Add message counting and date-bounded scrapes

poc/src/media.ts

@@ -14,6 +14,7 @@ import {
     ScrapeEngine,
     ScrapeStore,
     scrapeDir,
+    snowflakeFromIso,
     resolveDbRoot,
     type MediaKind,
     type MediaRow,
@@ -22,12 +23,14 @@ import {
 } from "../../media-scrape/src/index.js";
 import { PuppetBackend } from "../../media-scrape/src/puppet-backend.js";
 import {
+    countMessages,
     loadSlopbotConfig,
     slopbotFetcher,
     slopbotStatus,
     SlopbotBackend,
     startSlopbot,
     type SlopbotFetcher,
+    type SlopbotRest,
     type SlopbotStatus
 } from "../../media-scrape/slopbot/index.js";
 import type { CoreClient } from "./core-client.js";
@@ -51,6 +54,10 @@ export interface StartScrapeInput {
     confirm?: boolean;
     /** Absolute destination for downloaded files. Null keeps them under the scrape dir. */
     downloadDir?: string | null;
+    /** From-date as ISO; the walk stops below it. */
+    after?: string | null;
+    /** To-date as ISO; the walk starts just below it. */
+    before?: string | null;
 }
 
 export interface MediaErrorEntry {
@@ -295,6 +302,8 @@ export class MediaService {
 
         try {
             const mode = input.mode ?? "links";
+            const afterId = input.after ? snowflakeFromIso(input.after) : null;
+            const beforeId = input.before ? snowflakeFromIso(input.before) : null;
             let backend: ScrapeBackend;
 
             if (input.backend === "puppet") {
@@ -314,12 +323,28 @@ export class MediaService {
                         : this.puppetBackend();
             }
 
+            // Best effort: one search request gives the total for the range.
+            let estimatedMessages: number | null = null;
+            if (backend.name === "slopbot" && input.guildId && this.bot) {
+                estimatedMessages = await countMessages(this.bot.rest as SlopbotRest, {
+                    channelId: input.channelId,
+                    guildId: input.guildId,
+                    afterId,
+                    beforeId
+                });
+            }
+
             return this.engine.start({
                 channelId: input.channelId,
                 guildId: input.guildId ?? null,
                 mode,
                 backend,
-                downloadDir: input.downloadDir ?? null
+                downloadDir: input.downloadDir ?? null,
+                afterId,
+                beforeId,
+                fromDate: input.after ?? null,
+                toDate: input.before ?? null,
+                estimatedMessages
             });
         } catch (error) {
             this.recordError("scrape", error instanceof Error ? error.message : String(error));
@@ -373,6 +398,34 @@ export class MediaService {
         };
     }
 
+    /** Discord search total for a channel or date range. Slopbot (guild) channels only. */
+    async countMessages(input: {
+        channelId: string;
+        guildId?: string | null;
+        after?: string | null;
+        before?: string | null;
+    }): Promise<{ count: number | null; reason?: string }> {
+        if (!input.guildId) {
+            return { count: null, reason: "not available for DMs or group chats" };
+        }
+
+        const available = await this.ensureSlopbot();
+        if (!available || !this.bot) {
+            return { count: null, reason: "slopbot is not connected" };
+        }
+
+        const count = await countMessages(this.bot.rest as SlopbotRest, {
+            channelId: input.channelId,
+            guildId: input.guildId,
+            afterId: input.after ? snowflakeFromIso(input.after) : null,
+            beforeId: input.before ? snowflakeFromIso(input.before) : null
+        });
+
+        return count === null
+            ? { count: null, reason: "search unavailable or rate limited" }
+            : { count };
+    }
+
     private recordError(source: string, message: string): void {
         this.errors.unshift({ at: new Date().toISOString(), source, message });
         if (this.errors.length > 50) this.errors.length = 50;