fix: keep Telegram AI extract result on metadata errors (#1045)

worker/src/admin_api/e2e_test_api.ts

@@ -29,7 +29,7 @@ const receiveMail = async (c: Context<HonoCustomType>) => {
     if (!getBooleanValue(c.env.E2E_TEST_MODE)) {
         return c.text("Not available", 404);
     }
-    const { from, to, raw } = await c.req.json();
+    const { from, to, raw, ai_extract_result } = await c.req.json();
     if (!from || !to || !raw) {
         return c.text("from, to and raw are required", 400);
     }
@@ -54,7 +54,21 @@ const receiveMail = async (c: Context<HonoCustomType>) => {
         reply: async () => { state.replyCalled = true; return { messageId: '' }; },
     };
     const { email: emailHandler } = await import('../email');
-    await emailHandler(mockMessage, c.env, { waitUntil: () => {}, passThroughOnException: () => {} });
+    const aiExtractEnvOverrides: Partial<Bindings> = {
+        ENABLE_AI_EMAIL_EXTRACT: true,
+        AI: {
+            run: async () => ({ response: ai_extract_result })
+        } as unknown as Ai,
+    };
+    const env = ai_extract_result
+        ? { ...c.env, ...aiExtractEnvOverrides }
+        : c.env;
+    const executionContext: ExecutionContext = {
+        waitUntil: () => {},
+        passThroughOnException: () => {},
+        props: {}
+    };
+    await emailHandler(mockMessage, env, executionContext);
 
     return c.json({
         success: !state.rejected,

worker/src/email/ai_extract.ts

@@ -11,6 +11,7 @@ import { getBooleanValue, getJsonSetting } from "../utils";
 import { CONSTANTS } from "../constants";
 import { Context } from "hono";
 import type { AiExtractSettings } from "../admin_api/ai_extract_settings";
+import type { ExtractResult } from "../models";
 
 // AI Prompt for email analysis
 const PROMPT = `
@@ -144,7 +145,7 @@ async function extractWithCloudflareAI(
  * @param env - Cloudflare Workers environment bindings
  * @param message_id - The email message ID
  * @param address - The recipient email address
- * @returns Promise<void>
+ * @returns Promise<ExtractResult | null>
  */
 export async function extractEmailInfo(
     parsedEmailContext: ParsedEmailContext,
@@ -208,18 +209,21 @@ export async function extractEmailInfo(
 
         // If extraction found something useful, save it to database
         if (result.type !== 'none' && result.result) {
-            const metadata = JSON.stringify({
-                ai_extract: result,
-                extracted_at: new Date().toISOString()
-            });
+            try {
+                const metadata = JSON.stringify({
+                    ai_extract: result,
+                    extracted_at: new Date().toISOString()
+                });
 
-            // Update the raw_mails record with metadata
-            await env.DB.prepare(
-                `UPDATE raw_mails SET metadata = ? WHERE message_id = ?`
-            ).bind(metadata, message_id).run();
+                // Update the raw_mails record with metadata
+                await env.DB.prepare(
+                    `UPDATE raw_mails SET metadata = ? WHERE message_id = ?`
+                ).bind(metadata, message_id).run();
 
-            console.log(`AI extraction completed for ${message_id}: ${result.type}`);
-            return result;
+                console.log(`AI extraction completed for ${message_id}: ${result.type}`);
+            } catch (e) {
+                console.error('AI extraction metadata save error:', e);
+            }
         }
         return result;
     } catch (e) {
@@ -227,12 +231,3 @@ export async function extractEmailInfo(
         return null;
     }
 }
-
-/**
- * Type definition for extraction result
- */
-export type ExtractResult = {
-    type: 'auth_code' | 'auth_link' | 'service_link' | 'subscription_link' | 'other_link' | 'none';
-    result: string;
-    result_text: string;
-};

worker/src/models/index.ts

@@ -208,3 +208,9 @@ export type RawMailRow = {
     metadata?: string;
     created_at?: string;
 }
+
+export type ExtractResult = {
+    type: 'auth_code' | 'auth_link' | 'service_link' | 'subscription_link' | 'other_link' | 'none';
+    result: string;
+    result_text: string;
+}

worker/src/telegram_api/telegram.ts

@@ -10,11 +10,10 @@ import { sendTelegramAttachments } from "./tg_file_upload";
 import { bindTelegramAddress, deleteTelegramAddress, jwtListToAddressData, tgUserNewAddress, unbindTelegramAddress, unbindTelegramByAddress } from "./common";
 import { commonParseMail } from "../common";
 import { resolveRawEmail } from "../gzip";
-import { RawMailRow } from "../models";
 import { UserFromGetMe } from "telegraf/types";
 import i18n from "../i18n";
 import { LocaleMessages } from "../i18n/type";
-import type { ExtractResult } from "../email/ai_extract";
+import type { ExtractResult, RawMailRow } from "../models";
 
 
 // Helper to get messages by userId
@@ -76,60 +75,54 @@ const COMMANDS = [
     },
 ]
 
-const getAiExtractLabel = (
+const formatAiExtractForTelegram = (
     msgs: LocaleMessages,
-    type: ExtractResult["type"]
+    aiExtract?: ExtractResult | string | null
 ): string => {
-    switch (type) {
-        case "auth_code":
-            return msgs.TgAiExtractAuthCodeMsg;
-        case "auth_link":
-            return msgs.TgAiExtractAuthLinkMsg;
-        case "service_link":
-            return msgs.TgAiExtractServiceLinkMsg;
-        case "subscription_link":
-            return msgs.TgAiExtractSubscriptionLinkMsg;
-        case "other_link":
-            return msgs.TgAiExtractOtherLinkMsg;
-        default:
-            return msgs.TgAiExtractResultMsg;
+    if (!aiExtract) {
+        return "";
     }
-}
 
-const parseAiExtractMetadata = (
-    metadata: string | undefined | null
-): ExtractResult | null => {
-    if (!metadata) return null;
     try {
-        const parsed = JSON.parse(metadata);
-        const result = parsed?.ai_extract;
-        if (
-            result
-            && typeof result.type === "string"
-            && result.type !== "none"
-            && typeof result.result === "string"
-            && result.result
-        ) {
-            return result as ExtractResult;
+        if (typeof aiExtract === "string") {
+            const metadata = JSON.parse(aiExtract);
+            aiExtract = metadata?.ai_extract;
         }
     } catch (error) {
         console.warn("Failed to parse AI extraction metadata", error);
-    }
-    return null;
-}
-
-const formatAiExtractForTelegram = (
-    msgs: LocaleMessages,
-    aiExtract: ExtractResult | null | undefined
-): string => {
-    if (!aiExtract || aiExtract.type === "none" || !aiExtract.result) {
         return "";
     }
-    const label = getAiExtractLabel(msgs, aiExtract.type);
-    const displayText = aiExtract.type !== "auth_code" && aiExtract.result_text
-        ? ` (${aiExtract.result_text})`
+
+    if (!aiExtract || typeof aiExtract !== "object") {
+        return "";
+    }
+
+    const labels: Record<Exclude<ExtractResult["type"], "none">, string> = {
+        auth_code: msgs.TgAiExtractAuthCodeMsg,
+        auth_link: msgs.TgAiExtractAuthLinkMsg,
+        service_link: msgs.TgAiExtractServiceLinkMsg,
+        subscription_link: msgs.TgAiExtractSubscriptionLinkMsg,
+        other_link: msgs.TgAiExtractOtherLinkMsg,
+    };
+    const label = labels[aiExtract.type as keyof typeof labels];
+    const result = typeof aiExtract.result === "string"
+        ? aiExtract.result.replace(/\s+/g, " ").trim().slice(0, 600)
         : "";
-    return `${msgs.TgAiExtractResultMsg}\n${label}: ${aiExtract.result}${displayText}\n\n`;
+    if (!result) {
+        return "";
+    }
+
+    if (!label) {
+        return "";
+    }
+
+    const resultText = typeof aiExtract.result_text === "string"
+        ? aiExtract.result_text.replace(/\s+/g, " ").trim().slice(0, 120)
+        : "";
+    const displayText = aiExtract.type !== "auth_code" && resultText && resultText !== result
+        ? ` (${resultText})`
+        : "";
+    return `${msgs.TgAiExtractResultMsg}\n${label}: ${result}${displayText}\n\n`;
 }
 
 export const getTelegramCommands = (c: Context<HonoCustomType>) => {
@@ -369,9 +362,8 @@ export function newTelegramBot(c: Context<HonoCustomType>, token: string): Teleg
         const raw = mailRow ? await resolveRawEmail(mailRow) : undefined;
         const mailId = mailRow?.id;
         const created_at = mailRow?.created_at;
-        const aiExtract = parseAiExtractMetadata(mailRow?.metadata);
         const { mail } = raw
-            ? await parseMail(msgs, { rawEmail: raw }, queryAddress, created_at, aiExtract)
+            ? await parseMail(msgs, { rawEmail: raw }, queryAddress, created_at, mailRow?.metadata)
             : { mail: msgs.TgNoMoreMailsMsg };
         const settings = await c.env.KV.get<TelegramSettings>(CONSTANTS.TG_KV_SETTINGS_KEY, "json");
         const miniAppButtons = []
@@ -443,7 +435,7 @@ const parseMail = async (
     parsedEmailContext: ParsedEmailContext,
     address: string,
     created_at: string | undefined | null,
-    aiExtract?: ExtractResult | null
+    aiExtract?: ExtractResult | string | null
 ) => {
     if (!parsedEmailContext.rawEmail) {
         return {};