rankHoursLinks } from "./recipe-product-BrDKmDXb.mjs";
-import { t as exportWacz } from "./wacz-Cog-hnVy.mjs";
-import { n as exportWarc } from "./warc-FLgjirrM.mjs";
+import { a as readAmount, c as locationMatches, d as readVenue, f as venueReadingsOf, i as productOf, l as namesVenue, n as pickProduct, o as readProduct, r as pickVariant, s as VENUE_LIMITS, t as PRODUCT_LIMITS, u as rankHoursLinks } from "./recipe-product-DVj9oHdJ.mjs";
+import { t as exportWacz } from "./wacz-t-JDRIMf.mjs";
+import { n as exportWarc } from "./warc-Ci4ygka2.mjs";
 export { ALLOW_ALL, BOT_PAGE, CAPTURE_SCHEMA, CHUNK_LIMITS, CRAWL_LIMITS, CaptureLog, Crawler, DAEMON_DEFAULTS, DAEMON_STATE_SCHEMA, DAYS, DEADLINE_NOTE, DEFAULT_POLICY, DEFAULT_SECRET_PARAMS, DEFAULT_TOKEN, DIFF_LIMITS, DISCOVER_LIMITS, DocIndex, EgressRefused, Extractors, FAST_RANK_SHARE, FIELD_KINDS, FileBlobStore, HELD_BACK, HYBRID_DEFAULTS, KEY_DIRECTORY_PATH, KEY_DIRECTORY_TYPE, MemoryHostStateStore, NO_EMBEDDER, NO_OLLAMA, NO_PLAYWRIGHT, NO_TRANSFORMERS, OLLAMA_MODELS, PRODUCT_LIMITS, RECIPE_LIMITS, RESERVED_RECIPES, RenderRefused, SITEMAP_LIMITS, SqliteHostStateStore, TRANSFORMERS_MODELS, VALIDATOR_MEMO, VECTORS_SCHEMA, VENUE_LIMITS, VERSION, VectorStore, WATCH_LIMITS, WATCH_SCHEMA, WatchStore, archiveExtractor, archivedDocId, bodyEncoding, browserMissing, catchUpIndex, changeCommand, checkFields, chunkText, chunkTokensOf, compareHours, crawlSite, createHybridSearch, createRuntime, createWebBotAuthSigner, daemonStateFile, decide, decodeBody, defaultUserAgent, defineRecipe, describeWeek, diffDocument, diffTexts, disagreements, discoverSite, durationMs, egressLookup, estimateTokens, exportWacz, exportWarc, extractDocument, extractPage, feedDate, fetchAndIndex, fetchPage, findDocuments, formatDiff, fromCompact, fromSchemaOrg, fromText, fuseRanks, heldBackMeta, isHeldBack, isLoopback, isMailbox, isPdf, isoTime, keyDirectory, locate, locationMatches, makeRecord, namesVenue, nextValidatorMemo, ollamaEmbedder, openCrawler, parseRobots, pickProduct, pickVariant, playwrightRenderer, productOf, rankHoursLinks, readAmount, readDaemonState, readDocument, readJson, readPdf, readProduct, readSitemap, readVenue, realClock, recordOverride, registerRecipe, reindex, runRecipe, runWatch, scheduleWatches, startDaemon, toHoursWeek, transformersEmbedder, typesOf, unitVector, venueReadingsOf, verifyQuote, verifySignature, versionAtTime };

package/build/{open-9oUXhGCK.mjs → open-DWzixXcq.mjs}

@@ -2,15 +2,15 @@ import { t as __exportAll } from "./rolldown-runtime-8H4AJuhK.mjs";
 import { n as FileBlobStore, t as CaptureLog } from "./log-CmBWcyQm.mjs";
 import { r as playwrightRenderer } from "./render-CKE2LILH.mjs";
 import { r as checkedRenderer } from "./browser-56dBBLec.mjs";
-import { i as durationMs, n as DEFAULT_TOKEN, r as VERSION, t as BOT_PAGE } from "./version-cPnAph6k.mjs";
-import { A as DEADLINE_NOTE, C as fold, D as snippetOf, E as parseQuery, N as sha256$1, P as stableStringify, a as fetchPage, c as isHeldBack, f as reindex, i as fetchAndIndex, k as Crawler, o as findDocuments, p as verifyQuote, r as catchUpIndex, t as Extractors, u as readDocument, y as archivedOf } from "./runtime-z6x0pWSN.mjs";
+import { i as durationMs, n as DEFAULT_TOKEN, r as VERSION, t as BOT_PAGE } from "./version-BBVugNVm.mjs";
+import { A as DEADLINE_NOTE, C as fold, D as snippetOf, E as parseQuery, N as sha256$1, P as stableStringify, a as fetchPage, c as isHeldBack, f as reindex, i as fetchAndIndex, k as Crawler, l as isoTime, o as findDocuments, p as verifyQuote, r as catchUpIndex, t as Extractors, u as readDocument, y as archivedOf } from "./runtime-CmAOsSPn.mjs";
 import { MemoryHostStateStore, nextValidatorMemo, realClock } from "./politeness.mjs";
 import { matchesPattern } from "./robots.mjs";
 import { n as canonicalUrl, r as scopeTest, t as cacheKeyOf } from "./urls-BHFeSyEK.mjs";
 import { o as decodeBody, t as extractPage } from "./html-DkQUAE_d.mjs";
-import { n as diffDocument, r as diffTexts } from "./diff-ByOuHmq-.mjs";
-import { r as extractDocument } from "./extract-Ccgs1ERp.mjs";
-import { o as registerRecipe } from "./recipe-BODo4plI.mjs";
+import { n as diffDocument, r as diffTexts } from "./diff-DxhmN3bo.mjs";
+import { r as extractDocument } from "./extract-DmNeGMHX.mjs";
+import { o as registerRecipe } from "./recipe-BQX7GI1z.mjs";
 import { createHash, randomUUID } from "node:crypto";
 import { dirname, join } from "node:path";
 import { existsSync, lstatSync, mkdirSync, realpathSync, statSync } from "node:fs";
@@ -2807,10 +2807,12 @@ const createHybridSearch = (runtime, options) => {
 			}
 			const busy = working.get(sha);
 			if (busy !== void 0) {
-				shared.push(busy.then((made) => {
+				const after = busy.then((made) => {
 					store.point(embedder.id, id, sha, now);
 					return made;
-				}));
+				});
+				after.catch(() => void 0);
+				shared.push(after);
 				continue;
 			}
 			waiting.set(sha, {
@@ -3634,8 +3636,8 @@ var WatchStore = class {
 	}
 	/**
 	* This method adds a page, or changes the options of its watch and keeps its state. A new entry
-	* is due at once. A changed entry is due one `every` after its last check. A bad option throws
-	* a `RangeError`.
+	* is due at once. A changed entry is due one `every` after its last check. With `keepSchedule`,
+	* it keeps the time of its next check. A bad option throws a `RangeError`.
 	*/
 	add(url, options, now) {
 		const { key, target, host } = targetOf(url);
@@ -3653,11 +3655,32 @@ var WatchStore = class {
          ON CONFLICT(url) DO UPDATE SET target = excluded.target, host = excluded.host,
            every_ms = excluded.every_ms, render = excluded.render, confirm = excluded.confirm,
            scope = excluded.scope, labels = excluded.labels, on_behalf_of = excluded.on_behalf_of,
-           next_at = CASE WHEN watches.last_at IS NULL THEN watches.next_at
-                          ELSE watches.last_at + excluded.every_ms END`).run(key, target, host, every, render, confirm, options.scope ?? null, labels, agents, now, now);
+           next_at = CASE WHEN ? = 1 OR watches.last_at IS NULL THEN watches.next_at
+                          ELSE watches.last_at + excluded.every_ms END`).run(key, target, host, every, render, confirm, options.scope ?? null, labels, agents, now, now, options.keepSchedule === true ? 1 : 0);
 			return entryOf(this.row(key), now);
 		});
 	}
+	/**
+	* This method changes the time of the next check of a page, and nothing else. It gives the
+	* changed entry, or undefined when the list does not have the page. A bad time, or a time before
+	* one floor after the last check, throws a `RangeError`. A check in progress sets the time again
+	* when it ends.
+	*/
+	reschedule(url, options, now) {
+		const { key } = targetOf(url);
+		const at = Date.parse(isoTime("at", options.at));
+		return this.step(() => {
+			const row = this.row(key);
+			if (row === void 0) return void 0;
+			const earliest = row.last_at === null ? -Infinity : row.last_at + this.floorMs;
+			if (at < earliest) throw new RangeError(`the next check must be at ${iso(earliest)} or later, ${String(this.floorMs / 6e4)} minutes after the last check, so a watch cannot hammer a site, not ${iso(at)}`);
+			this.stmt("UPDATE watches SET next_at = ? WHERE url = ?").run(at, key);
+			return entryOf({
+				...row,
+				next_at: at
+			}, now);
+		});
+	}
 	/** This method removes a page. It gives false when the list does not have the page. */
 	remove(url) {
 		const { key } = targetOf(url);
@@ -3797,6 +3820,7 @@ const runWatch = (runtime, source, options = {}) => {
 					render: row.render,
 					maxAgeMs: 0,
 					onBehalfOf: agents,
+					job: options.job,
 					signal: abort.signal,
 					scope: row.scope ?? void 0,
 					labels,
@@ -4302,6 +4326,9 @@ const openCrawler = (options) => {
 			async list() {
 				return (await watchStore()).list(clockNow());
 			},
+			async reschedule(url, rescheduleOptions) {
+				return (await watchStore()).reschedule(url, rescheduleOptions, clockNow());
+			},
 			run(runOptions = {}) {
 				return runWatch(runtime, watchStore, {
 					...runOptions,
@@ -4329,4 +4356,4 @@ const openCrawler = (options) => {
 //#endregion
 export { NO_EMBEDDER as A, DISCOVER_LIMITS as C, readSitemap as D, feedDate as E, DocIndex as O, crawlSite as S, SITEMAP_LIMITS as T, estimateTokens as _, open_exports as a, SqliteHostStateStore as b, WatchStore as c, VectorStore as d, createHybridSearch as f, chunkTokensOf as g, chunkText as h, openCrawler as i, FAST_RANK_SHARE as k, runWatch as l, HYBRID_DEFAULTS as m, defaultUserAgent as n, WATCH_LIMITS as o, CHUNK_LIMITS as p, isMailbox as r, WATCH_SCHEMA as s, createRuntime as t, VECTORS_SCHEMA as u, fuseRanks as v, discoverSite as w, CRAWL_LIMITS as x, unitVector as y };
 
-//# sourceMappingURL=open-9oUXhGCK.mjs.map
+//# sourceMappingURL=open-DWzixXcq.mjs.map