fix(stats): stop counting duplicate typeset subtitle lines (#191)

src/types/stats-http-contract.ts

@@ -18,6 +18,7 @@ import type {
   SessionTimelinePoint,
   StatsAnkiNoteInfo,
   StatsCoverImagesData,
+  StatsDuplicateLineCleanupResult,
   StatsExcludedWord,
   StreakCalendarDay,
   TrendsDashboardData,
@@ -31,6 +32,13 @@ export type StatsTrendRange = '7d' | '30d' | '90d' | '365d' | 'all';
 export type StatsTrendGroupBy = 'day' | 'month';
 export type StatsMineMode = 'word' | 'sentence' | 'audio';
 
+/** Body of `POST /api/stats/maintenance/duplicate-lines`. */
+export interface StatsDuplicateLineCleanupRequest {
+  dryRun?: boolean;
+  /** Null means every recorded line, whatever its age. */
+  lookbackDays?: number | null;
+}
+
 export interface StatsSessionKnownWordsTimelinePoint {
   linesSeen: number;
   knownWordsSeen: number;
@@ -125,6 +133,7 @@ export interface StatsJsonResponseMap {
   vocabulary: VocabularyEntry[];
   excludedWords: StatsExcludedWord[];
   setExcludedWords: StatsOkResponse;
+  duplicateLineCleanup: StatsDuplicateLineCleanupResult;
   wordOccurrences: VocabularyOccurrenceEntry[];
   sentenceSearch: SentenceSearchResult[];
   kanji: KanjiEntry[];
@@ -178,6 +187,9 @@ export interface StatsHttpClient {
   getVocabulary: (limit?: number) => Promise<VocabularyEntry[]>;
   getExcludedWords: () => Promise<StatsExcludedWord[]>;
   setExcludedWords: (words: StatsExcludedWord[]) => Promise<void>;
+  cleanupDuplicateLines: (
+    options?: StatsDuplicateLineCleanupRequest,
+  ) => Promise<StatsDuplicateLineCleanupResult>;
   getWordOccurrences: (
     headword: string,
     word: string,

src/types/stats-wire.ts

@@ -82,6 +82,29 @@ export interface StatsExcludedWord {
   reading: string;
 }
 
+/** One animation burst the duplicate-line cleanup found in the stats database. */
+export interface StatsDuplicateLineSample {
+  videoId: number;
+  videoTitle: string | null;
+  text: string;
+  /** Events recorded for this run, including the one that is kept. */
+  frames: number;
+  removedLines: number;
+  startMs: number;
+  endMs: number;
+}
+
+export interface StatsDuplicateLineCleanupResult {
+  dryRun: boolean;
+  lookbackDays: number | null;
+  scannedLines: number;
+  burstGroups: number;
+  removedLines: number;
+  removedWordOccurrences: number;
+  removedKanjiOccurrences: number;
+  samples: StatsDuplicateLineSample[];
+}
+
 export interface StatsCoverImage {
   contentType: string;
   dataUrl: string;