@heripo/research-radar 6.0.3 → 6.2.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -23,7 +23,7 @@ Previously reported service metrics were $0.2–1 per issue and 15% CTR. These a
23
23
 
24
24
  - Type-safe TypeScript with strict interfaces
25
25
  - Provider pattern for swapping components (Crawling/Analysis/Content/Email)
26
- - 74 active crawling targets across heritage agencies, museums, academic societies, filtered at runtime by robots.txt
26
+ - 73 active crawling targets across heritage agencies, museums, academic societies, filtered at runtime by robots.txt, plus newly published journal issues injected by the application
27
27
  - OpenAI GPT-6 defaults for every LLM task, with per-task model overrides
28
28
  - Built-in retries, chain options, preview emails
29
29
 
@@ -212,21 +212,30 @@ deadlines }` from a successful refresh skips extraction only if a fresh fetch
212
212
 
213
213
  Pass host-selected verified actions to
214
214
  `renderVerifiedDeadlineReminders(reminders, publicationDate)` and place its result
215
- in `templateOptions.deadlineRemindersMarkdown`. Today's and upcoming active
216
- deadlines/events within seven days take priority; ongoing events through a known
217
- end date fill any remaining slots. Each group is ordered by date and time, with
218
- at most five entries overall. It preserves supplied periods and times
215
+ in `templateOptions.deadlineRemindersMarkdown`. Active deadlines/events from
216
+ D-3 through today take priority; ongoing events through a known end date fill
217
+ any remaining slots. Each group is ordered by date and time, with at most ten
218
+ entries overall (`DEADLINE_REMINDER_LEAD_DAYS`, `DEADLINE_REMINDER_LIMIT`). It preserves supplied periods and times
219
219
  without inventing 23:59. The heading is **다가오는 일정·마감**. Today's entries say
220
220
  **오늘 마감**, **오늘 개최** or **오늘 공개**, while ongoing events say **진행 중**.
221
221
  An event's start time is not a cutoff: only its stated end time expires it during
222
222
  the end date. Without an end date, it is listed only through its start date. The provider appends it to saved Markdown and HTML after
223
- generation, so reminders add no LLM context. The host owns D-7/D-1 scheduling,
224
- repeat suppression and publication history. Empty sections are omitted.
223
+ generation, so reminders add no article-generation context. The host owns
224
+ reminder scheduling, repeat suppression and publication history. Empty sections are omitted.
225
225
 
226
226
  Prefer `templateOptions.verifiedDeadlineReminders` with freshly verified historical
227
227
  actions. The provider merges these with source-grounded deadlines from articles
228
- actually linked in the generated body, sorts them, and displays at most five
229
- eligible deadlines/events in one linked section. Deduplication includes the
228
+ actually linked in the generated body and displays at most ten eligible
229
+ deadlines/events in one linked section. `generateNewsletter()` asks the
230
+ importance model to choose those rows from every eligible entry: it answers with
231
+ candidate indices only, favours actions readers can take, keeps D-1 to D-3
232
+ alongside today's entries and drops the same event announced by several
233
+ sources. Titles, dates and links are always rendered from the verified rows,
234
+ and an error, a 60-second timeout or an unusable answer falls back to the first
235
+ ten by date. A short selection is filled in date order, skipping the rows the
236
+ model named as duplicates. Set
237
+ `recentReminderCount` on a reminder to tell the selection how many issues in
238
+ the past week already listed it. Deduplication includes the
230
239
  entry type, so a registration cutoff cannot replace an attendance schedule.
231
240
  Event identity also includes the start/end dates and times, preserving separate
232
241
  occurrences with the same title and kind. Deadline action deduplication is unchanged.
@@ -254,6 +263,7 @@ formats, omit it for the default, or pass an empty string to suppress it.
254
263
  | `templateOptions` | Default or KRAS newsletter branding and Markdown sections (see below) |
255
264
  | `promptProvider` | Core `PromptProvider` overriding the package's own LLM prompts entirely; omit to use the package's prompts, which fall back to core's defaults |
256
265
  | `excavationReportSource` | Supplies 국가유산청 발굴조사 보고서 entries from the application; that board is then never crawled. Omit to crawl it as before |
266
+ | `journalIssueSource` | Supplies newly published journal issues (such as KCI) from the application, one article per issue. Omit and that target collects nothing |
257
267
  | `previewNewsletter` | Fetch a saved `Newsletter` and send it through a supplied core `EmailService` |
258
268
 
259
269
  `previewNewsletter` requires `fetchNewsletterForPreview: () => Promise<Newsletter>`, `emailService` (with `send(message)`), and `emailMessage` (core `EmailMessage` without `subject`, `html`, or `text`). The core fills those three fields and skips preview delivery when no newsletter was created. Ensure the callback fetches the issue saved by this run.
@@ -331,20 +341,24 @@ The optional `contentGeneration: { provider, apiKey, model? }` compatibility pat
331
341
 
332
342
  | Group | Active | Commented out |
333
343
  | ---------- | -----: | ------------: |
334
- | News | 57 | 1 |
344
+ | News | 56 | 1 |
335
345
  | Business | 5 | 0 |
336
346
  | Employment | 12 | 0 |
337
- | **Total** | **74** | **1** |
347
+ | **Total** | **73** | **1** |
338
348
 
339
- Only the excavation status board stays commented out, as low-value fragmented data. Boards that a site's robots.txt restricts are configured normally and refused at runtime by the robots.txt check, so the configuration does not have to track each site's policy by hand. With current policies 14 of the 74 targets are refused.
349
+ Only the excavation status board stays commented out, as low-value fragmented data. Boards that a site's robots.txt restricts are configured normally and refused at runtime by the robots.txt check, so the configuration does not have to track each site's policy by hand. With current policies 14 of the 73 targets are refused.
340
350
 
341
351
  Three targets are read from data.go.kr open APIs instead of scraped: 나라일터 (`PblJobService`) and 알리오 (`recruitment`) under Employment, and 나라장터 (`BidPublicInfoService`) under Business. Each is a single round of requests, and `publicDataApiKey` supplies the service key — omit it and they answer with an empty list without making a request. The boards carry every public-sector vacancy and procurement notice in the country, so both are narrowed before analysis. The job boards use `src/crawling/heritage-job-filter.ts`, whose exclusions are role nouns and hold up as patterns; roughly 2% of postings survive, about 2.6 a day. 나라장터 instead uses `src/crawling/heritage-triage.ts`, which asks a cheap model about 100 titles per request: heritage vocabulary cannot be settled by substring in Korean, and the deterministic filter remains behind it as the fallback when a batch fails.
342
352
 
353
+ Newly published journal issues are not crawled. The application supplies them through `journalIssueSource` — heripo-web reads them from its KCI paper catalog — and the `KCI_학술지_신간` target serves them from that source, so its URL is never requested. Each issue (권호) becomes one article: its `uniqId` is the issue's `externalId` and its URL the issue's `url`. `usedArticles` carry only the application's article id and URL, so the application records which issues a saved newsletter introduced through the `uniqId` it stored in `saveCrawledArticles`, or through the issue URL. The article body, which scoring sees, lists every paper with its authors, keywords and an abstract truncated to 300 characters, after the keywords several papers share. Without a source the target answers with an empty list. This replaces the former crawl of the 국립문화유산연구원 헤리티지:역사와 과학 and 보존과학연구 boards.
354
+
355
+ The generation model never writes about these issues. `ContentGenerateProvider` takes them out of its candidates and renders a `📖 새로 나온 학술지` section itself, placed before `📑 사업·입찰 공고` (or `📌 마무리` when there is no procurement section). Each issue links to its KCI table of contents and shows, as a table with their authors, the papers its readers would want to read, most interesting first; the rest are collapsed into a count and their topics. The criterion is the readers' interest, not academic discipline — a 역사학 paper can be left out and a non-archaeology one shown — and the readers are 한국고고학회 회원 in KRAS mode. The importance model only chooses and orders paper indices and names the topics of the papers it leaves out — titles, links and authors come from the injected data. If the selection fails or times out, every paper is listed; if the run cannot load the issues, the articles go to the model as ordinary candidates. The held-aside articles are appended to `usedArticles`, so they are recorded with the issue even though the model never saw them. As a result the title, briefing and 마무리 do not mention journals, and journal issues do not count toward the publication criteria.
356
+
343
357
  Crawling fetches pass through that check (`src/crawling/robots.ts`) before reaching the network: a disallowed request is refused rather than sent, and a missing or unreachable robots.txt allows it. `robotsExemptOrigins` in `src/config/index.ts` lists origins exempted from the check, each with its reason.
344
358
 
345
359
  Sources include the Korea Heritage Service, National Research Institute of Cultural Heritage, National Research Institute of Maritime Heritage, Korea Heritage Agency, Korea Association of Archaeological Heritage, archaeological societies, and national museums.
346
360
 
347
- [src/parsers/](./src/parsers/) contains 23 organization-specific parser modules plus shared date and URL utilities. List parsers return `ParsedTargetListItem[]` (title, date, detail URL, date type, and optional source ID); detail parsers return Markdown `detailContent` and attachment/image flags. Parsers can be synchronous or asynchronous. KRAS, Yeongnam Archaeological Society, and maritime heritage sources use additional API requests for client-rendered content.
361
+ [src/parsers/](./src/parsers/) contains 24 parser modules (23 organization-specific, plus the injected journal issue adapter) plus shared date and URL utilities. List parsers return `ParsedTargetListItem[]` (title, date, detail URL, date type, and optional source ID); detail parsers return Markdown `detailContent` and attachment/image flags. Parsers can be synchronous or asynchronous. KRAS, Yeongnam Archaeological Society, and maritime heritage sources use additional API requests for client-rendered content.
348
362
 
349
363
  `CrawlingProvider` uses a maximum concurrency of 5 and wraps the supplied fetch to route KRAS public detail URLs to its detail API, while retaining public URLs in article metadata. When constructing your own pipeline, use the provider's fetch together with its target groups.
350
364
 
@@ -356,7 +370,7 @@ import { getSourceList } from '@heripo/research-radar';
356
370
  const groups = getSourceList(); // [{ id, name, sources: [{ id, name, url }] }]
357
371
  ```
358
372
 
359
- `createCrawlingTargetGroups(customFetch?)`, `getSourceList()`, `contentOptions`, `newsletterConfig`, `llmConfig`, and `researchRadarPromptProvider` are public exports. `ExcavationReport` and `ExcavationReportSource` are exported as types. The package also exports the three provider classes, `DateService`, `TaskService`, and their public configuration/dependency types through [src/index.ts](./src/index.ts).
373
+ `createCrawlingTargetGroups(customFetch?)`, `getSourceList()`, `contentOptions`, `newsletterConfig`, `llmConfig`, `researchRadarPromptProvider`, and `JOURNAL_ISSUE_LIST_URL` (the journal issue target's URL, for finding the issue articles an application saved) are public exports. `ExcavationReport`, `ExcavationReportSource`, `JournalIssue`, `JournalIssuePaper`, and `JournalIssueSource` are exported as types. The package also exports the three provider classes, `DateService`, `TaskService`, and their public configuration/dependency types through [src/index.ts](./src/index.ts).
360
374
 
361
375
  ## Email templates
362
376
 
@@ -488,7 +502,7 @@ Use repeatable `--skip-target=<id-or-name>` or `--skip-target <id-or-name>` to e
488
502
 
489
503
  **Output**: Console table summary + compact text summary for CI integrations.
490
504
 
491
- **CI**: [.github/workflows/parser-health-check.yml](./.github/workflows/parser-health-check.yml) runs daily at 08:00 UTC (17:00 KST), or manually, on an `org-linux` runner with a 30-minute job timeout. It skips the two KHS excavation report/site-open targets. For crawling targets, the health-check composes the same fetch as production — robots.txt gate, then the KRAS detail adapter — so disallowed boards are reported as skipped rather than failed: 16 skipped and 55 checked with the current configuration. The detail check tries up to three list items, so one unreadable post at the top of a board does not fail the target. Slack notifications require the `SLACK_BOT_TOKEN` secret and `SLACK_ALERT_DEV_CHANNEL` repository variable. Forks need a matching runner and notification configuration to use this workflow unchanged. The CLI also writes GitHub Actions outputs and a job summary when their environment variables are present.
505
+ **CI**: [.github/workflows/parser-health-check.yml](./.github/workflows/parser-health-check.yml) runs daily at 08:00 UTC (17:00 KST), or manually, on an `org-linux` runner with a 30-minute job timeout. It skips the two KHS excavation report/site-open targets. For crawling targets, the health-check composes the same fetch as production — robots.txt gate, then the KRAS detail adapter — so disallowed boards are reported as skipped rather than failed: 17 skipped (including the injected journal issue target, which has no live board) and 53 checked with the current configuration. The detail check tries up to three list items, so one unreadable post at the top of a board does not fail the target. Slack notifications require the `SLACK_BOT_TOKEN` secret and `SLACK_ALERT_DEV_CHANNEL` repository variable. Forks need a matching runner and notification configuration to use this workflow unchanged. The CLI also writes GitHub Actions outputs and a job summary when their environment variables are present.
492
506
 
493
507
  ## 🤝 Contributing
494
508
 
package/dist/index.d.ts CHANGED
@@ -7,8 +7,24 @@ interface VerifiedDeadlineReminder {
7
7
  deadline: ArticleDeadline;
8
8
  url: UrlString;
9
9
  verificationStatus: 'verified';
10
+ /**
11
+ * Published issues in the past week that already listed this action.
12
+ * Host-supplied history; it only informs the reminder selection prompt.
13
+ */
14
+ recentReminderCount?: number;
15
+ }
16
+ /** Days before the date an action first becomes eligible (D-3 through D-day). */
17
+ declare const DEADLINE_REMINDER_LEAD_DAYS = 3;
18
+ /** Rows the unified deadline section displays. */
19
+ declare const DEADLINE_REMINDER_LIMIT = 10;
20
+ /** A reminder that passed every display guard, with its normalized link. */
21
+ interface EligibleDeadlineReminder {
22
+ reminder: VerifiedDeadlineReminder;
23
+ /** Calendar days from publication to the start date; negative while ongoing. */
24
+ days: number;
25
+ url: string;
10
26
  }
11
- /** Deterministic deadlines/events, no LLM context. Host owns selection and history. */
27
+ /** Deterministic deadlines/events, no LLM context: the first rows by date. */
12
28
  declare function renderVerifiedDeadlineReminders(reminders: VerifiedDeadlineReminder[], publicationDate: IsoDateString, now?: Date): string;
13
29
 
14
30
  /**
@@ -110,6 +126,77 @@ interface ExcavationReport {
110
126
  * Called at most once per generation run.
111
127
  */
112
128
  type ExcavationReportSource = () => Promise<ExcavationReport[]>;
129
+ /**
130
+ * One paper in an injected journal issue.
131
+ */
132
+ interface JournalIssuePaper {
133
+ /** Paper title. */
134
+ title: string;
135
+ /** Author names in publication order. */
136
+ authors: string[];
137
+ /** Author keywords. Blank and repeated keywords are ignored. */
138
+ keywords: string[];
139
+ /**
140
+ * Abstract shown to the scoring and generation models, in the original
141
+ * language when the source has one. Truncated before it reaches them.
142
+ */
143
+ abstract?: string | null;
144
+ /** Public page for the paper, such as its KCI article page. */
145
+ url: string;
146
+ }
147
+ /**
148
+ * A newly published journal issue (권호) supplied by the application.
149
+ *
150
+ * Each issue becomes one article whose body lists every paper it carries, so a
151
+ * twenty-paper issue is scored and introduced once rather than twenty times.
152
+ */
153
+ interface JournalIssue {
154
+ /**
155
+ * Stable identifier for the issue, such as
156
+ * `{journalId}|{year}|{month}|{volume}|{issue}`. It becomes the article's
157
+ * `uniqId` and reaches `saveCrawledArticles`, so the application can store
158
+ * it with the article. `usedArticles` carry only the article's own id and
159
+ * URL, so that stored value (or the issue `url`) is how the application maps
160
+ * a saved newsletter back to the issues it introduced. Keep it independent
161
+ * of the paper list, or a later snapshot of the same issue looks like a new
162
+ * one.
163
+ */
164
+ externalId: string;
165
+ /**
166
+ * Public page readers follow from the newsletter. It is also the article's
167
+ * URL, so it must be stable for the issue: previously saved issues are
168
+ * recognized by it and not saved again.
169
+ */
170
+ url: string;
171
+ /** Journal name, e.g. "헤리티지:역사와 과학". */
172
+ journalName: string;
173
+ /** Publication year. */
174
+ publishedYear: number;
175
+ /** Publication month, 1-12. */
176
+ publishedMonth: number;
177
+ /** Volume (권). Empty when the journal numbers issues only. */
178
+ volume: string;
179
+ /** Issue number (호). Empty when the journal numbers volumes only. */
180
+ issue: string;
181
+ /**
182
+ * Date the issue became available to the application, in ISO format
183
+ * (YYYY-MM-DD, Korean time) — for example when its papers were first
184
+ * collected. Used as the article's registered date.
185
+ * @example "2026-10-28"
186
+ */
187
+ registeredDate: IsoDateString;
188
+ /** Every paper in the issue, in the order to list them. */
189
+ papers: JournalIssuePaper[];
190
+ }
191
+ /**
192
+ * Supplies newly published journal issues from the application, such as those
193
+ * collected from KCI.
194
+ *
195
+ * The journal issue target has no live board behind it: when this is omitted
196
+ * the target collects nothing and makes no request. Called at most once per
197
+ * generation run.
198
+ */
199
+ type JournalIssueSource = () => Promise<JournalIssue[]>;
113
200
  /**
114
201
  * Repository interface for tag management
115
202
  */
@@ -346,6 +433,17 @@ interface NewsletterGeneratorDependencies {
346
433
  * it to keep crawling the board as before. No other target is affected.
347
434
  */
348
435
  excavationReportSource?: ExcavationReportSource;
436
+ /**
437
+ * Supplies newly published journal issues from the application (optional),
438
+ * such as those collected from KCI.
439
+ *
440
+ * Each issue becomes one article listing its papers. The target has no live
441
+ * board, so omitting this leaves it empty without making a request. The
442
+ * article's `uniqId` is the issue's `externalId` and its URL the issue's
443
+ * `url`; store either when articles are saved to record which issues a
444
+ * saved newsletter introduced.
445
+ */
446
+ journalIssueSource?: JournalIssueSource;
349
447
  /**
350
448
  * data.go.kr service key for the 나라일터 and 알리오 job boards (optional).
351
449
  *
@@ -386,6 +484,15 @@ interface NewsletterGeneratorDependencies {
386
484
  */
387
485
  declare function generateNewsletter(dependencies: NewsletterGeneratorDependencies): Promise<string | number | null>;
388
486
 
487
+ /**
488
+ * Board URL of the journal issue target.
489
+ *
490
+ * There is no list page behind it: the adapter below answers it from the
491
+ * application's injected issues and never sends it. It points at KCI because
492
+ * that is where the issues come from, and it is what the source list shows.
493
+ */
494
+ declare const JOURNAL_ISSUE_LIST_URL = "https://www.kci.go.kr/kciportal/main.kci";
495
+
389
496
  /**
390
497
  * Generates a welcome email HTML string with CSS inlined via juice.
391
498
  *
@@ -512,6 +619,78 @@ declare class AnalysisProvider implements AnalysisProvider$1 {
512
619
  update(article: ArticleForUpdateByAnalysis): Promise<void>;
513
620
  }
514
621
 
622
+ /**
623
+ * Chooses which eligible reminders the unified section displays.
624
+ *
625
+ * Returns a subset of the given rows in their original display order.
626
+ */
627
+ type DeadlineReminderSelector = (rows: EligibleDeadlineReminder[], publicationDate: IsoDateString) => Promise<EligibleDeadlineReminder[]>;
628
+ type DeadlineReminderSelectorOptions = {
629
+ /** Same model that scores article importance. */
630
+ model: LanguageModel;
631
+ /**
632
+ * Milliseconds before the selection gives up and falls back. It runs while
633
+ * saving a fully generated issue, so a stalled provider must not hold it.
634
+ * @default 60000
635
+ */
636
+ timeoutMs?: number;
637
+ /** Reports a selection that fell back to the first rows by date. */
638
+ onFallback?: (reason: string, candidateCount: number) => void;
639
+ /** Reports a model selection and its token usage. */
640
+ onSelected?: (info: {
641
+ candidateCount: number;
642
+ selectedCount: number;
643
+ /** Rows added by date order because the model chose too few. */
644
+ filledCount: number;
645
+ usage: unknown;
646
+ }) => void;
647
+ };
648
+ /**
649
+ * Builds the model selection that replaces "first rows by date".
650
+ *
651
+ * Several sources often announce the same 학술대회, and a busy day can hold
652
+ * fifteen same-day bids, so the earliest rows are a poor proxy for what readers
653
+ * need. The model judges relevance and duplicates; everything that decides
654
+ * whether a row may appear at all stays in {@link eligibleDeadlineReminders}.
655
+ *
656
+ * It never fails the newsletter: an error, a timeout, an empty answer or
657
+ * unusable indices fall back to the first rows by date. A selection shorter
658
+ * than the rows available is filled in date order, skipping named duplicates.
659
+ */
660
+ declare const createDeadlineReminderSelector: (options: DeadlineReminderSelectorOptions) => DeadlineReminderSelector;
661
+
662
+ /** Memoized access to one run's injected journal issues. */
663
+ type JournalIssueLoader = () => Promise<JournalIssue[]>;
664
+
665
+ /** Which papers of an issue the table shows, and which it summarizes. */
666
+ type JournalPaperSelection = {
667
+ /** Paper indices shown in the table, most interesting first. */
668
+ featured: number[];
669
+ /** Papers collapsed into a count and topic summary, in paper order. */
670
+ others: {
671
+ index: number;
672
+ topic: string | null;
673
+ }[];
674
+ };
675
+ /**
676
+ * Picks the papers the newsletter's readers would want to see, or returns null
677
+ * when no selection could be made — the section then lists every paper.
678
+ */
679
+ type JournalPaperSelector = (issue: JournalIssue) => Promise<JournalPaperSelection | null>;
680
+
681
+ /**
682
+ * Injected journal issues are rendered by the provider rather than written by
683
+ * the model: each is a table of up to twenty titles and author lists, which
684
+ * the model would have to copy exactly.
685
+ */
686
+ type JournalIssueRendering = {
687
+ /** The run's injected issues, matched to candidates by URL. */
688
+ load: JournalIssueLoader;
689
+ /** Picks the papers readers would care about; omitted, every paper is listed. */
690
+ select?: JournalPaperSelector;
691
+ /** Reports that the issues could not be loaded, so the model got them. */
692
+ onUnavailable?: (reason: string) => void;
693
+ };
515
694
  /**
516
695
  * Content generation provider implementation
517
696
  * - LLM-based newsletter content generation
@@ -522,15 +701,21 @@ declare class ContentGenerateProvider implements ContentGenerateProvider$1 {
522
701
  private readonly articleRepository;
523
702
  private readonly newsletterRepository;
524
703
  private readonly templateOptions?;
704
+ /** Chooses displayed reminders; omitted, the first rows by date display. */
705
+ private readonly selectDeadlineReminders?;
706
+ private readonly journalIssues?;
525
707
  private _issueOrder;
526
708
  private readonly deadlineRemindersMarkdown;
709
+ private heldJournalIssues;
527
710
  readonly model: LanguageModel;
528
711
  readonly fetchRecentTitles?: ContentGenerateProvider$1['fetchRecentTitles'];
529
712
  /** HTML template with markers for title and content injection */
530
713
  htmlTemplate: HtmlTemplate;
531
714
  /** Newsletter brand name (defaults to config, can be overridden via constructor) */
532
715
  newsletterBrandName: string;
533
- constructor(model: LanguageModel, articleRepository: ArticleRepository, newsletterRepository: NewsletterRepository, templateOptions?: NewsletterTemplateOptions | undefined, brandName?: string);
716
+ constructor(model: LanguageModel, articleRepository: ArticleRepository, newsletterRepository: NewsletterRepository, templateOptions?: NewsletterTemplateOptions | undefined, brandName?: string,
717
+ /** Chooses displayed reminders; omitted, the first rows by date display. */
718
+ selectDeadlineReminders?: DeadlineReminderSelector | undefined, journalIssues?: JournalIssueRendering | undefined);
534
719
  /** LLM temperature setting for content generation */
535
720
  temperature: number;
536
721
  /** Subscribe page URL */
@@ -565,6 +750,8 @@ declare class ContentGenerateProvider implements ContentGenerateProvider$1 {
565
750
  }): Promise<{
566
751
  id: string | number;
567
752
  }>;
753
+ private renderJournalIssues;
754
+ private renderDeadlineReminders;
568
755
  }
569
756
 
570
757
  type HeritageBidCandidate = {
@@ -597,7 +784,7 @@ declare class CrawlingProvider implements CrawlingProvider$1 {
597
784
  customFetch?: typeof fetch;
598
785
  /** Crawling target groups configuration */
599
786
  crawlingTargetGroups: CrawlingTargetGroup[];
600
- constructor(articleRepository: ArticleRepository, customFetch?: typeof fetch, excavationReportSource?: ExcavationReportSource, logger?: AppLogger, publicDataApiKey?: string, bidTriage?: HeritageBidTriage);
787
+ constructor(articleRepository: ArticleRepository, customFetch?: typeof fetch, excavationReportSource?: ExcavationReportSource, logger?: AppLogger, publicDataApiKey?: string, bidTriage?: HeritageBidTriage, journalIssueSource?: JournalIssueSource);
601
788
  /**
602
789
  * Fetch existing articles by URLs to avoid duplicate crawling
603
790
  * @param articleUrls - URLs to check
@@ -793,5 +980,5 @@ declare function fetchResearchRadarDeadlineSource({ source, customFetch, publicD
793
980
  /** Only explicit, grounded results are verified. Missing dates are never cancellation. */
794
981
  declare function refreshResearchRadarDeadlines(config: ResearchRadarDeadlineRefreshConfig): Promise<ResearchRadarDeadlineRefreshResult>;
795
982
 
796
- export { AnalysisProvider, ContentGenerateProvider, CrawlingProvider, DateService, TaskService, contentOptions, createCrawlingTargetGroups, extractResearchRadarDeadlines, fetchResearchRadarDeadlineSource, generateNewsletter, generateWelcomeHTML, getSourceList, llmConfig, maximumImportanceScoreByDomain, newsletterConfig, refreshResearchRadarDeadlines, renderVerifiedDeadlineReminders, researchRadarPromptProvider, robotsExemptOrigins };
797
- export type { ArticleRepository, ContentGenerationConfig, ContentOptions, DeadlineVerificationFailure, ExcavationReport, ExcavationReportSource, NewsletterConfig, NewsletterGeneratorDependencies, NewsletterModelConfig, NewsletterRepository, NewsletterTemplateOptions, PreviewNewsletterOptions, ResearchRadarDeadlineExtractionConfig, ResearchRadarDeadlineRefreshConfig, ResearchRadarDeadlineRefreshResult, ResearchRadarDeadlineSource, ResearchRadarDeadlineSourceOptions, ResearchRadarDeadlineSourceResult, SourceGroup, SourceItem, TagRepository, TaskRepository, VerifiedDeadlineReminder, WelcomeTemplateOptions };
983
+ export { AnalysisProvider, ContentGenerateProvider, CrawlingProvider, DEADLINE_REMINDER_LEAD_DAYS, DEADLINE_REMINDER_LIMIT, DateService, JOURNAL_ISSUE_LIST_URL, TaskService, contentOptions, createCrawlingTargetGroups, createDeadlineReminderSelector, extractResearchRadarDeadlines, fetchResearchRadarDeadlineSource, generateNewsletter, generateWelcomeHTML, getSourceList, llmConfig, maximumImportanceScoreByDomain, newsletterConfig, refreshResearchRadarDeadlines, renderVerifiedDeadlineReminders, researchRadarPromptProvider, robotsExemptOrigins };
984
+ export type { ArticleRepository, ContentGenerationConfig, ContentOptions, DeadlineReminderSelector, DeadlineReminderSelectorOptions, DeadlineVerificationFailure, EligibleDeadlineReminder, ExcavationReport, ExcavationReportSource, JournalIssue, JournalIssuePaper, JournalIssueSource, NewsletterConfig, NewsletterGeneratorDependencies, NewsletterModelConfig, NewsletterRepository, NewsletterTemplateOptions, PreviewNewsletterOptions, ResearchRadarDeadlineExtractionConfig, ResearchRadarDeadlineRefreshConfig, ResearchRadarDeadlineRefreshResult, ResearchRadarDeadlineSource, ResearchRadarDeadlineSourceOptions, ResearchRadarDeadlineSourceResult, SourceGroup, SourceItem, TagRepository, TaskRepository, VerifiedDeadlineReminder, WelcomeTemplateOptions };