Wire the media archive into the daemon: command, service, and API

poc/src/puppet.ts

@@ -110,6 +110,24 @@ export interface PuppetAudio {
     stop(): void;
 }
 
+/** The scrape state a chat command reads. */
+export interface PuppetScrape {
+    id: string;
+    channelId: string;
+    state: string;
+}
+
+/** The media actions a chat command can trigger. */
+export interface PuppetMedia {
+    startScrape(input: {
+        channelId: string;
+        guildId: string | null;
+        mode: "links" | "download";
+    }): Promise<{ id: string }>;
+    stopScrape(id: string): void;
+    listScrapes(): PuppetScrape[];
+}
+
 interface PollState {
     channelId: string;
     authorId: string;
@@ -130,6 +148,15 @@ const WHITELIST_HELP = [
     "  whitelist remove @user      Remove a user"
 ].join("\n");
 
+const SCRAPE_HELP = [
+    "scrape media — archive this channel's media.",
+    "Usage:",
+    "  scrape media            Index links (default)",
+    "  scrape media download   Also download attachment bytes (admin)",
+    "  scrape media status     Show this channel's latest scrape",
+    "  scrape media stop       Stop it (admin)"
+].join("\n");
+
 function clamp(value: number, min: number, max: number): number {
     return Math.min(Math.max(value, min), max);
 }
@@ -164,7 +191,8 @@ export class Puppet {
 
     constructor(
         private readonly core: PuppetTransport,
-        private readonly audio?: PuppetAudio
+        private readonly audio?: PuppetAudio,
+        private readonly media?: PuppetMedia
     ) {
         try {
             const state = JSON.parse(readFileSync(STATE_FILE, "utf8")) as PersistedState;
@@ -531,6 +559,8 @@ export class Puppet {
             await this.handleWhitelistCommand(command, author.id, tier, channelId);
         } else if (command?.name === "speak") {
             await this.handleSpeakCommand(command, author.id, channelId);
+        } else if (command?.name === "scrape") {
+            await this.handleScrapeCommand(command, author.id, channelId, guildId, tier);
         } else if (POLL_ENABLED && command?.name === "poll") {
             await this.handlePollCommand(author.id, channelId, text);
         }
@@ -671,6 +701,73 @@ export class Puppet {
         }
     }
 
+    private async handleScrapeCommand(
+        command: ParsedCommand,
+        authorId: string,
+        channelId: string,
+        guildId: string | undefined,
+        tier: TrustTier
+    ) {
+        const [kind, sub] = command.args;
+        const knownSub = sub === undefined || sub === "download" || sub === "status" || sub === "stop";
+
+        if (kind !== "media" || sub === "help" || !knownSub) {
+            return this.respond(authorId, channelId, "scrape", SCRAPE_HELP);
+        }
+
+        if (!this.media) {
+            return this.respond(authorId, channelId, "scrape", "The media archive is not available.");
+        }
+
+        if (sub === "status") {
+            const scrape = this.media.listScrapes().find(entry => entry.channelId === channelId);
+            return this.respond(
+                authorId,
+                channelId,
+                "scrape",
+                scrape
+                    ? `Media scrape ${scrape.id}: ${scrape.state}.`
+                    : "No media scrape for this channel yet."
+            );
+        }
+
+        if (sub === "stop") {
+            if (!hasTier(tier, "admin")) {
+                return this.respond(authorId, channelId, "scrape", "You do not have permission for this command.");
+            }
+
+            const scrape = this.media
+                .listScrapes()
+                .find(entry => entry.channelId === channelId && ["queued", "running", "paused"].includes(entry.state));
+            if (!scrape) {
+                return this.respond(authorId, channelId, "scrape", "No active media scrape for this channel.");
+            }
+
+            this.media.stopScrape(scrape.id);
+            this.actionLog.append({ type: "media", action: "stop", channelId, scrapeId: scrape.id });
+            return this.respond(authorId, channelId, "scrape", `Stopping media scrape ${scrape.id}.`);
+        }
+
+        const mode = sub === "download" ? "download" : "links";
+        if (mode === "download" && !hasTier(tier, "admin")) {
+            return this.respond(authorId, channelId, "scrape", "Download mode needs the admin tier.");
+        }
+
+        try {
+            const scrape = await this.media.startScrape({ channelId, guildId: guildId ?? null, mode });
+            this.actionLog.append({ type: "media", action: "scrape", channelId, mode, scrapeId: scrape.id });
+            return this.respond(authorId, channelId, "scrape", `Media scrape started (${mode}): ${scrape.id}.`);
+        } catch (error) {
+            this.actionLog.append({ type: "error", scope: "media", channelId, message: String(error) });
+            return this.respond(
+                authorId,
+                channelId,
+                "scrape",
+                `Media scrape failed: ${error instanceof Error ? error.message : String(error)}`
+            );
+        }
+    }
+
     /* Polls */
 
     private async handlePollCommand(authorId: string, channelId: string, text: string) {